Transforming Science in South Africa

Transforming Science in South Africa

This page intentionally left blank

Transforming Science in South AfricaDevelopment, Collaboration and Productivity

R. SooryamoorthyUniversity of KwaZulu-Natal, South Africa

© R. Sooryamoorthy 2015© Arthur L. Stinchcombe for the foreword 2015

All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission.

No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages.

The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988.

First published 2015 by PALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010.

Palgrave is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries.

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library.

Library of Congress Cataloging-in-Publication Data

Sooryamoorthy, R. Transforming science in South Africa : development, collaboration and productivity / R. Sooryamoorthy, University of KwaZulu-Natal, South Africa. pages cm Summary: “Science is the cornerstone of development. As the connection between scientific advancement and development becomes firmer, efforts are directed towards strengthening the scientific system. This is increasingly relevant and indispensable for countries on the path of scientific progress. Collaboration has been accepted as a key factor in scientific advancement, and the effects of collaboration are often manifested in the productivity of scientists. This book explores how science in South Africa has grown due to collaboration over the course of its colonial, apartheid and democratic regimes. It provides a comprehensive analysis of the role of collaboration in science and its relation to communication, networks and the productivity of scientists. In giving a detailed account of the concept of scientific collaboration, the South African model presented in this book has great significance not only for other African countries but also for developing nations generally. Transforming Science in South Africa: Development, Collaboration and Productivity will be of interest to anyone who wants to know how science works nationally and internationally in the contemporary world”— Provided by publisher.

1. Science and state—South Africa. 2. Research—South Africa—Imprints. 3. Authorship—Collaboration. I. Title.

Q127.S6S66 2015338.968’06—dc23 2015002144

Softcover reprint of the hardcover 1st edition 2015 978-1-137-49306-4

ISBN 978-1-349-50472-5 ISBN 978-1-137-49307-1 (eBook)

DOI 10.1057/9781137493071

To the loving memory of

K. N. D. Kurup and K. Devakiamma

This page intentionally left blank

Contents

List of Tables and Map ix

Foreword by Arthur L. Stinchcombe xi

Preface xiii

About the Author xv

List of Abbreviations xvi

1 Introduction 1

2 Science in Africa and in South Africa: A Historical Review 11 Science and Africa 11 Science in South Africa 19

Under European colonialism: 1652–1948 20During apartheid: 1948–94 34In the new South Africa: 1994 and after 45Conclusion 53

3 Scientific Collaboration: Towards Conceptual Clarity 57 Significance and relevance 57 Conceptual components 64

Motivations, determinants and origins 66Forms 68Disciplinary nature 71Institutional structure and cultural antecedents 72Benefits and rewards 73Productivity 75Trust 77Communication 79Collaboration effectiveness 80Challenges within 80

4 Research Publications of South African Scientists, 1945–2010 85

Co-authorship 87 Data and method 89

viii Contents

South African publications, 1945–2010 91Features 92Appendices 95

5 Publications through Collaboration 101 Collaboration 104 Partnering countries 108 Sectoral combinations 113 Subjects and citations 116 Collaborative versus non-collaborative research 122 Conclusion 128 Appendix 130

6 Scientific Research in South Africa 135 Scientists and academics 135 Research activities 141 Research projects and facets of collaboration 144 Collaboration versus non-collaboration 149 Predicting collaboration 153 Conclusion 156

7 Communication, Professional Networks and Productivity 159 Productivity and collaboration 163 Communication and collaboration 173 Professional contacts and communication 186 Communication, collaboration and productivity 190 Conclusion 194

8 Collaboration Experience: Portrait of an Eminent Scientist 197 Collaboration and publication productivity 209

9 Science and a Model for Scientific Collaboration 215 Scientific collaboration 215 The South African model 221 The scientific system 224

Notes 231

References 239

Index 259

ix

List of Tables and Map

Tables

2.1 R&D personnel in South Africa, 2009–2010 47

4A.1 Publications of South African scientists, 1945–2010 95

4A.2 Partnering countries of South African scientists, 1945–2010 96

5.1 Publication details of South African scientists, 1975–2010 105

5.2 Region-wise location of partners of South African scientists, 1975–2010 109

5.3 Country-wise location of major partners of South African scientists, 1975–2010 110

5.4 Sector of South African scientists and their partners, 1975–2010 114

5.5 Major disciplines/subjects of South African publications, 1975–2010 117

5.6 Citations of publications by disciplines/branches, 1975–2010 120

5.7 Highest count of citations of publications by disciplines, 1975–2010 121

5.8 Collaborative and non-collaborative publications, 1975–2010 123

5.9 Domestic and international collaboration 126

5.10 Regression of citation on collaboration 128

5A.1 Size of collaboration and subjects, 1975–2010 130

6.1 Respondents of survey, 2007–2008 137

6.2 Professional activity 142

6.3 Research and collaboration 145

6.4 Collaboration versus non-collaboration 150

x List of Tables and Map

6.5 Regression of collaboration on background factors 154

6.6 Regression of collaboration on professional factors 155

7.1 Productivity by sector 164

7.2 Productivity and collaboration 165

7.3 Productivity and type of collaboration 167

7.4 Regression of productivity on background factors 169

7.5 Regression of productivity on collaboration and professional factors 171

7.6 Email and Web use by sector 179

7.7 Email, Web use and collaboration 182

7.8 Professional contacts and sector 184

7.9 Professional contacts and collaboration 189

7.10 Productivity 192

7.11 Co-publication productivity 193

7.12 Regression of network structure on collaboration, productivity and email network 193

Map

2.1 South Africa and its provinces 45

xi

Foreword

Sooryamoorthy’s quantitative analysis of national and international col-laboration can be summed up, for our purposes here, in a few princi-ples that determine the big structure of scientific collaboration in South Africa. Scientists are more valuable as collaborators if they (1) know a lot (by, for instance, having more advanced degrees), (2) teach more advanced students (or are more advanced if they are students), (3) speak the same language (English, for example) or a closely related one (such as German and Dutch), and (4) can spend a long time working in the same place as the collaborator, so they can learn more. They are more valuable also if they have more resources (research equipment and mate-rials like, for instance, seeds) to work on, more useful data already gath-ered and calibrated in a well-understood system, access to publishing in prestigious journals, money for fellowships or visiting professorships, and affiliation to a university that is organized to facilitate collabora-tion. Much informal collaboration is incidental to advanced education (because the preparation to teach advanced courses forces one to keep up to date; reading one another’s work is a pervasive kind of collabo-ration, hardly ever measured) as well as to forming relationships with students and postdoctoral students and knowing their virtues and weak-nesses. And, finally, a simple point—a collaboration over one year pre-dicts a continuing one in the next year, and the collaborators will be better prepared in the second year to work together.

A still more condensed version is that scientific collaboration, formal and informal, takes place more at the top of any scientific profession. The ‘invisible college’ of the elite of a scientific profession is a gigantic ongoing international seminar, where teaching, publication, conferenc-ing and joint work in the labs and in the field provide intellectual assis-tance to each other, enabling people on the top to do better work and have it better understood and used in further work. Co-authorship and joint project proposals are the most visible peaks of everyone trying to get to the top so they can do better work and so stay on the top to do better work next time again. The better one collaborates, the better sci-entist one becomes, so the better the collaboration the next time.

Sooryamoorthy is particularly concerned that Africa and Africans as a whole have started near the bottom more recently than rich countries, though South Africa and the Mediterranean countries started scrambling

xii Foreword

toward that giant seminar earlier than some other countries. Barriers such as language, poverty, and access to fewer universities are harder to break down in some countries than others. This gives clues to the dif-ficulties of entering a system where those that are already present get more from collaboration, and so have the advantage again next time.

But we know from both kinds of data that the key portal into that big system is the institution of the university, with freedom of research and welcoming of collaboration. Intellectual property and keeping your knowledge to oneself do not grow into the great seminar at the top that is designed to increase one’s competence rapidly. Mass educa-tion, especially in the sciences, is a start, of course. For one thing, it will force us all to try to educate science teachers for children, who are equipped to answer all kinds of questions from curious children, and secondary teachers who know more than the adolescents, and profes-sors who know more than the postgraduate students. There should also be enough seminars in postgraduate education so that the students will be useful to their collaborators and the collaborators can in turn be use-ful to them. It is essential to have advanced students as well as mass education to keep professors and researchers reading the literature pub-lished. This will also keep us trying to become people who will be good collaborators for foreigners or people from other African universities, whose work will be improved by coming to universities in Africa, just as they improve the work of our own universities. But it is a hard scramble to the top, and a harder scramble in the poorer parts of Africa.

We scholars need to learn from Sooryamoorthy that we need to help others in their work, so that our own work will improve as well as theirs. We will then be more persuasive when we need new resources, new stu-dents, and new visitors from elsewhere.

Arthur L. StinchcombeNorthwestern University, USA

xiii

Preface

This book is about science and scientific collaboration; it is also about productivity and communication technology. This is my tribute to South Africa, the country from which I derived my strength and knowl-edge to write this book. In the last ten years that I spent on this work—with some interruptions and the deaths of my beloved parents-in-law (to whom this book is dedicated)—the land gave me the energy to sus-tain this effort.

The National Research Foundation (NRF) of South Africa under its various research programmes gave me funds to undertake a number of studies that finally culminated in this book. I received support from my university, the University of KwaZulu-Natal, in the form of grants and the required infrastructure.

Over two hundred scientists and academics graciously gave me their valuable time for rather long interviews, which laid the empirical foun-dation of this study. Patricia Berjak allowed me to make a research film on her and gave a long interview that sheds light on some aspects of the central theme of the book. A number of investigators and research assistants—Lee-Ann Inderpal, Hajra Yunus, Robin Shirley, Nkosikhona Nala, Mzwandile Makhoba, Sayida Dawood—played their parts well. Renjini carefully entered a huge number of bibliographic records on to a data management software that formed the content of chapters 4 and 5. An earlier version of chapter 9 was published in the South African Journal of Science which is acknowledged. I thank Wesley Shrum for bringing me back to this field. Though my collaboration with him ended in 2006, I continued to work in this exciting field on my own and with the support of NRF and my university. Since then a number of my papers have appeared in journals such as Scientometrics, South African Journal of Science, and Technology in Society. In 2007–08, I conducted an extensive face-to-face survey of 204 scientists in South Africa, which forms the empirical foundation of two chapters of this book. The book also con-tains analysis of a huge number of bibliometric records, another area that I have ventured into since 2007.

I am grateful to Arthur Stinchcombe for reading what follows and for writing the foreword. The book would not have been finished but for the gentle pushing and encouragement that came regularly from Geoff Waters, my colleague-turned-friend. I recall my parents, who would be

xiv Preface

happy to know about this book. My gratitude goes to my wife, Renjini, and my son, Dakshin, for their support. Beth O’Leary, Andrew James, Holly Tyler and Dominic Walker at Palgrave Macmillan professionally and promptly handled the project. Dharmendra Sundar Devadoss and his team efficiently managed the production of the book. The construc-tive suggestions of the anonymous reviewers helped improve the book.

Ngiyabonga Kubobonke Abantu! Thank you everyone!

R. Sooryamoorthy

xv

About the Author

R. Sooryamoorthy is Professor of Sociology at the University of KwaZulu-Natal in South Africa. He has taught at the Acharya Nagarjuna University and Loyola College of Social Sciences (both in India), the University of Calgary (Canada), and the Lulea University of Technology (Sweden). Publications include Science in Participatory Development (co-author) and NGOs in India: A Cross-sectional Study (co-author).

xvi

List of Abbreviations

AAAS American Association for the Advancement of ScienceADSL Asymmetric Digital Subscriber Line AIDS Acquired Immune Deficiency SyndromeARC Agricultural Research CouncilASSAf Academy of Science of South AfricaBAAS British Association for the Advancement of Science BRICS Association of Brazil, Russia, India, China and South AfricaCCTA Commission for Technical Cooperation in Africa South of

Sahara CGIAR Consultative Group on International Agricultural ResearchCSA Scientific Council for Africa South of Sahara CSIR Council for Scientific and Industrial ResearchEBED Inter-Africa Bureau of Epizootic DiseasesESO European Southern ObservatoryEU European UnionFAO Food and Agricultural OrganizationFRD Foundation for Research DevelopmentFTE Full-Time EquivalentGDP Gross Domestic ProductGERD Gross Expenditure on Research and DevelopmentGNP Gross National ProductHBU Historically Black UniversityHCO Harvard College Observatory HESS High-Energy Stereoscopic SystemHIV Human Immunodeficiency VirusHSRC Human Sciences Research CouncilIBAH Inter-African Bureau of Animal HealthIBSS International Bibliography of the Social SciencesICT Information and Communication TechnologyIITA International Institute of Tropical Agriculture ILARD International Laboratory for Research on Animal DiseasesIPR Intellectual Property RightsISI Institute for Scientific InformationJCR Journal Citations ReportJSPS Japan Society for the Promotion of Science

List of Abbreviations xvii

MINTEK Council for Mineral TechnologyMRC Medical Research CouncilNASA National Aeronautics and Space AdministrationNDP National Development PlanNEPAD New Partnership for African DevelopmentNRDS National Research and Development Strategy NREN National Research and Education NetworkNRF National Research FoundationNSI National System of InnovationNSID National Science Indicators DatabaseOECD Organization for Economic Co-operation and

DevelopmentR&D Research and DevelopmentSAAAS South African Association for the Advancement of ScienceSABS South African Bureau of StandardsSADC Southern African Development CommunitySALT Southern African Large TelescopeSANCOR South African National Committee on Oceanographical

Research SAPSE South African Post Secondary EducationSARCCUS Southern African Regional Commission for the

Conservation and Utilization of Soil SKA Square Kilometre ArraySRCA Scientific Revealed Comparative AdvantageTHRIP Technology and Human Resources for Industry

ProgrammeUN United NationsUNDP United Nations Development ProgrammeUNESCO United Nations Educational, Scientific and Cultural

OrganizationWHO World Health OrganizationWMO World Meteorological Organization

1

Society is supporting this structure and paying for it more and more because the results of his [the scientist’s] work are vital for the strength, security, and public welfare of all. With everything said to be depending on him, from freedom from military attack to freedom from disease, the scientist now holds the purse-strings of the entire state.

Derek J. De Solla Price (1963)

Science is a productive force in contemporary society (Price, 1965), entwined inseparably with development. The role that science and tech-nology now play in development is not a matter of contention. Price (1965) underlined the use of science and technology to achieve the social goals of society, while Weber discussed its relationship with the economy.1 Science, as Price (1963) observed nearly half a century ago, is a crucial but very expensive part of human activity and a major segment of a nation’s economy. Beginning from the training of an individual into a scientist to the building of the capacity to do science—laboratories, equipment, material, resources, organization, administration, communication and travel—science costs money. In short, it is an expensive investment.

Science gets a generous share of the budget of most countries. A log-linear relationship between the size of the national scientific effort and the gross national product (GNP) is reported (Price, 1969 cited in Frame, 1979). Similarly, a discernible tie between a country’s gross economic status and its ability to support indigenous scientific activity has also been observed (Frame, 1979). China and India are two contemporary examples. Remarkable progress has been achieved in science and tech-nology in these countries, and consequently, their economies are on a fast track.2 Large economies in the world invest proportionately more

1Introduction

2 Transforming Science in South Africa

in research and development (R&D), and by virtue of it, they become the largest players in the production of world science (May, 1997). The US, a world leader in scientific output with a 35 per cent share, spends 2.5 per cent of its gross domestic product (GDP) on R&D (May, 1997).3 Production, as a rule, depends on investment. The same applies to science as well. The more you invest, the more you produce, and the greater the growth.

Science produces knowledge which in turn generates wealth. This is an unremitting process. Science is not limited to the production of knowledge and wealth alone. It is more than that. Science is a human activity that involves interaction between individuals, often from het-erogeneous backgrounds, with characteristic traits. Understanding sci-ence, therefore, means understanding the people who are engrossed in it. In this research, several facets of science are taken into account and studied carefully. What matters here is how scientists work, associate with their peers and produce knowledge.

The emergence of a new mode of knowledge production, Mode 2, is bringing about fundamental changes in the ways in which scientific, social and cultural knowledge are created today. Expansion of research and education systems has brought into being a method of performing research that is different from the discipline-based activity which has dominated science for so long (Gibbons et al., 1994). This kind of knowl-edge production is carried out in non-hierarchical and heterogeneous organizational forms that entail a wider, more temporary and heteroge-neous mix of players collaborating on a problem (Gibbons et al., 1994).4

It is the intention of the sociology of science to study the influence of myriad social processes that occur in the production of scientific knowl-edge (Cole and Phelan, 1999). One fundamental aspect is to learn how social processes affect the construction of the cognitive content of sci-ence (Cole, 1992; Merton, 1938). Working together in what is termed ‘collaboration’ is one such mode. Scientific collaboration is a process in the production of scientific knowledge that is capable of determining both the content and direction of science in society.

Scientific collaboration, as is evident from the literature, is indeed a rewarding undertaking. It is by no means unproblematic. Collaboration is enjoyable and at the same time frustrating. It bestows benefits as well as losses on the collaborators. It is a mixture of everything—knowledge, recognition, publication, visibility, fame, fulfilment, stress, disagreement, trust, conflict, rewards and challenges.

Towards the end of my interview with a scientist, I asked his considered view on collaboration. He was at the threshold of his retirement after

Introduction 3

30 years of research, most of which was collaborative. After pausing for a while to reflect he replied candidly, ‘I would rather work alone than in a team of collaborators.’ Although this response of the scientist (that he would rather have worked alone) contradicted the enthusiasm with which he talked about his collaborations, those of us who have been involved in collaborative activity can sympathize with this response. However, many like Levine and Moreland have expressed pleasure in col-laboration: ‘[O]ver the years, we have worked on many joint projects, and the process has been immensely rewarding on both professional and personal grounds’ (Levine and Moreland, 2004: 170). Collaboration is undeniably an intricate process, not amenable to easy and precise meas-urements of costs and benefits. However, how can one make sense of its significance in science and scientific advancement? Measuring the extent and type of collaboration is one way to understand its relevance. Assessing the outcomes of collaboration is yet another way to study its profundity. Generally, the outcomes are manifest in the expansion of net-works with fellow scientists, the extensive use of communication tech-nologies and, most importantly, the productivity of scientists. This book seeks to investigate these in the South African scientific system and how these are relevant for other societies.

Collaboration, networking and communication among the scien-tific community are expanding (Gibbons et al., 1994; Ziman, 1994). In science and technology studies, collaboration has its own niche and is favoured by many as a fascinating area of investigation. It has been examined from different angles and perspectives and certainly in varied contexts. Katz and Martin (1997) summarize the issues that are studied under four broad categories of measurement of collabo-ration: factors in the formation of alliances; sources of collaboration; role of communication—physical and social proximity; and the effects on productivity. It is therefore a beneficial exercise to delve into the relevance of these factors in a specific context such as South Africa. Collaboration might occur in close physical proximity or at a distance. What are the collaborative propensities of scientists in South Africa, in terms of associating with their peers in neighbouring or distant countries? Is there something that nurtures social proximity, drawing on the historical linkages a country had established and maintained with other countries in the past? South Africa has distinctive but pecu-liar phases in its history—colonial, apartheid and democratic. Science would have passed through these political straits, not necessarily in a linear and undeviating fashion. The following chapters explore these issues in detail.


Due to the complex nature of human interactions that can take place amongst collaborators during the course of the process, it is not easy to understand the precise nature of those interactions using conventional methods (Katz and Martin, 1997). Here one needs to consider the essen-tial components of collaboration. The conceptual components may have varying levels of functioning in a country like South Africa. Some might be completely irrelevant and out of place, while others may not. The disentangling of the concept, however, takes us into the phenom-enon of scientific collaboration, and its specific meaning in countries such as South Africa that has turned its attention seriously to science and technology. We will consider these in general for conceptual clarity and to set the backdrop for the examination of collaboration in South Africa.

Scholars focus on specific facets of collaboration in order to grasp definite aspects of the process. This is crucial to our understanding of the impacts collaboration has on science and scientific growth. When scientific efforts become more and more a team activity, one would expect a tangible change in the way science is conducted and knowl-edge is generated. It would be of interest to examine how collaboration facilitates scientific production as opposed to the way science is done without collaboration. In other words, the focus should be on collabo-ration and its effects on scientific productivity. Collaboration, at the same time, is influenced by certain other key factors. These include the professional networks scientists build up and maintain, especially if these are going to change their productivity. These professional con-tacts in the digital age rely on the access to and use of various means of information and communication technologies (ICTs) for their initia-tion and maintenance.

Scientific collaboration is studied from two planes: the institutional and the individual. The former looks at the institutional components in collaboration, namely, organizational structure, management, admin-istration, resources, policies and preferences that facilitate or hinder partnerships between institutions. The institutional aspects of scientific collaboration can have historical origins. For instance, in societies like South Africa, science has a legacy and a historical past that could influ-ence the current approach to scientific processes like collaboration. The individual level, on the other hand, is micro in approach and looks more closely at the individual researchers and at those factors that are found in the entire span of the initiation and the implementation of scientific alliances. Undoubtedly, this approach of putting the research activities of scientists under the microscope reveals many unknown dimensions

Introduction 5

of collaboration. This is precisely the approach in this book: to learn lessons directly from the scientists themselves.

A set of operationalized variables (Corley et al., 2006) concerning the discipline, location, size and productivity is employed in such individual, partner-centred analyses. Of these variables, productivity in particular can offer new insights into the effectiveness of collaboration and also into whether or not it is desirable from the point of advancing science. Studies with institutional foci generally consider the subject at two levels: the structural level (formal structural arrangements of interaction) and the coordination level (behavioural rules governing the interactions), both of which have potential shortcomings (Landry et al., 1996). Landry et al. (1996) predict three shortcomings in the industry– university col-laboration context. One, it is easier to collect information about formal rather than informal structures, which might lead to underestimating the industry–university collaboration that is channelled through infor-mal and quasi-informal structures. Two, the identification of the struc-tural diversity of the formal arrangements provides little information about the intensity of the collaboration or the number of important joint decisions made by the partners. Three, paying attention exclu-sively to the diversity of the structures means collaboration takes place between structures rather than between individuals. The overlapping of factors between the institutional and individual determinants of col-laboration cannot be ruled out as there are intertwining variables that are relevant in both modes of inquiry. In addition, the impact of con-textual factors (geographical proximity, discipline, organizational struc-ture and levels of coordination) in collaboration is ambiguous (Peters and Fusfeld, 1983 cited in Landry and Amara, 1998). While considering the collaborative enterprises of scientists, one cannot afford to lose sight of the institutional and national structures that play a decisive role in science and collaboration. This dimension is also examined in this study.

Melin (2000) inquires why researchers collaborate and co-produce, what motivates them to collaborate, and what affects investigating the interaction, feelings and conditions within the research team. Bozeman and Corley (2004) examine the individual facets of collaboration in their study of 451 scientists and engineers in academic research centres.5 Does collaboration lead to increased productivity, serving as a motivating force for South African scientists? This question is explored in the chapters which follow.

Scholars are conscious that there is really a dearth of theory to under-stand new collaboration modes (Corley et al., 2006; Wagner, 2005). Corley et al.’s (2006) theoretical framework explains the relationship


among the epistemic norms of the disciplines represented in collabora-tion, the organizational structure of the collaboration and the level of collaboration success. This is an institutional level analysis of collabo-ration. This theory suggests that large-scale, multidiscipline, inter-insti-tutional collaborations need a high level of development, either in the epistemic development of the disciplines in the collaboration in ques-tion or in the organizational structure of the collaboration. The epistemic domain refers to the internal workings of research communities, namely, the norms and practices of research, research agenda-setting, incentives and rewards, while the organizational domain pertains to how the work-ings of organizations are made to enhance the work of research com-munities such as inter-institutional collaborations. This theory assists in comprehending the intricacies of institutional alliances—why there are more (or less) research alliances between institutions within the coun-try (domestic collaboration) than with institutions outside the country (international collaboration). This is a pertinent point in the context of South Africa and is examined in this book.

Transaction cost theory explains the costs of coordinating negotia-tions on collaborative research objectives, and on choices of resources and resource use in regard to the size of structural arrangements (Williamson, 1996 cited in Landry and Amara, 1998). This approach to the institutional structuration of collaboration takes into account various institutional structures and the reasons why researchers are persuaded to organize collaborative research in research institutes and others in research teams or outside the formal structures of institutions (Landry and Amara, 1998). This approach has two aspects, according to Landry and Amara (1998): ex ante and ex post cost. Ex ante costs are the costs of actions and tasks required to establish a research contract for collabora-tion. The costs involve the joint decision-making process of the research-ers about the research objectives, preparation of proposals for funding, work plans, methodology, use of financial and human resources, equip-ment and data, and preparation of publications. Ex post costs include those incurred in coordinating, monitoring and enforcing the contrac-tual promises of research outputs.

Transaction cost emphasizes that the contribution to an institutional arrangement depends on the benefits the researcher draws from collabo-ration. The prediction of this theory is that when costs incurred by par-ticipants in collaboration are higher, then they seek outside structures for collaboration. This is somewhat similar to the cost–benefit approach, according to which alliances materialize when benefits exceed costs (Harrigan, 1985, cited in Gulati, 1998). Similar to this is the economic

Introduction 7

approach which, in a researcher–industry collaboration scenario, rests on the premise of ‘return on investment’, taking into consideration the resources invested and the returns obtained from collaborative enter-prises. The limitation of this approach, as Belkhodia and Landry (2007) rightly note, is that collaboration cannot be reduced to agreeable eco-nomic measures nor explained only by monetary terms. This is true. In a real sense, collaboration is much more than a professional behaviour in that it transcends the immediate economic concerns and benefits. Collaboration depends on economic calculations as well as personal fac-tors, and therefore, collaboration determinants cannot be derived solely from economic theories that simplify a complex phenomenon (Belkhodia and Landry, 2007).

Economic approaches to the study of collaboration shed light on monetary aspects and economic gains as motivators of collaboration. In these approaches hardly any attention is paid to other factors that lure researchers to collaboration (Stephen et al., 2005 cited in Belkhodia and Landry, 2007). Rather, collaboration is seen as a personal choice driven by attributes of the researcher’s field and as a choice determined by eco-nomic calculations (Belkhodia and Landry, 2007). This book analyses how far this is true in South Africa.

In the competitive force approach, collaboration is viewed as a means to shape competition by improving an institution’s comparative compet-itive edge (Porter, 1980, 1985). In this competitive strategy, the firm takes an offensive or defensive position against its competitors or influences them in its favour (Hagedoorn et al., 2000). Hamel’s (1991) theory of competitive collaboration has components of collaborative logic, unit of analysis, underlying processes, success determinants and success metrics.

A theoretically satisfying explanation for collaboration has not yet been achieved (Katz and Hicks, 1997). Most of the existing studies on collaboration deal with motivations, mechanisms, and costs and ben-efits while paying little attention to its impact on scientific productivity (Landry et al., 1996). Comprehensive studies of scientific collaboration per se in any African country are rare, the exceptions being Duque et al. (2005) and Sooryamoorthy et al. (2007). Developing a model of scientific collaboration that applies not only to South Africa but also to several other countries that share the characteristics of South Africa is intended to be the final outcome of the series of studies conducted in South Africa and presented in this book.

Amongst developing countries, South Africa is reckoned to be one of the Third-World research powers bracketed along with India, Argentina and Brazil (Alabi, 1989 cited in Jacobs and Ingwersen, 2000). Of the 54


countries in Africa, South Africa has a strong legacy of scientific collabora-tion. Within the African continent, South Africa is emerging as a regional hub of collaboration (Wagner and Leydesdorff, 2005a).6

This study is set against the background of the scientific community in South Africa, a nation which has been undergoing rapid change. South Africa, on the southern tip of the African continent, extends latitudinally from 22° to 35° S and longitudinally from 17° to 33° E. In an area of 1,219,090 km2, it borders Namibia, Botswana, Zimbabwe, Mozambique and Swaziland. The Atlantic and the Indian oceans wash the South African coast on the west, south and east for some 3,000 km. Divided into nine provinces including the Western Cape, the Eastern Cape, KwaZulu-Natal, the Northern Cape, Free State, North West, Gauteng, Mpumalanga and Limpopo, the country has a varied landscape and climate and is rich in flora and fauna that are closely tied to the everyday life of South Africans.

In this book, an attempt is made to examine science, scientific collabo-ration, productivity and its associated factors among scientists in higher learning institutions and research institutes in South Africa. The central argument in this book is that South Africa, with its tradition of scientific collaboration, has influenced the current scientific system, and this col-laboration is positively changing the productivity of the players in the system. In particular, the major concerns in this study begin from the history of scientific collaboration under three major landmarks of South Africa—colonial rule, apartheid and the new democratic dispensation. It then takes us to the scientific activity of scientists. The study makes use of the data on publication records of South African scientists from 1945 to 2010, and detailed analyses of publications during the sampled years between 1975 and 2010. Based in large measure on a representative sample, it also examines their research, partnerships, productivity, use of ICTs, professional networks and other related individual aspects of col-laboration, collecting data directly from scientists. Most of the variables that are relevant in the literature find a place in this analysis. To comple-ment this, qualitative data were also collected.

Broadly, this book makes a thorough examination of science, scientific collaboration in South Africa and its linkages with productivity, profes-sional networks and communication.

Specifically, this study is centred around the following key questions.

Does South Africa have a legacy of scientific collaboration? If so, what lessons can be deduced from a historical past of scientific collaboration?Does the collaborative past impact on the current collaborative ten-dencies in the country?

Introduction 9

What major elements constitute the concept of scientific collaboration?What prominent forms of collaboration have been adopted by scien-tists in South Africa to advance their productivity?What other means such as professional networks and communication are commonly used by scientists for scientific advancement?How do the collaborators differ from non-collaborative scientists in terms of publication productivity, professional networks and contacts, and in the use of communication media, especially email and the Internet?How do we predict the collaborative propensities of scientists? Is dis-tance a factor in scientific collaboration?What mode of scientific collaboration is the most desirable to have a positive effect on productivity? In other words, can we develop a theory of scientific collaboration that applies to countries like South Africa, based on the evidence collected, analysed and presented in this book?

This study draws on material from a variety of sources: historical docu-ments, archival data, bibliometrical records of publications, qualitative interviews and face-to-face surveys of scientists working in universities and research institutes in the province of KwaZulu-Natal in South Africa.

This is the plan of the book: Being a country with a complex past, South African science has passed through historical trajectories. Portrayed in chapter 2 are the science and scientific collaborations in the country in the three major political periods of colonial rule, apartheid and democ-racy. This gives a glimpse of the backdrop of South African science, in contrast to other African countries, and sets the scene for understanding the themes taken up in the following chapters. Scientific collaboration as a concept demands elaboration for it is defined, explained and used in a variety of ways. In order to provide a holistic picture of this con-cept, chapter 3 attempts to first disaggregate its components and then to show how they are welded together in it. This discussion is based both on material drawn from the literature and on primary data obtained from respondents who have been doing collaborative research. Contributing to the understanding of scientific collaboration, this adds an empirical and realistic dimension to the concept as it happens in modern science today.

Co-publication is a product of collaboration. How do South African sci-entists collaborate with scientists from other countries and what charac-teristics can be inferred from their collaborative output? Chapters 4 and 5 are devoted to the analysis of a large number of bibliometric records on the publications of South African scientists stored in the ISI Web


of Science. This part of the book offers a window to the scientific activity of South African scientists and its collaborative proportions since 1945.

What do the scientists in South Africa do? This question takes us to sci-entists and academics working in the higher education institutions and research institutes in the country. In this examination one can see how far their research activities are collaborative and identify the predictors of collaboration. Chapter 6 focuses on these matters.

Chapter 7 is about communication, networks and productivity. The analysis presented in this part of the book brings out some interesting interrelationships in collaboration. This chapter illustrates how collabo-ration affects productivity and how it is connected to communication among scientists. Deduced from this primary data is the typical nature of South African science and scientific collaboration. We have a very interesting professional sketch of an outstanding scientist of high inter-national repute in chapter 8, where Patricia Berjak talks about her profes-sional life as a scientist and a collaborator. The concluding section takes a condensed view of scientific collaboration, presents a theoretical model of South African collaboration and looks at the country’s scientific sys-tem. Needless to say, the findings of the studies and the model presented in this book have important significance and applications for other coun-tries in Africa and elsewhere.

11

Science and Africa

Despite holding rich resources of minerals, metals and oil, most of the African continent remains poor. Forty-five per cent of the population of sub-Saharan Africa is extremely poor (Burns et al., 2006). Education indicators in Sub-Saharan Africa are also well below the average of devel-oped nations. The region provides higher education to just 3.5 per cent of the college-age population as against 60 per cent in developed nations (Zeleza, 2002). Science and scientific research in Africa need to be looked at against the background of this grim reality. Worthington, in his mon-umental work Science in the Development of Africa (1958), captures the situation of science in Africa:

In the 1920s, there were few scientists and not much was done for them . . . In the 1930s conditions were beginning to improve, but nearly all science was on a territorial and isolated basis . . . In the 1940s many organizations took shape, especially designed to enable scientific men and women to do good work. In the 1950s the terri-torial and regional barriers are breaking down through inter-African cooperation. By the 1960s we may see African science taking its full and proper place in the development of the continent. (Worthington, 1958, cited in Keay, 1976: 88)

In this passage, Worthington touches upon two key points: science was receiving a place in the development of Africa, and scientific collabora-tion within the continent was beginning to take shape. Both are indis-pensable to the growth and development and progress of science and people.

2Science in Africa and in South Africa: A Historical Review


Science in African countries is not homogenous: it varies in charac-ter, form, focus, strengths and application. This heterogeneity makes any credible generalization intricate. Fundamentally, African science is a mixed set of research systems of varying size, human and physi-cal resources, specializations and governing structures (Tijssen, 2007). And the potential for research in Africa is not evenly distributed among its countries (Gaillard, 1992). Strewn over the wide but disparate plant, animal and human landscapes, scientific research in Africa is quite illu-minating and holds great potential for the world scientific community (Sooryamoorthy, 2010b).

This subsection of the chapter looks at the contacts Africa had in the realm of science and examines how these contacts later materialized into collaborative efforts, leading to the joint production of scientific knowledge. The research aimed at gathering and presenting evidence with regard to whether there were shared interests in collaborative alli-ances on the continent that would have facilitated future collaborative enterprises. The peculiarity of science in Africa does not isolate it from the rest of the world of science. Since the beginning of the 17th century, Western scientists and scholars have frequented the continent on sci-entific expeditions and explorations, amassing a wealth of new knowl-edge. These voyages were chiefly meant to study the tropical diseases that were widespread in the region. Constituting a team of entomolo-gists, zoologists, a bacteriologist and a botanist, the Harvard Medical School dispatched its first expedition to Africa to investigate tropical dis-eases then prevalent in the region (Science, 1926). The expedition of the Prussian Academy of Sciences in Berlin made valuable contributions to the zoological knowledge of the continent (Plug, 2003). Apart from the investigations into the possibilities for fishing in the region, the work of this team later led to the publication of the five-volume Zoologische und Anthropologische Ergebnisseeiner Forschungsreise in westlichen und zen-tralen Sűdafrika (1908–1928). Specimens of plants, rare species included, were collected by curious Western travellers to Africa. These were then shipped outside Africa, to Britain, France, Germany, Denmark and Sweden. The traits and properties of these specimens were eventually documented (Keay, 1976).

European countries—Britain, France, the Netherlands, Belgium and Portugal—that had colonies in Africa ran their research machinery in several locations on the continent. For instance, under the Colonial Development and Welfare Act, 1940, the British government promoted and financed research in its colonial territories (Smith, 1967). Britain and France promoted research in their own colonies, but they were

Science in Africa and in South Africa 13

dissimilar in their emphasis, administration and funding. This differ-ence was obvious in several respects. The base of colonial research for Britain was in Africa but not for France. Britain’s focus was in applied research, while the French expected more long-term effects from the research in their African colonies. Most of the British laboratories in their colonies in Africa were located in Africa, but the French labora-tories were divided between France and Africa. Funding for the British research activities came from both the British and African governments along with the levies from colonies, while the French government sup-ported it entirely for their centres in Africa. British centres allowed both local and regional control of administration, while France preferred remote control from France (for more on this, see Smith, 1967). For both countries, their centres of research in Africa effectively served their sci-entific necessities. Britain, for instance, had an immediate need to dis-cover the causes of and cures for tropical diseases before they turned to crops and animals (Smith, 1967). These endeavours drew scientists from abroad to Africa to take advantage of the rich resources available in the land for investigations on an extensive range of theoretical and practical problems (Dillon, 1966). Being the original incidents of scientific inter-action between Africa and the west, these were later to become the steps towards more concrete scientific collaboration.

To tap into the high potential available in African countries, foreign countries vied with one another in establishing scientific institutions across Africa. To cite a sample, the Institute of Tropical Meteorology in Kenya was established in 1960 jointly with the Munitalp Foundation in the US and the Ghana Academy of Science and Learning to promote scientific knowledge and advancement of science and learning (Science, 1960). Collaborative links between European countries and African nations thus commenced in earnest.

Collaborative alliances that emerged occasionally during this period have ultimately led to solutions to problems that the African countries confronted. A major international collaboration in locust control in the 1920s yielded desired results. This collaboration was led by Boris Uvarov, a Russian entomologist, who was earlier asked to investigate the dev-astating locust problem in Southwest Asia (Keay, 1976). Two agencies, the Commission for Technical Cooperation in Africa South of Sahara (CCTA)1 and the Scientific Council for Africa South of Sahara (CSA), pro-moted inter-country collaboration in Africa. Scholars like Lord Hailey, author of African Survey (1945), stood for scientific cooperation between nations and disciplines for the development of Africa. With headquar-ters in Kenya, the Inter-Africa Bureau of Epizootic Diseases (EBED)2 was


constituted in 1952 by six member countries (Belgium, France, Portugal, Southern Rhodesia, the UK and South Africa) of the CCTA, to have a permanent centre for interchange, coordination and dissemination of technical knowledge on animal diseases (Jansen, 1977). The Southern African Regional Commission for the Conservation and Utilisation of Soil (SARCCUS) stood for regional cooperation.3 In order to encourage contacts with African academies of science, universities, professional societies, laboratories and individual scientists, the National Academy of Sciences in the US reconstituted its Africa Science Board in 1965. The Board in its agenda had two prime objectives: the promotion of coop-erative programmes and the development of scientific institutions; and the collection of data about natural resources that are useful for plan-ners (Dillon, 1966). The contacts were maintained through multiple channels such as correspondence, visits to Africa by the board members, meetings in Africa and African participation in international scientific organizations and programmes. As part of this, the president of the acad-emy visited universities, academies of science, scientific institutions and research centres in Nigeria, Ethiopia, Kenya and South Africa (Dillon, 1966). New forms of scientific activity were to spring from a series of workshops with African countries that the board organized in alliance with the CSA. Evident from the above details is the growing interest in forging cooperation among scientists, both within and beyond the continent. Driven by the need for growth, institutions in Africa found a means in scientific cooperation.

International organizations located elsewhere also found it beneficial to associate with African nations. With the political independence of African countries, which began in 1957, the functions of many erst-while organizations including the CSA became redundant. This vacuum was filled soon by world organizations, namely, the UN, WHO, Food and Agriculture Organization (FAO) and the World Meteorological Organization (WMO), charting new areas of collaboration between African nations and the rest of the world. Building on the CSA’s work, the FAO moved on with the inter-African collaboration in locust control and in agriculture (Keay, 1976). Well-equipped centres of research— the International Centre for Research in Agroforestry (Nairobi, 1977), the International Institute of Insect Physiology and Ecology (Nairobi, 1970) and the African Academy of Sciences (Nairobi, 1985)—were to take shape (Keay, 1976).

Collaborative efforts alone were not adequate for the African nations to develop their scientific systems. A majority of the African countries had only small scientific communities and were not in a position to


support training in scientific and technical skills to prevent the ero-sion that had occurred in their research capacity (Eisemon and Davis, 1992). In 1960, at the time of political independence for many African countries, only 9 per cent of the African population was literate, and this rose to 50 per cent in the next 30 years (Zeleza, 2002). The uni-versities in British Africa produced only about 150 graduates in agri-culture, and in francophone Africa, there were less than six graduates (McKelvey, 1965 cited in Eisemon and Davis, 1992). In 1964, the size of the trained research personnel in 41 independent countries (do not include South Africa, Mozambique and Angola) in Africa was limited to only 2,834 in 669 research institutions, just six per cent of what was actually required (Odhiambo, 1967). Subject-wise there were 1,406 scientists in agricultural and food sciences, 566 in earth and space sciences, 400 in medical science, 226 in biological sciences, 163 in mathematics and physics, 71 in industrial research and two in fuel and power. As for research institutes, there were 355 for agriculture and food research, 102 for earth and space research, 87 for biological research, 72 for medical research and 14 for industrial, fuel and power research (Odhiambo, 1967). What was required at this point in time were programmes that could strengthen the scientific systems by grow-ing the strength of trained staff.

The number of schools and universities built in Africa during cen-turies of colonial rule was very limited, and only increased in the post-independence era. Following a surge in higher education in most African countries in the 1960s, there had been an increase of nine per cent in the number of scientists later in the 1970s (Gaillard, 1992). In the post-colonial period, some African countries focussed their attention on the development of science, conceding that science has real rem-edies to many maladies, some inherited from colonial rule. But there were a host of problems to deal with at this time. Gaillard (1992) sum-marizes them as: lack of technicians and managers at universities and public research institutions, diminishing funds for research, imbalances between human and financial resources, heavy dependence on foreign aid for research funding, poor pay for researchers and technical person-nel and high rates of turnover in research staff.

In the 1970s, many of them relied on foreign scientists and support, but the leaders and scientific communities in some of these countries preferred to proceed with their own independent initiatives, despite the constraints they had on funds, scientific literature and equipment (Seaborg, 1970). The policies that emerged in this connection were therefore shaped by their concern for shortages of trained staff in science


(Eisemon and Davis, 1992). Kenya and Nigeria were among the first to do this. They had to deal with the deficits of the colonial era in the absence of opportunities for scientific training and to obtain independence from European scientific manpower. This created policies to achieve some kind of self-sufficiency in science through advanced scientific training and prioritized research activities (Rabkin et al., 1979). It was Nigeria rather than Kenya that vigorously expanded its scientific training oppor-tunities, reducing its dependence on foreign scientists and training, and creating local alternatives to reliance on Western scientific communities (Rabkin et al., 1979). In general, there was an emphasis on a modern scientific approach in several African countries, manifested in their edu-cation system (Krige, 1997). The institutions meant to foster science and research were, however, not free from certain inhibitive factors such as state politics, policies, international donors (Zeleza, 2002) and shifting objectives and priorities.

Several African countries felt the need for cooperation and recognized its relationship with growth and development. In the wake of the politi-cal freedom obtained from colonial regimes, the collective initiative of African countries via joint meetings and conferences gave new impetus to science and collaboration. The UNESCO conference on the organiza-tion of research and training in Africa at Lagos in 1964 brought together 28 countries and recommended the promotion of science and technical research in these countries. In the new light of sovereignty this con-ference made the governments accept the value of increased scientific research and strike a balance between fundamental and applied research in their own countries (Smith, 1967).

Scientific cooperation among the African nations in the 1960s, as reported at the Conference of Ministers of African Member States Responsible for the Application of Science and Technology in Dakar in 1974, was not very impressive. The newly independent nations were consolidating themselves, and the lack of contacts between scientists, government functionaries and political leaders affected the momentum of collaboration (Keay, 1976). Later in 1973, a Consultative Group on International Agricultural Research (CGIAR) was formed jointly by the FAO, the United Nations Development Programme (UNDP), the Ford Foundation, the Rockefeller Foundation and some Western countries. CGIAR supported at least three international agricultural research cen-tres in Africa—the International Institute of Tropical Agriculture (IITA), Ibadan; the International Laboratory for Research on Animal Diseases (ILARD), Nairobi; and the International Livestock Centre for Africa, Ethiopia—which opened up collaborative possibilities in desired areas.


In Egypt, for example, the extent of collaboration in agricultural science steadily increased after 1960 (Farahat, 2002).

The scientific emphasis in Africa largely revolved around medicine, agriculture and biology (Arvanitis et al., 2000; Seaborg, 1970),4 three major fields of scientific enquiry with immediate practical applications to African countries. Among these branches of science, medical science was more internationally oriented and brought in relatively more inter-national funding and partnerships than the rest. In countries such as Kenya, there has been a strong concentration of international research in medical and life sciences. In the publication profile, these sciences dominate with a share of 61 per cent against 44 per cent for the world (Tijssen, 2007).

International collaboration, as mirrored in the publication productiv-ity of scientists preserved in prominent databases, seems to be common in the fields of biomedical research, biology, earth and space science, and physics (Narváez-Berthelemot et al., 2002; Tijssen, 2007).5 The obvi-ous reason for this can be found in the scope for biomedical research like HIV/AIDS in the African context. Western countries looked for coopera-tion with Africa where some of the best centres of scientific excellence such as the Immunology Biotechnology Laboratories (Cameroon) are located.

African science had some structural dependency on the scientific sys-tem of advanced countries. Many countries on the continent are strug-gling to sustain their scientific activity amidst scarcity of funds and the brain drain (Narváez-Berthelemot et al., 2002). Often, the best aca-demics and researchers leave their home countries for lucrative posi-tions abroad. One estimate is that about 30,000 PhD holders of African descent live and work outside their home countries (Hassan, 2001).

Recent studies (Tijssen, 2007, for example) show a definite decline of science in this most fertile region for scientific research. Africa lost 11 per cent of its share in world science since its peak period of produc-tion in 1987 and sub-Saharan science lost 31 per cent (Tijssen, 2007). This decline is not recurred in the absolute number of publication pro-ductivity but in percentage. Tijssen’s (2007) reasons for this decline include the lack of willingness to invest in scientific infrastructure, ina-bility to retain scientific workers in universities, laboratories and insti-tutes, low pay and dull career prospects. In 2000, Africa’s share of the worldwide publication output was just 1.4 per cent. For Sub-Saharan Africa, the percentage had slipped to less than 1 per cent since the mid-1980s.6 The citation impact of African science is far below the inter-national average. The grounds for this downslide can be found in the


functioning of higher education institutions and research institutions. Universities in Africa, by and large, are teaching rather than research institutions (Zeleza, 2002). This affects the research output, scientific growth and development and of course the prospects for collabora-tion. The science departments of the best universities in Africa of the 1960s and 1970s—the University of Lagos (Nigeria), the University of Dar es Salaam (Tanzania), the University of Accra (Ghana) and the University of Khartoum (Sudan)7—have since declined and are unable to meet their basic responsibilities and functions (Hassan, 2001). This is attributed largely to the obsolete and inadequate facilities and a lack of proper incentive structures for career development and scientific quality (Tijssen, 2007). In many African universities research is carried out typi-cally for consultancies to subsidize the living of the staff who work there (Habib and Morrow, 2006). The innovation system in Africa suffers from structural weaknesses and research capabilities and is characterized by the absence of an integrated regional collaborative network (Toivanen and Ponomariov, 2011). Lagging behind other regions (Tijssen, 2007), several African countries need an overhaul of their scientific research systems. A turnaround is possible with proper utilization, retention and allocation of resources that are available to them locally. Often, this is a question of priorities and the preference for scientific advancement. The African Union declared 2007 as the year of scientific innovation, underscoring the importance of science for Africa. Nonetheless, the sci-ence produced in Africa is neither mediocre nor irrelevant. As reported in a scientometric analysis by Pouris and Pouris (2009), Africa during 2000–04 produced 68,945 publications which is 1.8 per cent of the total world publications. The recent momentum in the growth of science in Africa is partly due to scientific collaborations (Irikefe et al., 2011).

Despite having a weaker scientific system in the continent in general and in certain countries in particular, some countries have recognized the potential for growth and development through scientific coopera-tion. Contacts with the Western world were primarily and initially in medical, agricultural and life sciences that opened doors for cooperation between African and Western scientists. There were partners from coun-tries in Europe and the US with whom scientific cooperation was insti-tuted in specific areas. Alongside this, regional cooperation within the countries in the continent continued to take shape and grew. Attempts made by international organizations such as UNESCO, FAO and UNDP to promote scientific research emphasized the need for scientific coop-eration in the region. Although many African countries badly required cooperation with the outside developed world for their scientific growth,


it was also in the interest of the scientists in developed countries who saw Africa as a fertile ground for groundbreaking research in many scien-tific fields. In the later years these cooperative endeavours materialized into concrete collaborations. Studies have indicated that knowledge that is being produced by African scientists is largely in collaboration with their partners from outside the region. For instance, Boshoff’s (2009) analysis showed that 80 per cent of the research papers produced in Central Africa was jointly with a partner from outside the continent. Also significant in this analysis is that African countries maintained their colonial connection, producing 46 per cent of the papers with European scientific partners. At least four key points can be distilled from the above information on science in Africa: that Africa received scientific community from abroad with the intention of cooperating in fields that required urgent attention—medical, agricultural and life sciences in particular; that these cooperative initiatives were mutually beneficial, if not completely symbiotic in nature and form; that many of these efforts were later translated into scientific collaboration leading to joint production of knowledge; and that Africa maintained its link-ages from its colonial past to the current times, working with partners in Europe and the US.

The questions Jan Hofmeyr had asked in 1929 at the annual meeting of the British and South African Associations for the Advancement of Science held in South Africa are still relevant. He asked: ‘What can Africa give to science? What can science give to Africa?’ (Tilley, 2011).

Science in South Africa

South Africa, according to the World Bank classification, is the only African country to be placed among the scientifically proficient countries along with Spain, Brazil, Cuba, some of the former Eastern European countries and India (Narváez-Berthelemot et al., 2002). The place of South Africa on the science map of the continent is unassaila-ble. Scientific research in Africa, according to a scientometric analysis by Pouris and Pouris (2009) for the period 2000–04, is concentrated in the two countries of South Africa and Egypt, which jointly produce more than 50 per cent of Africa’s publications. A number of African coun-tries stand to gain from the scientific advancement that South Africa has accomplished over the years. In the recent years (2000–10) South Africa’s world share of publications reached an all-time high, advancing its inter-national ranking to the 33rd position in 2010 (Pouris, 2012a). Among the Southern African Development Community (SADC) countries South


Africa’s position is prominent. South Africa is credited with 79 per cent of the total publications of all 15 SADC countries in 2004–08 (Pouris, 2010).

Drawing on historical details, the following sections present a review of the state of science in South Africa under the three major periods of European colonialism: 1652–1948, 1948–94 (Apartheid) and since 1994. The intention is to examine both the legacy of scientific contacts in distinctive political historical periods of the country and how these have influenced the current collaborative forms and trends in the South African scientific system.

Under European colonialism: 1652–1948

Since the second half of the 17th century, South Africa has been in regular contact with other Western countries, many of them leaders in world science. This began with the first European settlement by Jan van Riebeeck at the Cape of Good Hope in 1652 when the Dutch East India Company started its expansion of the Cape.8 It was not the land but the sky that attracted the first scientist to South Africa. An observatory was soon to be established in 1751 (Talbot, 1977). Scientists from other countries visited the Cape for professional reasons and also as tourists. South Africa received them with open arms, fostered interactions with them and opened the doors of scientific research and discoveries. For any inquisitive mind South Africa was a land of opportunities. This wel-coming approach of the country was very encouraging for international scholars to come to South Africa. The rich flora was unique enough to entice botanists to collect plant specimens as early as 1652. John Burchell (1781–1863), who landed in Table Bay in 1810 gathered over 40,000 specimens from all over the region before he returned to England and published his Catalogus Geographicus (Talbot, 1977). The oldest marine fauna of Lower Devonian age, the oldest rocks (Rogers, 1929), perma-nent and enduring mineral wealth and the pleasing weather were irre-sistible. The marine life of South Africa drew close attention and study. The early accounts of these attempts could be seen in the publications of Old and New East Indies and the Present State of the Good Hope (Gill, 1905). The first comprehensive list of the Cape fauna running into 45 pages on mammals, 22 pages on birds, 24 pages on fishes and 20 pages on snakes, insects and other animals was published in 1719 (in German, and later in 1727 in Dutch) by Peter Kolb (Forbes, 1977).

Early on, individuals with an interest in science began to settle in the Cape Colony, contributing to the advancement of science. Pieter Potter travelled from Amsterdam in 1655. He was the first land-surveyor


and cartographer at the Cape to do surveys and draw diagrams of the land (Forbes, 1977). In 1685, Father Tachard and five other Jesuits made astronomical observations to ascertain the longitude of the Cape. Caspar Commelin (1667–1731), a Dutch doctor and a professor of botany, stud-ied medical plants in the land and published his findings in 1703. Carl Thunberg (1743–1828), a Swedish botanist, arriving in Table Bay in 1772 travelled around for an intensive investigation of the flora, collected plants from the Cape and published Prodromus Plantarum Capensis and Flora Capensis (Forbes, 1977; Plug, 2007). He later earned the title ‘the Father of Cape Botany’. William Duckit (1768–1825), an English agricul-turist, settled in the Cape in 1801. In 1802, Petrus Truter (1775–1867), a physician and judge, along with William Somerville (1771–1860), navigated an expedition into the interior. Captain Dugald Carmichael (1772–1827), a Scottish surgeon, who was in the team that captured the Cape from the Dutch stayed in the colony and studied plants (Plug, 2006). Joseph Mackrill (1762–1820), an English medical practitioner with experience in the West Indies and the US, decided to settle in the Cape Colony in 1806. Like Dugald, Joseph was fond of medicinal plants. A well-known bacteriologist, Robert Koch, was invited from Berlin to initiate research in immunization. Robert Broom, an Australian doctor pursuing interests in fossils, arrived in 1897 to become a leading pal-aeontologist; he was later responsible for changing the face of South African palaeontology (Cluver and Barry, 1977). The interests of inter-national scholars in the specific areas of science were carried forward in collaborative research and joint scientific publications in later years. Further details are provided in chapter 4.

The Royal Observatory at the Cape was the first scientific institution to be established in South Africa. This was in 1820. Under the leader-ship of Andrew Smith, the most important scientific expedition to the interior in recorded history was conducted, resulting in a publication on several marine species (Day, 1977; Naudé and Brown, 1977). Andrew’s work also led to the establishment of the South African Museum in Cape Town in 1825, one of the first to be opened outside Europe.9 The museum, opened by Lord Charles Somerset, has its antecedents in a small museum established by Willem Adriaan Van der Stel in Cape Town in the 18th century (Naudé and Brown, 1977). Andrew Smith is credited with the spirit behind the establishment of the South African Institution in 1829 to promote research in geography, natural history and the gen-eral resources of South Africa (Talbot, 1977). His work, Illustrations of the Zoology of South Africa in particular, brought him the honour ‘the Father of South African Zoology’.


Rapid developments in plant sciences, mycology in particular, ensued under the leadership of Christiaan Hendrik Persoon (1762–1836), whose publication of Synopsis Plantarum in 1805 was a major breakthrough in the field (Plug, 2005). Plant pathology grew with the contributions of the Welsh pathologist Illtyd Pole Evans (1879–1968) (Plug, 2005). The South African School of Forestry was established in 1906 in Tokai. Soon, in 1911, a Central Department of Agriculture came into being to devote specialized attention to a range of scientific areas—veterinary research and services, sheep and wool services, dairying, entomology, agrostol-ogy, botany, plant pathology, pedology, tobacco and cotton cultivation, viticulture, chemical services and dry land farming (Joubert, 1977). Veterinary science, for the obvious reasons of its bearing on the popu-lation and economy, matured much faster than any other branch of science. European veterinarians were appointed in the early 1870s, and South Africa continued to obtain them from Britain, Ireland, Switzerland and Germany (Brown, 2005). In 1920, the government constituted its veterinary faculty at the University of Pretoria, the first in Africa (Brown, 2005). Simultaneously, medical science in South Africa made much headway. Its contribution to the detection, prevention and treatment of respiratory diseases was highly rated (Hofmeyr, 1929b). At this time South Africa entered into a new branch of science in ecology through its study on the veld (Brown, 2005). The study of geology seized the imagination of many with seminal publications such as the Introduction to the Geology of the Cape Colony (1905, by Arthur W. Rogers), The Geology of South Africa (1905, by Frederick H. Hatch and George S. Corstorphine) and the Catalogue of Printed Books, Papers and Maps relating to the Geology of South Africa (1905, by Maria Wilman).

Professional associations were formed,10 and their members met fre-quently, adding to the growth of scientific inquiry. Societies such as the South African Literary Society (1824, to cater to the sciences and to encourage reading and enquiry), the South African Literary and Philosophical Society (1825, for the cultivation of science and literature), the South African Institution (1829, for investigating the geography, natural history and general resources of South Africa), the South African Literary and Scientific Institution (formed after merging the South African Institution and the South African Literary Society in 1829), the South African Philosophical Society (1877, for the progress of science and publication of research results, which became the Royal Society of South Africa in 1907), the Cape Society of Engineers, the Chemical Metallurgical and Mining Society, the Cape of Good Hope Veterinary Medical Society, the Victoria College of Scientific Society (1901) and the


South African Association for the Advancement of Science (1902) were among them (Gevers, 2001; Hall, 1977; Hofmeyr, 1929a; Juritz, 1915; Plug, 2001, 2005).

Prior to 1900, the meetings of scientists held under the auspices of pro-fessional organizations in South Africa were not infrequent. The Royal Society of Northern Antiquaries met in April 1885 (Science, 1885). The South African engineers held their meeting in 1901 at the instance of a British-born engineer, Theodore Reunert (1856–1943). Later, this turned out to be an occasion for engineers as well as other scientists represent-ing fundamental and applied science to meet and share their scientific explorations (Plug, 2002). The subjects of these meetings ran from min-erals, agriculture, veterinary science, plant and animal diseases, chemi-cal science, biochemistry, nutrition to mining technology (Juritz, 1915, 1916a, 1916b, 1916c, 1917, 1919). The annual meetings of the South African Association for the Advancement of Science (SAAAS, founded in 1902), and the South African Association of Analytical Chemists (1913), witnessed the range and magnitude of scientific research in South Africa. The first annual congress of the SAAAS showcased a spectrum of scien-tific topics, ranging from atmospheric electricity to language that gave a glimpse of the research being carried out in Africa.11 SAAAS, adjudged as a successful organization,12 functioned effectively to meet its objec-tives of advancing scientific inquiry and promoting contacts between individual scientists and institutions of science (Dubow, 1995; Plug, 2002). The conference of this association on diseases of cattle and other animals in South Africa in 1903 brought together scientists from south-ern African territories. SAAAS owes its origin to the influence of the British Association, founded in 1831, and continued to have interest in the ideas and approaches of the scientific communities in the UK and the US, from where a number of its members had been recruited (Rich, 1990). Meetings and conferences aside, some associations granted research funds to scientists, published papers through their outlets of scientific journals and appointed special committees for specific pur-poses. Manifest in the concern to address the ‘neglect of science’ (Juritz, 1917) were the serious efforts of scientific associations to nurture science in schools, hoping to raise a future generation of scientists for South Africa. Ensuring participation of scholars from overseas, these meetings set the scene for interaction and association with the international sci-entific community.

In the early 1900s the scientific public in South Africa remained small. Science still depended on imperial connections. As in several other African countries, the colonial legacies were influential in the scientific


ties of South Africa (Narváez-Berthelemot et al., 2002). This explains why South Africa published more scientific publications with scholars originating from countries with which it had an imperial scientific con-nection. Britain, partly because of the political and colonial connection with South Africa and partly owing to its interest in South Africa on geographic, climatic and scientific grounds, was zealous in sustaining contacts with South African scientists. Under the auspices of the British Association for the Advancement of Science (BAAS), the leaders of British science visited South Africa, first in 1905 and then in 1929. In their first meeting a contingent of 385 members including leading representatives of science arrived in South Africa. This lent SAAAS, which hosted the event, a rare opportunity to hold the largest scientific conference in the country (Plug, 2005).

The BAAS visit brought together people from diverse branches of science—mathematics, physics, chemistry, geology, zoology, geogra-phy, engineering, physiology and botany. As part of this event, papers were presented and lectures delivered in different parts of the country on themes such as fly-borne diseases, atmosphere, radioactivity, metal, mining and astronomy (Science, 1905a, 1905b); and geological excur-sions were conducted. The government was hospitable to the team of scientists (Lomas et al., 1905). This hospitality made it possible for the country to officially commence scientific contacts with the outside world. To commemorate the meeting, a volume entitled Science in South Africa presenting the state of science in South Africa was released. As the first large conglomeration of scientists of all hues and nationalities, rep-resenting numerous branches of and specializations in science, it turned out to be a major event in establishing scientific contacts. The meet-ing opened the gates of collaborative research in the country, which was later to be become more prominent in the magnitude of the joint production of knowledge. Following this, quite a few collective ven-tures were initiated. A Russian astronomer and a member of the visiting team, Johan Backlund (1846–1916), asked for assistance with an inter-national programme to study the variation in latitude. In response to his request, the Transvaal Observatory in Johannesburg cooperated with the programme, which continued for a few years with the participation of South African, British and Russian scientists (Plug, 2006). Such con-tacts and working together with scientists from abroad had an effect on the formation of professional scientific bodies. Some of them were founded at the instance of foreign nationals. John McCrae (1875–1960), who became the first president of the South African Association of Analytical Chemists, was from Britain (Plug, 2003). Like these moments


of confluence, South African scientists had other occasions to be close to the international scientific community. This was both within the continent and outside, through visits, interaction, sharing and presenta-tions. They went abroad to study, to conduct collaborative research or to attend conferences, allowing them to orient with the developments in their respective fields of science, which eventually led to the creation of new professional contacts—significant for future collaborations. To cite a few cases, a veterinary bacteriologist, William Robertson (1872–1918), was sent to Southern Rhodesia to study a new stock disease, East Coast Fever. Another bacteriologist, Arnold Theiler, visited laboratories and colleges in Europe and represented the Government of the Transvaal at the International Veterinary Congress held at Budapest in 1905 (Plug, 2005).

Professional organizations—a few in number then—and the state were eager to have regular interface with representatives of world science. A substantial number of South African professional organizations main-tained contacts with their international counterparts (Joubert, 1977). These were soon to assist in institutional collaborations.

By engaging in the activities outlined above, professional organizations such as SAAAS and BAAS were probably unintentionally sowing the seeds for at least three forms of collaboration. One, it placed scientists in touch with each other within their own institutions and those in the country—domestic collaboration. Two, by bringing researchers from southern African countries, it was providing a platform for intercontinental col-laboration. Three, by recruiting professionals from overseas and showing interest in their scientific approaches, it paved the way for international collaboration.

Governmental support for interactions with overseas scientists was forthcoming. How these forms have further developed and grown to substantial levels of collaboration can be seen in chapters 4 and 5, which analyse the publication records of scientists. At the International Geological Congress held in Pretoria in 1929, the Association of African Geological Survey was constituted. The Association collaborated closely with the CSA and produced an international geological map of Africa (Keay, 1976). From then on, collaboration both as a concept and practice received general approval (Jansen, 1977). South Africa kept its links with international organizations for the two-way exchange of research results and expertise. In disease control research, at Onderstepoort Institute in particular, collaboration with a number of countries in Africa has been mutually beneficial. In agriculture, South Africa was a member of inter-national organizations such as the International Office of Epizootic


Diseases, the International Seed Testing Association, the International Dairying Federation and the International Wine Office (Joubert, 1977). These activities brought South Africa to the centres of world science, and established closer contact with the international community.

Possibilities in astronomy triggered international attention and col-laboration. This, in the subsequent years, encouraged international scholars to produce scientific papers in collaboration with South African scientists and steered the growth of astronomy as a discipline in the country. The geographical location of South Africa was well suited for astronomy. When Abbe de Lacaille, in 1752, made his first extensive and accurate observations of the stars of the Southern Hemisphere in the Cape of Good Hope, he had scientific reasons for choosing this loca-tion (Gill, 1905). Being the birthplace of astronomy in the Southern Hemisphere (Gill, 1905), the Cape offered the astronomer delightfully clear skies and cloudless nights (Hofmeyr, 1929b). All the most inter-esting sky objects are in the south and South Africa is one of the few areas from where observation of these objects is possible (Evans et al., 1972). For this reason and also because of the accompanying scientific openings, several countries queued up to set up their own observato-ries in South Africa, furthering shared scientific interests, common goals and close interactions. This was critical for the growth of the science of astronomy as well. British optical astronomy would have been practi-cally non-existent but for its operations in South Africa (Evans et al., 1972). Many countries, the US, the UK, France, Germany and Holland included, sent in their astronomers to the land. A German astronomer, Peter Kolbe, arrived in the Cape in 1705 to study the geographical posi-tion of the Cape and the distance to the Moon, in collaboration with European astronomers (Plug, 2005; Science, 1925, 1929b). Dynamic interactions with scientists from abroad, with persistent support from scientists, associations and governments, contributed in no small meas-ure to the foundation of international collaboration ventures in South Africa.

The early 1890s saw British scientists arriving in sizable numbers, both to develop and exploit natural resources in different parts of the country (Plug, 2003). Scientists from elsewhere also took part in scientific pur-suits with South Africans and produced results of great scientific value. In the discovery of a practical method of immunization for the rinder-pest cattle disease, which devastated South Africa for quite some time, it was the efforts of a German scientist that finally bore fruit. Robert Koch from Berlin arrived in South Africa in 1896 and found two methods of immunization for this cattle disease (Bruce, 1905b). There were other


foreign scientists who were also involved in this research (Bruce, 1905a). Rinderpest was successfully eliminated by 1905. Charles Rousselet, a British marine scientist, collected samples of more than 156 species from South Africa and published Contribution to Our Knowledge of the Rotifiers of South Africa in 1906 (Plug, 2006). These joint efforts for a common purpose were rewarding for both parties.

The arrival of the first astronomer at the Cape has been recorded in 1821 (Plug, 2006). A distinguished astronomer from England, Sir David Gill (1843–1914), who since 1879 served as the director of the Royal Observatory at the Cape of Good Hope, was the man behind the geo-detic survey of South Africa for some 26 years (Plug, 2006). William Sclater (1863–1944), another Englishman, served as the director of the South African Museum, which during his directorship achieved promi-nence as a leading institution in southern Africa (Plug, 2006). Robert Lehfeldt (1868–1927), came from London to take up the position of a professor of physics. A British agriculturist, Frank Smith (1864–1950) was appointed as the Agricultural Advisor to the Transvaal Government in 1902; on his recommendation the Department of Agriculture was constituted (Plug, 2002). A Cambridge trained geologist, Maria Wilman (1867–1957), joined the South African Museum (Plug, 2002). In 1907, George Daniell (1864–1937) from England became the first medical person to be appointed in the country as an anaesthetist. James Drury (1875–1962), a taxidermist at the South African Museum in Cape Town, was a Scot (Plug, 2007). A German physician and naturalist, Martin Lichtenstein (1780–1857), was asked by the administration to undertake research into smallpox, to develop a vaccine. Canadian agronomists and entomologists served the Department of Agriculture of the Orange River colony.

The colonial legacy of South Africa was to bring in foreign-born sci-entists to serve in key positions in its scientific institutions. South Africa remained dominated by Britain and Western Europe and depended on them for education and training in science (Schaffer, 1977). Britain is the country that produced the largest number of scientific publications from South Africa. The list of such foreigners who contributed to the growth of science in South Africa is a long one. A sample of them is listed below, highlighting the prominence of international exposure in South African science.

Arnold Theiler (1867–1936), instrumental in the inception of the Onderstepoort Veterinary Institute, was a Swiss bacteriologist. Francis Kanthack (1872–1961), a British civil engineer, landed in Cape Colony in 1906 and later became the Director of the Irrigation Department of


the Union of South Africa (Plug, 2006). Hermann Bohle (1876–1943), also from England, came in 1906 to become the first professor of elec-trotechnics at the South African College (later the University of Cape Town) in Cape Town and worked extensively for the next three dec-ades on the practical applications of electricity (Plug, 2006). A Scottish geologist, Andrew Young (1873–1937), joined as a professor at the South African College, while his brother Robert Young (1874–1949) went to the University of Witwatersrand (Plug, 2002). Henry H. W. Pearson (1870–1916) from Cambridge University took up the post of a profes-sor at South African College in 1903 (Plug, 2003). Joseph Burtt Davy (1870–1940), an Englishman with working experience in the US, served as an Agrostologist with the Transvaal Department of Agriculture in 1903 (Plug, 2003). In the same year, an entomologist, C. B. Simpson (1876–1907), from the US joined the Transvaal Government (Plug, 2003). Steward Stockman (1869–1926) from India succeeded Simpson. Herbert Ingle from Leeds (UK) was made the chief chemist of the Transvaal Department of Agriculture in 1903. Fermour Rendell (astron-omer), Arthur Hodgson (physicist) and Ernest Warren (zoologist) were among the numerous others who came from overseas to serve as scien-tific personnel in South Africa (Plug, 2003).

One more underlying dimension of the participation of international scientists is to be discerned in the ensuing scientific phases of South Africa. This shows the confidence South Africa had in these overseas scientific professionals. Scientists who came from other countries made major contributions to their respective fields and made every conscious effort to build their own branches of science. Scientific activities can flourish only in an environment where it is valued, supported and respected. The South African system had this advantage in its colonial period, which, as we will see in the coming chapters, was carried forward in the ensuing phases of its political system. The contributions of these scientists who have made South Africa their home for life and work were enough to bring forth a positive and encouraging attitude among South Africans and native scientists. This confidence is quite obvious in the way South Africa is forging international alliances today. If a country is keen on tying up with international partners, as is being done today in the country, it should have some historical context that is rooted in pleasant experiences and confidence.

Parallel to these developments in scientific contacts, association and collaboration, educational and scientific institutions emerged in differ-ent parts of the country. The educational reforms of 1839 promised the evolution of a proper university, separated from the elementary classes


of schools (Naudé and Brown, 1977) that in due course ensured a regular supply of scientists. The University of the Cape of Good Hope, the first university in the country, was formed in 1873 and became the University of South Africa. The University of Cape Town, the first English-speaking university, was founded in 1916, taking its origin from the South African College, which was originally established in 1829. It is also the largest South African university (Greenberg, 1970a; Naudé and Brown, 1977).

Traditionally, the universities in the country were teaching institu-tions (Smith, 1956, cited in Pouris, 2006a) that evolved from colleges. In the early years, science departments in these institutions were slow to develop and suffered frequent setbacks (Naudé and Brown, 1977). At the new University of Cape Town, research was lagging while it was trying to overcome the academic activities of a school from which it had meta-morphosed. Science was still in its infancy, and the scientific community was small. The University of Cape Town, for example, at its inception had 25 professors, 24 lecturers, 15 assistants and 469 undergraduate stu-dents (Naudé and Brown, 1977). It operated around one university and a few colleges, 33 professors, 27 graduates in pure and applied science, a few others who worked in the government departments, two astronomi-cal observatories (one in Cape Town and another in Johannesburg) and some scientific societies. With the addition of more colleges and univer-sities, science was set to take-off in the following years.13

After 1915, the university system expanded, drawing its faculty from outside the country (Dubow, 1995). In 1916, the Union of South Africa14 passed the Bill creating a federal and two single-college univer-sities (Juritz, 1916c). Eventually, the scope of science in South Africa began to grow. By 1929, the country had three single-college teaching universities, a federal university with six constituent colleges, 134 pro-fessors, 275 graduates, four more observatories (totalling six) housing sophisticated telescopes and refractors, and more scientific societies of engineers, chemists, biologists, botanists, astronomers, geographers, ornithologists, veterinarians, pharmacists, horticulturists and econo-mists. Added to the list were the South African Institute of Electrical Engineers, the South African Institution of Engineers, the Cape Chemical Society, the Cape Society of Civil Engineers, the South African Chemical Institute, the Botanical Society of South Africa, the South African Biological Society, the Astronomical Society of South Africa, the South African Geographical Society, the South African Ornithologists’ Union, the Transvaal Veterinary Medical Association (which later became the South African Veterinary Association), the Institute of Mine Surveyors, the Pharmaceutical Society of the Orange River Colony, the Durban and


Coast Horticultural Society, the Natal Society and the South African Economic Society (Hofmeyr, 1929a; Plug, 2003, 2004).

These developments apart, consolidation of scientific departments in the civil service took place. Two prominent institutions—the Institute of Veterinary Research at Onderstepoort (an institution sought after by scientists from other countries for its expertise) in 1908 and the South African Institute for Medical Research at Johannesburg—were founded (Hofmeyr, 1929a). Located near Pretoria, the former has shown the world how the institutionalization of veterinary knowledge takes place with the incorporation of the knowledge of African communities and settler farmers (Brown, 2005). A milestone in veterinary research, the institute at Onderstepoort helped develop an international scientific culture in the country (Brown, 2005). The South African Institute for Medical Research was created in 1912 to conduct research, especially in human diseases, to elucidate their causes and devise methods of diagno-sis. It also performed studies in applied bacteriology, pathology, parasi-tology and pharmacology (Naudé and Brown, 1977). Later, in 1949 the South African Council for Scientific and Industrial Research’s (CSIR’s) Committee for Medical Research was set up (Brock, 1977; Lister, 1929). Promoting scientific research through its liberal policy, the state con-stituted a Research Grant Board in 1919 to advise it in scientific affairs and to distribute grants to individual researchers (Hofmeyr, 1929a). Businesses, as in the contemporary research terrains of the country, joined hands with the government to participate in scientific endeavours by both starting new and consolidating the existing research institutions and departments. This was also vital for the business sector to improve their technologies and to serve their interests. The mining industry is a prime case. An institute of medical research was created jointly by the government and the mining industry (Science, 1929a). In agreement with the Triple Helix I model, research activities in South Africa during this time encompassed the state, the industry and academia. Prestigious research centres15 were to be formed and many of them received inter-national accolades in the years of their existence. These efforts worked well to strengthen certain disciplines in South Africa that helped attract scholars from abroad. When we examine the robust nature of these dis-ciplines and collaborative efforts, a clear link is visible in recent years. Medical science is a case in point (Sooryamoorthy, 2010a).

South Africa had a clear edge over other countries in its capacity and mastery of skill in some branches of science. To an extent, this is main-tained today, as pronounced in its productivity of scientific scholars. Its feats in science had been well recognized by other countries on the


continent as early as 1910 (Brown, 2005). By 1925, South Africa had made tremendous progress in its knowledge base of rocks, diseases in plants, animals and humans (Hofmeyr, 1929a; Juritz, 1919) and of weather conditions. Geological research caught the imagination of scholars with the setting up of first a geological survey office at the Cape of Good Hope in 1895, and later a Geological Survey in 1910 (Naudé and Brown, 1977). Archaeology took root in the South African soil and flour-ished after 1923, characterized particularly by its phases of institution-alization in the earlier period and its re-emergence as a discipline after a spell of neglect (Shepherd, 2003). Besides, the discipline secured political patronage, as evident in the constitution of the Bureau of Archaeology in 1935, which later became the Archaeological Survey and a directly funded branch of the civil service. The oldest hidden sediments of rocks and abundant wealth of extinct animal remains assisted palaeontology to keep pace with other cognate disciplines.16 Raymond Dart’s discovery of Australopithecus africanus (1924) was a huge leap for South African sci-ence. Born in Brisbane, Australia, Dart came to South Africa and became a professor of anatomy at the University of Witwatersrand and an inter-nationally respected figure.

It is clear that the headway in science was stimulated by immediate practical concerns of the times as well as economic growth. Many of the scientific institutions, including the Onderstepoort Veterinary Research Institute, grew out of such exigencies (Naudé and Brown, 1977). The Institute owed its origin to the outbreak of the rinderpest disease in the Transvaal province in 1896. The discovery of the vaccine for heartwater disease was one instance that proclaimed the independence and con-fidence of South African scientists in resolving scientific problems. In 1944, Petrus Johann du Toit, the South African Director of Veterinary Services, rejected a request from the US Department of Health for strains of the heartwater rickettsia for which the US was hoping to devise and manufacture a vaccine. The refusal of South African authorities was based on the fact that heartwater was an African disease and their con-viction that the task of solving the problems connected to the disease should be first of all entrusted with African scientists (Onderstepoort Archives, cited in Brown, 2005). The Onderstepoort Veterinary Institute made several groundbreaking discoveries. The success of Onderstepoort Institute has culminated in the establishment of similar research insti-tutes in other parts of the country as well (Naudé and Brown, 1977). These achievements brought South Africa to the centre stage of toxi-cological research (Brown, 2005). Through these strides, South Africa was showcasing its scientific capabilities and its potential for scientific


research, which drew international scholars to associate with it. All these significantly contributed towards forging linkages with the international community. Concurrently, South Africa was taking firm steps towards internationalization of its science.

In the field of medicine, South African scientists concentrated on dis-eases that were causing havoc in the region. Not much has been recorded on indigenous science and knowledge. The efforts of some scholars, including anthropologists, such as Good (1987), Khumalo (2001), Krige (1974), Schapera (1966), Schapera and Farrington (1933) and Sindiga et al. (1995) are of importance in this regard. Immediacy of problems did not compel them to seek assistance from abroad. Geologists, marine biologists, mathematicians, economists (Hofmeyr, 1929a), agriculturists and anthropologists contributed to scientific progress. Agriculture sci-entists, with the support of the government, were in the forefront of research concerning dryland farming, the generation of better varieties of crops and grasses, and the elimination of crop pests (Brown, 2005). South Africa’s scientific efforts in agriculture were evident in more ways than one. Maize production in South Africa, to cite one exam-ple, almost quadrupled between 1910 and 1950 with the introduction of hybrid maize cultivars, artificial fertilizers, mechanized production methods, effective pest and disease control measures and improved cul-tivation practices (Joubert, 1977). The sugar industry is a case in point. The production figures of sugar rose consistently between 1907 and 1915 (Juritz, 1917). Hofmeyr (1929a), suggesting a measure of scientific growth, pointed out that in 1927 the value of the products South Africa exported increased five-fold from the year 1906. The value of the prod-ucts amounted to £5,928,000 and £27,815,000 in 1906 and 1927 respec-tively. The discovery of a new kind of primate in a limestone fissure at Taung was a remarkable feat in paleoanthropology (Tobias, 1965). South Africa was thus thrust into prominence in the eyes of the scientific world (Hofmeyr, 1929a).

The most significant feature of science in South Africa during the period 1905–29 was its shift from the exotic to the native nature of its scientists (Hofmeyr, 1929a). The founding of SAAAS in 1902 was only a beginning of this South Africanization. The regular meetings of SAAAS had a ramifying effect on the growth of science and its popularity in the country. Professionalization of science teaching in schools and uni-versities began to happen (Rich, 1990). Rated highly in international science, South Africa developed its own channels for the dissemination of science, including peer-reviewed journals.17 The multidisciplinary journal of the SAAAS, now called the South African Journal of Science,


with a reasonable impact factor in the ISI database, is adjudged a great South African scientific achievement by the international scientific com-munity (Gevers, 2003). The ISI index ranks this journal as the eighth among the 19 international multidisciplinary journals and the only one in Africa and in the southern hemisphere (Gevers, 2003). In the launching of scientific publications, government departments were con-spicuous. The Department of Agriculture published periodicals such as Farming in South Africa and Crops and Markets for the benefit of farmers and officers. Recognized as of great national value, science became South Africanized while attracting the attention of the world scientific com-munity (Dubow, 1995; Hofmeyr, 1929a). South Africa scored fairly well on this point at this time of its scientific trajectory. From the point of view of collaboration this was imperative for South Africa in the subse-quent years.

In response to the activities of scientists and the recognition science had received by then in the scientific circles, nationally and interna-tionally, and among the public, the government stepped in to support science through a series of measures. Three years before the start of the apartheid era, a scientific advisor to the prime minister was appointed in 1945. This was to formulate plans for the establishment of an organiza-tion to advise the government on the full use of the natural resources and to coordinate scientific research in the country. Drawing on the experiences of similar institutions elsewhere in the world and adopting their best practices, the CSIR18 was established. A major step in scientific research in the country, the CSIR paved the way for accelerated growth in the crucial sectors of science and technology. It was the aim of the Council, among others, to provide both the government and industry with basic facilities of research through the creation of well-equipped national laboratories for fundamental and applied research across the country. It was also the objective of the Council to promote research in industries through a mechanism of inducement (remission of taxa-tion on research, financial assistance and access to facilities in national laboratories), to ensure a steady flow of trained researchers (through postgraduate bursaries and research grants) and to ensure the rapid uti-lization of the research results, locally and internationally (Naudé and Brown, 1977). The effects of these initiatives were to be seen in the com-ing years, although under a distinctive and biased political system.

In summary, South African science gained during the colonial period. Some branches of science—pathology, veterinary science, medical sci-ence, astronomy, geology, palaeontology and mineralogy, for instance—more than other branches found the right climate for their genesis and


growth and the consolidation of their position in the subsequent years. South Africa also drove towards internationalization of its science that furthered the interests of international scholars. Through its achieve-ments South Africa demonstrated to the world its scientific potential and this attracted international scholars. Strategically, the support of the professional organizations, scholars and the colonial government/s made it possible for interaction and exchange with foreign scholars. These were in future to transform into different forms of scientific col-laboration—domestic, intra-continental and international—as well. Scholars from Europe, the US, Canada, Australia and Nordic coun-tries found South Africa a land of scientific opportunities. Landing in the country and choosing it as their home, they offered leadership to develop South African science by establishing scientific institutions, departments, museums and observatories, helping to form professional associations and compiling volumes of scientific information. Resultant confidence in them enabled South Africa to develop a positive attitude towards them. The structures created during this colonial period—CSIR, for instance—became solid pillars for scientific research and interna-tional alliances in the country.

The subsequent chapters (4 and 5) examine how these colonial ties and this legacy have made a lasting impact on the contemporary scien-tific associations with the international community.

During apartheid: 1948–94

Both the nature and character of science are often influenced by the political organization and ideology in which science and the scientific system are nested. In the case of South Africa science was nurtured and grew and developed in a politically segregated terrain for quite a long period of time. This distinguished the South African scientific system as different from others where a similar form of political organization did not exist. Started in 1948 and officially ended in 1994, apartheid in more ways than one commanded the science and scientific system of South Africa. This section examines how science in the apartheid era performed, with a particular focus on collaboration.

Due to the inherent preferences and advantages accorded a particular race, the national and international communities did not approve of many of the scientific policies and programmes of the apartheid govern-ment. The pressures from outside the borders of South Africa were rather strong. In such circumstances it was not easy for the apartheid govern-ment to develop its own scientific system without support. In response to the pressure and approach taken by the international community


against the apartheid political regime, South Africa strove for its inde-pendence in science and moved towards the South Africanization of science (Sooryamoorthy, 2010b). This was also reflected in the leader-ship role it assumed in Africa. As was evident in the colonial period, in several realms of science the leading role of South Africa was signifi-cant. The country formed the South African National Committee on Oceanographical Research (SANCOR) in 1956. Following this, with the intention of conducting research in the production of animal fibres, the South African Wool Textile Research Institute was established along with the Animal and Dairy Science Institute for dairy research (Joubert, 1977). These were soon to become the preferred areas of expertise for the country.

South Africa built its first nuclear reactor, which used natural ura-nium, in 1960. Soon South Africa was able to evolve its own unique and new process of uranium enrichment. However this ‘unique and new process’ has not been without contention (R. G., 1974). South Africa’s advancement in this area concerned the powerful nations, who feared South Africa’s resources and ability to build nuclear bombs. South Africa maintained that it was to improve its export income from processed ura-nium rather than raw uranium. At this time South Africa was planning to build a commercial-scale enrichment plant with or without foreign finance and assistance (Gillette, 1975). Scientists, physicians and aca-demics were brought in to produce chemical and biological weapons (Gould and Folb, 2002). South Africa was able to build six-and-a-half nuclear bombs which were similar to the one that was used in Hiroshima (Harris et al., 2004). They were but soon to be dismantled. This part of the history of South Africa’s nuclear past is not fully known to the public due to the destruction of records and documents (Harris et al., 2004). Named Project Coast, the chemical and biological warfare programme was initiated in 1981. But the South African scientific community was ill-equipped to advance this harmful and misdirected programme fur-ther (Gould and Folb, 2002). In these ways, where it could advance with-out the support of the outside world, South Africa was attempting to build its scientific strengths.

A further factor that boosted South African science was the intro-duction of a funding formula for research publications of scholars in universities. This was in 1987, and it continues today in varying and amended forms. There were at least three formulas for this objective of funding: the Holloway formula, the van Wyk de Vries formula and the South African Post Secondary Education (SAPSE) formula. The second and third formulas have retained the major components of the first and


added new features and parameters to suit the changing academic envi-ronment in South Africa. The SAPSE formula was later revised in the late 1980s and early 1990s (Steyn, 2002; Steyn and Vermeulen, 1997 cited in Steyn, 2002). These measures inspired scholars to publish their research in national and international journals. Naudé and Brown (1977) noted that publications became the only criterion for appointments and pro-motions in universities in early 1960s. The value of publications is still high in such appointments and in promotions in tertiary institutions. The government subsidy to universities is also based on the research out-put of scholars. As will be seen in the next three chapters, all these meas-ures had an effect on the scientific output of South African scientists.

The ‘framework autonomy’ was introduced in 1988. Its purpose was to serve statutory councils such as Council for Scientific and Industrial Research (CSIR), Human Sciences Research Council (HSRC), Council for Mineral Technology (MINTEK), Foundation for Research Development (FRD), South African Bureau of Standards (SABS), Medical Research Council (MRC) and Agricultural Research Council (ARC) to indepen-dently manage their institutions (IDRC, 1993). The autonomous MRC was formed in 1969 out of the Committee for Research in Medical Sciences. Technikons were established in 1978 along the lines of British polytechnics although they were accorded a research mandate only in 1983 (Marais, 2000).

The apartheid government was determined to support science for both its economic benefits and also for building technical manpower, defence research and industrial development (Sooryamoorthy, 2010b). But this support was not enough in an internationally hostile environment in the apartheid era. Langer (1967) maintained that the ideology and prac-tice of apartheid meant much more to the government than science and scientific development. The apartheid government thus sought to con-trol scientific activities through several means and practices.

The UNESCO report for the UN Special Committee on apartheid19 in 1962 unequivocally recorded the way the apartheid government exerted pressure on scientific societies to segregate the membership in their organizations along racial lines. Fourteen such societies that had been receiving government subsidies were instructed by the Minister of Education to prevent mixed membership in their societies (Sooryamoorthy, 2010b). The societies that did not conform to this edict were warned of economic reprisals and sanctions (Langer, 1967).

The government continued to coerce the societies to toe the line, despite their criticism of this interference. At least half (seven societies) did not budge, while one decided not to reapply for support in view of


the strictures and the remaining six opted for a ‘wait and see’ approach (UNESCO, nd, cited in Langer, 1967: 1387). This, in other words, did not mean that all the societies were uniformly in favour of mixed member-ship in their associations. Conflicts erupted in many of these profes-sional societies, resulting in splinter groups in some cases, while others maintained mixed membership at a nominal level. In response to the UNESCO inquiry into the racial mix of membership, the South African Institute of Physics responded that they had three non-white members (two Africans and one Asian), the Dental Association had eight (all Asians), the Genetic Society and the Royal Society of South Africa had no non-whites. The South African Psychological Association had split, with a separate group which favoured exclusion. The case with the asso-ciation of sociologists was similar. In 1983, some members of the South African Association of Archaeologists left the association in protest against not condemning apartheid and other forms of discrimination that spread across higher education and learning domains (Shepherd, 2003). Archaeology, however, had taken a pragmatic approach to the apartheid policies of the state. Archaeology, benefitting from state patronage, functioned separately from the society and disengaged from the turbulence of the times (Shepherd, 2003).

Scientists who raised their voices against the regime risked being either banned or imprisoned (Hoffman and Cox, 1971). At the same time, the state tactically dealt with some well-known scientists although they were critical of the government and its policies. As one scientist remarked: ‘I am persona non-grata with the government because of my political views . . . but they [the government] don’t touch my research grants or travel funds . . . I’ve got an international reputation’ (reported in Greenberg, 1970b: 263).

Facilities for joint racial meetings at universities were not generally permitted under apartheid (Sooryamoorthy, 2010b). Race was the pre-dominant parameter that decided access to higher education (Marais, 2000), like many other privileges. University education was rigidly restricted on racial grounds under the Extension of University Education Act in 1959, which is a derivate of the Separate University Education Bill of 1957. Except in special cases non-white enrolment was prevented under this Act. Instead, a system of ‘university colleges’ which were not equiva-lent to the universities for the whites was introduced for non-whites (Sooryamoorthy, 2010b). As Greenberg (1970b) noted, in 1967 only three universities catered to the entire black population of 12.7 million as against ten fully fledged universities for the 3.6 million white popula-tion. Particularly in the fields of science, this provided extremely limited


higher educational opportunities for the black population. Apartheid was trying to create a system in which science and scientific activities (as also the benefits derived from it) were exclusively advantageous for a particular race. This naturally prevented the chances for collaborative enterprises with the scientific communities between other races and the outside community.

The black universities were known as Historically Black Universities (HBU). These were the University of Fort Hare (1916), the University of North for the Tsonga, Sotho, Venda and Tswana peoples (1959), the University of Durban-Westville for Indians (1960), the University of the Western Cape for Coloured, Griqua and Malay students (1960), the University of Zululand for Zulu and Swazi students (1960), the Medical University of South Africa for black students (1976), and universities in the apartheid era ‘Bantu homelands’ of Bophuthatswana (1979), Transkei (1977), QwaQwa (1979) and Venda (1979). The University of Fort Hare, the first university to serve the black population, was formed in 1916, but its student enrolment remained small from the outset (Nordkvelle, 1990). Meant to serve the racially disadvantaged, these institutions restricted them from receiving full advantages of science and lagged behind the other white/English universities.

HBUs in general were meant to provide vocational training at the undergraduate level. Only a few masters and doctoral students (one per cent of the total number of degrees and diplomas awarded in 1993, compared to over ten per cent by the white universities) gradu-ated through these institutions (Marais, 2000). The number of white graduates was much higher than the black graduates, who did not have equal opportunities for higher education in their own country. Quoting NUSAS (1951) Nordkvelle (1990) reported that out of a population of nine million blacks there were only 1,000 university students in 1947 as against 20,000 students for a white population of 2.5 million (a ratio of 0.11:1,000 for blacks and 8:1,000 for whites). In 1986, there were 147,697 white students for a population of 4.8 million and 93,753 black students to serve a population of 27.79 million. The result was a ratio of 3:1,000 for blacks and 30:1,000 for the whites (Nordkvelle, 1990). The disparity was thus huge. These figures indicate that the ratio was favourable to white students by ten times. Discrimination prevailed not only in terms of access to higher education and teacher–student ratio but also in terms of the cost to government. The average cost of educa-tion for a white student was eight times higher than a black student and the student–teacher ratio was 19:1 for the whites and 41:1 for blacks (Omond, 1985 cited in Nordkvelle, 1990). As regards the ethnicity of


the teachers, 92 per cent of the academic staff were whites, and in black universities no more than 35 per cent of the academic staff were blacks in 1984 (Nordkvelle, 1990).

Racial discrimination was evident in the admission to medical sciences as well. Black students were not allowed to join traditionally white medi-cal schools, although in the 1980s this changed under certain condi-tions. Even those who were fortunate to enter the medical sciences could not complete their training as they were barred access to white patient hospitals for their clinical training (Lawrence, 1941 cited in Perez et al., 2012). Studies of those black medical students have confirmed that the quality of medical training of black students was adversely affected by racial discrimination that was manifest in the way tutors were allocated, access to hospitals for training, accommodation in residences and exam-inations (Perez et al., 2012).

University appointments were tainted by racial preferences and seg-regation. Universities continued to advertise for academic positions on the basis of race. In one of such instances, the University of Natal (which later merged with an HBU, the University of Durban-Westville to form the University of KwaZulu-Natal) advertised in Science, for a position exclusively for whites (Hoffman and Cox, 1971). As far as science educa-tion was concerned in both the colonial and apartheid periods little or no science was taught to blacks. This resulted in few schools offering science to blacks (Yoloye, 1995 cited in Khumalo, 2001).

Segregation was not confined to university education. It was extended to other fields of scientific research as well as to resource allocation. The white English-language universities received far larger research grants than the black universities. This trend persisted even during the demise of apartheid. In 1991–92, the HBUs received only seven per cent of the total research and development (R&D) expenditure of the entire univer-sity sector (Marais, 2000).

The environment characterized by deterrence, discrimination, racial preferences and conflicts had a decidedly negative impact on the free-dom of thought and the intellectual pursuits, imagination and creativ-ity of scientists (Sooryamoorthy, 2010b). This restrictive climate did not entirely frustrate some serious scholars. They managed to maintain their contacts and links with the scientific world outside the country and moved on with their research (Sooryamoorthy, 2010b).

The impact of apartheid on the South African scientific system was to be seen outside the borders of the country. South Africa had been either excluded or has resigned from several international organiza-tions (UNESCO, cited in Langer, 1967). In the early years of apartheid


governance, South Africa found itself isolated from other sub-Saharan countries in science and technology (Keay, 1976). In 1955 the coun-try withdrew its participation in UNESCO in response to the UNESCO publications on racial matters in South Africa. The Commission for Technical Cooperation in Africa South of the Sahara (in 1962), the FAO (in 1963), the Council for Science in Africa (in 1963) and the Economic Commission for Africa (in 1963) were some of the other international organizations from which South Africa exited. These were construed as acts of retaliation for the stance taken by these organizations against segregated rule in South Africa. Definitely, these were not in the interests of forging international collaborations.

A more drastic step on the part of South Africa was yet to come. It declared the withdrawal of its world-renowned institute of Onderstepoort Veterinary Research Institute, which was serving as a reference centre for certain diseases typical to Africa, from the FAO. Again, by leaving the CSA in 1963 South Africa prejudiced other countries in the region (Sooryamoorthy, 2010b). UNESCO perceived this as a great loss to the CSA in Africa because South Africa was no longer making contribu-tions to several fields of scientific research—in the treatment of diseases, research into low-cost housing, road technology, surveying and photo-grammetry, precise surveying methods in mining, psychometrics, nutri-tion and telecommunications—which were otherwise available to the member countries (UNESCO, cited in Langer, 1967: 1,387). This was not healthy for South Africa either, from the point of view of international alliances.

WHO expelled South Africa (Greenberg, 1970a). In 1980, the General Assembly of the UN adopted a number of resolutions including one on an academic boycott of South Africa. Some countries such as Canada, France, Federal Republic of Germany, the UK and the US voted against, while the Nordic countries abstained from voting on one of the reso-lutions (35/206 E). In 1981, the Nordic countries agreed to work on a scheme to deal with apartheid in the cultural, sports and academic fields, which led to measures to detach these fields from South African univer-sities (Nordkvelle, 1990). These had serious negative consequences for both South Africa and the world of science (Sooryamoorthy, 2010b). In contrast to the colonial period in which science grew with the participa-tion of international scholars, apartheid brought down the shutters on potential association with international scholars and world science.

Increasingly, South Africa was becoming alienated from international science platforms. This was not a healthy sign for South Africa or for its science. Although South Africa had several outstanding scientific


achievements, its losses were considerable. This was more damning than anything else, if viewed from the perspective of collaboration. Some of the leading scientists left the universities of South Africa in protest (Sooryamoorthy, 2010b). The University of Cape Town took a serious blow when 25 of its faculty resigned in 1961. The University of Natal lost 35 academics for the same reason. Several other universities had simi-lar experiences. As UNESCO reported, apartheid played a definite role in academics leaving mainly from the English universities (Greenberg, 1970b; UNESCO, cited in Langer, 1967). The government was not inter-ested in those who did not want to support its science and higher educa-tion policies rooted in racial segregation. The same approach was shown to those international scholars who wanted to come to the country to conduct research. In April 1988, members of the American Association for the Advancement of Science (AAAS) and three other organizations were denied travel visas to the country as they were to enquire into the medical care provided to political detainees in South Africa. This fol-lowed the death of an anti-apartheid political leader, Steve Biko, who died in police custody (Marshall, 1988). One would also find that the professional organizations which had done astounding work in receiv-ing and associating with foreign scholars were no longer in the same position to advance their collaborative interests for the growth of sci-ence in the country.

Those promising and talented academics, researchers and students who applied for passports for foreign study or travel were given a one-way exit visa, meaning that they could not return to the country after their business overseas (UNESCO, cited in Langer, 1967). The regime feared that they would have been exposed to a different political and scientific culture while they were abroad and would influence others in the country if they returned. Clearly, the regime was not prepared to grow science through international exposure that might undermine its separatist policies of governance.

From outside the borders of South Africa, the international sci-ence community supported the academic isolation of South Africa in order to make a forceful impact on its racist policies and programmes (Sooryamoorthy, 2010b). This support was firmly grounded in the ideal of internationalism under which scientific knowledge is created for the benefit of all (Zachariah and Sooryamoorthy, 1994) and not exclusively for the benefit and use of some (Hoffman and Cox, 1971). One of the norms of scientific knowledge is communism, according to which any new scientific idea introduced by anyone is to be considered collectively owned knowledge (Sztompa, 1986 cited in Nordkvelle, 1990).


From 1960 until the late 1980s, science in the country was caught in a detrimental environment from both within and outside. Despite the immigration of scientists who looked for jobs in South Africa, there was a virtual atrophy in the international exchange of experts and in access to overseas scientific facilities (Marais, 2000). Due to the inten-sity of the academic boycott in the late 1980s, the CSIR could neither recruit foreign scholars nor ensure a free supply of researchers through visits (Nordkvelle, 1990). International journals, a strong arm of the scientific body, joined in the fight against segregation in science insti-tuted by the apartheid regime. They declined to accept South African contributions and avoided South African participants in professional conferences (Nordkvelle, 1990). Marked as a ‘closed off’ period, from the mid-1980s to 1994, this weakened the publication output of South African scientists (Ingwersen and Jacobs, 2004; Jacobs and Ingwersen, 2000).20 Science in the country suffered during this ‘closed off’ period. Intellectual exchange with the international community turned out to be hard and impractical for scientists who aspired to gain international experience in scientific tools and methods and opportunities that would have led up to continued association with the international community.

Nordkvelle (1990) argued that the academic boycott had only a lim-ited effect and success and there has been a movement of visitors to and from South Africa while their publications were carried in international journals. It is difficult to agree or disagree with this viewpoint for lack of reliable facts and figures. While the boycott existed, some South Africans travelled abroad as exiles and gained education and training, many in science subjects (Sooryamoorthy, 2010b). When the political situation was reorganized from apartheid to democracy, those in exile overseas had reason to return with their newly learned skills and knowledge. The unintended consequence of this boycott, however, was that universities remained insulated from the changes that affected those in other parts of the world (Pouris, 2006a).

The efforts aimed at isolating the country, both by those within the country and the international community, did not deter South Africa from progressing scientifically. Notwithstanding the pressures of political turmoil and racial tension, South Africa continued to build its research and training infrastructure and scientific capacity (Marais, 2000). As Greenberg (1970a) recorded, South Africa developed its capabilities in science, technology and medicine and in some other areas exception-ally well. The country had been recognized for its advanced medical knowledge. The first heart transplant in human history by Christiaan Neethling Barnard (in 1967, at Groote Schuur Hospital in Cape Town)


was a remarkable feat for the country. The science and technology policy of the government stressed the creation of a system to promote scientific culture in the country, increase scientific manpower for R&D and also to raise the standards of R&D. It also sought to raise the standards of R&D and quality of research programmes (Marais, 2000). Recognizing the connection between science and economic growth, the govern-ment (and industry) invested in technology to the tune of about USD 70 million a year on R&D, which included 18 per cent allocated for basic research alone (Greenberg, 1970a). South Africa spent about half of its R&D funds on agricultural and veterinary research (Greenberg, 1970a). The R&D expenditure in 1979–80 was ZAR 310 million, which was 0.64 per cent of the GDP (Nordkvelle, 1990).

South Africa managed to expand its scientific base. The wealth of the country’s resources was something which other countries found quite tempting. For many countries the ban and boycott were more of a political necessity and a diplomatic strategy than a genuine reaction to apartheid and support to anti-apartheid activists. Even before the boycott and opposition came into being, South Africa had laid a strong foundation for its science (Sooryamoorthy, 2010b), beginning from the colonial period. Further progress, even without the necessary technical and financial assistance from other countries, was not difficult for South Africa. The rich mineral reserves in different parts of the country paved the way for collaborative research, as in the case of gold production. The expertise of South Africa in the extraction and processing of minerals was soon adopted by countries such as the US, Peru and the Philippines (Bunt, 1977).

Given the country’s remarkable advancement in scientific research, scientists from developed countries, and especially from the US and the UK, sought positions in South Africa (Greenberg, 1970b). Remember this was again at the peak of apartheid. The networks South Africa had with leading research institutes in Western Europe and the US were mutu-ally beneficial (Greenberg, 1970a). International institutions such as the National Aeronautics and Space Administration (NASA), the National Institute of Health and the Atomic Energy Commission were but some with which South Africa collaborated at this time. Built in 1961, the Radio Space Research Station of NASA in Johannesburg provided space-tracking and deep space instrumentation facilities intended to make a great impact on scientific training and technological advancement in the country (Greenberg, 1970a). The training programmes at this sta-tion were beneficial to secondary school graduates (Greenberg, 1970a). The NASA-funded Jet Propulsion Laboratory at California Institute of


Technology trained the scientists of the CSIR in the US (Greenberg, 1970a). The tie-up in space technology, while enhancing the national prestige of the country, also provided the chance to acquire the techni-cal know-how to staff and maintain its ground stations and exposed its research and technical staff to cutting-edge equipment and techniques (Greenberg, 1970a). The examination of whether these linkages in the apartheid period have been lost or carried forward in contempo-rary South Africa has revealed evidence that is positive, as detailed in chapters 4 and 5. The US continues to be the top collaborator of South Africa in knowledge production.

South Africa’s relations with foreign science and technology assisted the country in its geophysical and metallurgical research (Greenberg, 1970a). In the presence of the typical and oldest dated rocks, American geologists came forward to collaborate with researchers at the University of Witwatersrand, which also had research collaboration with the Colorado School of Mining Engineering (Greenberg, 1970a).

There were 6,000 research workers in South Africa in 1970 (Greenberg, 1970a). Before long the country realized that there was a dearth of scien-tists (South African Journal of Science, 1977).21 One measure to resolve this, as John Pratt suggested, was to build up and produce more and better research in universities (South African Journal of Science, 1977). However, matters changed. The number of scientific societies and organizations grew. In 1981, South Africa had some 47 learned societies, 53 research institutes and a large number of research centres attached to universities (The World of Learning, 1981 cited in Nordkvelle, 1990).

Overall, science did not appear to suffer greatly despite the troubled apartheid times, but scientific collaboration with the external scientific community did. South Africa was poised to become a sub-metropolitan place for science (Brown, 2005) and achieved authority in a number of scientific realms. Compared with many other African countries, South Africa’s scientific institutions grew rapidly with the production of solid original work in science (Naudé and Brown, 1977). However, as Murphy (2011) noted while reviewing the nuclear power project of South Africa, the exclusion of the African majority from the national science and technology scenario in the apartheid period has blunted the technologi-cal edge of the country.

The apartheid system acted on science in South Africa in several ways. First, it tried to sway its benefits and advantages towards one single race. This division, however, did not last long beyond the demise of apartheid in 1994. In the new democratic South Africa, as exposed in the next section of this chapter, the division has been discarded, broadening the


scope of science that benefits all. Second, strong structures of scientific institutions were built purposefully when the international community was largely non-supportive of the segregated regime and its scientific policies. Some branches of science such as medical sciences and mining engineering flourished. This put South African science on the track of advancement and progress. Third, the expanse of scientific collabora-tion in contrast to that under the colonial era receded.

In the new South Africa: 1994 and after

A new political era dawns in South Africa (Map 2.1). Ending years of struggle South Africa embraced democracy. Nelson Mandela (1918–2013) constituted the first democratic government in South Africa. As expected and desired changes took place, and the transformation unfolded, new policies were put in place and approaches to problems had to be revisited. As in other realms, science was not insulated from

Map 2.1 South Africa and its provinces

Limpopo

Mpumalanga

GautengNorth West

Free State

Northern Cape

Eastern Cape

Western Cape

KwaZulu-Natal

NAMIBIA

BOTSWANIA

SOUTH AFRICA

ZIMBABWE

AtlanticOcean Indian

Ocean

N

200 km


the effects of this political transformation. The spirit of collaboration with the past was not lost. It is interesting to see whether there is any connection or disconnection with the scientific past and how far science of the colonial and apartheid era impacted on contemporary science in the country.

Merton (1938 [1970]) and Price (1963) discerned that often a linear relationship exists between the number of scientists and the amount of scientific knowledge produced in a country. There are three sectors that are currently active in scientific research in South Africa: government, business and private. These sectors often cooperated with each other, promoting institutional domestic collaboration. State-owned corpora-tions, science councils, universities, technikons and domain-specific research organizations are part of the government. A higher participa-tion was recorded in higher education in South Africa (Gultig, 2000). It had one of the largest higher education systems in Africa. Continuing on a comparative note Gultig (2000) reported that in the mid-1990s, more than 20 per cent South Africans in the age group of 20–24 were enrolled in higher education institutions as against 28 per cent in the UK, 12 per cent in Brazil and 4 per cent in Nigeria. To cater to a populace of 51,770,560 (Census 2011), the higher education sector of the country is distributed over 23 universities with a total student count of 938,201 as of 2011 (Department of Science and Technology, 2013).22 The 2009–10 figures indicated that South Africa had 30,891 full-time equivalent (FTE) research personnel (researchers, technicians and other support research staff) in its sectors of R&D (Table 2.1). Morever, 88 per cent of the total R&D expenditure in 2004–05 went to natural science, technology and engineering (Department of Science and Technology, 2006), which in 2009–10 reduced to 87 per cent (Department of Science and Technology and HSRC, 2013). Professional scientific associations had increased from 4 to 68 in 2007.23

Research endeavours in South Africa enjoyed international recognition (Sooryamoorthy, 2010b). This applied to international collaborations as well. Several South African scholars who are internationally renowned figures, stayed in the forefront of their own fields of research (Vaughan et al., 2007). The country has produced a few Nobel laureates as well. In 2002, Sydney Brenner shared the Nobel Prize in Medicine.24 Aaron Klug won the Nobel Prize in 1982 for his outstanding work in chemistry. Numerous instances of groundbreaking research have occurred in the fields of medicine, veterinary science and plant science. The country has retained its position in astronomy, geology, ecology and veterinary sci-ence. In the publication analysis by Pouris (2003) based on the records

Tabl

e 2.

1 R

&D

per

son

nel

in

Sou

th A

fric

a, 2

009–

2010

Ca

tego

ry

Bu

sin

ess

ente

rpri

seG

over

nm

ent

Hig

her

ed

uca

tion

*N

on-p

rofi

tSc

ien

ce c

oun

cils

Tot

al

Per

cen

t

Hea

d

cou

nt

FTE

#

Hea

d

cou

nt

FTE

#

Hea

d

cou

nt

FTE

#

Hea

d

cou

nt

FTE

#

Hea

d

cou

nt

FTE

#

Hea

d

cou

nt

FTE

#H

CFT

E#

Res

earc

her

s8,

366

6,05

9.5

986

680.

428

,552

10,6

14.2

224

187.

52,

669

2,25

1.5

40,7

9719

,793

.168

.664

.1Te

chn

icia

ns

5,36

23,

612.

650

935

6.8

2,11

557

9.8

7663

.71,

381

1,17

9.4

9,44

35,

792.

215

.918

.8O

ther

s4,

488

2,35

2.6

1,08

586

6.7

1,72

567

6.4

8058

.61,

876

1,35

1.8

9,25

45,

306.

015

.617

.2

Tota

l18

,216

12,0

24.6

2,58

01,

903.

932

,392

11,8

70.4

380

309.

75,

926

4,78

2.7

59,4

9430

,891

.310

010

0

Perc

enta

ge30

.638

.94.

36.

254

.438

.40.

61.

010

.015

.510

010

0

Not

es: #

Full

-tim

e eq

uiv

alen

t. T

his

rep

rese

nts

th

e ti

me

the

staf

f m

embe

r sp

end

s at

an

in

stit

uti

on a

nd

on

a s

taff

pro

gram

me.

If

som

eon

e w

as a

t an

in

stit

uti

on

for

six

mon

ths

s/h

e w

ould

get

an

FT

E va

lue

of 0

.5. *

In

clu

des

doc

tora

l an

d p

ostd

octo

ral

stu

den

ts.

Sour

ce: D

epar

tmen

t of

Sci

ence

an

d T

ech

nol

ogy

and

HSR

C (

2013

). ‘N

atio

nal

Su

rvey

of

Res

earc

h a

nd

Exp

erim

enta

l D

evel

opm

ent

(200

9/10

Fis

cal

Year

)’.

Pret

oria

, Sou

th A

fric

a: D

epar

tmen

t of

Sci

ence

an

d T

ech

nol

ogy

and

HSR

C, p

. 32.


in the ISI database, a larger share of South African publications in world science was found in ecology and environment (1.18% in 1990–94 and 1.14% in 1996–2000), earth sciences (1.19 and 1.12%) and plant and animal sciences (1.75% and 1.53%). HIV/AIDS is one area where South African scientists have made outstanding research discoveries. Scientists at the University of KwaZulu-Natal have earned international recogni-tion in their path-breaking research in HIV/AIDS. Researchers at the University of Witwatersrand recently developed an innovative technique to perform CD4 cell-count testing on HIV patients. A whole-genome sequence of the South African strain F11, which is believed to have great significance in tuberculosis research, was released (South African Journal of Science, 2005). Many more can be added to this list. These were not just small steps for South African science. Significant accomplishments in several fields of science made the international community to view South Africa more seriously and with renewed interest. As much as South Africa was interested in collaborative enterprises with the outside world, both developed and developing nations were keen to establish scientific projects jointly with South Africa.

South Africa shares pre-eminence with Australia, Canada and New Zealand in world science research on natural resources (May, 1997). There is a high level of specialization in the country (Burns et al., 2006). This works very well, both for its own scientific growth and international association in scientific research. The country is the home to some inter-nationally recognized journals. Among them are the South African Forestry Journal, Philosophical Papers, Quaestiones Mathematicae, and African Journal of Range and Forage Science, South African Medical Journal and the South African Journal of Science. These journals also functioned as an effective instrument for scientific interaction and collaboration with the interna-tional community. Many scholars within the country and abroad found new openings for association through common research interests. The Scientific Revealed Comparative Advantage (SRCA) for South Africa from 1981–2001 showed consistent specialization in geology/petroleum and mining engineering, general and internal medicine, veterinary medicine and animal health, animal sciences and aquatic sciences (Sooryamoorthy, 2010b). In this parameter of SRCA, South Africa in 2001 demonstrated its lead in three fields of geology/petroleum and mining engineering, ani-mal sciences and entomology/pest control (Albuqerque, 2003, cited in Kahn, 2007). In relation to the ratio of the agricultural R&D to the agri-cultural share of the GDP, the investment made in agriculture research is high by world standards (Kahn, 2007).


The scientific production of a country refers to a set of characteris-tics: the number of scientific discoveries and the number of people mak-ing those discoveries (Ben-David, 1960). In the international journal literature South Africa is the largest country in Africa accounting for more than 31 per cent of Africa’s total publication output for the period 2001–04 (Tijssen, 2007). Tijssen’s study (2007) covered 32 African coun-tries in which South Africa was followed by Egypt, Morocco, Tunisia, Nigeria, Kenya and Algeria in the number of publications. During this period of 2001–04, South Africa produced 14,809 papers, Egypt 9,895 papers, Morocco 3,535, Tunisia 2,857, Nigeria 2,309, Kenya 2,067 and Algeria 2,028 (Tijssen, 2007). Pouris and Pouris (2009) reported that 88 per cent of the utility patents of Africa during 2000–04 originated in South Africa. This was not the case in the domain of intellectual prop-erty. Although not precise, comparisons are revealing. South Africans have received about 100 patents per year in the US, which is 2.5 patents per million population, compared to 776 patents per million for Japan (Department of Science and Technology, 2002).

South Africa’s contribution to world science, in terms of the number of publications listed in the ISI database, reached its highest percentage of 0.67 in 1987, which in 2000 declined to 0.49 per cent (Pouris, 2003). Since 2000 there has been a significant improvement in its position in world science (Pouris, 2012a). The analyses of the ISI database showed that during the last two decades the scientific output of South Africa had grown at a compound rate of 2.4 per cent in contrast to 3.4 per cent for international science (Pouris, 2003). As King (2004) reported,25 South Africa is the only African country that appears in the science citation rank order,26 appearing in the 29th position, while China and India take the 19th and 22nd positions respectively.27 King’s study (2004) of 31 countries including those of the G8 countries and EU illustrated that they account for more than 98 per cent of the world’s highly cited papers while the remaining 162 contributed less than 2 per cent. The countries covered in this study are Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland, France, Germany, Greece, India, Iran, Ireland, Israel, Italy, Japan, Luxembourg, the Netherlands, Poland, Portugal, Russia, Singapore, Spain, South Africa, South Korea, Sweden, Switzerland, Taiwan, the UK and the US. These facts and figures are vital for South Africa if seen through the lens of scientific collaboration. The improved standing of South Africa in the world of science had a magnetic effect, drawing scientists and nations to work with South Africa. It also made things easier for South Africa to conclude scientific deals with nations that it preferred.


Turning to the R&D scene in 1990, the apartheid government spent 0.61 per cent of the GDP on science and technology, most of which was for military purposes (Cherry, 2010). In 1994, the share of the GDP on R&D declined to 0.7 per cent from 1.1 per cent in 1990 (Habib and Morrow, 2006). Comparable figures for other countries show the differ-ence: 3 per cent for Japan and 1 per cent for Australia (Marais, 2000). In 2006, it was 0.95 per cent for South Africa (Cherry, 2010), which dropped to 0.76 per cent in 2011–12. The National Development Plan (NDP) called for more investment in R&D. In response to this, the gov-ernment is targeting 1.5 per cent of its GDP for R&D by 2019. There have been cutbacks on R&D spending by the private sector as well (Department of Science and Technology, 2002). The relocation of some of the major South African companies overseas, accompanied by the relocation of their R&D wing and an increasing rate of disinvestment in R&D by some small companies (Kaplan, 2004), adds to the difficulties of R&D. Kaplan (2004) summarizes the trend: in 1985–86, the expendi-ture on R&D financed by business was 99 per cent; in 1991–92 it was 89 per cent, and in 1997–98 it fell to 80 per cent.

The Gross Expenditure on R&D (GERD) in the region of below one per cent of the GDP (0.87%) does not augur well for South Africa. GERD has decreased by 81 million rands between 2008–09 and 2009–10 (from 20.955 billion to 21.041 billion) (Department of Science and Technology and HSRC, 2013). This does not allow the country to catch up with the current levels of growth in world science (Sooryamoorthy, 2010b). South Africa fares well when compared with other developing countries on the GERD, but this does not justify complacency (Sooryamoorthy, 2010b). The target of 1.5 per cent of the GDP by 2019 would be very significant for the development of science in the country.

Clearly, there is disparity in expenditure across sectors, obviously due to differences in the needs and aims of the individual sectors that fund research. Engineering sciences received the single largest share (24% and 22% respectively for the two periods) of research funding from all sec-tors, but most of its funding came from the business sector. Medical and health sciences followed with a share of 15 per cent in 2004–05, which in 2009–10 had increased to 17 per cent. Social Sciences and humanities received a slightly increased allocation, rising from 12.4 to 13 per cent over these two periods (Department of Science and Technology, 2006; Department of Science and Technology and HSRC, 2013).

The larger share of the publication productivity (90%) of South Africans, as indexed in the ISI database, came from the higher educa-tion sector, most of which is from the five largest research universities


of Cape Town, KwaZulu-Natal, Pretoria, Stellenbosch and Witwatersrand (Kahn, 2007).

Science Councils, the strong arm of scientific research in the country, utilize their monies on the contemporary needs of the society. Spending in 2004–05 was 53 per cent on economic development (unclassified, plant production and plant primary products, animal production and animal primary products, energy, minerals, manufacturing, construc-tion, transport, information and communication technology, and natu-ral resources); 16.2 per cent on society (unclassified, health, education and training, social development and community services); 7.3 per cent on environment; 15.4 per cent on the advancement of knowledge (unclassified, natural sciences, technologies, engineering, social sciences and humanities); and 8.1 per cent on defence research (Department of Science and Technology, 2006). Both the number of researchers and the R&D expenditure over this period increased for the Science Councils in the country. The total count of researchers rose from 1,545 in 2004–05 to 2,251 in 2009–10. The R&D expenditure grew by 73 per cent over these two periods (Department of Science and Technology, 2006; Department of Science and Technology and HSRC, 2013), with the highest increase recorded for the CSIR.

The Academy of Science of South Africa (ASSAf) is now a statutory organization under the Bill of the Parliament passed on 26 October 2001. Originally founded in 1996, ASSAf’s declared mission is to work for the ‘highest level of scientific thinking in the service of the nation and to be an instrument for forming considered scientific opinion’ (Gevers, 2001). The statutory status permits the Academy to control the science system of South Africa, which consists of national science academies, and the Academy helps to establish links with other academies in other countries (Sooryamoorthy, 2010b). The South African Academy for Science and Art is another important entity. Founded in 1909, and initially named de Zuid-Afrikaanse Akademie voor Taal, Letteren en Kunst, the South African Academy for Science and Art is a multidisciplinary organization with representatives from all scholarly cultures and focuses on science, technology, the arts and Afrikaans (Marais, 2000). These organizations currently play an active part in facilitating, encouraging and launching collaborations between scholars and institutions, both within the coun-try and the outside world.

South Africa in 1996 became the first country in the developing world to employ the National System of Innovation (NSI) framework to design an integrated system for the performance and management of innova-tion (Kaplan, 1999 cited in Kaplan, 2004). The country has initiated


a number of specific programmes to develop its scientific strength. The Innovation Fund, Technology and Human Resources for Industry Programme (THRIP), the Competitiveness Fund, the Support Programme for Industrial Innovation, Competitive Grants (for rated and unrated Scientists) and Blue Skies Research Funding are some of them. The national R&D strategy is expected to be the way forward for strengthen-ing science by fortifying institutional and individual research capacities and building global networks (Sooryamoorthy, 2010b). Institutional and individual collaborations are an essential component of many of these programmes.

Science and technology in the post-apartheid era was at first managed under the Ministry of Science, Technology, Arts and Culture, which is now under the Department of Science and Technology. A better future for science is now envisaged. The scientific system is undergoing a tran-sition, which is also visible in the changing demographic profile of the researchers (Department of Science and Technology, 2006). Discernible clearly in the data we have collected and presented in chapter 6 is the demographic transformation that is underway in the country.

The Royal Society of South Africa has produced a discussion docu-ment that explores the areas of research that the country needs to consider for its future (Ellis, 1994). Water supply, agriculture and food supply (fish, animals, birds and plants), energy supply (storage, use and conservation), mineral resources (location, extraction and processing), work (processing, manufacturing and services), environment (assess-ment, processes and restoration), pollution, waste disposal and sanita-tion, transport (energy use, efficiency and cost), health (preventative and curative care), education and information technology and planning are the key areas.

Between 1994 and 2002, the funding South African science sourced internationally rose substantially (Government of South Africa, 2006). Efforts to attract international participation in several ventures are now bearing fruit. Amongst them are the construction of the High-Energy Stereoscopic System (HESS) observatory in Namibia; the Southern African Large Telescope (SALT), which is a multimillion rand joint project with Germany, Poland, the US, New Zealand and the UK in the Northern Cape; winning the bid to host the European Developing Countries Clinical Trials Partnership; and South Africa’s bid to be the site of the Square Kilometre Array (SKA) radio telescope (Sooryamoorthy, 2010b). The state level international agreements have been signed in the fields of material science, manufacturing technology, biotechnology, environ-mental management, natural resources and minerals, medical research,


public health, engineering science, agriculture, mathematics and sci-ence education, and in advancement of technologies (Sooryamoorthy, 2010b). The Department of Science and Technology is now keen to play a leadership role in its cooperation with SADC and BRICS (Association of Brazil, Russia, India, China and South Africa) initiatives (Department of Science and Technology, 2013). What is more obvious today is that South African collaboration is not just confined to some developed nations or with the nations with which South Africa had colonial and apartheid ties but also with the developing nations in Africa, Asia and Latin America. More concrete evidence to substantiate this is presented in chapters 4 and 5.

Conclusion

South African science has drawn its strength, capacity and direction from its long colonial past of nearly three centuries. It is primarily the science of the colonialists and the settlers, transplanted in a gifted land that possessed the vital ingredients for scientific research, interaction, exchange and collaboration. Endowed with rich natural resources and a divergent flora and fauna in a congenial climatic ambience, it was a land of opportunity for the European settlers. For the scientists, bestowed with inquisitiveness and talent, South Africa offered immense possi-bilities for invention, innovation and discovery in fields ranging from the natural sciences to astrophysics. These led to active participation and association of South African scientists with their peers beyond the borders.

The colonial past of the country under the Dutch and the British has had its consequence on scientific collaboration in South Africa. Documenting the history of science in South Africa, Cornelis Plug (2003) rightly pointed out that colonization facilitated a net flow of expertise in two directions, benefiting both the colonizers and the colonized, and as a result starting new systematic work in several branches of science. The colonial government(s) both encouraged and invited scientists and academics to the land and supported their scientific activities, which were indispensable for the country’s progress and development. As part of the settlement and survival strategy in a new habitat, the government found it pressing to seek scientific remedies for diseases that affected humans and their cattle. In these circumstances of immediate neces-sity and urgency, veterinary and clinical sciences and research related to plants and agriculture flourished. The result was the advancement of these sciences over others even in contemporary South Africa.


The element of collaboration with the international community in these endeavours was not neglected.

If not unparalleled in the history of science largely due to the colonial legacy, the focus on specific branches of science and their consequent growth in South Africa are rather distinctive. India, for instance, was also a European colony. Its science and scientific culture were thousands of years old and soundly founded, not brought to it by the colonialists. India was and continues to be a source of several scientific discoveries. Its centuries-old scientific tradition and the value the society attached to it (in mathematics, physics and traditional medicine) did not warrant the implantation of a scientific culture as did that of South Africa under its colonial regime.

South Africa held its scientists and their work in high esteem. This ten-dered a supportive function for the country to achieve some of its great-est landmarks and current levels of standards in scientific research. This gave the country a head start in its scientific endeavours (Schaffer, 1977). South Africa has come a long way and yet has to cover more in the future. Linkages with the outside world, mostly with Europe, which had existed in the past are maintained and are getting stronger as with Asia, Africa and Latin America. South Africa participated, with other African nations in regional and international collaborations and rendered timely leadership. South Africa, along with France, Britain, Belgium, Portugal and Southern Rhodesia (now Zimbabwe), founded the CSA in 1950 for interstate cooperation and to encourage collaboration (Keay, 1976). This was made possible by the strength, position and proficiency in certain crucial areas of science which South Africa had at that time.

The government has taken international collaboration and cooperation more seriously as a means to strengthen its scientific system. Its programme of International Cooperation and Resources initiated by the Ministry of Science and Technology in 2011 aims at developing, promoting and managing international relationships, opportunities and agreements to strengthen the NSI. This programme is intended for the exchange of knowl-edge, capacity and resources between regional and international partners.28

South Africa has entered into agreements with countries in the EU for mutual benefits. It has earned an observer status at the Organization for Economic Co-operation and Development (OECD) Committee for Science and Technology Policy (Department of Science and Technology, 2002). The R&D survey conducted by the Department of Science and Technology (2006) gives a snapshot of the collaboration between local and foreign firms. Of the 165 firms that responded to this survey, 68 per cent have confirmed their collaboration with local universities, 48 per cent with


other local firms in the country and 44 per cent with science councils (Department of Science and Technology, 2006). About 7.3 per cent of South Africa’s R&D (2001–02) were from abroad (Blankley and Kahn, 2004), and this grew to 10.7 and 12.1 per cent in 2007–08 and 2009–10, respectively (Department of Science and Technology and HSRC, 2013).

Collaboration with foreign institutions is carried on more vigorously now in South Africa. In the 1970s the country had collaborative research associations with several universities, research institutes and govern-ment departments in the US, including the University of Minnesota, MIT, the University of California at Berkeley, the US Department of Agriculture and the National Science Foundation. Many of these con-tacts are being maintained in the new, democratic South Africa. The country is now playing a leading role in SADC and establishing coop-eration with countries that are part of BRICS. These relations are mutu-ally valuable and beneficial. Collaborative research is an accepted part of the job profile of academics in universities in South Africa. In several South African universities it is the current norm. As the vice-chancellor of Rhodes University (established in 1904) remarked of his university, many researchers are currently taking part in a growing number of inter-national collaborations (Woods, 2005).

The above review reveals some key dimensions of South African sci-ence. The country benefitted from its contacts with the outside world. Connecting with its colonial past, and to a lesser extent the apart-heid era, South Africa maintains its staunch belief in domestic, intra- continental and international collaborations. Based on past experience, it continues to emphasize that scientific policies are firmly grounded in exchange and associations with the world of science. Concomitantly, South Africa learnt how to be powerful on its own. It did so mainly in specialized areas where it could, attracting international powers in sci-ence and improving its own ability to settle collaborative deals. South Africa recognizes cooperation as an important element in its science policy. This worked well in the past and in the colonial period in par-ticular. Moving beyond its imperial and apartheid connections, South Africa is now increasing its associations with neighbouring countries in Africa and the BRICS countries. The degree of collaboration with these countries, as shown in the bibliometric analysis presented in chapters 4 and 5, is growing steadily. The legacy of the past led to different forms of collaboration: institutional, domestic, regional and international. The past continues to influence indigenous contemporary science.

The next chapter explores the concept of scientific collaboration and its components.

57

Scientific collaboration has been defined in several ways, not always pre-cisely. One reason why it cannot be defined clearly is that its compos-ite components are all not easily comprehensible and measurable. One can identify as many components as possible in it. In this chapter, the crucial components of the concept in the literature are explored and presented in two sections. The first section reviews the importance of collaboration in science today. In the second section some of the major parts of scientific collaboration are dealt with, broken into elementary components.

Significance and relevance

Modern science is irrefutably global and internationally minded (Dienel, 1999). It is no longer an individual enterprise but a team activity. People in science try and apply innovative techniques, methods (Peterson, 1993) and equipment, and continually update these with the intention to produce new knowledge while making their own distinctive marks in their fields of choice and interest. In a profession, members are the possessors and custodians of a special branch of knowledge acquired by long and assiduous study (Bush, 1957) which is unlikely to be accom-plished single-handedly. The inherent nature of scientists to be creative and productive inspires them to look towards newer openings that will take them to the threshold of precocious advancements in their area of research. Working together in science is therefore an acceptable norm and practice for the scientific fraternity. Teamwork, called collaboration, is thus a means to sustain one’s creativity and imagination.1 It can also be fun to do research together as collaborators have remarked about their own experiences (Melin, 2000).

3Scientific Collaboration: Towards Conceptual Clarity


Science is fundamentally a collaborative enterprise (Finholt and Olson, 1997) and scientific problem solving is now part of collective enterprises (Fujimura, 1988). Science does not, or cannot, function in isolation or in seclusion although certain circumstances sometimes warrant it. Science has been increasingly collaborative in the past few decades (Porac et al., 2004; Wagner and Leydesdorff, 2005b) and has become a prevailing trend (Kundra and Kretschmer, 1999), attracting more and more people to this modus operandi. A feature of science is its intense use of network-ing and collaboration (Ziman, 1994).

Being a socio-cognitive practice, science has social reasons to stimulate collaboration (Bozeman and Corley, 2004; Stern, 1996 cited in Melin, 2000). The fascinating growth of science and its increasing complexities (Jeffrey, 2003; Katz and Martin, 1997)2 has spawned collaboration as an essential means to unravel the hidden mysteries of Nature. The inter-sectoral nature of new technologies, along with the cross-fertilization of scientific disciplines and their interrelationships, further accelerate collaborative attempts (Hagedoorn, 1993).3 The literature on science is replete with references to scientific collaboration.4 Deciphering the structure of DNA, one of the greatest discoveries of science, involved the tireless work of numerous scientists. In high energy physics human brains are in league. A sample of recent discoveries in science is ample proof that research alliances between organizations and between indi-viduals work effectively in resolving intricate scientific problems. Two important discoveries may be cited. The development of a scientific model for the cure of Alzheimer’s disease saw the participation of 34 sci-entists, two biotech companies, one pharmaceutical company, a univer-sity, a laboratory and a non-profit research institute. Another research endeavour meant to determine the gene that is susceptible to breast and ovarian cancer had 45 scientists from a biotech firm, two medical schools from two countries, a pharmaceutical company and a govern-ment research laboratory (Powell et al., 1996). It is not unusual in recent times for the Nobel Prize to be awarded to two or more persons for their collective work. Nobel Prizes being awarded to two or more persons for a single project had increased from 14.8 per cent in the 1900s to 60 per cent during 1970–90 (Hafernik et al., 1997). All these suggest that col-laboration is now a strong and indomitable force in science.

Known under several names and forms, scientific collaboration5 has a long history, perhaps as old as science itself. Collectively done, in the same or distant locations—invisible college6 or collaboratories7—it is for a specific purpose and goal, through the pooling of and mutual exchange of resources by the participants. In eliminating errors that

Scientific Collaboration 59

might occur in the individualistic mode of scientific inquiry, collabo-ration has an advantage over the former in that more than a single individual is available to check the conduct of research and its findings (Browne, 1936). Perpetually, scientists look beyond their territories of disciplines and geographical locations, seek others who work in similar areas and establish contacts and networks to realize their dreams. This happens across continents, between north and west, south and east, in all directions and locations and at all times. Universities, research insti-tutes, laboratories, hospitals, museums, industries and government are part of this growing activity.

Collaboration, originating from different continents, regions and countries via the medium of technologies, produces a blend of knowl-edge, products and solutions (Walsh and Maloney, 2002) and it is stimulated by the potential for accessing advanced equipment, data and funds (Katz and Martin, 1997). Expensive equipment required for research makes collaboration an essential requisite for many disciplines. Scientists seek collaborative opportunities when they need access to new and expensive equipment which they or their institutions cannot afford or when a particular skill is required to complete a research experiment. The complex and expensive equipment-dependent nature of science and technology (S&T) makes collaboration worth pursuing to do more with less (Bozeman and Boardman, 2003a).

At the macrolevel, countries hunt for collaborative partners, prompted by the growing specialization of science and the cost factor in con-ducting research and experiments (Luukkonen et al., 1992; Stichweh, 1996). Firms consider collaboration as a mechanism to avail technical opportunities in research (Hicks et al., 1996). The responses of research and development (R&D) managers in European and Japanese firms, as reported by Hicks et al. (1996), reveal that research collaborations are motivated by research efficiency and technical opportunities. Research efficiency implies reduction of costs and risks through sharing them with partners and avails the advantage of cross fertilization and synergy. Accessing technical opportunities, on the other hand, is helping in the recruitment of high-quality researchers, acquiring skills, accessing tech-nology and knowledge, entering into networks and building networks in scientific circles. In view of the mutual benefits on the development and technological front, countries and regions promote scientific collabora-tion between industries, universities and research institutes (Etzkowitz and Leydesdorff, 2000).

There are strong arguments in favour of collaborative undertak-ings between institutions in which industries are also involved.


Inter-institutional structures (Frame and Carpenter, 1979; Landry and Amara, 1998) for sharing resources and production of knowledge lead to institutional collaboration. Traoré and Landry (1997) believed that collaboration originates from strategic, organizational and operational reasons. Often scientific collaboration depends on the differences in scientific resources in terms of R&D expenditure (Acosta et al., 2011). Scientific alliances can lower the cost for participating institutions, gain more in terms of results, monitor developments in technologies, lower the risk of duplication in inventions (Rosenberg and Mowery, 1990), avail technological opportunities and advance scientific under-standing and techniques (Klevorick et al., 1995). Inter-institutional col-laboration, according to Bozeman and Boardman (2003a), takes place because of the limited resources and the potential to bring in a mix of specialties and disciplines. It has a critical role in developing scientific and technical human capital (Bozeman and Corley, 2004),8 and it is a means to enhance effectiveness. These advantages draw both small and large firms closer to work jointly on specific projects combining their resources—equipment, machines, material, skills and technical know-how. At the same time, industries have their preferred areas that encour-age alliances.9 In the pharmaceutical industry, medical professionals and scientists work together very effectively (Bush, 1957).

Industries are not only active in inter-institutional collaborative research but also contribute assiduously in the circulation and diffusion of knowledge through scientific publications (Godin, 1996).10 Industrial firms have built up their research capacities and extended linkages with universities and research institutions (Meyer-Krahmer and Schmoch, 1998), and academics are favourably disposed to this move (Lee, 1996). A national survey of about 1,000 faculty members of research-intensive universities in the US has indicated that the US academics were more favourably disposed towards closer university–industry collaboration in the 1990s than in the 1980s (Lee, 1996). Any university–industry collab-oration is subjected to a host of factors namely, institutional culture and structural differences, infrastructural disparities, locational proximity, preferences in areas of research and cost–benefit considerations. Western countries, via mechanisms such as grant programmes, enhance scien-tific collaboration between university and industry (Landry and Amara, 1998).11 University–industry collaboration in the US has been growing in the 1990s as is clear from the swelling funds and the production of academic publications (Bozeman, 2000).12

A substantial hike in the number of collaborations between different institutions and also in the number of scientists and their publications


is evident (Price, 1963). International scientific collaboration has been growing not only in volume but also in relevance (Luukkonen et al., 1992; Yoshikane and Kageura, 2004), doubling between 1973 and 1981 in several fields of science (Frame and Carpenter, 1979; Luukkonen et al., 1992).13 The past three decades in the US were termed the ‘era of inter-institutional research collaboration’ (Corley et al., 2006). In particular, during 1981–99, the rate of domestic institutional collabora-tion between different universities and between universities and firms in the US more than doubled while foreign collaborations increased fivefold (Adams et al., 2005).14 More in this shift was seen when the US moved its preference away from decentralized, small-scale, individual-initiated projects to centralized, large-scale, multidisciplinary research (Bozeman and Boardman, 2003b). At the university level too funds are being earmarked in the US for inter-university collaborative research by prominent funding agencies. The National Science Foundation, for instance, through its S&T centres and its sponsored engineering research centres, supports this mode of collaboration (Porac et al., 2004). In Canada, programmes are designed at the governmental level under which the eligibility of research grants is made conditional to collaborative research between research institutes and research teams (Landry and Amara, 1998).

The European Commission increasingly supports collaboration between countries (Katz and Martin, 1997).15 In the 1980s, European companies were more aggressive in seeking research links with outside organizations (Hicks et al., 1996).16 Evident in joint publications, col-laboration between the European Union and other countries, industrial-ized countries in particular, had drawn an upward curve between 1985 and 1995 (Georghiou, 1998).17 Collaborative research between European and Japanese firms increased sharply in the 1980s (Hicks et al., 1996).18 Japan’s S&T policies and appropriate structures assist collaborations between universities, industries and government laboratories (Wen and Kobayashi, 2001). Joint research with the private sector, commissioned research and cooperative research are the systems that encourage col-laboration. The Japan Society for the Promotion of Science (JSPS), a quasi-governmental organization also plays a key role in promoting col-laborations. Its system of joint research with the private sector, started in 1983, helps researchers at national universities to conduct joint research in their laboratories.19

Sweden is changing the way research had been conducted in the country, as seen in the Swedish R&D policy (1997–09), which advocates that academics seek collaboration with organizations. Collaboration


between Sweden and other countries has progressed, resulting in the advancement of Swedish industrial scientific research (Okubo and Sjöberg, 2000). Sweden’s internationalization of scientific research, in comparison to other industrialized countries, is very high; more so in the case of firms which collaborated with international competencies in science. This internationalization developed faster when the firms participated in research than when research was conducted by universi-ties and other public institutions (Okubo and Sjöberg, 2000). Private companies in Sweden get integrated with national and international academic networks for collaboration, funds flowing from them to uni-versities to intensify scientific collaborations (Okubo and Sjöberg, 2000). In Latin American countries, international collaboration is reliant on size20 (Gómez et al., 1999): the size of the country is inversely related to international collaboration, and the size of the country is related to participation in bilateral collaboration. The size of the national R&D systems and geographical distances also govern the network of collabo-ration (Anuradha and Urs, 2007).

Institutions encourage tie-ups that are now politically supported, apparently due to their implications for technological progress and development. Governments advocate international collaborations because it is a political objective and a matter of foreign policy (Gupta and Dhawan, 2003; Luukkonen et al., 1992). Political changes in Eastern Europe led Western scientists to collaborate with their counterparts in the East for stronger political and cultural ties (Katz and Martin, 1997). Russian scientists sought collaboration with the West when the government support for science sank in the early 1990s (Wilson and Markusova, 2004) as a direct consequence of the political changes that the country had undergone. Sometimes scientific collaboration between countries that are fundamentally hostile happens. For instance, scien-tific and technical collaboration between France and Germany took place in several fields when they were foes during 1860–1950 (Dienel, 1999). Scientists in both countries put aside the prevailing hostility between the two countries and maintained close personal contacts (Dienel, 1999). During the apartheid era, South Africa had similar link-ages with other countries that politically and economically ‘boycotted’ South Africa. Despite political conflicts, Israeli scientists were open to collaboration with Palestinian scientists (Nature, 2002).

In the UK, as reported by Hicks and Katz (1996) in their examination of co-authored publications during 1980–90, collaboration of scientists with more than one institution has risen steadily. They predicted that if


this trend continues, the share of collaborative articles will exceed the share of non-collaborative articles around the turn of the century. The production of co-authored papers, an anticipated outcome of collabo-ration and a proxy index of collaborative research, has shown remark-able increase in the last few decades (Hicks and Katz, 1996; Melin, 2000).21 Wagner and Leydesdorff’s (2005a) analysis shows that co-authored papers doubled between 1990 and 2000, accounting for 15.6 per cent of all the papers published, with an expansion of core coop-erating countries from 37 to 54. In the US, according to the National Science Board (2004, cited in Walsh and Maloney, 2007), the percent-age of scientific papers that had two or more authors increased from 48 to 62 per cent in 2001. In 1994, the UK, the second-largest pro-ducer of scientific publications in the world, had 26 per cent of their publications produced through international collaborations (May, 1997). Discipline-wise, the share of co-authored publications in phys-ics expanded in the second half of the 19th century, and in the first decades of the 20th century for mathematics (Wagner-Döbler, 2001). Corroborating this is Kim’s (2006) recent findings from the study of physicists in South Korea that suggest that multilateral collaborations (with other countries) have increased considerably in the last 20 years. To check the scenario in South Africa, a large amount of data on co-authored publications by South African scientists will be examined in the following chapters.

Funding agencies that support research opt for interregional or inter-national collaborative projects rather than individual projects located in a single institution or in a single country. Governmental agencies and private foundations formulate policies that aim at accelerating inter-institutional collaboration (Rosenfeld, 1996). The National Research Foundation (NRF), the national agency endowed with the responsibility of supporting research and training postgraduates in South Africa, asks for the nature and significance of inter-institutional collaboration of the projects that are submitted to it for funding.

Implicit in the above account is that collaboration has come to stay as an unavoidable and preferred path that is central to the advance-ment of science and knowledge production. Increase in communication both in international air traffic and international telephone calls (Hicks and Katz, 1996)22 augment this scientific process directly and indirectly. To carry on with research alliances, laboratories have transposed into collaboratories. Collaboration is now a widely accepted and recognized course of action in scientific research.


Conceptual components

Collaboration, in common parlance, is viewed as a voluntary coopera-tive agreement between individuals or between organizations. Scientific collaboration is a behavioural association to share knowledge, skills and expertise for the completion of the mutually agreed goals (Genuth et al., 2000; Sonnenwald, 2007). It is an important component of sci-ence, technology and innovation policy (Pouris and Ho, 2014). It is human behaviour that facilitates sharing and completing tasks together (Olson et al., 2007). Collaboration is a multifaceted concept. Subsumed under it is a string of elements extending from geographical location, process, partners, division of work, coordination and benefits to chal-lenges. Broadly and generally, these components can be grouped into personal, academic and organizational. Interpersonal relationships, moti-vations and trust are personal elements, while knowledge, skills, expe-rience, expertise and creativity form part of the academic dimensions. Organizational elements in joint ventures encompass institutional fea-tures, leadership, coordination, administration, management, commu-nication and ability to deal with conflicts. A more precise breakdown of the elements of collaboration, covering mostly the structural dimen-sions rather than the personal, is seen in Hackett (2005) for whom the components are: (i) extent (measured as a distribution over substan-tive, social or geographic space, or over time); (ii) intensity (the fre-quency or significance of interaction among persons, places or units of time); (iii) substance (the aims and content of collaborative work); (iv) heterogeneity (the variety of participants, purposes, languages); (v) velocity (the rate at which results are produced, analysed, inter-preted and published); and (vi) formality (ranging from contractual arrangements among nations or organizations to handshake agree-ments and unstated understanding among friends and acquaintances). Chompalov and Shrum (1999) identify seven structural dimensions in institutional collaboration: project formation, magnitude, interde-pendence, communication, bureaucracy, participation and technologi-cal practice.23

Although not always defined in precise terms (Katz and Martin, 1997), the concept of collaboration merits conceptual clarity. It is working closely with others to produce new scientific knowledge or technology (Bozeman and Corley, 2004; Katz and Martin, 1997) or a ‘class of activ-ity requiring the active and reciprocal involvement of two or more peo-ple for the achievement of some joint aim’ (Watts and Monk, 1998), or a system of research activities by several actors in a functional way


(Laudel, 2002). The constituent elements of form, goal, duration, part-ner, location, context, leadership, organizational structure and use of modern technologies including information and communication tech-nologies (ICTs) make collaboration a complex phenomenon that can-not be understood from a single perspective. Not easy to measure and predict without clear indicators, collaboration is assessed in terms of rate and its relation to certain key variables. The rate of collaboration falls off strongly with geographical distance and the differences in language, history and culture (Hicks et al., 1996) or increase with the stock of R&D, private control and encouragement through professional awards (Adams et al., 2005).24 Scientists in countries who publish a higher per-centage of papers are likely to collaborate with institutions in the same country as a larger share of the potential partners work in the same country (Hicks et al., 1996).

Collaboration is a collective enterprise but it is more than the sum of individual participants, particularly in matters of knowledge that it produces jointly (Buber, 1970 cited in John-Steiner et al., 1998). For many, internalization of skills that are scarce will be a priority in collaboration (Fox and Faver, 1984; Hamel, 1991), though collabora-tion is not a means to compensate for the lack of skills (Powell et al., 1996). Scientific collaboration reportedly increases the effectiveness of research, raises its quality (Adams et al., 2005; Fox and Faver, 1984; Stephen and Levin, 1987) and builds the scientific capability of scien-tists (Oliver, 2004).

Collaboration is the other side of competition in scientific activi-ties. Collaboration and competition are not mutually exclusive but two aspects of the same concept (Atkinson et al., 1998). Some prefer col-laboration to stressful competition, accepting competition as morally unworthy and socially unimportant (Hagstrom, 1974). In collabora-tive circles—a primary group of like-minded individuals—people share similar occupational targets through fairly long periods of association (Farrell, 2001). A partner in a collaborative endeavour is identified by his/her role in it, his/her active involvement in all major activities con-cerning it and the decisive contribution he/she has to make towards its fullness. The commitment, in regard to shared resources, power and talent (John-Steiner et al., 1998), by the partners has to be recognized. It is aligned with the idea of equal participation, responsibility and rep-resentation in a comfortable and friendly environment (Pérez et al., 1998). Scientists collaborate with others also when they are quite strong in their own areas of research, and expect the same kind of strength in other partners.


Motivations, determinants and origins

Manifested in a spectrum of dimensions, scientific collaboration is driven by a host of causes: specialization of science (Hardwig, 1991; Melin, 2000; Stichweh, 1996), internal differentiation of disciplines (Stichweh, 1996), ever-growing complexities in science and disciplines in particular (Frame, 1979), cross-fertilization of disciplines (Beaver and Rosen, 1979), data (Wagner, 2005),25 access to sophisticated expensive equipment (Frame, 1979; Wagner, 2005), pooling of resources, tal-ents, skills and knowledge (Beaver, 2001), obtaining mutually benefi-cial results by merging the scientific assets of the partners (Porac et al., 2004), the desire for researchers to enhance their professional visibility (Beaver, 2001; Luukkonen et al., 1992; Wagner, 2005), career advance-ment (Beaver, 2001; Luukkonen et al., 1992), improved productivity (Barjak, 2006; Beaver, 2001), professionalization of scientists (Beaver and Rosen, 1978; Katz and Martin, 1997)26 and scientific institutes (Beaver and Rosen, 1978), changing patterns of funding, advancement in com-munication technologies, or reduction of isolation (Beaver, 2001). This list is not exhaustive.

Collaboration is a preferred choice when costs of conducting research are escalating (Katz and Martin, 1997). For marine scientists and geolo-gists, expensive equipment is a close companion. In order to share the facilities and equipment and to reduce costs, attempts are made to force researchers to communicate and collaborate (Ziman, 1994).

Luukkonen et al. (1992), in their analysis, find that the high level of international collaboration that exists in several scientific fields such as astronomy, oceanography and atmospheric and space science is a result of the high cost of equipment and resources—telescopes and observatories—involved. Markusova et al. (1999 and 2000 cited in Wilson and Markusova, 2004) report that external collaboration was greater in areas such as physics and space research. In the same way, funds are crucial in scientific research and their availability often dic-tates collaboration. The likelihood of researchers to collaborate will tend to increase when the research budgets increase (Belkhodia and Landry, 2007). To Luukkonen et al. (1992) the less developed the scientific infra-structure, the higher the tendency for international collaboration in areas of co-authored publications.

At the individual level, a gamut of motivations comes into play. Melin’s study (2000) identifies increased knowledge, that is, gaining knowledge or getting access to methods and equipment, as a compel-ling reason. Bozeman and Corley (2004), in a study of scientists and engineers in the US, have used 13 factors to this end, entailing an


entire range of motivations. Chief among them refer to acquaintance, previous collaboration, reputation of the collaborator, skills and work ethic.27 Collaboration is viewed as a self-organizing system in which the selection of partners and the location is made by the researchers themselves (Wagner and Leydesdorff, 2005b), suggesting the personal, idiosyncratic dimensions of collaboration. Knowledge sharing between partners becomes easier when they have prior experience in associat-ing with each other and profound experience in knowledge domains (Porac et al., 2004). For some, collaboration itself is an incentive and provides a built-in system of support (Hafernik et al., 1997). The incen-tives are sometimes relative and some researchers have higher incen-tives to collaborate than others (Landry et al., 1996). The likelihood of collaboration increases with prior collaboration experience between partners (Balakrishnan and Koza, 1993). A number of collaborations are launched on the edifice of previous contacts, networks and joint work-ing experience. There are cases in which the originator had known oth-ers previously or had worked together on some other projects (Bozeman and Boardman, 2003a) or had long-time friendships (Melin, 2000). Prior collaboration experience with a partner can accelerate collaboration (Katila and Mang, 2003).28

Scientists and researchers hold the freedom to associate or not to asso-ciate with others. There are individuals who would not collaborate with anyone under any circumstances (Bush, 1957). Some, like Nobel lau-reates, are very prolific collaborators. Laureates are apt to collaborate more often with other laureates, laureates-to-be and those who are dis-tinguished and highly productive (Zuckerman, 1967). When members of a profession have a better understanding about what others are doing, have a genuine interest in others’ subjects or have the conviction that joint efforts would lead to the creation of new knowledge that would not be otherwise possible, the chances for collaboration are greater. This, in other words, implies that collaboration can take place under formal and informal settings. Formal and informal structures give rise to different forms of partnership, with an obvious preference for the latter. It is natu-ral that most of the time collaborations begin at informal personal levels of interaction before they are taken to higher levels. In the US, about 90 per cent of the research partnerships with other firms and universi-ties from a sample of the US manufacturing sector were informal (Link and Bauer, 1989). Hagstrom (1964) reports the correlation between the frequency of informal contacts and the frequency of collaboration.

Socio-technical conditions are relevant in inspiring collaboration that occurs from distant locations. The determinants of collaboration vary


according to the contextual factors, the individual benefits derived from the association (Landry and Amara, 1998) and disciplines (Belkhodia and Landry, 2007; Melin, 2000). Geopolitical aspects, geographical prox-imity (or non-proximity), history, language and cultural tradition affect collaboration between countries at the macrolevel (Gupta and Dhawan, 2003; Luukkonen et al., 1992).

For Olson and Olson (2000), working together at a distance is yet another kind for collaboration. Identifying four conditions such as com-mon ground, coupling of work, collaboration readiness and collabora-tion technology readiness, Olson and Olson (2000) note that groups with high common ground and loosely coupled work, with readiness for both collaboration and collaboration technology will have a fairly good chance to succeed in remote work.

Studying large-scale multi-organizational collaborations, Genuth et al. (2000) report that the formation of such collaborations can be construed as the interplay of factors, namely, the interpersonal context, the donor context, the sectoral context and the home-organization context. By forming large-scale multi-organizational collaborations, sci-entists create a context which, they believe, would serve them better than others (Genuth et al., 2000). Emerging from Hamel’s (1991) study are three broad determinants of learning outcomes: intent, transparency and receptivity. Intent is the initial propensity of the institution to view collaboration as an opportunity to learn; transparency is openness to each partner and therefore the potential for learning; and receptivity is a partner’s capacity for learning and absorptiveness (Hamel, 1991). While accepting the fact that collaboration has salutary effects on the partners by way of sharing and internalizing the skills of the other, this might not be always the case. In this sharing and learning of knowledge, all the partners cannot expect to be equally adept.29 Wagner and Leydesdorff (2005b) group the factors that lead to the growth of international col-laboration into those relating to diffusion of scientific capacity and to the interconnectedness of scientists.30 The underlying assumption and also motivation—inter-institutional alliance included—is that the accu-mulation of knowledge from organizations can build the intellectual mass necessary for scientific breakthroughs (Porac et al., 2004).

Forms

Collaboration has travelled a long way from its traditional forms that in many ways had become redundant and dysfunctional. In tradi-tional forms, teamwork was rather limited to peers, teachers and stu-dents (Hagstrom, 1964).31 Characteristically, in the modern forms of


collaboration scientists have greater dependence and a complex division of labour (Hagstrom, 1964) among themselves. The variants of these new forms differ.

Katz and Martin (1997) classify the variants broadly into intra- and inter-collaboration between individuals, groups, departments, institutions, sectors and nations, which can be homogeneous (either the intra- or the inter-form) or heterogeneous (mixture of intra or inter). Maienschein’s (1993) three categories, based on the reasons for scientists to associate, are: to promote an efficient division of labour, to enhance credibility and to build community. Shrum and Morris (1990, cited in Chompalov and Shrum, 1999) categorize scientific structures consisting of institutional collaboration in terms of three dimensions of economy (collective or pri-vate beneficiary of the outcome), size and complexity, and epistemology (degree of uncertainty). The tradition and culture of discipline, academic and intellectual background and distance between locations can also determine the forms of collaboration (Melin, 2000). Using bibliometric records Kim (2006) studied both symmetrical and asymmetrical research collaboration on the premise of the extent of participation of the partici-pating scientists. In symmetrical collaboration, scientists from different countries enter into a more or less equivalent relationship as against the asymmetrical type in which the transfer of knowledge takes place one to another rather than in a more or less equal fashion.

Categorization is made on the grounds of the geographical loca-tions of the participating scientists as well. Collaboration takes place at the local, national, regional and international sites. Local collabora-tion involves local participants from within the province; the national type has scientists from outside the province but within the country; the regional takes collaborators from the region (like Africa or Asia) but outside the country; and the international type involves scientists from countries outside the region associating with each other. If the place of activity is considered, scientific collaboration can be domestic (local) or international. Put simply, ‘domestic’ is between scientists within the same locale, while ‘international’ is between scientists in developing countries or between scientists in developed and developing countries (Sooryamoorthy et al., 2007). Domestic can be further classified into regional or national.

Laudel (2002) groups collaboration into those involving a division of labour characterized by a shared research goal and a division of creative labour; service collaboration in which the research goal is set by one of the collaborators alone; collaboration wherein there is a provision of access to research equipment; and collaboration for the transmission of


know-how. Examining 53 collaborations in physics and allied disciplines, Genuth et al. (2000) deduce five types. In the first type, dominant sector/conventional collaboration, members of similar organizations are con-vinced that they have excellent research potential if they work together. This type is formed without pressure from the organizations but with a dominant sector. The second type, the dominant sector/unconven-tional, is formed when scientists and their organizations have to work together to revive their scientific prospects with a dominant instigat-ing sector and a clear source of funds for collaboration. Entrepreneurial funding collaboration—the third type—begins when an accomplished scientist has fears about being competitive in his/her field in the aging facilities of the home organization. These formations arise in a situa-tion wherein neither the government nor the home organization re-capitalizes the facilities, leading to the grouping of a few organizations which can share the costs and the benefits. Obviously, there is no ready source of funds for these collaborations, but they will be able to man-age from self-funding, philanthropies and other sources. The business-as-usual collaborations, the fourth form, are conceived when scientists with a vision succeed in selling the vision to other scientists who have the requisite competencies to materialize the vision. In this genre, as dis-tinct from the previous forms, no single sector dominates, nor have the collaborators worked together earlier. The externally brokered collabora-tion, the last type in this classification, takes a different route to its ori-gin. Here, scientists having a successful career with a secure position in a secure organization seriously consider a brilliant research opportunity that demands collaboration. New facilities that require people to use them effectively or funding prospects might promote this form. These formations are less likely to be created on prior relations because of the interpersonal competition in selecting partners.

Chompalov et al. (2002), referring to the same set of institutions studied by Genuth et al. (2000), came up with a four-category taxon-omy of the inter-organizational and managerial mechanisms of multi- institutional collaboration in physics and allied sciences on a spectrum of bureaucratic to participatory types. This classification of manage-ment of collaboration into bureaucratic, leaderless, non-specialized and participatory reveals the internal functioning of collaboration forms. Bureaucratic collaborations under this categorization have features of Weberian democracy—hierarchy of authority, written rules and regu-lations, formalized responsibilities and division of labour. Similar to this type is non-specialized collaborations but with relatively less for-malization and differentiation, while in participatory collaborations,


bureaucratic features are obviously absent (Chompalov et al., 2002). Leaderless collaboration has formally organized and highly differenti-ated structures that ensure that private interests are not stamped on the collaboration and that appropriate people stay focused on specialized tasks. As the study brings out, leadership and managerial issues merit careful examination in collaborations. Clearly evident from the study is the absence of a scientific leader in a considerable number of organiza-tions, missing out the opportunity to give an inspiring leadership, both intellectually and administratively, to the collaborators.

Disciplinary nature

A decisive factor in collaboration is the discipline. As Bozeman and Boardman (2003a) rightly point out, collaborations develop at differ-ent rates, contingent upon the nature of the science, goals of science and disparities in scientific fields and disciplines. Subject and area of specialization are always a strong determinant in the rate of collabora-tion. Certain fields such as biotechnology and space science invite more collaboration than other branches. Bibliometric studies (for instance, Frame and Carpenter, 1979) indicate that biochemistry, earth/space science, physics and mathematics have predominantly higher levels of international collaboration. Collaborations are decisive for develop-ment and growth in certain disciplines. Some disciplines have a pref-erence for collaboration. Discipline-wise, collaboration preference, as Belkhodia and Landry (2007) indicate, shows that the respondents in earth sciences, engineering, chemistry, computer sciences and life sci-ences are more positive towards collaboration than those working in other areas. Natural sciences, engineering and medical sciences are in the forefront of collaborative research. Biotechnology, as an example, looks for research alliances for its maintenance, development and sur-vival (Oliver, 2004);32 and as a result, the researchers in this field tend to publish more and obtain patents more frequently (Blumenthal et al., 1986).33 Certain disciplines, such as physics (various branches of special-ization), biochemistry, space science, chemistry (some specializations at least), oceanography and geology, normally require collaborative efforts for further development. Hagstrom’s (1964) interviews with the faculty members at the University of California reveal that a considerable per-centage of academics work in groups in the disciplines of physics (97%), chemistry (88%) and experimental biology (62%). Luukkonen et al.’s (1992) study of collaboration hinted that the disciplines of earth and space, mathematics, physics, biomedicine, biology, chemistry, engineer-ing and technology and clinical medicine have most collaborations.


True to this, Adams et al. (2005) have found rapid growth in alliances within universities in agriculture, biology, chemistry and psychology.

Institutional structure and cultural antecedents

The culture of scientific collaboration is both a trait and a capability. This is acquired over time in response to developments in science and to the organizational preferences to take advantage of the growth of science through partnerships. Bozeman and Boardman (2003a) stress the need for a good management and organizational culture to suit the collaborating institutions. The organizational requirements for early stage research collaborations and for research centres are typically dis-similar (Corley et al., 2006). In most cases, in the early stages of collabo-rations, there would be a provision for external resources, agreements about sharing these resources and stipulated conditions for accessing them. In fully articulated research centres, one can expect to have a hierarchy, multiple resources, administrative system, apparatus for the allocation of common pool resources, multiple professional and organi-zational roles, authoritative plans and objectives, performance standards and diverse stake holders (Corley et al., 2006). In a cumulative manner, each of these gives rise to an institutional culture when collaborative alliances come into play.

The institutional ambience favouring collaborative enterprises per-meates down to scholars. This, in turn, enthuses scholars to make use of the available opportunities for joint ventures in their own field of interest and specialization. In order to cultivate this institutional environment, a set of structures, including supportive administrative machinery and an incentive system, need to be in place. Bureaucratic delays and indifference to joint initiatives will not help but rather deter scientists from entering into partnerships. In a suspicious and dampened environment of scientific activity, as is the case in some institutions, col-laborative culture cannot take root. Initiatives at the institutional level by way of identifying niche areas of research, seeking prospective tie-ups with other institutions, creating supportive structures at the administra-tive and human resource levels and developing an appropriate reward system—including funding, facilities and career advancement—foster a collaborative culture.

In some institutions collaboration is recognized as part of their declared vision and mission, and fitting structures are created to meet this goal within the given institutional framework. A regular admin-istrative set up—a collaborative research cell, for instance—within the organization can inspire, support and monitor as well as incorporate


lessons learned from previous collaborations. Scientists will have to syn-thesize the tenor of this collaborative spirit, abetted by the institutional apparatus of their organizations. On the other hand, existing norms and structures of institutions would dissuade aspiring scientists from form-ing collaborations. Genuth et al.’s study (2000) has shown that scientists who wished to be part of a multi-organizational collaboration had to secure approval from their own organizations and warns that the opera-tion of such collaborations become an arena where broader conflicts are played out.

Benefits and rewards

Why do researchers prefer collaboration to working alone? Interviewing 195 collaborators, Melin (2000) reports that, for the majority, the ben-efits are increased knowledge and improved quality, which would not have been possible if the scholars had worked alone. Partners learn from everyone else in the team and bring in ‘different’ aspects of the research in question and contribute to the achievement of a higher quality of research. Collaboration has an enriching effect on the collaborators—personally and professionally (Hafernik et al., 1997).34 While fostering professional development (Clark et al., 1998), learning is integral to collaboration. Commencing with the conception of the project, learn-ing continues through to its completion. Contributions and sharing of knowledge among the associates are therefore useful benefits for the partners, although they are largely reliant on the degree of participa-tion and professional experience. Researchers are attracted to structures that not only generate adequate funding for their projects (Landry and Amara, 1998) but also advance knowledge and productivity (Belkhodia and Landry, 2007). Melin’s (2000) respondents were clear that the quality of the work would rise when more people were involved in interacting and discussing the research problem. To be precise, these collaborators acknowledged that their partners have a special competence (41%), spe-cial data or equipment (20%) or social reasons of friendship and past collaboration experience (16%), while mutual exchange of ideas and thoughts were more rewarding for some others (Melin, 2000). Added to this list are the alluring benefits such as publications, working in teams, updating skills, gaining new expertise, widening the knowledge hori-zon, professional contacts and networks.

In a similar vein, Bozeman and Boardman (2003a) in their study of two institutions, acknowledge the place of incentives for the partners in alli-ances. The issue of incentives, for them, is centred on its three-pronged dimension of incentive sufficiency, incentive alignment and incentive


compatibility. There is an incentive sufficiency when the partners see that the alliance will result in a valued outcome. Incentive alignment is about the alignment of the incentives of the partners in which these add up to the outcomes of the project. Incentive compatibility is the ability of one party to achieve its objectives without obstructing the ability of the other party to achieve its objectives (Bozeman and Boardman, 2003a). In some South African universities, collaboration is valued in terms of the performance of academics.

There is also a flip side to the returns of collaboration. The view is that the benefits that accrue to the participating researchers are not always equal and it is a source of tension between partners (John-Steiner et al., 1998). The argument is also that there are always transaction costs that might offset the benefits of research partnerships (Landry and Amara, 1998). Hagedoorn et al. (2000) summarize the incentives of research partnerships as transaction costs and strategic management.35 The incen-tives are to: minimize transaction costs (including technical knowl-edge), avoid high costs of internalizing the activity, share R&D costs, pool risks, improve the competitive position or co-opt competition, learn from partners, transfer technology, increase efficiency and energy through network and acquire new skills and capabilities. Collaboration can turn into a mechanism that could facilitate the effective transfer of knowledge to enhance capabilities (Hagedoorn et al., 2000), skills, com-petencies and efficiency (Fox and Faver, 1984; Hargens, 1978) and enter new areas of scientific inquiry (Dodgson, 1991).

Co-authored papers are the benefits of collaboration. International papers reportedly have a higher level of attractiveness and impact than the national ones (Benavent-Pérez et al., 2012; Glänzel et al., 1999, Sooryamoorthy, 2009a). They, in comparison to single-author papers, are highly cited in science (Narin, 1991, cited in Wagner and Leydesdorff, 2005b) and judged more favourably (Presser, 1980) with a higher acceptance rate (Lawani, 1986).36 If citation rates of co-authored publications are any indication, collaboration increases the quality of research and publications (Goldfinch et al., 2003). Increasing the num-ber of authors, countries and institutions increases the expected citation rates of publications.37

Likewise, the impact factor of co-authored papers is higher than that of the indigenous papers (Basu et al., 1999 cited in Basu and Aggarwal, 2001). It has also been acknowledged, as shown in Katz and Hicks’s study (1997) of the publications of UK scientists (life sciences, natural sciences, engineering and material sciences, and interdisciplinary sci-ences), that the impact of the papers in any chosen discipline is higher


if there is a component of collaboration of some kind; and the impact is the highest when the publications involved collaboration of foreign institutions. The impact is increased exponentially with the number of collaborating authors from the same institution and linearly with the increasing number of domestic and foreign institutions.

The influence of a reward system in international collaboration has not received much attention (Wagner and Leydesdorff, 2005b). Institutional preferences for collaborations filter down to individual sci-entists if they are linked to an acceptable reward system. In South African higher-learning institutions publication productivity is valued and an incentive system operates effectively to encourage it. By way of grant-ing productivity units, the incentive system brings in research money to researchers. Since the introduction of a formula in 1984, research money is paid out to the university academics and scientists who pub-lish papers in South African Post Secondary Education (SAPSE) approved peer-reviewed journals. The SAPSE list consists of the journals that appear in major abstracting/indexing databases such as the ISI and the IBSS. This is actually the basis on which the universities in South Africa obtain government funding. Each journal article brings about 80,000 rands (Vaughan et al., 2007); this has now increased to over 100,000 rands. In some universities, a part of this government funding is given to the authors who are affiliated to South African institutions, although the amount paid out to the academics varies and some institutions do not share these funds with the researchers who have SAPSE publications. In 2012, some universities in South Africa paid up to ZAR 24,000 to each single-author publication in approved journals. Given the undeniable effect of productivity on careers, researchers are inspired to seek out pos-sible means of collaborative alliances that would maintain or increase their productivity. While accepting this as a reasonable incentive system to reward research productivity, Vaughan et al. (2007) present the other side of it by pointing out that it has become a most perverse incentive, mitigating against long-term high-quality research and encouraging South African researchers to publish inconsequential papers in the least demanding journals. This is true. Some South African universities are looking at the quality of publications based on the h-index of authors.

Productivity

Scientific collaboration produces tangible results. Increased productivity is an inspiring force behind the intention to associate. Studies (Duque et al., 2005; Lee and Bozeman, 2005; Pao, 1982; Price and Beaver, 1966; Sooryamoorthy and Shrum, 2007; Sooryamoorthy et al., 2007;


Sooryamoorthy, 2013a; Zuckerman, 1967) have underlined the close relationship between collaboration and productivity although it is not always straightforward for individual, institutional and research reasons. Because of the division of work (i.e. collection, experiments, analysis and writing up) and the number of people involved in joint projects, partners have reasons to believe that they are able to produce more papers than they would have done single-handedly (Melin, 2000).

Productivity comprises publications, citations of papers and patents produced as a result of collaboration (Irvine and Martin, 1985; Landry and Amara, 1998; Lee and Bozeman, 2005; Pelz and Andrews, 1966). Pertinent in collaboration is the adage to ‘publish or perish’ (Luukkonen et al., 1992). The index of the visible outcome of collaboration is the number of co-authored papers produced. Landry and Amara (1998) assert that an increase in the intensity of collaboration is associated with an increase in the number of publications that come out of the alliances.

Since the publication of the first collaborative paper in 1665, the pro-duction of co-authored papers has increased tremendously (Luukkonen et al., 1992). Within a short span of eight years, between 1986 and 1994, joint publications of authors from more than three countries have increased to 25 per cent from 17 per cent (Okubo and Sjöberg, 2000). Large variation in the rates of international co-authorship and collaboration between countries has also been recorded (Luukkonen et al., 1992). In developing countries, this has been highlighted by a pio-neering study of Duque et al. (2005), who reported the conditions that make the relationship between collaboration and productivity problem-atic and undermine the collaborative benefits of new information and technologies.

Publication productivity as a measurable variable can quantify the outcome of any research collaboration. Glänzel and de Lange’s (2002) examination of international links of countries have concluded that the ratio of the number of international links and international papers is roughly proportional to the ratio of full and fractional publication counts. In another measure, employed to determine the success of col-laboration, productivity is a major variable. The degree of collabora-tion (Subramanyam, 1983) is measured by the ‘ratio of the number of collaborative research papers to the number of research papers pub-lished in the discipline during a certain period of time’.38 Confirming the effect of scientific collaboration on the productivity of participants, Lee and Bozeman (2005) establish that while the number of collabora-tors is a predictor of publishing productivity in the normal count of papers published in peer-reviewed journals, it is not so in the fractional


count of the papers of the collaborators. Among researchers in the natural sciences and engineering researchers in Canadian universities and government agencies, the likelihood of them collaborating, as Belkhodia and Landry (2007) report, increases their productivity. Basu and Aggarwal (2001) confirm the interrelationship between interna-tional collaboration, productivity and the impact factor in their study of Indian science.

Again, variance can be observed in the productivity of partners in col-laboration, subject to the fields of research. Landry et al. (1996) predict that the partners in engineering, natural sciences and social sciences produce more in collaborative settings than those in the humanities. At the same time, there are two contrasting views on the effects of col-laboration on productivity: collaboration does not increase scientific productivity and it really increases productivity (Landry et al., 1996). In spite of all these arguments and counterarguments, for and against, collaboration which is nourished in an affable environment of mutual trust and recognition has been recognized as a major driving force in productivity (Smalheiser et al., 2005).

Trust

A major element that makes things easier in collaboration is the trust the partners build up amongst themselves over the years of association. Collaboration and trust are mutually reciprocal processes, one fostering the other (Tschannen-Moran, 2001). Trust, a functional prerequisite for the continuance of harmonious social relationships (Lewis and Weigert, 1985), is the foundation on which the scientific enterprise is erected (Alberts and Shine, 1994). Trust could be raised on the structure of prior knowledge about the partners, social ties and past experience of work-ing together. It binds the partnership firmly, allaying the fears of suspi-cion, unreliability and untrustworthiness about the partners. As Arrow (1974: 23) defines it, ‘trust is the reliance one can have on others and works as a lubricant in a social system.’ It is the whole fabric of research (Elizabeth Neufeld [nd], cited in Hardwig, 1991). A lack of trust is, in other words, a barrier to effective collaboration (Powell et al., 1996). Collaboration continues to rest on trust between partners although it is sometimes misplaced (Goldstein and Friedhoff, 1988; Pérez et al., 1998).

Many (Alter and Hage, 1993; Chin et al., 2002; Kramer and Tyler, 1995; Olson and Olson, 2000; Shrum et al., 2001) acknowledge trust as an imperative in collaborations. Without trust, long-lasting collabora-tion is a myth. Collaboration might not endure for long in a suspicious environment where the partners have lost their mutual trust and faith.


In scientific activity, people trust others who are honest and sincere in fulfilling their commitments and do not take excessive advantage of others when there is a chance (Cummings and Bromiley, 1996). Low levels of trust are attributed to unpredictable and irregular patterns of communication (Jarvenpaa and Leidner, 1999). Social communication and communication of enthusiasm can facilitate trust in the early stages of team work, while in the later stages predictable communication and substantial and timely responses are central (Jarvenpaa and Leidner, 1999). Previous acquaintance and contacts, even if sporadic, can be the first step in building trust between potential collaborators.

It is not sufficient to get a trusted associate at the beginning of the collaborative process; the associate should be a reliable person who is steadfastly committed to the agreed goals, methodologies, meetings39 and deadlines for the entire period of the project. Trust becomes an essential prerequisite particularly at the time of the final stage of the project, not allowing the data and results to leak out to the competing teams (Melin, 2000).

Sonnenwald (2003, 2007) speaks of cognitive and affective trust. The cognitive focuses on judgements of competence and reliability, and the affective is centred on interpersonal bonds among partners. In the event of a high level of cognitive trust and a low level affective trust, con-straints on monitoring research build up (Sonnenwald, 2003). Subjected mainly to interactional dynamics between partners, the degree of trust tends to be inconsistent. Binding trust and unity assuage differences and disagreements, preventing it from assuming the form of serious chal-lenges that could threaten the execution of the project.

Trust works in relation to the size of the team as well. Collaborators favour a small team with a minimum number of members who can accomplish the tasks with the essential skills or expertise. No more, no less. This might not be the case with multi-institutional alliances. In a study of 23 multi-institutional collaborations (involving three or more organizations) in physical science, Chompalov and Shrum (1999) reported an average of 39 participants per collaboration with a mean number of six institutions. Smallness indeed regresses the scope for tension, disagreement and conflict, helps the team leader to keep the integrity and solidarity of the members and keeps trust from waning. Changes in the size of the team, either by the entry of a new member during the course of collaboration or the exit of a member midway, takes its toll on trust. The arrival of a new member after the project has started without an immediate need for the knowledge or skills the new member


is expected to bring in might not be unanimously welcomed. For the sake of the leader, the existing partners may permit it to happen without objecting to it straightaway. However, objections will return in a cumu-lative fashion when other differences and disagreements crop up. Trust is the casualty in such instances.

Communication

How do the collaborators located in distant places keep in contact with-out disruption or delay? This points to another component of collabo-ration, namely, the use of a reliable media of communication for both regular and intermittent contacts. No alliance can exist without proper channels of communication. Instances that highlight this central link-age between effective communication and successful collaborations abound in the literature and will be explored in the following chapters. Deficient communication channels interfere with efficient coordina-tion of responsibilities, integration of the phases of the research and the incessant transfer of information. The use of the technologies in this respect tends to vary. A number of options exist to choose from: teleph-ony, fax, voicemail, audio/video conferencing and various methods on the Internet. From conventional means of the face-to-face mode to the modern ICTs, scientists have passed quite a few milestones to keep their communication alive.

Modern communication technologies are competent in creating a conducive environment for collaboration. Collaboration technologies can result in productive uses and revolutionize collaboration if the users are motivated to take advantage of the medium (Olson and Olson, 2000). It is unrealistic to expect similar and uniform means of communication in all the locations of the collaborators. Diverse forms of technology are available in centres where the collaborators are positioned. The caveat that Olson and Olson (2000) provide is that it is not worth introducing remote technologies in institutions unless they have a culture of sharing and collaboration.

The mere presence of communication technologies is no guarantee for successful collaborative initiatives. As shown in some recent stud-ies (Narváez-Berthelemot et al., 2002), international collaboration in the most productive African countries, such as South Africa, Egypt and Nigeria, is lower than in some other less productive countries like Ivory Coast, Morocco, Senegal and Tanzania. If ICT is a key variable in pro-ductivity, some of these countries, such as South Africa and Egypt, have relatively developed ICT infrastructures (Jensen et al., 2007).


Collaboration effectiveness

The success of collaboration, at the institutional and individual levels, is gauged by a set of variables. In a general sense, it measures the effective-ness of common incentives to collaborate, manage and administer the projects, and the outcomes of such collaboration (Corley et al., 2006). Successful alliances could enhance the research effectiveness of an institu-tion (Bozeman and Boardman, 2003a). When a collaboration structure pro-duces more benefits per given cost, it is more efficient (Landry and Amara, 1998). Efficiency, as Landry and Amara (1998) demonstrate, is manifest on the three planes of the researcher’s level, the collective structural level and at the national level—conditional to the choices of cost/benefits. They argue that the efficiency of the individual participant will increase with the size of institutional structures, and therefore, efficiency at the indi-vidual level can be maximized by organizing collaborative research within large formal structures at the cost of maximizing transaction costs and incentives. There are three sets of variables in this model that explain the efficiency of institutional structures: publication assets, coordinating costs and characteristics of the collaborative research context.

The concept of collaboration effectiveness, as put forward in a model by Bozeman and Boardman (2003a), considers the attributes of col-laborating individuals, institutions, collaboration and the processes in the assessment of collaborative effectiveness. The factors that affect collaborating individuals include heterogeneity, incentives, acquaint-ance, distribution and role. On the institutional front, the attributes are resources, structure/design, organization culture and collaboration role. Whereas the attributes of collaboration and processes comprise plan-ning/mutability, fit (goals, collaborators), Intellectual Property Rights (IPR) rules for knowledge use, management and leadership, communica-tion, representation model, assessment, mission/scope and role of exter-nal stakeholders. This model does not take into account factors such as publication productivity and co-authored publications that stem from collaborative research partnerships. Effective collaborations do not rule out the challenges they confront.

Challenges within

Productive outcomes and positive aspects aside, collaboration for many a scientist is not an easy course of action to follow. Recurrent in the litera-ture are also cases of failed or inconclusive collaborations. Collaboration consumes a great amount of time (Fox and Faver, 1984; Katz and Martin, 1997), money and resources, and entails non-trivial problems associated with coordination and communication (Porac et al., 2004). Milliken and


Martins (1996) call it a ‘double-edged sword’. There is always a trade-off in collaboration. The quality of collaboration relies largely on the quality of collaborators and the extent to which they complements each other without working at odds (Bozeman and Boardman, 2003a). Knowledge sharing and production, a certainty in most alliances, is arduous, and many collaborations have reportedly failed, not realizing the expecta-tions of their associates (Fisher et al., 2002, cited in Porac et al., 2004). Crucial to the success of the alliance are knowledge generation and the sharing of resources (Corley et al., 2006), and skills and outcomes; they are not imposed upon the partners. Collaboration is not dictatorship, and the essence of it is really the suppression of the instincts for pecking order (Bush, 1957). A pecking order in collaboration operates against the spirit of cooperation and equality. It is unlikely that partnerships would survive if they are imposed on the individuals under any structural con-ditions. Contrasted with this is institutional collaboration between two or more institutions, which is to be viewed through a different lens.

Joint projects work in a time-bound manner for their completion, but might be extended beyond the initially set limits of time and targets. Due to this time factor, projects are to be sustained for a long period of time, contingent on the resources in hand and other institutional prescriptions. During this prolonged stage, it is only natural that chal-lenges to the originally designed plans, objectives and work agreements arise. Given the heterogeneous character and background of the part-ners, long-lasting research activity is not something that can be easily executed. Consistent attempts to identify those challenges and adopting the best means to address them (Sonnenwald, 2007) can solve some of those difficulties that arise.

Competition and conflicting views (Atkinson et al., 1998) between part-ners are to be anticipated in collaborations as they are natural to human beings. Geographically and institutionally dispersed collaborations have more problems of coordination and problem solving (Setlock et al., 2005 cited in Walsh and Maloney, 2007) than the rest. Where individual researchers are led to collective activities, verbal understanding and agree-ment guide the whole process, which might last for years. If the partners get on well and are able to deal with the issues pertaining to the profes-sional and personal aspects of collaboration, it would finally reach its frui-tion. For an extensive period of research association, written agreements between partners serve a great purpose in thrashing out the problems that might surface unexpectedly. In the absence of written agreements or guidelines, the handling of conflicts on any facet of collaboration is a mission. This applies even to the scheduling of a meeting of partners.


Studies (for instance, Chompalov et al., 2002) have drawn attention to the pertinence of formal contracts, of roles and assignments and of the procedures for decision making in institutional collaborations. In multi-institutional collaborations with teams of researchers, where dif-ferentiated tasks and responsibilities are performed by every one of the collaborators (Chompalov et al., 2002), clarity of roles for each of them is vital. In short, defined roles assist the collaborators to assume their core programme responsibilities (Bozeman and Boardman, 2003a). An understanding on agreed roles, responsibilities and tasks of the partners, preferably in a written form, lends clarity to what everyone is expected to do to achieve the common goals and to help run the project without any avoidable hitches, administrative setbacks and conflicts. But, owing to the complexity of the project and the unsure nature of all the tasks and skills that could be envisaged beforehand, this might not be a viable prescription in most situations. However, because of the advantage of avoiding disagreements and conflicts during the course of the project, it is worth making the effort in this regard. As opposed to individual collab-oration, certain matters have to be considered in institutional collabora-tion: research vision, goals, tasks, organizational leadership, use of ICTs, intellectual property and legal issues are important (Olson et al., 2007).

Inherent in collaborations are the problems that surround interper-sonal and professional relationships among partners. A partner can eas-ily delay or even prevent publication of the findings in the name of credit, order of authors, property rights and preference for a particular journal (Smalheiser et al., 2005). Besides these, collaborations are not discharged because of the negative facets revolving around the issues of unethical practices, particularly when partners come from diverse contexts such as developed and developing countries. In the partner-ship of research involving collection of primary data or clinical test trials—between scientists in developed and developing countries—the obligatory ethical procedures that are mandatory in developed countries may not be observed, conveniently forgotten or flouted in developing countries. To maintain the spirit of collaboration among the associates and to keep the process going, all of them are required to provide and receive resources (Sonnenwald, 2007). When collaboration entails part-ners from developed and developing country locations, ones who have access to and control over funds tend to be the more powerful partner. In such associations, one might need knowledge (or data for that matter) and have the resources while the other partner has the proximity to this knowledge but does not have the requisite resources. Scientists from developed countries seek research partners who have this proximity, and


for those in the developing countries, it is not an overriding priority in the given structures in which s/he is a part. Collaboration is initiated in such instances due to the possibilities for new knowledge and the publi-cation of it for one partner and monetary advantages for the other; but it has to be beneficial to both parties in the end. As is clear from studies (Glänzel et al., 1999), collaboration is as advantageous for the developed world as for the developing world.

Research alliances sometimes become powerful lobby groups, compe-tent to influence research policies and research priorities in their favour (Sonnenwald, 2007). The obvious consequence of this is that individual researchers who prefer to work alone find accessing grants competitive, hard and seldom successful. As noted earlier, institutional preference for partnerships runs parallel to the interests of the individual research-ers who can realize their research goals without association of any sort. For the individual researchers, if they are juniors and beginners in their career, this puts them in a situation wherein their career graph gets stuck for want of independent research and the impending publications out of it. In a way, this is a double bind position for the budding scientists. First, because of their stature in the discipline as novices, it is hardly possible to attract collaborative offers from senior scientists unless the latter are lured by some absolutely necessary data or equipment. Second, when institutions prefer team research to single individual studies, the chances for juniors to secure funds for their projects are slim.

The extensive history of collaboration does not imply that stud-ies of it also have a lengthy past; they are of recent origin, particularly those of multi-institutional alliances involving more than three insti-tutions and their scientists (Chompalov and Shrum, 1999). As is the case of any maiden attempt, such studies are not complete, or cannot be expected to be, in their coverage of the dimensions of collaboration wherein humans and institutions interact and cooperate for a specific purpose. As Chompalov and Shrum (1999) note, most of these studies are micro-social in focus, case study in their approach and descriptive in presentations. Furthermore, the existing studies are incomplete exami-nations of the structural characteristics of collaboration, lacking in gen-eralizability and narrow in their focus on single institutions rather than multi-institutions, fail to consider relevant factors (communication, division of labour, technology and size) in their order of importance and neglect the relationships between properties of collaboration and their outcomes (Chompalov and Shrum, 1999). In the next chapter the scientific research of South African scientists through their publications is examined.

85

Scientific research findings are released to the public domain via several outlets. Peer-reviewed journals, monographs and books are among them. Apart from what these researches produce (discoveries, for instance), scholars study research as well. One method of investigating and under-standing the research of a particular group of scientists (scientists in an institution or in a country) is to examine the publications the scientists have produced. These publications, if they are carried in peer-reviewed journals, are usually stored and preserved in databases that could be accessed and analysed. The ISI Web of Science, PASCAL and SCOPUS are some of the journals that are widely used for the analysis of the trends and patterns of scientific research (Glanzel, 2002; Glover and Bowen, 2004; Gomez et al., 1999; Harsanyi, 1993).

The study of scientific publications takes to the realm of bibliometrics. As a tool for assessing and mapping the state of science and in the study of collaboration, bibliometrics is of immense use (Arvanitis et al., 2000: Gómez et al., 1999; Subramanyan, 1983). Studies on research collabora-tion have focused mostly on industrialized countries, and scientometric studies on collaboration in Africa and South Africa, in particular, are in an embryonic stage (Pouris and Ho, 2014. So far only very few biblio-metric studies have been undertaken about science in Africa, and recent ones are hard to find (Arvanitis et al., 2000). This is more so in the case of South Africa (Ingwersen and Jacobs, 2004; Jacobs and Ingwersen, 2000; Pouris, 2003), which is a scientifically active and productive country in the African continent. There are a few exceptions such as the study of Narváez-Berthelemot et al. (2002) which analysed scientific produc-tion, institutional participation and international collaboration in 15 of the most productive African countries during 1991–97, specifically

4Research Publications of South African Scientists, 1945–2010


examining the fields of clinical medicine, biomedical research, biology, chemistry, physics, earth and space science, engineering and technology and mathematics. South Africa contributed heavily to all these fields, medicine and mathematics in particular, according to this study.

Pouris (2003), analysing publications drawn from the ISI Web of Science for a 20-year period (1980–2000), noted that though the num-ber of publications of South Africans increased in absolute numbers, the rate of growth has not been on a par with the international growth. In another scientometric paper, Pouris (2012b) examined the research per-formance of South African scientists for a ten-year period between 2000 and 2010. As a result of the incentives introduced by the government the research outputs have improved significantly. Pouris’ (2012b) study concluded that South Africa’s world share of publications in 2010 was beginning to reach the highest contribution improving its international ranking by two positions to the 33rd position. For another 20-year period (1986–2006), Onyancha and Jacobs (2009) studied the nature of the capacitation of research in the natural sciences in the country. This study reported an unclear and mixed pattern of growth in different nat-ural sciences disciplines in the country. Disciplines of biology, chemis-try, geology, biochemistry and physics maintained a dominant presence in the country during this period of analysis. Jacobs (2008) inquired the patterns of research collaboration in the natural and applied sciences in South Africa for the period 1995–2003. They found that the publication output of South African scientists was greatly influenced by their col-laboration endeavours.

Using a bibliometric analysis of the publications of South African scientists for about three decades (1975–2005) Sooryamoorthy (2013b) examined the growth, trends and patterns in the production of scientific publications in the natural sciences in South Africa. This analysis presents some dimensions of South African publications and the relationship between publications and collaboration. Ingwersen and Jacobs (2004), choosing some disciplines, reported that there had been a decline in the number of South African publications in certain fields during 1986–90. But there has been an increase in the absolute citation impact of South African publications during 1989–93 in all the chosen fields of analysis.

Collaboration often leads to the production of co-authored scientific publications. Co-authorship provides a clear and active indicator of col-laboration of partners,1 offers a window to the patterns of collabora-tion within academic communities (Newman, 2004) and indicates the magnitude of scientific activities across countries (Guan and Ma, 2007).

Research Publications of South African Scientists, 1945–2010 87

Katz and Martin (1997) considered this not more than a partial indicator of collaboration. The history of science records people like Paul Erdős who published 1,401 papers, a good number of them with his colleagues (Newman, 2001).2 This chapter looks at the publication records of South African scientists, more specifically at joint publications, to under-stand their collaborative research patterns and features for the period 1945–2010.

Co-authorship

Co-authored papers in science have risen exponentially in the last few decades. The analysis of the ISI-indexed journals by Science Watch (1994) showed that the number of multi-author papers had escalated sharply since 1991, although there was a fall in the number of papers with more than 50 authors. There is no limit to the number of co-authors who can join together in the production of a single paper. The largest reported number of authors who worked to write a single paper is 1,681 (Newman, 2001). Price (1963), using the data from Chemical Abstracts, recounted that in 1900 more than 80 per cent of the papers were single-author and since then the proportion of multi-author papers had increased mani-fold. At this rate, he predicted, single-author papers would be extinct by 1980. His prediction did not come true, but the number of multi-author papers continued to rise. Price (1963) also observed that three-author papers were growing more rapidly than two-author papers, and four-author papers more rapidly than three-author. In 1990, the mean number of authors per paper, based on the ISI-indexed journals, was 2.6, which rose to 3.6 in 2003, compared to a decline of single-author papers to 25 per cent from 38 per cent during the same reference period (Science Watch, 2004).

As a spin-off of collaborative research, co-authored publications have reasons for their massive growth. The most cited ones are ever- increasing professionalization, specialization (Bush and Hattery, 1956; Jewkes et al., 1959), the complexity and interdependency of science, and multidis-ciplinary and joint efforts for the analysis and solution of problems. The desire for fame, acceptance, recognition, popularity, visibility and to increase one’s productivity also inspires scientists to co-publish. Evidence suggest that the total credit given by the scientific community to all the authors of a co-authored paper is greater, on average, than the credit allocated to the author of a single-author paper (Nudelman and Landers, 1972 cited in Katz and Martin [1997]). Visibility and citation are other major attractions. Internationally co-authored papers are cited


up to twice as often as single-country papers (Goldfinch et al., 2003; Narin and Whitlow, 1990 cited in Katz and Martin, 1997).

Scholars have advocated and/or extensively used co-authored publi-cations as a measure to understand research collaboration and relation-ships and also in the study of scientific productivity.3 Edge and Mulkay (1976) were among the first to use co-authorship in the study of spe-cialties (Katz and Martin, 1997; Stokes and Hartley, 1989). A cautionary note on this point is in order. As Frame and Carpenter (1979) suggested, the proportion of international co-authorships depended on the basic and applied nature of the field of science and was more common in the natural sciences than in the social sciences (Moody, 2004). Multiple authorship has made a remarkable growth in some disciplines, before the advent of machines that required teams to conduct experiments together (Price, 1965b). Presenting differing levels of co-authorships in disciplines, Meadows’ (1974) calculations showed that multiple author-ship was 83 per cent in chemistry, 70 per cent in biology and 15 per cent in mathematics.

Co-authorship symbolizes mutual intellectual and social influ-ence (Stokes and Hartley, 1989). Price (1963) considered multi-author publications as a measure to understand changes in collaboration. Co-authorship is not rated equally well in all branches of knowledge; it is devalued in the humanities and social sciences (Hafernik et al., 1997). Ervin and Fox (1994) reported that in the tenure and promotion process, preference was given to single-author papers rather than to multi-author publications.

A caveat is in order when indexing databases are relied upon for the study of African science. The international scientific literature underes-timates the real situation of African research and scientific capabilities (Tijssen, 2007). To take a sample, South African science is not adequately represented in the Science Citation Index (SCI) database. Of the 253 jour-nals approved by the Department of Education of the Government of South Africa,4 only 19 were listed in the SCI index of 2004 (Tijssen, 2007). The analysis of citations received during 2001–04 by 166 South African journals, including SCI-listed and non-SCI-listed journals, reveals that the non-SCI journals accounted for 48 per cent of all references within the SCI-listed literature to South Africa’s co-authored research articles (citation data refer to cited publications published during 1980–2004) (Tijssen, 2007). Apart from this is the removal of South African jour-nals from the citation index. During 1993–2004, the number of South African journals in the index dropped drastically from 35 to 19 (Tijssen,


2007). In the Journal Citations Report (JCR) of the ISI in 2002, there were only 17 South African journals, which is however 90 per cent of the African journals indexed in the JCR (Pouris, 2005). Tijssen (2007) also found that non-SCI-listed South African journals do have a considerable citation impact on world science.

Data and method

The online SCI Expanded (1945–present) database of the ISI Web of Science was used for the analysis that is presented in this chapter. Widely used in bibliometric analyses,5 the database provides generous information for the study of publications. Limitations aside, the SCI in relation to several other bibliographic databases has the advantage of a wide coverage of recognized, citation-based and widely read scien-tific journals. SCI contains high-quality published research output and citations (Hicks and Katz, 1996), indexed on the basis of certain strict citation criteria, which enables reliable analyses. Bibliographic records of co-authored publications in the SCI can supply a few key aspects of collaboration, namely, the distance, number of collaborators and papers, the degree of clusters (Newman, 2001),6 location (country of collaborators), sector (university, research institute, industry, govern-ment, hospital),7 proxy year of collaboration, subject and discipline of the partners and the type of collaboration (internal, external or inter-national). The database nevertheless has been criticized for its bias towards the English language basic research in industrialized countries (May, 1997; Velho and Krige, 1984) and its questionable coverage (May, 1997; Velho and Krige, 1984) of scientific publications produced in other parts of the world. Where English is not the first language, schol-ars publish in their native languages, which are not always covered in the SCI. There are other reasons too. As Fuenzalida (1971, cited in Velho and Krige, 1984) believed, scholars may choose not to publish in advanced countries’ journals due to their anti-imperialist and national-ist sentiments. Jagodzinski-Sigogneau et al. (1982, cited in Velho and Krige, 1984) also made a similar observation that in peripheral coun-tries publishing in one’s language was a way of ensuring the independ-ence of one’s science from the hegemony of the centre. Despite all this, it has been accepted that all high-quality papers are published in English (King, 2004). English being a common language in South Africa, the scientific community largely publish in English and has a preference for international journals.


Another issue, as Frame (1979) raises, is its under-representation of publications produced in least-developed countries. Of the 7,681 jour-nals listed in the Science edition and/or Social Sciences edition of the Thompson’s Journal Citation Reports 2004, only 0.3 per cent (23 journals) is of African origin (Tijssen, 2007). Researchers in Third-World countries fail to publish their research or do not publish in international journals (Jacobs and Ingwersen, 2000). Also, as noted by Arvanitis et al. (2000), a substantial amount of scientific research does not reach the publication destination as African specialists (contracted to do research) spend much of their time writing up reports to the neglect of papers for scientific journals. Some countries have their own databases. China has its own databases of the Chinese Science Citation Database, the China Scientific and Technical Papers and Citations Database and the Chinese Social Science Citation Index (Negishi et al., 2004).

The data in the ISI is stored under several categories8,9 including articles, notes and reviews. This analysis is restricted to ‘articles’ and ‘reviews’ as the focus is only on research papers written by South African scholars.10 First, the individual records of all the ‘articles’ and ‘reviews’ published from 1945 until the end of 2010 were retrieved by entering ‘South Africa’ in the address box of the database. It returned, in two stages, with 115,447 (111,097 articles and 4,160 reviews) records with at least one South African author in every record. These records, covering all the available years until 2010, supplied the details of the publication trend across various years, disciplines, journals and partnering countries. In the second stage of analysis eight years—1975, 1980, 1985, 1990, 1995, 2000, 2005 and 2010—were chosen as the representative sample years for a deeper analysis (dealt with in chapter 5).

Details of the records of all the ‘articles’ and ‘reviews’ published by South African scholars and indexed in the SCI database were then obtained from the database. Full bibliographic records of these arti-cles and reviews, among others, included the names of all authors, titles of the papers, source (journal name, volume and page number), year of publication, language, type (article, review), key words, affili-ation addresses of all authors, subject category and times cited. From these basic variables many more new variables such as international collaboration, domestic collaboration, internal-institutional collabora-tion (within South Africa), external-institutional collaboration (within South Africa) and fractional count of papers were created along with some additive measures for sectors and the countries of the co-authors.


These variables were then coded and manually entered into a computer programme for further statistical analysis.

In this chapter, the focus is on the nature of the research publications of South African scientists for the period 1945–2010. The features of the South African publications are examined in terms of their subject areas, journals in which these publications appeared, partnering countries and collaboration.

South African publications, 1945–2010

The objective of this section is to examine the major traits of scientific activity in South Africa as revealed in the publications of scientists in the country, captured in the ISI database from the beginning of the period (1945) that permits any bibliometric study. First, the anal-ysis is made around the size of South African science over the last few decades with its trends and patterns, collaboration (participating countries) and scientific activity that can be inferred from the focus of disciplines and specializations. The huge number of bibliographic records (115,447) does not permit the reprocessing of the publication details of every single record. The analyse function of the database to classify the records by year, countries of authors, journals and subjects was therefore relied on.

For reasons not clearly known, there were no papers authored by South African scholars11 in the period from 1945 until 1965. There can be two plausible explanations for this (Sooryamoorthy, 2009a). One, the publications of South African authors started appearing in this database only after 1966. Two, South African authors began publishing their arti-cles in those journals that are being indexed in the database only after 1965 (Sooryamoorthy, 2009a). Until 1971, the number of publications of South African authors lingered around a single digit figure, before it jumped to 93 in 1972 and 1,169 in 1973. From then on, the productivity of South African scientists has been on a steady path until it moved from 1,156 to 3,045 scientific papers (excluding reviews) in 1987 (Table 4A.1). This rapid growth has been reported by other researchers as well (Pouris, 2003, for instance). It began falling in 1988 with 2,897 papers; the total number of papers did not cross the 3,000 mark until 1995, and since then it has not fallen back to below 3,000 papers a year. This finding can be corroborated by that of other studies. Jacobs and Ingwersen’s (2000) analysis of 1981–96 SCI data revealed that the total publication output of South African scholars had declined since 1991. Working


on the PASCAL database Arvanitis et al. (2000) found that, although South Africa was the main scientific producer in Africa in 1997 ( contributing 28% to the total scientific papers) the annual growth rate for South Africa (1991–96) had declined by 5 per cent. The increase in the world’s scientific output, during 1981–94, was 3.7 per cent per year, which means a doubling time of 19 years (May, 1997). The greatest growth rates—10 per cent per year—have been witnessed in scientifi-cally emerging countries such as Hong Kong, China, Singapore, South Korea and Taiwan (May, 1997).

In 1989, the total output of South African scientists was reduced by 13 per cent of the previous year. The average decline during 1989–95 was 9.63 per cent. The percentages of decline from 1988 were: 13.08 (1989), 11.8 (1990), 7.87 (1991), 8.44 (1992), 10.7 (1993), 9.52 (1994) and 6.03 (1995). During 1989–95, the decline is clearly linear (Table 4A.1). This downturn in the publication productivity of South African scientists in certain areas after 1989 is attributed to a ‘brain drain’ fuelled by the political turmoil in the apartheid times. During 1994–97, a total of 24,196 professionals left the country for the UK, the US, Australia and Canada, and the annual emigration was 56 per cent higher than that of the period 1989–94 (SANSA, cited in Ingwersen and Jacobs, 2004). A good number of doctorates in the disciplines of medicine (43%), science (26%) and engineering (25%) migrated to other countries (Lutjehams and Thompson, 1993 cited in Ingwersen and Jacobs, 2004).12 The impending change of the political system from apartheid to democracy did cast a spell of uncertainty about the future of scientists in South Africa. Political disorder and instability can affect the productivity of scientists, as it happened in Russia. Political changes and the conse-quent economic changes in Russia in the early 1990s had their effects on science (Wilson and Markusova, 2004). Between 1991 and 1993, the output in Russia was reduced by 20–24 per cent, and its recovery to the previous level of output took another nine years (Wilson and Markusova, 2004).

Features

The dataset used in this analysis demonstrates numerous subjects and research areas. For South African scientists their subject areas ranged from medicine to communication, with general internal medicine being the most productive discipline of South African science, garnering about 9 per cent of the total number of publications for the period of analysis. Close to it is chemistry with 8 per cent of publication records. Others in descending order were: environmental sciences (6.1%), engineering


(5.85%), physics (5.76%), plant sciences (5.42%), zoology (4.62%), biochemistry and molecular biology (3.71%), science, technology and others (3.68%), mathematics (3.48%), agriculture (3.36%), veterinary sci-ences (2.92%), surgery (2.83%), marine and freshwater biology (2.79%), astronomy and astrophysics (2.74%), geology (2.62%) and other areas of research.

South African scientists publish their research, either single-handedly or with partners, in national and international journals. Some of the journals with international standing in subjects in which they publish are edited from South Africa. The South African Medical Journal, South African Journal of Science and South African Journal of Botany are examples. All the first eight journals that South African authors preferred for pub-lication of their research and have the most number of publications originate from South Africa. Seven per cent of the total publications of South African scholars was published in the South African Medical Journal. The South African Journal of Science carried another 2.4 per cent of South African research publications. The largest number of publica-tions for the period 1945–2010 appeared in journals published from South Africa. These journals including South African Medical Journal (7.2%), South African Journal of Science (2.4%), South African Journal of Botany (1.5%) and Water SA (1.2%) are internationally known and are impact journals.

South African scientists work with people from a large number of countries (Table 4A.2). Of the 115,447 publications the country pro-duced post-1945, 61,559 (53%) were collectively authored with other countries. On a comparative note, the most productive countries in Latin America, such as Brazil, Argentina, Mexico and Chile, have an international collaboration rate below 35 per cent in their joint publica-tions (Gómez et al., 1999).

As elaborated in chapter 2, scientific collaboration in South Africa has had strong roots in the past, which is still present in the post-apartheid South Africa. South Africa has become a regional hub of collaboration as Wagner and Leydesdorff (2005b) infer from their analysis of co-authored papers in 1999–2000. This points either to the increasing propensity of South African scientists to associate, or to zealous foreign scientists who want to establish ties with South African scientists for joint research initiatives.

South African partnership with overseas scholars is dominated by a few countries. Most of the scientists (10.35% of the total number of publications) who worked with South African scientists are from the US, followed by England (6.15%), Germany (3.65%), Australia (2.9%),


France (2.39%), Canada (2.1%), the Netherlands (1.69%), Italy (1.23%), Switzerland (1.19%), Belgium (1.16%), Scotland (1.3%), Sweden (0.95%), Spain (0.94%), Japan (0.92%), Israel (0.78%), India (0.71%) and China (0.71%). This confirms the findings of some other previous studies. Narváez-Berthelemot et al.’s (2002) research into the 15 most productive African countries revealed that institutions in the US were the principal collaborators with African countries, followed by France and the UK. Colonial ties, as Zitt et al. (2000) demonstrated, come into play in part-nership. This is evident among South African scientists who keep their ties with their former colonial colleagues intact through shared research enterprise. A recent analysis by Pouris and Ho (2013) using ISI data for 2007–11 showed that the major research partners of African scholars were from the US, France and the UK.

It is worth examining total publications of countries as captured in the ISI Web of Science for the period 1975–2013. Only science publi-cations, which were drawn from the SCI Expanded (1945–present), are included. Among the 23.24 million publications over the period, South Africa’s contribution was 0.57 per cent. This can be contrasted with Nigeria, another major player in science on the African continent, who had a share of 0.17 per cent. One-third of the publications (33%) was produced by the scientists in the US. Some other top publishers are also shown. Among the BRICS countries (Brazil, Russia, India, China and South Africa), China tops the list with the highest percentage (6%) of the total publications for this period, followed by India (3%), Russia (2%) and Brazil (1.5%). China is in the league of other powerful nations such as Canada (5%), England (7%), Germany (6%) and France (5%). How did South Africa perform in 2013 in regard to its scientific publi-cations? In 2013 South Africa produced 0.7 per cent of the total world publications, which was more than the share of Nigeria (0.14%). China is the best player among the BRICS countries, garnering 17 per cent of the publications in 2013. India follows with 4 per cent, and Brazil and Russia have a share of about 3 per cent each. The US had produced 26 per cent of the total publications, compared to its share of 33 per cent in 1975–2013.

Having seen the key characteristics of the research publications of South African scientists from the ISI Web of Science, it is appropriate to have a closer look at them. This would assist in an in-depth study of the collaborative dimensions of these publications. The analysis of collabo-ration with regard to a representative sample of these publications pro-vides a better understanding of South African research. The next chapter is devoted to this examination.


Table 4A.1 Publications of South African scientists, 1945–2010

Year Articles Reviews Total

1945–65 – – –1966 7 – 71967 6 – 61969 3 – 31971 8 – 81972 92 1 931973 1,156 13 1,1691974 1,255 19 1,2741975 1,219 17 1,2361976 1,336 42 1,3781977 1,719 48 1,7671978 1,884 36 1,9201979 1,842 25 1,8671980 1,835 28 1,8631981 1,833 25 1,8581982 1,936 48 1,9841983 2,032 73 2,1051984 2,102 66 2,1681985 2,340 69 2,4091986 2,658 64 2,7221987 3,045 55 3,1001988 2,897 54 3,1511989 2,683 56 2,7391990 2,719 60 2,7791991 2,852 51 2,9031992 2,831 54 2,8851993 2,753 61 2,8141994 2,768 83 2,8511995 2,849 112 2,9611996 3,016 84 3,1001997 3,180 90 3,2701998 3,256 83 3,3491999 3,448 112 3,5602000 3,217 133 3,3302001 3,401 144 3,5452002 3,522 146 3,6682003 3,518 156 3,6742004 3,768 163 3,9312005 3,520 170 3,6902006 4,467 236 4,7032007 4,883 269 5,1522008 5,241 373 5,6142009 5,795 436 6,2312010 6,205 405 6,610

Total 111,097 4,160 115,447

Appendices


Table 4A.2 Partnering countries of South African scientists, 1945–2010

Countries/territories Publications Per cent

US 11,991 10.35England 7,131 6.15Germany 4,234 3.65Australia 3,373 2.91France 2,772 2.39Canada 2,388 2.06Netherlands 1,960 1.69Italy 1,431 1.24Switzerland 1,386 1.20Belgium 1,345 1.16Scotland 1,190 1.03Sweden 1,106 0.95Spain 1,091 0.94Japan 1,062 0.92Israel 912 0.79India 819 0.71China 819 0.71Poland 737 0.64Denmark 706 0.70Austria 660 0.57Brazil 656 0.57New Zealand 649 0.56Norway 607 0.52Russia 597 0.52Nigeria 514 0.44Zimbabwe 508 0.44Kenya 456 0.39Federal Republic of Germany 404 0.35Namibia 375 0.32Argentina 370 0.32Finland 340 0.29Hungary 336 0.29Czech republic 335 0.29Mexico 327 0.28Chile 323 0.28Portugal 316 0.27Ireland 299 0.26Wales 271 0.23Greece 244 0.21South Korea 233 0.20Turkey 228 0.20Botswana 222 0.19Tanzania 202 0.17Uganda 199 0.17


(Continued )


Ethiopia 193 0.17Thailand 187 0.16Taiwan 181 0.16Malawi 179 0.15North Ireland 176 0.15Pakistan 172 0.15Saudi Arabia 172 0.15Cameroon 164 0.14Zambia 161 0.14Ukraine 141 0.12Romania 130 0.11Egypt 129 0.11Singapore 128 0.11Armenia 122 0.11Mozambique 118 0.10Slovakia 109 0.09Uruguay 104 0.09Colombia 99 0.09Bulgaria 94 0.08Ghana 92 0.08Croatia 91 0.08Malaysia 87 0.08Iran 82 0.07Swaziland 77 0.07Peru 74 0.06Slovenia 69 0.06Indonesia 67 0.06United Arab Emirates 62 0.05Lesotho 59 0.05Oman 59 0.05Philippines 56 0.05Benin 55 0.05Morocco 54 0.05Senegal 53 0.05Madagascar 52 0.05Cote d’ Ivoire 48 0.04Sudan 48 0.04Gabon 46 0.04Vietnam 46 0.04Venezuela 45 0.04Bangladesh 44 0.04Sri Lanka 43 0.04Tunisia 41 0.04Gambia 38 0.03


Table 4A.2 (Continued)


Lebanon 38 0.03Reunion 38 0.03Algeria 37 0.03Hong Kong 37 0.03Burkina Faso 35 0.03Lithuania 33 0.03Yugoslavia 32 0.03Ciskei 28 0.03Congo 28 0.03Estonia 28 0.03Kuwait 28 0.03Mali 28 0.02Cuba 27 0.02Rwanda 26 0.02Belarus 25 0.02Serbia 24 0.02Iceland 22 0.02Costa Rica 21 0.02USSR 21 0.02West Indies 21 0.02Ecuador 20 0.02Panama 20 0.02Republic of Georgia 20 0.02Transkei 20 0.02Zaire 20 0.02Angola 19 0.02Cyprus 19 0.02Mauritius 19 0.02Rhodesia 19 0.02Latvia 18 0.02Malagasy Republic 18 0.02Luxembourg 15 0.01Guatemala 12 0.01Jamaica 12 0.01Papua New Guinea 12 0.01Trinidad and Tobago 11 0.01Brunei 10 0.01Czechoslovakia 10 0.01Monaco 10 0.01Niger 10 0.01Qatar 10 0.01Azerbaijan 9 0.01Guinea 9 0.01Jordan 9 0.01



Laos 9 0.01Nepal 9 0.01Guadeloupe 8 0.01Seychelles 8 0.01West Germany 8 0.01Barbados 7 0.01Cambodia 7 0.01Serbia Montenegro 7 0.01Bahrain 6 0.01Central African Republic 6 0.01Eritrea 6 0.01German Democratic Republic 6 0.01Haiti 6 0.01Saint Kitts and Nevis 6 0.01Syria 6 0.01Bophuthatswana 5 0.004Dominican Republic 5 0.004Fiji 5 0.004Iraq 5 0.004Martinique 5 0.004Uzbekistan 5 0.004Burundi 4 0.003Netherlands Antilles 4 0.003The Vatican 4 0.003Albania 3 0.003Bhutan 3 0.003Bolivia 3 0.003Chad 3 0.003Guinea Bissau 3 0.003Guyana 3 0.003Macedonia 3 0.003Mauritania 3 0.003Micronesia 3 0.003New Caledonia 3 0.003Paraguay 3 0.003Sierra Leone 3 0.003Surinam 3 0.003Togo 3 0.003Venda 3 0.003Zimbabwe Rhodes 3 0.003Bahamas 2 0.002Bosnia Herzegovina 2 0.002Bundesrepublik 2 0.002French Guiana 2 0.002

(Continued )



Honduras 2 0.002Kazakhstan 2 0.002Libya 2 0.002Malta 2 0.002Myanmar 2 0.002Nicaragua 2 0.002S West Africa 2 0.002Senegambia 2 0.002St Lucia 2 0.002Yemen 2 0.002Andorra 1 0.001Antigua Barbuda 1 0.001Bermuda 1 0.001Bophuthatswana 1 0.001Cape Verde 1 0.001Central African Empire 1 0.001Comoros 1 0.001Deutsch Democratic Republic 1 0.001East Germany 1 0.001Equatorial Guinea 1 0.001Greenland 1 0.001Grenada 1 0.001Ivory Coast 1 0.001Kyrgyzstan 1 0.001Macao 1 0.001Maldives 1 0.001Moldova 1 0.001Mongolian People’s Republic 1 0.001Palau 1 0.001Solomon Islands 1 0.001Spanish Sahara 1 0.001United Arab Republic 1 0.001

Note: Both the old name and new name of some countries appear in the table, as they were found in the same way in the database.


101

Following the general analysis of all publications of South African sci-entists from 1945 to 2010 presented in the previous chapter, a closer examination of the relevant aspects of publications is warranted. For this purpose the publications of eight years have been sampled. The sampled years are 1975, 1980, 1985, 1990, 1995, 2000, 2005 and 2010, with the same class interval of five years. All the publications that come under these selected years are covered for the analysis.

Drawing on the inferences from the historical backdrop of scientific collaboration in South Africa in chapter 2 and the conceptual compo-nents identified in chapter 3, this chapter seeks to discover how collabo-ration, both generally and in its various specific forms, is embedded in the scientific research in the country. The central theme in this chapter is to investigate how collaboration, which is carried through from the past, is connected to the publication productivity of scientists in South Africa. Specifically, answers to the following questions are sought, using appropriate statistical procedures:

Has scientific collaboration as evident in the publication productivity of scientists in South Africa increased substantially over the years?Is there any correlation between collaboration and the publication productivity of scholars in South Africa?What forms of collaboration are predominant in South Africa?In terms of partnership in publications, who are the major partners of South Africa—old and new?Are there any disciplinary variations in scientific collaboration and the productivity of scientists?

The chapter also provides an analysis of some of the underlying themes such as partnering countries, sectors the authors belonged to, and

5Publications through Collaboration


subjects and citations that allow us to tie up with the broader theme of the book—how collaboration is connected to the productivity of scientists.

From the eight sample years (1975, 1980, 1985, 1990, 1995, 2000, 2005 and 2010), a total of 24,589 records of publications by South African scientists and their collaborating authors are drawn and used in the final analysis. Some of them had to be discarded for reasons of duplication or because they were non-science subjects.

Bibliometric studies that examine the publication records for the dimensions of collaboration of researchers have two preferred themes of enquiry: explanatory factors of collaboration and the effects of collabo-ration on productivity (Landry et al., 1996). This kind of bibliometric analysis yields new insights into collaboration. Katz and Martin (1997) list some of the advantages. Firstly, it is an invariant and verifiable data set allowing researchers to check and reproduce the results. Secondly, it is an inexpensive and practical method to quantify the dimensions of collaboration. Thirdly, the data set is huge in terms of the count of bibliographic records, permitting researchers to choose the right size of sample for analysis. The Web of Science database has the online facil-ity to do some basic statistical analyses of the selected bibliographic records, such as running frequency tables by author, year of publication, institutional address, country or region of the authors, publication and subject area. But it is not enough for our detailed analysis.

Co-authorships do not depict all collaborative relations but only a certain fraction of these (Laudel, 2002). Laudel’s study (2002) of 322 collaborations identified variants in research collaboration that are not covered in co-authorships. For instance, conventions in science in the listing of the names of co-authors on a paper are not always observed. The presence of the name of a co-author on a paper means that the person has contributed to it and had a role in any or all the steps involved in its writing. If the names of the co-authors are not in alpha-betical order, the convention is that the work was performed by the first author, superintended by the last author and with the assistance by those in between; but this is not complied with in all cases (Stokes and Hartley, 1989). Nobel laureates, to help their junior colleagues, prefer not to put their names on a paper to which they have made a contri-bution (Zuckerman, 1968). Sometimes, the senior scientist of a labora-tory is listed as a co-author, irrespective of his/her contribution to the paper (Stokes and Hartley, 1989). Due to such inconsistent practices of ordering and naming of co-authors, joint publication as a completely

Publications through Collaboration 103

valid measure of collaboration is not beyond question. Regardless of the degree of contribution of each author, these practices are common across disciplines and in varying degrees. In many co-authored publications, names appear for social reasons or because they are ‘ honorary coauthors’ (Hagstrom, 1965), or additional authors are added for research-external reasons (Melin and Persson, 1996). This is not unique to science disci-plines; the humanities and the social sciences too violate recognized norms, making it difficult to assess the real contribution of co-authors in a paper or the degree of their contribution to the production of the paper. Honorary authors are common too, their names appearing on the paper not because they worked for it at any stage of its production but rather for other extra-professional and non-academic reasons.

Estimation of the true number of distinct authors of a publication is not easy when there are two authors with the same name or when authors have identified themselves differently in different papers, by giving only the first initial sometimes and by using the full name at other times (Newman, 2001). All the co-authors of a publication cannot expect to be collaborative research partners either. Kim (2006) questions the validity of the assumption that all co-authored publications are the outcomes of research collaboration which, to him, might not always be the case; research collaboration sometimes would not result in concrete outcomes like co-authored publications.

Katz and Martin (1997) present two scenarios in the study of inter-national collaboration using bibliometric data. The first scenario is that in which the researchers from different countries collaborate in a single institution and the papers that come out of it have the name of only this institution and the country where the work was located. The second scenario is when the scientist has two institutional affiliations in two different countries. This is the problem with double addressing. When an author gives more than one affiliation address in co-authored papers, measurement of collaboration becomes imprecise and incorrect. This double addressing is common when authors are working in an institu-tion (as a visitor or spending a sabbatical) other than his/her parental institution. Both affiliation addresses appear in the publication showing institutional credit to both the institutions. Likewise, the exact nation-ality of the scientists cannot be ascertained correctly from the institu-tional affiliations of the authors in co-authored publications due to the mobility of scientists in the international scientific community (Frame and Carpenter, 1979; Kim, 2006).

These are some of the active debates in the field.


Collaboration

Collaboration of authors is measured using four variables as displayed in Table 5.1: the count of multiple authors, the average number of authors per paper, fractional count of papers and the degree of collaboration. A majority of the papers were co-written, either by South African scien-tists themselves or in collaboration with scientists from other countries. South African scientists chose to produce joint papers more than sole-authored papers. That 83 per cent of the papers were written by multiple authors shows that the intensity and proliferation of collaboration in the country is conspicuous (Table 5.1). In 1975 multiple papers accounted for 68 per cent of the total publications, which by 1985 increased to 74 per cent. In the following years, they formed 80% (1990), 83% (1995), 85% (2000) and 89% (2005) of the total publications. The 2010 figures showed the highest percentage (92%) of multiple-authored— collaborated—papers. The trend again, except for 1980, was one of growth.

Multi-authored papers, on average, have more than four authors (4.27) for the entire sample period. The average number of authors per paper in 1975 was 2.33 which rose to 7.09 to reach the highest average, in 2010. The variation across the years is significant in statistical tests (F = 14.759, df = 7). The fractional count of papers, an index of the size of collaboration, was 0.44 for these eight years. Specific to this segment of the data is that not only collaboration but also the size of collaboration has been growing in South Africa. A reduction in the fractional count of papers (one divided by the number of authors ) denotes an increase in the size of collaboration (in terms of the number of partners). In 1975, the fractional count was 0.58, which shrank to 0.34 in 2010 in keep-ing with the increased number of contributors per publication. Along with this, the degree of collaboration was measured counting the inci-dence of internal-institutional, external-institutional and international collaboration. A maximum score for this variable could be 3 when the publication involved all these three types of collaboration. The ANOVA test found significance between the selected years (F = 40.315, df = 7). The mean degree of collaboration ranged between 0.1 and 1.4, show-ing a consistent growth over the years. The average for all the years was 1.04, which means the presence of collaboration in the publica-tion is pronounced in the majority of the cases. Here too, the difference across the years is statistically significant (F = 588.369; df = 7). A previous study, which included the publications for the years of 2000, 2003 and 2005, reports that collaboration research in South Africa has been grow-ing steadily and scientists prefer collaborative to individualistic research

Tabl

e 5.

1 Pu

blic

atio

n d

etai

ls o

f So

uth

Afr

ican

sci

enti

sts,

197

5–20

10

Var

iab

les

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

Nu

mbe

r of

pu

blic

atio

ns

1,21

24.

91,

828

7.4

2,35

59.

62,

748

11.2

2,80

111

.43,

366

13.7

4,18

317

.06,

096

24.8

24,5

8910

0N

umbe

r of

aut

hors

***,a

Sin

gle

auth

or39

132

.362

834

.462

326

.554

920

491

17.5

502

14.9

479

11.5

518

8.5

4,18

117

Mu

ltip

le a

uth

or82

167

.71,

200

65.6

1,73

273

.52,

199

802,

310

82.5

2,86

485

.13,

704

88.5

5,57

891

.520

,408

83M

ean

num

ber

of a

uth

ors*

**,b

(F =

14.

579,

df =

7)

2.33

(1

.54)

2.33

(1

.67)

2.64

(2

.72)

2.83

(1

.85)

3.3

(6.4

5)3.

66

(7.0

)4.

55

(10.

23)

7.09

(5

3.7)

4.27

(2

7.36

)Fr

acti

onal

cou

nt

of

pap

ers*

**,b

(F =

331

.337

, df

= 7)

0.58

(0

.3)

0.6

(0.3

1)0.

54

(0.3

)0.

49

(0.2

8)0.

46

(0.2

7)0.

43

(0.2

7)0.

38

(0.2

6)0.

34

(0.2

4)0.

44

(0.2

8)

Mea

n n

um

ber

of

fore

ign

cou

ntr

ies

(F =

240

.315

, df

= 7)

0.1

(0.4

)0.

11

(0.3

9)0.

15

(0.4

5)0.

18

(0.5

2)0.

36

(1.1

)0.

21

(1.8

)0.

35

(1.2

)1.

05

(2.0

)0.

43

(1.4

)

Deg

ree

of c

olla

bora

tion

(F

= 5

88.3

69, d

f =

7)0.

75

(0.6

)0.

69

(0.5

)0.

79

(0.5

)0.

86

(0.5

)0.

87

(0.5

)0.

97

(0.5

)1.

06

(0.6

)1.

4 (0

.8)

1.02

(0

.6)

All

Sout

h A

fric

an a

uth

ors*

**,a

717

59.2

1,03

556

.61,

439

61.1

1,81

265

.91,

652

591,

713

50.9

2,20

752

.82,

496

40.9

13,0

7153

.2A

ny

type

of c

olla

bora

tion

***,a

821

67.7

1,20

065

.61,

732

73.5

2,19

980

2,31

082

.52,

864

85.1

3,70

488

.55,

578

91.5

20,4

0883

Dom

esti

c co

llabo

rati

on**

*,a73

460

.61,

051

57.5

1,47

562

.61,

879

68.4

1,73

261

.81,

967

58.4

2,24

753

.74,

241

69.6

15,3

2662

.3In

tern

atio

nal

co

llab

orat

ion

***,a

104

8.6

168

9.2

290

12.3

389

14.2

658

23.5

1,18

435

.21,

993

47.6

3,04

850

7,83

431

.9

Inte

rnal

-in

stit

uti

onal

co

llab

orat

ion

**,a

525

43.3

750

411,

005

42.7

1,33

948

.71,

234

44.1

1,38

541

.11,

503

363,

735

61.3

11,4

7646

.7

Exte

rnal

-in

stit

uti

onal

co

llab

orat

ion

***,a

284

23.4

345

18.9

558

23.7

625

22.7

543

19.4

687

20.4

931

22.3

1,74

728

.75,

720

23.3

Mu

lti-

cou

ntr

y co

llab

orat

ion

121

261.

432

1.4

652.

413

95

302

958

013

.91,

197

19.6

2,35

39.

6

Not

e: a C

hi-

squ

are

test

; b in

dep

end

ent

t-te

st. S

ig: *

p <

0.1;

**p

< 0

.05;

***

p <

0.01

.


(Sooryamoorthy, 2009a). Jacobs and Ingwersen (2000) too had arrived at similar findings.

Hicks and Katz (1996) examined collaboration by calculating the aver-age number of authors, institutions and countries in the production of a co-authored publication. Tijssen (2007) broadly classifies collaboration as international cooperation (two or more different countries includ-ing at least one African country) and domestic cooperation (one African country with two or more different main organizations). In this chapter, international collaboration is treated as the one in which at least one partner is from outside South Africa. Parallel to this is domestic collabo-ration in which all partners are South Africans, either from the same institution or from institutions within the country.

Kim (2006) uses a classification of no collaboration (single-author publication), internal-institutional collaboration (authors belonging to different divisions or departments of the same research institution), external-institutional collaboration (authors from two independent research institutions) and international collaboration (at least one for-eign country is involved). The numbers of both domestic and interna-tional collaborations for this sample were deduced from the data based on this classification. Domestic collaborations are composed of non-single-author publications in which all of the authors belong to South Africa. If any one author in a non-single-author publication comes from a foreign country, presumably it is an international collaboration. Domestic collaborations are again split into another level of internal-institutional and external-institutional collaborations. In the former, all authors are South Africans and from the same institution. In the external-institutional kind, all authors are South Africans but represent different institutions in the country.

As seen in domestic collaboration (Table 5.1), two-thirds (62%) of the South African scientists for all the selected years (1975–2010) were enthusiastic about working with their colleagues within the county. The percentage of domestic collaboration was more than half of the total publications for all the selected individual years with the lowest percent-age in 2005 (54%). The highest percentage of domestic collaboration was reported in 2010 (70%). In domestic collaboration, internal-institutional collaboration persists over external-institutional collaboration. Growth is obvious in internal-institutional than in external-institutional type as internal-institutional collaboration has increased from 43 per cent in 1975 to 61 per cent in 2010. In contrast, external-institutional col-laboration for the entire period was only 23 per cent of the publications. This is only half of the internal-institutional collaboration. Evident from


this is that South African scientists prefer collaborating with research partners within their own organization to those in other institutions in the country. The peak point in external-institutional collaboration, as in internal-institutional collaboration, was in 2010 (28%).

Increasingly, South African scientists have moved beyond the cor-ridors of their institutions to join hands with scientists from abroad. The proportion of international collaboration (of any collaboration) has risen to 50 per cent in 2010 from 9 per cent in 1975 (Table 5.1). For the period, 32 per cent of the publications had the participation of foreign authors. When this figure is compared with the earlier finding relating to the data for 1945–2005 (37%) provided in chapter 4, the increase is quite revealing. Although international collaboration in South Africa is not as much as domestic collaboration, the trend suggests that the former will get past the latter in a few years’ time. Read together— external-institutional and international collaborations—the trend is skewed in favour of non-local collaboration. South African scientists are looking for international partnerships, and at the same time, scholars from abroad increasingly associate with South African scholars. The structural and institutional support, as seen in previous chapters, should be play-ing a role here.

Reviewing the findings from chapter 2 regarding the historical back-ground of scientific collaboration can provide new insights. Data for the sample years can be more or less grouped into apartheid (1975, 1980, 1985, 1990) and post-apartheid (1995, 2000, 2005 and 2010) periods. In the early years of the first period (1975 and 1980), the percentage of multiple authors, in comparison to the later years, was low (68% and 66%). Seemingly this trend did not continue. After 1995 when the coun-try moved to the current democratic dispensation there was an increase in the percentage of multiple-authored publications and touched the highest points ranging from 83 to 92 per cent. The average number of authors per publication corresponds to this trend. Until 1990, the aver-age was less than 3 authors per paper. This increased to 3.3, 3.7, 4.6 and 7.1 in 1995, 2000, 2005 and 2010 respectively. Fractional count of papers also supported this trend. Implied here is the feature that after a lull in some years in the apartheid times, collaboration continued to prosper. It is also revealed that apartheid was not as promotive as the post-apartheid times for scientific collaboration.

The degree of collaboration too can be analysed under the two peri-ods. The degree of collaboration began to expand prominently since 1990, the closing years of apartheid. It touched an all-time high figure in 2010. This increase is not only in the number of collaborations and the


number of authors collaborating but also in the degree of collaboration and shows that the post-apartheid era provides an encouraging environ-ment for scientific collaboration.

Does this also apply to matters concerning international collabora-tion? As elaborated in the previous chapters, international collabora-tion did exist in the apartheid era. Amidst the general non-collaborative attitudes of the international community, South Africa produced pub-lications that involved the participation of international scholars. But they were not many until 1990. Up until 1990 South Africa produced an average of 11 per cent of the total number of papers that brought home international participation. When the political climate was changing in favour of democracy, international participation in South African sci-ence grew: 24 per cent in 1995 to 50 per cent in 2010. By 2010, South African collaboration with international scholars contributed to half of the total publications South Africa had produced. This emphasizes the indispensable nature of international collaboration in South African science.

Clearly, the colonial period prepared the ground for domestic and inter-national collaboration in science in the country that, as seen in the data referring to the second and third phases (apartheid and post- apartheid), is being continued today with intense vigour and enthusiasm.

Partnering countries

In the past, particularly in the colonial period, South Africa had sci-entific contacts with countries in Europe, Nordic countries, the US, Australia and Canada. Since then, these contacts have materialized into collaboration. The data on the publication records of South African sci-entists is a source to examine whether these contacts have grown into fruitful scientific research participation. The countries with which sci-entists associated, building up to the production of scientific research pieces, can also be elicited from this data set.

South Africa works with many countries and in all continents (Tables 5.2 and 5.3). The maximum number of foreign countries involved in a single co-authored paper was 98. This paper was published in 2010 in the subject area of general internal medicine. The paper, titled ‘Prevention of pulmonary embolism and deep vein thrombosis with low dose aspi-rin: Pulmonary Embolism Prevention (PEP) trial’, brought together 343 authors from 99 countries (including South Africa) and was carried in the journal Lancet (Vol. 355, No. 9212). It showed that aspirin could reduce the risk of pulmonary embolism and deep vein thrombosis.


Tabl

e 5.

2 R

egio

n-w

ise

loca

tion

of

par

tner

s of

Sou

th A

fric

an s

cien

tist

s, 1

975–

2010

Par

tner

s

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

Reg

ion

of p

artn

ers

(for

the

firs

t fi

ve a

utho

rs)

Euro

pe

6040

.396

40.3

151

37.9

246

41.3

484

45.2

994

46.9

1,72

347

.13,

660

42.6

7,41

444

.1N

orth

Am

eric

a69

46.3

107

4516

742

236

39.7

320

29.9

716

33.8

1,18

632

.42,

060

244,

861

28.9

Au

stra

lasi

a7

4.7

114.

621

5.3

467.

790

8.4

171

8.1

298

8.1

470

5.5

1,11

46.

6A

sia

32

31.

314

3.5

183

393.

610

55

220

663

97.

41,

041

6.2

East

ern

Eu

rop

e0

01

0.4

20.

52

0.3

504.

760

2.8

742

318

3.7

507

3La

tin

Am

eric

a0

01

0.4

112.

88

1.3

292.

79

0.4

722

322

3.8

452

2.7

Afr

ica

64

135.

57

1.8

132.

229

2.7

291.

450

1.4

935

10.9

1,08

26.

4M

idd

le E

ast

42.

76

2.5

256.

326

4.4

292.

736

1.7

361

179

2.1

341

2

Tota

l14

90.

923

82.

439

82.

459

53.

51,

070

6.4

2,12

012

.61,

659

21.8

8,58

351

.116

,812

100


Tabl

e 5.

3 C

oun

try-

wis

e lo

cati

on o

f m

ajor

par

tner

s of

Sou

th A

fric

an s

cien

tist

s 19

75–2

010

Par

tner

s

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

US

5637

.610

443

.714

335

.818

230

.527

625

.761

929

.21,

030

28.1

1,79

420

.84,

204

24.9

Engl

and

3120

.838

1654

13.5

9816

.415

414

.331

614

.942

711

.71,

027

11.9

2,14

512

.7G

erm

any

64

145.

945

11.3

477.

910

09.

320

29.

530

88.

450

15.

81,

223

7.3

Au

stra

lia

64

114.

619

4.8

406.

775

713

36.

326

57.

237

84.

492

75.

5Fr

ance

42.

74

1.7

82

294.

940

3.7

994.

724

96.

838

14.

481

44.

8C

anad

a13

8.7

31.

324

654

9.1

444.

197

4.6

156

4.3

266

3.1

657

3.9

Net

her

lan

ds

21.

37

2.9

82

81.

320

1.9

683.

212

93.

525

42.

949

62.

9It

aly

00

31.

39

2.3

203.

431

2.9

542.

510

22.

820

12.

342

02.

5B

elgi

um

21.

310

4.2

82

101.

721

247

2.2

126

3.4

169

239

32.

3Sw

itze

rlan

d2

1.3

31.

31

0.3

81.

321

239

1.8

711.

918

82.

233

32

Swed

en0

03

1.3

00

00

70.

740

1.9

601.

617

22

282

1.7

Ch

ina

00

00

10

00

00

320.

278

0.5

171

228

21.

7In

dia

20.

71

0.4

00

40.

77

0.7

200.

949

1.3

195

2.3

278

1.6

Spai

n0

00

00

03

0.5

242.

220

0.9

711.

915

91.

827

71.

6Ja

pan

10.

72

0.8

102.

59

1.5

222

472.

261

1.7

113

1.3

265

1.6

Scot

lan

d6

48

3.4

82

111.

813

1.2

391.

853

1.4

981.

123

61.

4Is

rael

32

62.

522

5.5

254.

228

2.6

281.

326

0.7

650.

820

31.

2B

razi

l0

00

02

0.5

40.

79

0.8

20.

137

111

61.

317

01

Ru

ssia

00

10.

40

00

020

1.9

361.

727

0.7

690.

815

30.

9


The study is based on a large size of nearly 20,000 sample patients from different countries. This research publication has received a citation count of 380 as on 3 October 2013.

To take the analysis forward, country names of the first five authors of each co-authored record were collected. The examination of all the authors from each record seems to be impractical from the point of data entry, large size of the sample and the broad range of authors (3,172). The average number of authors (4.27) justifies this rationale of the anal-ysis of the first five authors of each publication.

Obviously, collaboration has been proliferating across the regions. Since 2000, North American and European collaboration with South Africa grew in absolute numbers only and did not increase in the share of publications. The publications involving collaboration with North American scientists reduced from 29 to 21 per cent between 2000 and 2010. As for Europe, the figures were 15 and 12 per cent during the same period. In the case of Eastern European and Latin American countries, collaboration was not conspicuous. Collaboration with other African countries is not very significant in these publication records. The same trend has been reported by Mêgnigbêto (2013) in the analysis of the publications of 15 West African countries for the period 2001–10 where intra-regional collaboration was rather weak and negligible. A notable increase over the years is found in the number of collaborative papers with Asian countries (particularly China and India) as also with Latin American countries (Brazil, Chile and Argentina). Asian partnership since 1995 began to expand to the extent of reaching a twofold increase in 2010. Collaboration with countries in Eastern Europe and Latin America has also increased in terms of the number of research publica-tions produced but not in terms of the percentage of total publications. This phase, the end of apartheid era and the dawn of the new demo-cratic South Africa, is a landmark in the history of scientific collabora-tion in the country. There is room for further collaboration with some of these countries as South Africa is becoming an active player in the BRICS (Brazil, Russia, India, China and South Africa) alliance. Boshoff (2009), in his analysis, has similar comments to make. He noted the predomi-nance of North–South collaboration rather than South—South collabo-ration in African countries. Mêgnigbêto (2013) also observed the rising levels of collaboration of West African countries with South Africa.

A closer examination of some of the top collaborators shows the fea-tures of South African scientific collaboration, its linkages with past col-laborators and new linkages with countries in Asia and Africa. Persistently, Europe is the most preferred region for collaboration (Table 5.2) for South


African scholars. A total of 44 per cent of the publication output of South African scientists was made in association with European countries, mainly England, Germany, France, the Netherlands, Italy, Belgium and Switzerland (Table 5.3). Among the European countries, England has a pronounced presence in collaborative research with South Africa, a fea-ture that has existed since the colonization of South Africa. Combined for all three years (1975, 1980 and 1985), England’s contribution to South African collaborative output was 13 per cent as against 7 per cent from Germany, 5 per cent from France, and 3 per cent each from the Netherlands and Italy. These countries, as shown in chapter 2, have main-tained scientific ties with South Africa since the 17th century. This has also been confirmed by the previous analysis for the period of 1945–2010, as seen in chapter 4. Collaboration with Europe and Australasia has grown significantly in the number of publications while it continued to increase with North America.

North America follows Europe with one-third (29%) of the total out-put (Table 5.2). Viewed on a country basis, US scientists had associated in a quarter of the total publications, the largest share of any other coun-try of collaborators. Canada’s slice was just three per cent (Table 5.3). The US has always been the single most productive country amongst all other countries with the largest portion of collaborative publication with South Africans. Not surprisingly, this is a continuation of the ante-cedents of scientific contact with South Africa throughout the stages of colonialism, apartheid and after, as discussed earlier in chapter 2. Notably, until 1980 the US, England and Canada were the prominent partners producing publications with South African scientists. In 1975 the US was engaged in the creation of at least 38 per cent of total pub-lications for the year while England contributed to 21 per cent. Canada had 9 per cent. The number of papers South African scientists published in 1975 with US colleagues was 56. By 2010 the figure had increased severalfold to 1,794 publications. Looking at the year-on-year increase in percentage of the total South African publications, the range was 127–224 per cent. This, in other words, means that the publications were increasing twofold. On average, the increase over the years was about one-and-a-half fold. Since 1995, the increase in the number of collabora-tive publications (from 276 in 1995 to 1,794 in 2010) was more signifi-cant than the previous years. Scientific cooperation for the production of research publications with England did not grow as much as the US collaboration. The year-on-year increase was in the region of 123–241 per cent, with an average of 148 per cent. Between 1975 and 2010 the publication output of the South Africa–England association grew from


31 to 1,027. More significantly, a surge in the production was seen after 1995. In 1990, there were 98 publications. In 1995 joint publications had increased to 154. The increase in percentage terms since 1995 was more than twofold. Countries such as Germany and Australia (from 1980 onwards), France (from 1990), the Netherlands, Italy, Belgium and Switzerland (from 2000) became significant partners in scientific research with South Africa. In the case of South Africa–Germany collabo-ration, the move was from 6 papers in 1975 to 501 in 2010. The increase from 100 to 202 publications happened between 1995 and 2010, and again since 1995, the number of publications has been in three digits. Table 5.3 shows a clearly distinct pattern before and after 1995 for most of the countries with which South African scholars preserved their pro-fessional ties in the colonial and apartheid times.

Australia and New Zealand together contributed eight per cent of the total collaborative output, while the research tie-ups with Asian coun-tries such as India, China and Japan resulted in another three per cent of the research publications. Strikingly, South African scientists have least contacts with Eastern Europe (Russia and Poland for instance), Latin America (such as Brazil, Argentina and Chile), Middle Eastern countries (such as Israel and UAE), reporting a share in the region of two to three per cent. Partnership with African countries until 2005 remained as low as two to three per cent of the total publication output. Since then the increase has been phenomenal at eleven per cent in 2010. Focused efforts were taken recently in South Africa through several funding support to associate with countries in the African continent that were neglected for long.

Sectoral combinations

All major sectors—university, research institute, industry, government and hospital—are represented in the research publications of South African authors and their partners. The separated sectors of the first five authors (recalling that the average size is 4.27) from their affiliation addresses were coded for analysis. First, the sectors of all authors, then the South African authors and finally the non-South African authors for all sample years were examined separately and additively.

The higher education sector—universities and technikons—emerged as the predominant sector of research in South Africa. It produced, mostly universities rather than technikons, 74 per cent of all the papers in the sampled eight years (Table 5.4). Irrespective of the country (South Africa or others) and collaboration, universities and technikons generate

Tabl

e 5.

4 Se

ctor

of

Sou

th A

fric

an s

cien

tist

s an

d t

hei

r p

artn

ers,

197

5–20

10

Sect

or

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

Sect

or o

f al

l aut

hors

Un

iver

sity

/Tec

hn

ikon

1,48

558

.31,

727

58.4

2,63

764

.23,

574

69.5

4,09

174

.26,

295

77.5

8,21

077

.914

,307

78.3

42,3

2674

Res

earc

h i

nst

itu

te27

810

.940

113

.662

215

.173

814

.457

110

.41,

143

14.1

1,61

515

.32,

120

11.6

7,48

813

.1H

osp

ital

410

16.1

434

14.7

536

1345

18.

835

06.

339

94.

935

43.

465

03.

63,

584

6.3

Gov

ern

men

t31

712

.431

710

.721

35.

228

85.

635

66.

514

61.

818

01.

796

05.

32,

777

4.9

Ind

ust

ry56

2.3

792.

710

12.

591

1.8

144

2.6

138

1.7

176

1.7

238

1.3

1,02

51.

8

Sect

or o

f SA

aut

hors

Un

iver

sity

/Tec

hn

ikon

1,40

558

1,62

458

.62,

397

63.9

3,27

370

.33,

520

75.5

5,19

379

.16,

361

80.6

12,2

8580

.936

,058

75.3

Res

earc

h i

nst

itu

te26

510

.937

713

.657

615

.364

113

.844

39.

583

412

.797

512

.41,

571

10.3

5,68

211

.9H

osp

ital

398

16.4

416

1551

613

.742

79.

230

36.

531

24.

726

93.

450

63.

33,

147

6.6

Gov

ern

men

t29

612

.227

910

.117

34.

623

95.

126

65.

711

91.

814

41.

863

44.

22,

150

4.5

Ind

ust

ry58

2.4

742.

792

2.5

741.

612

82.

711

11.

714

41.

819

01.

387

11.

8

Sect

or o

f pa

rtne

rsU

niv

ersi

ty/T

ech

nik

on87

61.3

124

54.4

255

66.6

370

61.3

687

62.7

1,51

267

.32,

742

66.3

5,89

068

11,6

6766

.7R

esea

rch

in

stit

ute

1913

.430

13.2

5013

.111

218

.517

415

.947

821

.31,

064

25.7

1,30

815

.13,

235

18.5

Gov

ern

men

t16

11.3

4519

.742

1170

11.6

138

12.6

401.

863

1.5

895

10.3

1,30

97.

5H

osp

ital

1812

.723

10.1

277

355.

878

7.1

179

820

55

446

5.2

1,01

15.

8In

du

stry

21.

46

2.6

92.

317

2.8

191.

737

1.6

641.

512

01.

427

41.

6

Not

e: T

he

tota

l d

oes

not

tal

ly a

s th

e re

cord

s of

th

e fi

rst

five

au

thor

s of

all

, Sou

th A

fric

a or

non

-Sou

th A

fric

a, a

re n

ot t

he

sam

e.


the largest number of research publications. Gevers et al. (2006) refer to this sector as the one with a tremendous stake in publication. Following them is the research institute sector with 13 per cent of the total output. Three other sectors involved in research in descending order of contribution are hospital, government and industry. The con-tribution of the sector is destined to grow given the supportive funding formula adopted by the current government.

Although the share of the university sector is higher by one percentage point, the pattern generally remains the same for South African authors (Table 5.4). The university sector is leading with three quarters of the total output for all the years, followed by the research institute (12%), hospital (7%), government (4.5%) and industry (2%) sectors. Over the years, only the university sector has registered an increase, while the rest have either stabilized or declined in their proportion of contribution, particularly the hospital sector whose contribution has reduced to 3.6 per cent in 2010 from 16 per cent in 1975. No distinctive pattern is clear in the sector before and after 1995.

The affiliating sector of the collaborators presents a different order of size and share. Immediately after the university and research institute sectors come government, hospital and industry. Although university is a key sector for foreign collaborators (67%), research institutes contrib-ute to 19 per cent of the total collaboration. In comparison to the sectors for all authors and South African authors, the share of the research insti-tutes is higher by about five points for the collaborators. In other words, in the case of foreign collaborators in relation to all authors and South African authors, there is a drop of seven to eight per cent in the uni-versity sector, while there is an increase of five per cent in the research institute sector. This finding prompts the examination of both sectors in South Africa to find out why the research institutes in the country are not producing in the same way as the universities.

Universities in South Africa are the real contributors to scientific research. For the period from 1945 to 2010 (analysed in chapter 4), the first five institutions were universities with a combined share of 61 per cent. If the data for the recent sampled year 2010 is also con-sidered, it can be observed that nine universities that topped the list produced 83 per cent of the total scientific output for the entire period. They are the University of Cape Town (19%), University of Stellenbosch (14%), University of Witwatersrand (13%), University of Pretoria (12%), University of KwaZulu-Natal (11%), Rhodes University (3.8%), University of Johannesburg (3.5%), North West University (3.4%) and University of Western Cape (3%). The contribution of the research institutes was,


in contrast, modest. The Council for Scientific and Industrial Research (CSIR) (2.9%), and the Medical Research Council (1.4%) are examples.

The theoretical prism of the Triple Helix model applies here. The Triple Helix of university–industry–government affirms that the university can play an enhanced role in innovation in knowledge-based societies (Etzkowitz and Leydesdorff, 1998, 2000). In the Triple Helix I configura-tion the nation-state encompasses academia and industry and directs the relations between them. In the Triple Helix II there are separate institutional spheres (state, academia and industry) with strong borders dividing them and highly circumscribed relations among the spheres. This model entails a laissez-faire policy to reduce the role of the state in Triple Helix I. Triple Helix III is generating a knowledge infrastructure in terms of overlapping institutional spheres of academia, state and indus-try (creating trilateral networks and hybrid organizations), with each taking the role of the other and with hybrid organizations emerging at the interface (Etzkowitz and Leydesdorff, 2000). South Africa appears to be at this stage of the Triple Helix, linking with sectors in the production of knowledge through collaborative and non-collaborative research.

Subjects and citations

Broadly, five major branches of science (referring to the sampled years) have been classified in the database: natural sciences, health sciences, agricultural sciences, engineering sciences and applied technology, and social sciences and humanities. Natural sciences is the most pre-ferred branch of science in South Africa, supplying 43 per cent of all the research publications amongst the subjects. Close to it is the health sciences with a share of 33 per cent of the total count of publications. Agricultural sciences constituted 11 per cent while engineering sciences formed another 9 per cent. Among the sciences, excluding social sci-ences and humanities, the least researched branch is engineering. For quite a long time, South Africa has been in the forefront of research in medical and biological sciences (Sooryamoorthy, 2010a). The country still maintains the edge in these branches. For various practical reasons, these branches of science have consistently attracted scientists from abroad to work with South Africans (Table 5.5).

Some major disciplines/subjects are presented in Table 5.5 to under-stand the publication trends of South African scientists (also see Table 5A.1 for all subjects). General and internal medicine has recorded the highest number of publications (2013; 8.2%) for the period of analysis. The number of research publications in this field grew from 338 in 1975


Tabl

e 5.

5 M

ajor

dis

cip

lin

es/s

ubj

ects

of

Sou

th A

fric

an p

ubl

icat

ion

s, 1

975–

2010

Sub

ject

s

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

Med

icin

e, g

ener

al a

nd

in

tern

al33

827

.939

021

.442

418

.030

711

.214

85.

313

74.

110

52.

516

42.

72,

013

8.2

Plan

t sc

ien

ces

272.

232

1.8

913.

916

76.

115

25.

418

65.

517

64.

224

74.

11,

078

4.4

Mu

ltip

le d

isci

pli

nar

y sc

ien

ces

675.

576

4.2

103

4.4

913.

362

2.2

135

4.0

170

4.1

164

2.7

868

3.5

Bio

chem

istr

y an

d m

olec

ula

r bi

olog

y39

3.2

723.

986

3.7

812.

997

3.5

100

3.0

110

2.6

128

2.1

713

2.9

Ast

ron

omy

and

ast

rop

hys

ics

312.

636

2.0

492.

155

2.0

135

4.8

932.

810

42.

515

02.

565

32.

7V

eter

inar

y sc

ien

ces

70.

693

5.1

723.

178

2.8

541.

983

2.5

832.

098

1.6

568

2.3

Zool

ogy

90.

765

3.6

612.

689

3.2

531.

951

1.5

872.

199

1.6

514

2.1

Tota

l1,

211

4.9

1,82

57.

42,

353

9.6

2,74

811

.22,

798

11.4

3,36

713

.74,

183

17.0

6,09

624

.824

,581

100

Not

e: T

otal

mea

ns

all

the

dis

cip

lin

es t

hat

are

not

sh

own

.


to 424 in 1985 but in percentage terms, these were 28 and 18 per cent respectively for these two years. Between 1985 and 2010 the number of publication records had decreased substantially from 424 to 164. In par-ticular, the trend was one of decline from 1985 until 2005 before it recov-ered slightly in 2010 with 164 publications. One reason for this could be that the specialized subjects in this branch, such as surgery, rheumatol-ogy and haematology, are counted separately. Plant sciences, the second in the total number of publications, produced 27 publications in 1975. By 2010, the figure rose to 247. But the percentage of this branch of sci-ence to the total South African publications has not produced any con-sistent increase during the period. Astronomy and astrophysics displayed similar growth patterns in the number of records. These disciplines had strong support in the past, as recorded in chapter 2.

In Africa, as reported by Pouris and Ho (2014), research emphasis is on natural resources and medical fields. Referring to the analysis of four prominent databases including the National Science Indicators Database (NSID) and Institute for Scientific Information (ISI), for a 16-year period (1981–96), Ingwersen and Jacobs (2004) outlined that the growth rate in the publication productivity of South African scientists in physics, mathematics, astrophysics, chemistry, plant and animal sciences, and biochemistry was 48 per cent as against 36 per cent for the world. Pouris’ (2003) study found immunology growing by 80 per cent from 1990–94 to 1996–2000, agricultural sciences by 21 per cent, mathematics by 11 per cent, microbiology by 12 per cent, neurosciences by 15 per cent and pharmacology by 14 per cent while computer science, materials sci-ence and clinical medicine declined by 22–29 per cent. This trend has not been sustained during 1975–2010, as seen in the data presented here.

Where do the South African scientists choose to publish their research? This is an indication of the internationalization of South African sci-ence. During the sampled years of analysis between 1975 and 2010, South African authors published a total of 5,284 research papers in the journals that are based in South Africa. This constitutes 22 per cent of the total publications of South African authors. In 1975, it was 35 per cent of the total publications, which by 1995 reduced to 16 per cent. It fur-ther shrunk to 14 per cent in 2005 and then to 13 per cent in 2010. Since the apartheid phase, scientists turned more and more to overseas journals to publish their research findings, and the world scientific com-munity became more receptive to South African research. The closed-off period, discussed in the previous chapters, had its effect on the choice of the publication of research as well. This finding can be read with that of Mouton et al., (2006) who noted that in 1990, the heyday of


apartheid when academic isolation was rampant, only 36 per cent of the publications appeared in foreign journals, which by 2002 increased to 47 per cent. Mouton et al. (2006) also found that there has been an increase in the number of papers by South African authors published in ISI-listed journals though the number of South African journals in the ISI list remained constant. International journals that are indexed in prominent databases and are used internationally are attractive targets for South African authors (Gevers et al., 2006). In the absence of simi-lar data, comparison with the journals in which South African scholars published their research findings during the colonial and apartheid peri-ods is not possible.

The three major countries of origin of the journals in which the authors preferred to publish were the US, England and the Netherlands. The journals based in these countries carried 29, 26 and 10 per cent respectively of the publications. As shown above in the partnering coun-tries section, the research partners came mostly from the US, England and the Netherlands.

Citation analysis indicates how the production of scientific knowledge is used by the world scientific community (Rabkin et al., 1979), going beyond mere referencing in a paper to substantiate or disapprove of a view point or a scientific argument. Citations reflect recognition, visibil-ity, impact and utility (Pouris, 2006b; Vieira and Gomes, 2010) but is not free from problems (MacRoberts and MacRoberts, 2010). The number of citations per paper, called relative citation impact (RCI), is used as a measure to gauge the impact of a nation’s scientific output (King, 2004; May, 1997). Reviewing the average citation rates of the publications of South African authors in the ISI database, Barnard et al. (2012) counted the average citation rate for the period 1995–2008 as 74 per cent of the rest of the world. The h-index of Hirsch (2005) is another important measure to assess the impact of publications. Wohlin (2009) has intro-duced another one, w-index.

The average number of citations per South African paper for the period was 8.43 (S.D. = 9.5). The citation count was 17.2 (1975), 14.5 (1980), 13 (1985), 13.2 (1990), 13 (1995), 9.7 (2000), 2.7 (2005) and 2 (2010). As the count tends to decrease over the years, the count for the period since 2000 should be taken with caution. It is heavily dependent on the time the data are captured for these years. The average figures for the years are nevertheless useful. Health sciences in general have received the highest number of citations for the entire period 1975–2010 (Table 5.6). Following this are the natural sciences, social sciences and humanities, agricultural sciences, and engineering sciences and applied


technology in the descending order of the number of citations received for the period.

The highly cited papers for the period of 1975–2010 belonged to meteorology and atmospheric sciences, geochemistry and geophysics, gastroenterology and hepatology, rheumatology, genetics and heredity, oncology, biochemical research methods, endocrinology and metabo-lism, haematology, psychology, and cardiac and cardiovascular systems. Most of these highly cited subjects are in the discipline of health sci-ences. A detailed view of the highly cited subjects during 1975–2010 is given in Table 5.7. In the selected years between 1975 and 2010, the most cited subjects included plant sciences, multidisciplinary sciences, vet-erinary sciences, zoology, surgery, water resources, public environmen-tal and occupational health, physics (multidisciplinary), oncology and ornithology. The count for plant sciences, multidisciplinary sciences, veterinary sciences and zoology has recorded a steady increase over the years. The study of Jeenah and Pouris (2008) for the two periods 1995–2004 and 1996–2005 showed that there has been an improvement in the number of citations received by South African publications in most of the disciplines between the study periods. In their analysis the high-est increase in citations has occurred in disciplines such as immunology, social sciences, neuroscience and behaviour, microbiology, computer sciences, geosciences, clinical medicine and environment and ecology (Jeenah and Pouris, 2008). Ingwersen and Jacobs (2004), in agreement with the analysis provided here, pointed out that the obtained citations for South African publications paint a muddled picture of fluctuations in the five selected research fields of animal and plant sciences, bio-chemistry, chemistry, microbiology and molecular biology, and physics, between 1981 and 2000. This is also true with the data presented in this chapter.

Table 5.6 Citations of publications by disciplines/branches, 1975–2010

Branches of science N %Mean

citation S.D.

Health sciences 8,106 33.1 10.4 24.2Natural sciences 10,506 42.8 8.6 38.1Social sciences and humanities 939 3.8 7.8 13.7Agricultural sciences 2,737 11.1 6.1 18.2Engineering sciences and

applied technology2,267 9.2 3.8 8.9

Total 24,555 100 8.43 29.9


Tabl

e 5.

7 H

igh

est

cou

nt

of c

itat

ion

s of

pu

blic

atio

ns

by d

isci

pli

nes

, 197

5–20

10

Sub

ject

s

Yea

r

Tota

l19

7519

8019

8519

9019

9520

0020

0520

10

N%

N%

N%

N%

N%

N%

N%

N%

N%

Plan

t sc

ien

ces

272.

232

1.8

913.

916

76.

115

25.

418

65.

517

64.

224

74.

11,

078

4.4

Mu

ltid

isci

pli

nar

y sc

ien

ces

675.

576

4.2

103

4.4

913.

362

2.2

135

417

04.

116

42.

786

83.

5V

eter

inar

y sc

ien

ces

70.

693

5.1

723.

178

2.8

541.

983

2.5

832

981.

656

82.

3Zo

olog

y9

0.7

653.

661

2.6

893.

253

1.9

511.

587

2.1

991.

651

42.

1Su

rger

y7

0.6

211.

253

2.3

742.

757

254

1.6

421

320.

534

01.

4W

ater

res

ourc

es0

026

1.4

301.

333

1.2

381.

456

1.7

571.

490

1.5

330

1.3

Publ

ic, e

nvi

ron

men

tal

and

oc

cup

atio

nal

hea

lth

70.

63

0.2

140.

632

1.2

281

381.

177

1.8

131

2.1

330

1.3

Phys

ics,

mu

ltid

isci

pli

nar

y7

0.6

241.

332

1.4

331.

233

1.2

461.

449

1.2

380.

626

21.

1O

nco

logy

70.

623

1.3

301.

333

1.2

311.

145

1.3

370.

936

0.6

242

1O

rnit

hol

ogy

00

170.

925

1.1

120.

418

0.6

381.

133

0.8

530.

919

60.

8


As indicated in these citation data, some branches of science in South Africa enjoy an international reputation. References to and citations of research pieces do not appear in a vacuum. In order to get the atten-tion of the scientific community, these research outcomes should have the traits of credibility, authenticity, quality and replicability. South Africa has been deliberately strengthening its capacity in these areas of science—medical science and astronomy, for instance—since the colo-nial period.

Collaborative versus non-collaborative research

Returning to the central thread of collaboration and productivity, more analysis is required to find the way these two variables interact in South African science. This section is, therefore, devoted to the discussion on this aspect. Also explained are the additional dimensions—disciplinary and sectoral elements—of collaboration and productivity. Does collabo-ration influence the scientific output of scientists? If so, does it make any difference in the preference of disciplines? What sectoral differentiation could be seen between collaborated and non-collaborated publications? Before examining these questions and the related evidence shown in Table 5.8, the publication trends warrant repetition.

South Africa produced a total of 24,589 publications during the selected years (Table 5.1). There were 1,212 publications in 1975, which rose to 6,096 in 2010. This increase between 1975 and 2010 was fivefold (503%). The percentage contribution of each of the years to the total number of publications of the selected years shows a pattern of increas-ing contribution: 5 to 25 per cent for the period of analysis. The most significant increase in the share of publications occurred in 2010 (25%), jumping from a share of 17 per cent in 2005. Until 2005 the increase was modest—in the range of 5–14 per cent. In 1980 the publication records grew by 151 per cent over the 1975 figure. For the subsequent years the increase was 129 (1985), 117 (1990), 102 (1995), 120 (2000), 124 (2005) and 145 (2010) per cent over the respective preceding years. Clearly, the publication output of South African scientists has been unceasing since 1975. Science in South Africa in the apartheid era (1948–94) and in the current phase (1994–) has been moving on a stable path of growth. In the previous chapters it was noted that despite resistance to the apart-heid regime by the international scientific fraternity, South Africa toiled towards strengthening its own scientific system without much outside support and assistance. This is clearly demonstrated in the data set pre-sented here. During some years of the apartheid period, that is, 1975–1990, the output grew from 5 to 11 per cent. Then, in the new political


Table 5.8 Collaborative and non-collaborative publications, 1975–2010

Variables CollaborationNo

collaboration Total

N % N % N %

Any collaboration 20,408 83 4,181 17 24,589 100All South African authors***,a 12,754 96.2 497 11.9 13,071 53.2

Domestic collaboration***,a 15,244 74.7 82 2 15,326 62.3International collaboration***,a 7,774 38.1 60 1.4 7,834 31.9

Internal-institutional collaboration***,a

11,442 46.7 34 0.8 11,476 46.7

External-institutional collaboration***,a

5,667 27.8 53 1.3 5,720 23.3

Citations by discipline Mean S.D. Mean S.D. Mean S.D.

Mean number of citations***,b 8.54 31.1 7.85 19.9 8.43 29.5

Plant sciences 7.1 24.6 4.97 8.5 6.7 22.7

Multidisciplinary sciences 12.2 32.9 4.69 7.2 10.5 29.9Veterinary sciences 4.6 7.14 6.0 12.5 4.9 8.5Zoology 6.6 11.7 7.8 10.5 6.9 11.4Surgery 8.0 15.7 4.4 7.3 7.4 14.7Water resources 3.3 6.2 2.8 4.2 3.2 5.9

Public, environmental and occupational health

6.0 10.8 4.4 9.9 5.9 10.7

Physics, multidisciplinary 9.0 24.1 4.7 7.2 8.1 21.8Oncology 20.2 40.1 11.4 20.9 19.5 39.3Ornithology 4.0 8.0 5.6 7.9 4.6 8.0

Sector of authors (first five authors combined) N % N % N %

University/Technikon 39,348 93 2,978 7 42,326 74Research institute 6,869 91.7 619 8.3 7,488 13.1Industry 852 83.1 173 16.9 1,025 1.8Government 2,403 86.5 374 13.5 2,777 4.9Hospital 3,232 90.2 352 9.8 3,584 6.3Total 52,704 92.1 4,496 7.9 57,200 100

Notes: Sig: ***p < 0.01. a Chi-square test; b ANOVA test. Figures in the sector of authors refer to responses, not just cases of records.

environment that began in 1994, South Africa consolidated its science. A two-digit growth was to be experienced in 1995 and thereafter.

In order to ascertain whether this growth in publications was a result of collaboration, another segment of data will be examined to determine


whether collaboration during the period of analysis had a correspond-ing increase or growth.

The data confirm that South African scientists are in favour of col-laborative research to a great extent, with 83 per cent of them being involved in collaboration of some kind, domestic or international (Table 5.8). The rate of collaboration for the period was 96 per cent, that is, nine out of every ten papers was a product of collaboration. As seen earlier, both the number and percentage of collaborated publications grew significantly from 68 to 92 per cent. Parallel to this is the increase in the number of publications from 5 to 25 per cent (1,212 to 6,096 publications). The evidence suggests a strong concomitant relationship between publication productivity and collaboration.

The forms of collaboration that exist in the country are varied. When the collaboration of South African scientists was grouped into domestic and international collaboration, 75 per cent were domestic collabora-tions and 38 per cent were international collaborations. This means, about three-fourths of collaborators were either in the team of domestic or in the team of international partners. A further break-up shows inter-nal-institutional and external-institutional collaboration within the domestic type. Quite revealingly, internal-institutional collaborations outweigh external-institutional collaborations, showing the heightened interest of scientists to work with those in the same institution. In a previ-ous study Tijssen (2007) reported that the proportion of co- publications (with international partners) to total publications grew by 39 per cent in South Africa during the period 2001–2004. Comparable countries in this study (Tijssen, 2007) were Egypt with 39 per cent increase, Mali 85 per cent and Gabon 87 per cent. In domestically co-authored publica-tions, on the other hand, the proportion diminished in many African countries from 48 to 34 per cent when the worldwide trend was just the opposite (Tijssen, 2007). Jacobs’ (2008) analysis on the publication records of South African authors for the period 1995–2003 too showed that South African authors collaborate more frequently with interna-tional community than with themselves within the country. Another study (Sooryamoorthy, 2009b) is in agreement with the finding that international collaboration is preferred to domestic collaboration in South Africa.

As regards the relationship between subject and collaboration, the data are shown in Table 5.8. The second panel of Table 5.8 presents the citations of collaborated publications and non-collaborated publica-tions. A large majority of the scientific output in all subjects, at varying levels, has been the result of associated research. The mean number of


citations for all subjects for the selected years indicates that the publi-cations created out of collaboration have earned more citations than the papers published individually. This difference is statistically signifi-cant in the independent t-test (p < 0.001). A few of the selected subjects that received the highest number of citations show the same pattern. Plant sciences, for example, have recorded higher citations of collabo-rated papers in the discipline than of non-collaborated papers. For the multidisciplinary sciences, the difference was greater than for the other subjects by nearly threefold. Publications in oncology too had reported a higher rate of citations for its collaborated papers. Except for zoology and ornithology, collaborated papers attracted significantly more cita-tions than single-authored papers.

Sectoral contrast can also be noticed in the sectors of the first five authors. In the data, there were 83 per cent papers that were writ-ten jointly. This has increased to 92 per cent by sector (Table 5.8). Collaboration is favoured by all sectors but to a lesser extent by the industry and government sectors. For the non-collaborators, the sectors of government and industry are relatively more.

Domestic and international partners have markedly different charac-teristics. International collaboration brings in more people to the team than domestic collaboration–about two persons more (Table 5.9). This variation is statistically significant. In comparison to domestic collabo-ration, international collaboration is nearly double in size. While the average number of authors for all papers is about 4.3, the average num-ber of authors for internationally collaborated papers is 7.8, more than 80 per cent than the overall average. This factor is reflected in the fractional count of papers wherein domestic has a more fractional count than the international, which implies more participants in internation-ally collaborated publications. The citation figures also show that inter-nationally collaborated publications are cited more significantly than domestic collaborated publications. The average citation figures for internationally collaborated publications are also more than the average number of citations for all the papers. This difference is also significant in the independent t-test.

The number of collaborations varies from discipline to discipline. International collaborations, in the total count of papers in the field, are mostly in public, environmental and occupational health, oncol-ogy, plant sciences, zoology, multidisciplinary sciences, and veterinary sciences. Among the sectors, international publications are produced mostly at universities. This is true of domestic publications as well (Table 5.9). Internationally partnered publications that are generated


Table 5.9 Domestic and international collaboration

VariablesDomestic

collaborationInternational collaboration Total

Mean number of authors 4.1 (18.4) 7.8 (48.2)***,b 4.3 (27.4)Mean number of citations 7.5 (30.1)***,b 9.8 (29.2)***,b 8.4 (29.5)Mean fractional count of papers 0.34 (0.1)***,b 0.26 (0.2)***,b 0.4 (0.3)

N % N % N %

Citations by discipline***,a

Plant sciences 703 4.6 282 26.2 1,078 4.4Multidisciplinary sciences 473 3.1 213 24.5 868 3.5Veterinary sciences 395 2.6 108 19 568 2.3Zoology 274 1.8 138 26.8 514 2.1Surgery 267 1.7 30 8.8 340 1.4Water resources 258 1.7 28 0.4 330 1.3Public, environmental and

occupational health218 1.4 168 50.9 330 1.3

Physics, multidisciplinary 101 0.7 137 1.7 262 1.1Oncology 151 1 111 45.9 242 1.0Ornithology 105 0.7 9 14.1 64 0.3

Sector of authors (first five authors)University/Technikon 29,935 70.7 16,176 72.8 42,326 74Research institute 4,770 63.7 3,455 15.5 7,488 13.1

Hospital 2,734 76.3 970 4.4 3,584 6.3Industry 623 60.8 318 1.4 1,025 1.8Government 1,595 57.4 1,313 5.9 2,777 4.9

Total 39,657 69.3 22,232 38.9 57,200 100

Notes: a Chi-square test; b independent t-test. Sig: ***p < 0.01.

at research institutes are far fewer than those in domestic collaborated papers.

In some fields such as earth/space sciences and physics and, to a slightly lesser extent, in engineering/technology, as Frame and Carpenter (1979) observed, international institutional co-authorships take place more heavily than in other fields. Newman’s (2001) study indicated that experimental high energy physics had a staggering size of collabora-tion, an average of 173 collaborators per paper. In biomedical research, a much lower degree of clustering was found, that is, it is more common in biomedicine to start a collaboration with just two people (Newman, 2001). South African science follows this international pattern, as can be observed from this study.


In this data set a physics paper had 3,172 authors; the paper was ‘Search for New Particles in Two-Jet Final States in 7 TeV Proton-Proton Collisions with the ATLAS Detector at the LHC’) and wascarried in the Physical Review Letters (105; 16: 2010). There were another six papers that brought together more than 1,000 authors. As the data demon-strate, geochemistry and geophysics topped with the largest size of col-laboration. The mean number of authors for the papers in this discipline for the years of 1975–2010 was 311. Physics (particles and fields in par-ticular) comes next in the list with an average size of 108 partners in the papers the discipline had produced during the sampled years. Following these were physics (multidisciplinary) with an average size of 24 authors and astronomy and astrophysics with that of 16. Some of the subjects with the lowest size included engineering, and information science and library science.

The size of the team in collaboration has a role in determining the research outputs, that is, papers and citations. Smart and Bayer (1986) analysed the citation rates of single-authored and multiple-authored papers in three specified applied science fields for ten years. They found consistently lower citation rates for single-authored papers than mul-tiple-authored papers. Papers and citations increased correspondingly with the size of the team, while the role of institutional share in the production of papers was not very clear (Adams et al., 2005). Examining the correlation between citations and number of authors of papers from selected journals, Hsu and Huang (2011) reported that citations were the lowest for single-authored papers. Bartneck and Hu (2010) could not observe any kind of relationship between collaboration and cita-tion in their bibliometric analysis. Studying Finnish scientists for the period between 1990 and 2008, Puuska et al. (2014) found a positive relationship between international cooperation and citation in all the disciplinary groups. They also noted that domestic collaboration did not necessarily lead to a higher citation count. Gazni and Didegah (2011) acknowledged a positive correlation between the number of authors and citations of the Harvard publications they analysed. They also noticed that the publications that had international collaboration had received a larger normalized mean of citations than papers that had domestic collaboration (Gazni and Didegah, 2011). Persson (2010) had a different take on this. His analysis of international papers among 100 most cited papers in the chosen research specialties illustrated that international papers were not strongly represented among high impact papers. To see how far this is true with the data, Pearson’s correlation coefficient test was run for the number of authors and the number of citations; this


showed significant positive correlation (r = 0.39, p < 0.01). An increase in the number of authors (collaborators) was accompanied by a subsequent increase in the number of citations the paper received. South African publications, therefore, attract more citations depending on the number of authors involved in the production.

In the regression model presented in Table 5.10, a step further from the correlation analysis, five control variables (log converted number of authors and foreign countries, year of publication, and collaboration—internal-institutional, external-institutional and international) were used. Two variables were very significant and positively associated with the number of citations: number of authors and international collabora-tion. The year of publication was negatively correlated with the number of citation. This was clear from the previous analysis that the citation count showed a decreasing trend as the year progressed. The model explains 24 per cent of the variance (R2 = 0.249). It is quite clear that papers that were produced with international partners obtain a higher rate of citations in science. Domestic collaboration does not influence the number of citations a publication earns in its life.

Conclusion

Scientific productivity in South Africa seen through the scientific pub-lications in journals indexed in the Web of Science database since 1975 has progressed at a varying pace with definite ups and downs. The growth was substantial in the last year of analysis (2010) in particular. The publication productivity of South African scientists has increased substantially in the two political phases—apartheid and democracy. Most importantly, South African scientists worked collaboratively with

Table 5.10 Regression of citation on collaboration

Collaboration No. of citation

Number of authors 0.253***Year of publication −0.508***Number of foreign countries involved 0.005nsInternal-institutional collaboration in South Africa (1 = yes, 0 = no) −0.074***External-institutional collaboration in South Africa (1 = yes, 0 = no) −0.042***International collaboration (1 = yes, 0 = no) 0.045***R2 0.249N 24,493

Note: Sig: ***p < 0.01.


the international science community spread all over the world, produc-ing a sizeable share of joint publications. This proves a positive relation-ship between the number of publications produced and the extent of collaboration—both in terms of the number of collaborated publications and the number of partners. Although the direction of the relationship cannot be confirmed, evidence suggests that productivity and collabo-ration are positively correlated. In domestic collaboration, external-institutional rather than internal-institutional is the leading category of collaboration in the country. International collaboration is not lagging behind in the production of scientific publications and might exceed the domestic form in future, if the trend in 2010 is any indication. Major partners included countries that had maintained ties with South Africa in the past and new countries with whom alliances have been forged in the new political phase.

Some branches of science have been more productive and collabora-tive than some other branches. The disciplinary dimension in collabo-ration is supported in the case of certain disciplines. These branches of science had the opportunity to develop and grow during the colo-nial and apartheid times. Papers in these disciplines have appeared not only in international journals that originated in South Africa but also in those from other centres of world science. Some of them are high impact journals recognized internationally for their standards and quality. This worked well for the internationalization of South African science and in the growing interest of scholars leading to collaborations.

A microscopic view of the scientific output for the selected years has uncovered the nature and character of South African science, particu-larly its collaborative angles. Collaboration at domestic and interna-tional levels is flourishing in an encouraging scientific environment. The number of foreign partners, an index to the size of collaboration, is on an ascending trail. In this collaborative enterprise, all sectors—university, research institute, industry, hospital and government—take part actively, accumulating and disseminating the scientific wealth of South Africa to world science. Certain branches of sciences grew well in relation to some other branches, and collaboration was a decisive fac-tor in their growth. Collaboration also enabled South Africa to main-tain its edge that it has achieved over the years in some disciplines. It has brought more visibility to its science through the increased rate of citation and its impact on world science. Distance does not seem to affect collaboration. Most of the partners of South African scien-tists are from distant locations, not just from neighbouring African countries.


More revealing would be the examination of scientific research that South African scientists are currently engaged in. Primary empirical data, gathered directly from scientists in face-to-face interviews, would serve this purpose, and this is dealt with in the next chapter.

Appendix

Table 5A.1 Size of collaboration and subjects, 1975–2010

Subjects N Mean S.D.

Geochemistry and geophysics 85 311.0 3.3Physics, particles and fields 30 108.1 321.0Physics, multidisciplinary 262 24.0 216.1Astronomy and astrophysics 653 16.2 54.2Education, special 1 9.0 –Psychology, clinical 29 9.0 6.2Medical informatics 1 8.0 –Virology 94 6.9 5.0Immunology 436 6.6 5.7Genetics and heredity 180 6.1 6.5Infectious diseases 219 6.1 4.3Physics, nuclear 148 6.0 11.0Medical ethics 1 6.0 –Rehabilitation 1 6.0 –Substance abuse 28 6.0 4.9Oncology 242 5.9 5.6Meteorology and atmospheric sciences 68 5.7 9.0Rheumatology 51 5.7 6.7Nursing 29 5.6 5.6Geriatrics and gerontology 9 5.6 2.2Respiratory system 81 5.4 3.8Biology 251 5.3 4.7Haematology 100 5.1 3.4Demography 1 5.0 –Public, environmental and occupational

health330 5.0 3.7

Gastroenterology and hepatology 110 5.0 3.4Critical care medicine 88 4.9 3.9Clinical neurology 176 4.9 7.1Endocrinology and metabolism 140 4.9 2.4Health care sciences and services 54 4.8 3.1Microbiology 242 4.8 3.9Mycology 113 4.8 5.3Peripheral vascular disease 31 4.7 4.0Neurosciences 63 4.7 5.0Chemistry, medicinal 104 4.6 4.2


Instruments and instrumentation 105 4.6 4.2Cardiac and cardiovascular systems 247 4.5 3.5Engineering, biomedical 27 4.5 2.8Dermatology 63 4.5 5.1Toxicology 15 4.5 2.6Nutrition and dietetics 133 4.4 2.5Parasitology 113 4.4 3.0Psychiatry 76 4.4 3.2Paediatrics 181 4.3 3.0Ethics 14 4.2 4.5Pharmacology and pharmacy 215 4.2 4.1Biochemistry and molecular biology 713 4.1 4.2Multidisciplinary sciences 868 4.1 11.0Biochemical research methods 149 4.1 2.8Psychology 31 4.0 2.8Archaeology 3 4.0 2.0Biotechnology and applied microbiology 474 4.0 4.2Physiology 82 4.0 4.6Medicine, general and internal 2,013 4.0 14.6Obstetrics and gynaecology 211 3.9 2.5Urology and nephrology 96 3.8 4.7Health policy and services 9 3.8 1.9Medical laboratory technology 56 3.8 1.9Biophysics 15 3.7 3.6Pathology 65 3.7 2.2Engineering, aerospace 66 3.7 4.3Radiology, nuclear medicine and medical

imaging103 3.7 2.1

Evolutionary biology 39 3.7 2.9Education and educational research 22 3.7 2.1Medicine, research and experimental 88 3.7 2.7Andrology 26 3.6 1.6Physics, applied 68 3.6 2.6Electrochemistry 52 3.6 2.1Oceanography 63 3.6 2.6Sport sciences 75 3.6 1.8Geography, physical 47 3.6 3.1Physics, condensed matter 122 3.5 2.5Biodiversity conservation 208 3.5 2.9Nuclear science and technology 23 3.5 2.2Psychology, developmental 2 3.5 2.1Chemistry, inorganic and nuclear 378 3.5 1.8Cell biology 106 3.4 2.5Environmental sciences 255 3.4 2.4Rehabilitation 9 3.4 3.3

(Continued )



Fisheries 133 3.4 3.0Chemistry, applied 123 3.4 1.7Chemistry, organic 251 3.4 1.5Materials science, biomaterials 8 3.4 1.9Nanoscience and nanotechnology 24 3.3 2.3Developmental biology 20 3.3 1.7Psychology, biological 8 3.3 1.4Food science and technology 145 3.2 1.5Chemistry, multidisciplinary 487 3.2 2.1Surgery 340 3.2 2.2Agronomy 196 3.2 2.0Behavioural sciences 81 3.2 2.5Agriculture, multidisciplinary 133 3.2 1.5Geosciences, multidisciplinary 283 3.2 2.9Acoustics 30 3.2 1.5Ecology 663 3.2 5.8Anaesthesiology 79 3.1 1.8Crystallography 180 3.1 1.3Agronomy 25 3.1 1.4Anthropology 62 3.1 2.1Political science 123 3.1 1.5Chemistry, analytical 251 3.0 2.3Engineering, environmental 158 3.0 1.8Orthopaedics 98 3.0 2.0Agriculture, dairy and animal science 261 3.0 1.5Spectroscopy 58 3.0 1.6Chemistry, physical 299 3.0 1.5Energy and fuels 119 3.0 2.2Agricultural economics and policy 27 3.0 1.2Forestry 73 3.0 1.4Materials science, multidisciplinary 221 3.0 1.8Neuroimaging 5 3.0 0.7Transplantation 2 3.0 1.4Veterinary sciences 568 3.0 1.7Engineering, chemical 252 3.0 2.8Horticulture 66 3.0 1.8Dentistry, oral surgery and medicine 162 3.0 1.7Otorhinolaryngology 50 3.0 2.3Geology 123 2.9 1.5Emergency medicine 8 2.9 1.1Materials science, composites 34 2.9 1.2Plant sciences 1,078 2.9 1.8Anatomy and morphology 76 2.8 2.6Marine and freshwater biology 389 2.8 2.1Integrative and complementary medicine 11 2.8 1.8




Materials science, ceramics 20 2.8 1.8Agriculture, soil science 19 2.8 1.5Materials science, coatings and films 4 2.8 1.0Optics 62 2.7 1.6Physics, atomic, molecular and chemical 33 2.7 1.6Palaeontology 38 2.7 1.9Computer science, cybernetics 3 2.7 2.1Materials science, characterization and

testing6 2.7 1.4

Ophthalmology 64 2.7 1.9Political science 6 2.7 1.2Engineering, manufacturing 8 2.6 0.9Medicine, legal 29 2.6 2.1Computer science, hardware and

architecture18 2.6 1.2

Water resources 330 2.6 1.2Engineering, electrical and electronic 215 2.6 1.9Entomology 416 2.6 1.5Physics, fluids and plasmas 50 2.5 1.1Agricultural engineering 23 2.5 1.1Mineralogy 31 2.5 1.1Ethnic studies 2 2.5 2.1History and philosophy of science 4 2.5 2.4Materials science, textiles 14 2.5 1.3Reproductive biology 30 2.4 1.2Zoology 514 2.4 1.6Engineering, industrial 49 2.4 1.1Engineering, civil 96 2.4 2.2Geography 23 2.4 2.5Computer science, artificial intelligence 36 2.3 0.8Limnology 12 2.3 1.7Ornithology 196 2.3 1.5Thermodynamics 113 2.3 1.1Materials science, paper and wood 30 2.3 1.1Computer science, interdisciplinary

applications157 2.3 1.2

Construction and building technology 82 2.3 1.1Engineering, mechanical 112 2.2 1.0Engineering, geological 21 2.2 1.1Economics 7 2.1 1.1Operations research and management

science29 2.1 1.0

Mechanics 46 2.1 0.9Computer science, information systems 53 2.1 1.1Automation and control systems 55 2.1 0.8

(Continued )



Mathematics, applied 185 2.1 0.9Microscopy 21 2.1 1.0Physics, mathematical 24 2.0 0.7Computer science, software engineering 35 2.0 1.0Business 1 2.0 –Psychology, experimental 5 2.0 0.7Statistics and probability 73 2.0 0.9Telecommunications 1 2.0 –Computer science, theory and methods 47 2.0 1.0Metallurgy and metallurgical engineering 324 2.0 1.0Engineering, multidisciplinary 91 2.0 0.9Mathematics, interdisciplinary applications 8 1.9 1.4Materials science, textiles 355 1.9 1.0Communication 6 1.8 1.6Education, scientific disciplines 28 1.8 1.0Management 13 1.8 0.8Mining and mineral processing 17 1.4 0.6Psychology, social 10 1.4 1.0Applied linguistics 1 1.0 –Engineering, marine 1 1.0 –Environmental studies 1 1.0 –Imaging science and photographic

technology1 1.0 –

Information science and library science 2 1.0 0.0International relations 1 1.0 –Language and linguistics theory 1 1.0 –Law 3 1.0 0.0Psychology, multidisciplinary 3 1.0 0.0Robotics 1 1.0 –Social issues 1 1.0 –Sociology 1 1.0 –Total 24,581 4.27 27.37



135

Based on primary data gathered from scientists working in research institutes, in a university and in an agricultural college located in the province of KwaZulu-Natal, this chapter presents the features of scien-tific research as it exists in today’s South Africa. Specifically, the research projects of both collaborative and non-collaborative kinds of respond-ents, their distinctive collaborative facets and the factors that predict collaborative research in South Africa are discussed.

Scientists and academics

Provided in this and in the subsequent chapter is the report of an inten-sive investigation of collaboration, research communication and pro-ductivity of scientists and academics working in one of the provinces, namely, KwaZulu-Natal. KwaZulu-Natal is the second-largest populated province in South Africa, after Gauteng, with 19.8 per cent of the coun-try’s total population (10,267,300) but only 7.6 per cent of the country’s total area (Census, 2011). In this survey, carried out in 2007–08, 204 scientists and academics working in 16 teaching/research departments and 10 research institutes situated in 5 major centres—Cedara, Durban, Mount Edgecombe, Pietermaritzburg and Umhlanga—were interviewed. As part of the transformation that is underway in the higher educa-tion sector of the country, new institutions were formed, merging and incorporating small universities, which were formerly white or black universities, into larger institutions. Currently, there are 26 universi-ties in South Africa including one new medical university. As regards research institutions, South Africa has a number of statutory science councils that carry out research for social, scientific and technological development (Scholes et al., 2008). They inlcude the African Institute of

6Scientific Research in South Africa


South Africa, Agricultural Research Council, the Council for Scientific and Industrial Research (CSIR), the Council for Geosciences, the Human Sciences Research Council, the Medical Research Council, the Council for Mineral Technology, the Nuclear Energy Corporation of South Africa and the National Research Foundation. Each of these, except the National Research Foundation, operates through several research insti-tutes situated through the entire length and breadth of the country. A few of these—the Agricultural Research Council, the CSIR, the Council for Geosciences, and the Human Sciences Research Council—were cho-sen for our research institute sample. The academic sample was drawn from one of the three categories—traditional universities, comprehen-sive universities and technikons. This classification existed at the time of the study in 2007–08 but is no more relevant now.

Face-to-face interviews covered a total of 204 respondents from the selected departments of a university and an agriculture college (n = 141), and national and regional research institutes (n = 63). Attempts were made to interview all willing and available respondents in the depart-ments and institutes representing the fields of biology (15%), physics (15%), mathematics (14%), chemistry (10%), zoology (9%), agriculture (7%), engineering (2%) and others. They were all full-time academics or scientists on the permanent roll. Respondents on study leave, sabbatical and seconded to other areas were not considered eligible participants for the study. In this analysis the term ‘scientists’ is interchangeably used for both academics and researchers in the institutes, unless a distinction is warranted.

Two major sectors—university and research institute—represented approximately two-thirds and one-third respectively of the sample (Table 6.1). Gender is distributed in the same way— two-thirds men and one-third women. Women choose to work at research institutes rather than in the academia. They preferred to be scientists rather than aca-demics as 40 per cent of the total women respondents were employed in research institutes. Within sectors this gender differentiation was more pronounced: 36 per cent of academics as against 40 per cent of scientists in research institutes were women. Respondents were predominantly whites. Next to them were Indians, closely followed by Africans: 53 per cent of the respondents were whites and 20 per cent Africans. Not much deviation from this pattern is apparent within sectors. This proportion of race is to be contrasted with the country’s total population where whites formed only 9.5 per cent and Africans 79 per cent (at the time of data collection in 2007–08). Married people, a key variable in research productivity studies, were in the majority. That about one-third of the

Scientific Research in South Africa 137

Table 6.1 Respondents of survey, 2007–2008

Characteristics Academics Scientists Total

N % N % N %

Sector 141 69.1 63 30.9 204 100Gender*,a

Male 104 73.2 38 26.8 142 69.6Female 37 59.7 25 40.3 62 30.4

RaceWhite 77 54.6 32 50.8 109 53.4Indian 34 24.1 14 22.2 48 23.5African 25 17.7 16 25.4 41 20.1Others 5 3.5 1 1.6 6 3.0

Born in South Africa***,a 86 61.4 55 87.3 141 69.5Marital status*,a

Married 93 66.0 37 58.7 130 63.7Single 36 25.5 25 39.7 61 29.9Divorced 8 5.7 1 1.6 9 4.4Separated 2 1.4 0 0 2 1.0Widowed 1 0.7 0 0 1 0.5

Sector worked last***,a

University 74 54.3 16 25.8 91 45.5Private 11 8.0 6 9.7 17 8.5Government research institute 13 9.4 1 1.6 14 7.0Government Agency 6 4.3 3 4.8 9 4.5NGO 1 0.7 2 3.2 3 1.5Other 11 8.0 18 29.0 29 14.5No previous employment 21 15.2 16 25.8 37 18.5

Highest degree***,a PhD 95 67.9 13 20.6 108 53.2Master’s 36 25.7 18 28.6 54 26.6Bachelor’s 7 5.0 15 23.8 22 10.8Diploma 2 1.4 15 23.8 17 8.4Other 0 0 2 1.2 2 1.0

Current affiliation***,a

Lecturer 59 42.1 0 0 59 29.1Junior researcher/Researcher/

Scientist3 2.1 0 0 3 1.5

Senior researcher 5 3.6 43 68.3 48 23.6Professor 22 15.7 0 0 22 10.8Others in academic sector 6 4.3 0 0 6 3.0Senior lecturer 32 22.9 0 0 32 15.8Associate professor 13 9.3 0 0 13 6.4Others in research institute

sector0 0 19 30.2 19 9.4

(Continued )


Characteristics Academics Scientists Total

Discipline**,a

Natural sciences 78 55.3 25 39.7 103 50.5Life sciences 41 29.1 16 25.4 57 27.9Agricultural sciences 10 7.1 5 7.9 15 7.4Others 6 4.3 9 14.3 15 7.4Engineering 6 4.3 8 12.7 14 6.9

Field of specialization***,a

Biology 21 14.9 6 9.5 27 13.2Chemistry 14 9.9 31 17.5 25 12.3Physics 21 14.9 2 3.2 23 11.3Environmental science 19 13.5 2 3.2 21 10.3Mathematics 19 13.5 0 0 19 9.3Zoology 12 8.5 5 7.9 17 8.3Agriculture 10 7.1 5 7.9 15 7.4Marine science 4 2.8 8 12.7 12 5.9Engineering 3 2.1 7 11.1 10 4.9Biochemistry 3 2.1 0 0 3 1.5Geology 1 0.7 1 1.6 2 1.0Others 15 9.9 16 25.2 30 14.8

Highest degree from foreign countriesUK 10 7.1 1 1.6 11 5.4US 10 7.1 1 1.6 11 5.4Canada 5 3.5 0 0 5 2.5

Mean S.D. Mean S.D. Mean S.D.

Age (mean years)***,b 44.1 10.8 37.5 9.6 42.1 10.8Academic age (mean years)**,b 12.3 9.9 8.6 8.2 11.1 9.5Institutional experience 10.7 10.8 8.9 9.5 10.1 10.4Year first worked in the

organization (mean)**,b

1996.6 10.8 1998.5 9.3 1997.2 10.4

Year in which moved to South Africa (mean)*,b

1992 15.1 1987 11.0 1991.4 14.6

Years spent outside the country for higher education (mean)***,b

3.8 5.3 0.9 2.2 2.9 4.7

Years spent in a developed country (mean)***,b

3.9 6.0 1.1 2.6 3.1 5.3

Notes: a Chi-square test; b independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

Table 6.1 (Continued)


respondents were not born in South Africa means the science in the country, as seen in the previous chapters, still relies on foreign scientists. Those who moved to South Africa have been in the country for a con-siderable length of time, averaging 16 years. Scientists in the institutes arrived five years earlier than those in the university sector.

Among the participants there were graduates from universities around the world. One scientist arrived from Northern Ireland where he studied geography at the undergraduate level. Pursuing his interest in geogra-phy he earned his PhD before landing in South Africa in 1974. Another academic settled in South Africa 40 years ago in 1970, having been to different countries and studying science in a number of universities. He was born in Wales and had completed his honours and master’s there before venturing into Canada to do PhD at the Dalhousie University in Halifax. Before taking up his teaching job in South Africa he went to Britain for a year on a doctoral fellowship.

Universities, more than research institutes, attracted foreign scientists, the proportion being 87 and 13 per cent respectively. Nearly a quarter of the sample had been to other countries including the UK, the US, Australia, Canada, Sweden, Germany, Cuba, and Zimbabwe to obtain their highest professional degree, which they got after living there for an average of 2.9 years. University respondents stayed abroad longer than the institute respondents by three years. They had lived in a developed country for nearly five years.

The educational itinerary of an academic is illustrative of this exposure to Western science and points to the way the South African scientific system benefits from it. One scholar first completed his undergraduate degree in the National University of Ireland, which then operated from six different locations. His double honours majors were physics and chemistry. Moving to Queen’s University in Belfast, he did his PhD in a period of 3.5 years, specializing in the field of fluorescence. Then he flew to the United States where he could work with a Nobel laureate in chemistry.

South African science still works with scientists from abroad, who are active players in both universities and research institutes. Most of these scientists in the sample, like the ones mentioned in the following para-graph, had a good deal of exposure to the world outside Africa.

One biologist had done a master’s in molecular biology in Moscow. Having worked in the area of human tissues and cells, he shifted his atten-tion to plant sciences and worked for his PhD in seed physiology and seed germination. Following this he worked at the Bulgaria Academy of Sciences for about two years. Finally, he arrived in South Africa to do his


postdoctoral in plant physiology, studying how plants resist pathogens. Another biologist arrived from the UK, having obtained his honours degree in marine biology from the University College of North Wales and his PhD in Lancaster University. Before moving to South Africa, he worked as an environmental consultant for about a decade.

It happened the other way round as well, going from South Africa to the outer world as some participants did. Keen on historical geography, one scientist went to Queen’s University first to do his master’s and then a PhD.

The scientific system is seemingly strong in the presence of qualified researchers. More than half of them in this study had doctorates (mostly those who were in the university sector), while another one-quarter had master’s as their highest degree (Table 6.1). Specialization was in several disciplines: biology, agriculture, geology, chemistry, zoology, environ-mental science, physics, mathematics and others (in descending order of the number of people who were working in these fields). Biology, chemistry, physics and environmental science are the four popular sub-jects in the country. The sectoral difference in specialization was largely in subjects such as biology, chemistry, physics, marine science and engi-neering. Juniors (lecturers) were more in the university where senior academics also formed a considerable percentage of the sample. Of the three-fourths who had had previous jobs before taking up positions in the university and research institutes, 66 per cent had had similar expe-rience with science, coming from the same sectors. Put differently, they were not novices in science and scientific research but had experience and skill, some in rare areas of specialization and expertise. A seed physi-ologist, who returned to South Africa after completing her PhD in the UK, showed how specialized her knowledge was.

Since the political transformation of South Africa to a democracy in 1994, the country has been making efforts to balance its demographic and gender profile. Supported by legislation, this transformation has led to discernible changes in the number of women who take up jobs in many sectors of the economy. True to this changing proportion of gen-der in different activities, more and more women are now attracted to universities and research institutes. As evident from the data, this trans-formation towards equity in gender has happened more rapidly within the research sector than in the university sector.

These results can be compared with those from another similar research completed in 2004–05 in South Africa (Sooryamoorthy and Shrum, 2007). The percentage of white respondents in the sample insti-tutions had decreased from 69 to 53 per cent between 2004–2005 and 2007–2008. This denotes increases in other racial groups such as African


and Indian. It also reflects the demographic transformation the country has been going through in the post-apartheid phase. The corresponding figures for the two periods (2004–05 and 2007–08) showed that the per-centage of both African and Indian people involved in scientific research had increased. The gender divide among the sample population had not changed since 2004–05. The average age of the respondents had increased by two years between these two periods, which suggested that the population of academics and scientists in the country was aging.

Transformation in the workplace has been ongoing in South Africa post-apartheid. This is aligned with the national policy of equity and transformation to ensure that equity in employment is ensured cor-responding to the proportion of various ethnic/racial groups. This applied to the institutions studied in this research. The pace at which this required transformation happened differed between academic insti-tutions and research institutes in the sample. Research institutes, as is clear from the staff complement of the research institutes in the sam-ple moved faster on the path of equity than the academic institutions. Having more younger people, in terms of both institutional and aca-demic ages of the respondents, the institutes were successful in attract-ing new staff to change the demographic balance.

Since 2004–05, there has been an increase of about four per cent in South African-born respondents in the institutions studied. The research institutes in the sample had a higher proportion of South African-born scientists on their rolls, but they were less internationalized. On the other hand, South Africa attracted from other parts of the world acad-emicians who contributed more to the internationalization of the academic institutions than the research institutes.

Research activities

How do scientists allocate the time for professional activities? The respondents were asked to calculate the proportional allocation and uti-lization of their professional time for research, teaching, administration and other activities. In general and on average, 40 per cent of their time was fruitfully applied to research activities, 32 per cent to teaching, 19 per cent to administration and another 9 per cent to other profession-related activities (Table 6.2). In this pattern, an inter-sectoral difference was obvi-ous when it was statistically tested. Not surprisingly, scientists in research institutes earmarked more time for research than any other activities. No discrepancy was noticed in the share of time meant for the admin-istrative activities of university and institute respondents (no statistical


Tabl

e 6.

2 Pr

ofes

sion

al a

ctiv

ity

Act

ivit

y

Aca

dem

ics

Scie

nti

sts

Tota

l

NM

ean

(S.

D.)

NM

ean

(S.

D.)

NM

ean

(S.

D.)

Allo

cati

on o

f ti

me

Mea

n p

erce

nta

ge o

f ti

me

spen

t on

res

earc

h**

*,a13

634

.7 (

23.2

)63

50.9

(29

.1)

199

39.8

(26

.2)

Mea

n p

erce

nta

ge o

f ti

me

spen

t on

tea

chin

g***

,a13

841

.6 (

22.2

)56

7.7

(9.4

)19

431

.8 (

24.7

)M

ean

per

cen

tage

of

tim

e sp

ent

on a

dm

inis

trat

ion

134

19.5

(18

.3)

6317

.5 (

15.6

)19

718

.7 (

17.4

)M

ean

per

cen

tage

of

tim

e sp

ent

on o

ther

p

rofe

ssio

nal

act

ivit

ies*

**,a

122

4.2

(11.

5)59

23.9

(0.

18)

182

9.7

(19)

Cur

rent

sup

ervi

sion

Prof

essi

onal

sci

enti

sts

and

en

gin

eers

116

0.98

(4)

610.

9 (2

)17

70.

94 (

3.4)

Tech

nic

ian

s an

d fi

eld

wor

kers

116

1.6

(5.3

)61

2.9

(6.9

)17

72.

04 (

5.9)

Doc

tora

l st

ud

ents

***,a

122

1.44

(2.

2)59

0.01

(0.

18)

181

0.98

(2)

Post

grad

uat

e st

ud

ents

***,a

132

3 (4

.9)

610.

7 (2

.7)

193

2.2

(4.4

)N

on-t

ech

nic

al s

taff

112

0.5

(1.5

)60

0.9

(3.4

)17

20.

7 (2

.3)

Wor

k cl

osel

y w

ith,

to

disc

uss

rese

arch

pro

ject

sPr

ofes

sion

al s

cien

tist

s13

94.

6 (6

.2)

605.

4 (4

.5)

199

4.9

(5.8

)Te

chn

icia

ns*

*,a11

82.

1 (3

.8)

604.

8 (6

.1)

178

3.0

(4.8

)D

octo

ral

stu

den

ts**

*,a12

62.

5 (4

.2)

560.

8 (1

.4)

178

3 (4

.8)

Post

grad

uat

e st

ud

ents

***,a

132

4.2

(4.8

)59

1.5

(2.7

)19

13.

3 (4

.4)

Non

-tec

hn

ical

sta

ff**

,a11

71.

1 (2

.2)

582.

3 (3

)17

51.

5 (2

.6)

Not

es: a i

nd

epen

den

t t-

test

; S.D

. in

par

enth

eses

. Sig

: **p

< 0

.05;

***

p <

0.01

.


difference). In addition to teaching and administration tasks, the academ-ics who wanted to be research active tried hard to devote sufficient time to their research. In the current working environment in the sampled institutions, as some respondents remarked, it is rarely possible. For some of them research is the passion that drives them ahead in the profession.

It is vital for any scientific system to groom the next generation of its personnel for continuity and change. For South Africa, due to the short-age of skilled personnel not only in science but also in other fields, this is very important in building a strong regiment of scientific personnel in the country. A measure of the number of people—students and staff who are aspiring to work towards a research degree—the scientists guide, direct and supervise is, therefore, appropriate. Keeping this objective in mind, the respondents were asked about the number of professional sci-entists, technicians, field workers, non-technical staff, and doctoral and master’s students they supervised (Table 6.2). To supplement this infor-mation, their close interactions with professional scientists, technicians, non-technicians and students in research affairs were explored. These measures provided an idea about the rate at which scientific person-nel were recruited into the scientific system of the country. On average, scientists at the moment supervised less than one professional scientist or engineer, two technicians or field workers, less than one doctoral can-didate, more than two postgraduates and less than one non-technician. While technicians, field workers and non-technical staff were supervised mostly by scientists in research institutes, more doctoral and postgradu-ate students underwent training at the university under academics. Disparity between university and institute sectors was not evident in the supervision of professional scientists and engineers. But significant differences were noticed in t-test (p <0.01) in the number of doctoral and postgraduate students guided by academics and scientists.

A closely related collaboration measure is the number of professionals with whom scientists seriously discuss their research projects. In their daily academic routine, such discussions with colleagues and experts are ben-eficial and unavoidable. They regularly meet with other scientists, techni-cians, doctoral candidates, and postgraduate students as much as they meet non-technicians for this purpose. The respondents seemed to be very prolific on this count, deliberating their project matters with at least five professional scientists, three technicians, about three doctoral students, more than three postgraduates and more than one non-technical staff (Table 6.2). Sectoral variations were significant in the number of technicians (scientists consult more), doctoral and postgraduate students (academics have more of them) and in the number of non-technical staff (more for


the institute respondents). Like scientists and technicians, students form an integral part of research, in the science disciplines in particular.

Research projects and facets of collaboration

The range of research areas and topics that scientists currently study suggests that South Africa is a fascinating place for scientific investiga-tions. The respondents were engaged in a total of 1,834 projects. This list emerged when each respondent gave the details about his/her current projects. Extending from agriculture to space research, the areas of study covered agronomy, air quality, nuclear physics, astrophysics, plasma phys-ics, space science, coastal management, coral reef research, cryobiology, ecotoxicology, estuarine ecology, marine life, HIV/AIDS, wetland science, water management and malaria. Some of them are of particular relevance to South Africa and to Africa in general. Applied research dominates pure research in the knowledge production centres where the study was con-ducted, with direct implications for the economy. At the same time, scien-tists are drawn to advanced realms of scientific knowledge as well.

Scientists worked on more than a single project at a time, and a num-ber of projects ran simultaneously, often in collaboration with others. The respondents had been working with an average of six projects, academics with five and institute scientists with eight (Table 6.3). In addition, they directed an average of 3.8 projects (five for scientists and three for academics). A physiologist, one of the respondents from the academic sector, had multiple projects running concurrently. One of them was on desiccation tolerance and sensitivity, particularly, of seeds. The other project was general whole plant physiology in relation to the environment. At the same time he was also working on coastal dune systems. He worked with a wide range of collaborators in different fields.

Projects do not stem from one’s own discipline alone. Disciplinary boundaries are surpassed when creative scientists step into new territo-ries of knowledge, as in the experience of a geographer who became part of a team of chemists, meteorologists and physicists.

Eighty-five per cent of the respondents had collaborated with oth-ers in research, in an average of 5.02 collaborative projects (Table 6.3). As for the nature of collaboration, three-fourths of those who reported a current research project (139 out of 204) had domestic partner-ships, while 42 per cent had international linkages. The respondents who had ever collaborated were more by 10 percentage points for academics. International partnerships demonstrated a statistically sig-nificant difference between sectors. Scientists in research institutes were


(Continued )

Table 6.3 Research and collaboration

Project Academics Scientists Total


Mean number of research projects*,b

5.26 5.70 7.64 11.71 5.98 8.07

Mean number of projects directed

3.32 4.07 4.72 10.16 3.75 6.60

Mean number of collaborative projects#

5.17 9.50 4.68 5.83 5.02 8.56

Mean number of domestic collaborative projects#,***,b

1.33 1.08 0.86 0.86 1.19 1.04

Mean number of intercontinental (African) collaborative projects

0.18 0.55 0.06 0.25 0.14 0.48

Mean number of international collaborative projects#,***,b

0.80 0.94 0.32 0.67 0.65 0.89

Collaborated partners in career 15.4 18.38 20.4 27.81 16.8 21.51Collaborated years in career** 9.35 9.65 6.38 8.02 8.43 9.27Duration of all three

collaborative projects (in years)***

8.47 7.27 4.08 3.1 7.37 6.75

Partners in all three collaborated projects

15.13 10.43 19.0 7.71 15.77 10.1

N % N % N %

Ever collaborated*,a 124 87.9 49 77.8 173 84.8Domestic collaboration 101 71.6 38 60.3 139 68.1International collaboration***,a 71 50.4 14 22.2 85 41.7

First projectFirst project reporteda 128 70.3 54 29.7 182 89.2Collaboration 115 74.7 39 25.3 154 84.6Regional collaboration

(collaboration in KwaZulu-Natal)

66 75.0 22 25.0 88 57.1

National collaboration (collaboration elsewhere in South Africa)

33 68.7 15 31.3 48 31.2

International (within Africa) collaboration

11 73.3 4 26.7 15 9.7

International (outside Africa) collaboration

52 85.2 9 14.8 61 39.6

Domestic collaboration 85 73.9 32 82.1 117 75.9International collaboration 57 49.6 13 33.3 70 45.5




Partners 5.94 10.56 7.62 4.75 6.37 9.45Beginning year of the project 2002.60 5.08 2003.08 8.244 2002.72 6.03Duration of collaboration (year) 4.15 3.23 3.15 2.53 3.88 3.08

Second projectSecond project reported***,a 96 80.0 24 20.0 120 58.8Collaboration 85 82.5 18 17.5 103 85.8Regional collaboration


48 80.0 12 20.0 60 50.0


22 73.3 8 26.7 30 25.0


8 100.0 0 0 8 6.7


30 88.2 4 11.8 34 28.3

Domestic collaboration 63 79.7 16 20.3 79 65.8International collaboration**,a 35 89.7 4 10.3 39 32.5


Partners* 4.59 3.94 6.69 4.11 4.94 4.02Beginning year of the project 2002.43 5.61 2002.63 11.51 2002.5 6.86Duration of collaboration (year) 4.81 4.83 1.87 1.19 4.33 4.58

Third projectThird project reported**,a 56 83.6 11 16.4 67 32.8Collaboration 50 86.2 8 13.8 58 86.6Regional collaboration**,a


29 87.9 4 12.1 33 49.3


16 80.0 4 20.0 20 29.9


6 100.0 0 0 6 9.0


15 83.3 3 16.7 18 26.9

Domestic collaboration 40 87 6 13 46 68.7International collaboration 21 87.5 3 12.5 24 35.8

Table 6.3 (Continued )


predominantly in favour of domestic collaboration, while the academ-ics in the university preferred international collaboration.

Due to practical considerations for the analysis, the examination was limited to a maximum of three projects for each of the respondents. With the exception of six respondents, all mentioned a project. The newly appointed were yet to start their research projects. The average number of collaborative projects (not the total number of projects in their careers), international and domestic (maximum three) seen separately, suggests that domestic has an edge over international by about 0.5 percentage points. Inter-sectoral difference is salient, and it is statistically significant. The average number of international collaborative projects run by aca-demics was more than those by institute scientists. The reverse applied for domestic collaborative projects. For 90 per cent of the scientists and academics, there was a first project to provide the details (Table 6.3).

Eighty-five per cent of the first projects were collaborative but dissimi-lar: regional, national and international in form. Collaboration was ana-lysed in line with the classification followed in other chapters. Within the region—the province of KwaZulu-Natal—as well as within the country, the partnership is evident in the first project. In other words, as far as the first projects of the sampled respondents are concerned, collaboration was quite strong in domestic and international categories. A large major-ity (76%) reported domestic while 46 per cent had international par-ticipation in their first projects. Sectoral features of these collaborations are interesting. Domestic collaboration found greater favour with insti-tute scientists than with academics, whereas international collaboration was preferred by academics who tended to have a more international presence in their area of research. Going beyond this categorization of domestic and international, there are regional, national, continental and international categories. More than 50 per cent of these collaborative



Partners 4.72 4.92 4.67 2.65 4.71 4.61Beginning year of the project 2002.78 4.77 2004.86 1.57 2003.04 4.54Duration of collaboration (year) 4.37 4.33 2.29 1.38 4.11 4.13

Notes: a Chi-square test; b independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.# Maximum possible number of projects is three, as we have asked the respondents to give the details of their first three projects. The figures for domestic and international collaboration would not tally if the projects have partners from the province, within the country and outside the country.


projects (first projects) had participants drawn from the province, 33 per cent from the country, 10 per cent from the continent and 46 per cent from across the national boundaries. As a cryopreservation sci-entist observed, his research knew no boundaries, national or international. He was interested in plants that were used in South Africa and in Kenya for medicinal purposes. And these plants were fast becoming extinct. This was because people were unable to store the seeds that they produced,and when the seeds died, they merely removed them and the plants. He got funding from an international body on plant genetic resources in Italy and obtained material from Kenya, Tanzania and South Africa. His team worked to find out methods of storing the seeds in liquid nitrogen.

The first project of the sample respondents brought together an aver-age of six collaborators to the research. Clearly, the institute sector had an edge over the academic sector in the number of partners the project involved. Though the project had an average of six years since it origi-nated, the duration of collaboration was only four years. This happens when a project takes in partners in the later stages of the project rather than from its inception.

In the case of at least 59 per cent of the sample, second projects were ongoing, more so for academics (95%) than for institute scientists (38%) when the percentage was calculated within sectors. In the case of second projects, 86 per cent were in collaboration, with a difference between sectors. Amongst collaborative respondents (103), domestic collabo-ration applied to 66 per cent and international to 33 per cent. In the second projects as well, international collaboration can be segregated clearly for the two sectors, more for academics than scientists. Further classification of collaboration into regional (50%), national (25%), and international within Africa (7%) and international outside Africa (28%) can also be made for the second projects.

Second projects attracted about five partners, one less than the first projects, with significant difference between the sectors. Institutes, as in the first projects, had more participants than in the academic sector. These projects began about six years before the year of data collection and had four years of collaboration.

Just over 30 per cent reported a third project, more often by academ-ics (55%) than by institute scientists (17%). This is surprising given the professional character of research scientists, who were full-time research-ers and who usually had more than one current project. Collaboration was reportedly more for third projects than for the first two projects irre-spective of the sector. As noted in the previous finding, an increase in the number of projects offered increased the chances for collaboration.


Respondents with collaborative projects had both collaborative and international projects, 69 and 36 per cent, respectively. Again, academ-ics were more internationally oriented to partnerships, while institute scientists looked for more domestic alliances within the region and country. Regional collaboration applied to about 50 per cent of those who reported any collaborative third project, national collaboration was maintained by 33 per cent, international collaboration within Africa held 9 per cent, and collaboration outside the continent was established by 27 per cent. Sectorally, regional collaborators and international collabo-rators were more common among the academics (both within Africa and outside). National collaboration also showed variation. The diversity of collaboration for South African scientists substantiates the position the country held on the scientific map of the continent. South Africa’s lead-ership position in several realms of the science on the continent has been widely acknowledged. In the research of a young geophysicist, this ele-ment of leadership is very straightforward, as he revealed in the interview.

Many of the studies that the respondents carried out had to be done in trying conditions of getting sufficient funding, procuring costly equip-ment and attracting appropriate partners with necessary skills and exper-tise. A geologist revealed that the main funding in the late 1970s and early 1980s was for continental margin research. A great deal of work in terms of conducting cruises and collecting data was accomplished. It is the most expensive research one can conduct.

Collaboration versus non-collaboration

Having examined the research projects and their collaborative facets, the analysis now turns to a classification on the basis of the variables ‘any’ and ‘no’ collaboration. Presumably, it would help understand the distinctive background and professional characters of the respondents in terms of their research and to deduce the factors that facilitate or impede collaboration. The analysis is on the collaborative/non- collaborative dichotomy and in relation to a set of variables that have turned out to be relevant in the data. They include gender, race, age, native place of birth, year moved to South Africa, marital status, highest degree obtained, years spent outside South Africa for higher education, years spent in developed countries, year started working in the organization, current position, fields of specialization, sector last worked, number of research projects, directed projects, collaborative projects, interna-tional projects, domestic collaboration and international collaboration. Table 6.4 presents the characteristic differences between those engaged


Table 6.4 Collaboration versus non-collaboration

Variable CollaboratorsNon-

collaborators Total

N % N % N %

GenderMen 121 69.9 21 67.7 142 69.6Women 52 30.1 10 32.3 62 30.4

RaceWhite 97 56.1 12 11 109 53.4Indian 39 22.5 9 29.0 48 23.5African 32 18.5 9 29.0 41 20.1Coloured 2 1.2 1 3.2 3 1.5Others 3 1.7 0 0 3 1.5

Born in South Africa*,a 116 67.1 25 83.3 141 69.5


Mean age***,b 43.1 10.5 35.8 10.9 42.1 10.8Academic age**,b 11.82 9.6 7.04 8.5 11.1 9.5Institutional experience***,b 10.9 10.7 5.7 7.7 10.1 10.4Year in which the

respondent moved to South Africa***,b

1990.2 14.6 2005.2 3.0 1991.4 14.6

Marital statusMarried 115 63.7 15 48.4 130 63.7Single 46 26.6 15 48.4 61 29.9Divorced 8 4.6 1 3.2 9 4.4Separated 2 1.2 0 0 2 1.0Widowed 1 0.6 0 0 1 0.5

Highest degree***,a

PhD 107 61.8 1 3.3 108 53.2Master’s 40 23.1 14 46.7 54 26.6Bachelor’s 13 7.5 9 30.0 22 10.8Diploma 12 6.9 5 16.7 17 8.4Other 1 0.6 1 3.3 2 1.0


Years spent outside the country for higher education***,b

3.3 5.0 1.0 2.3 2.9 4.7

Years spent in the developed countries***,b

3.5 5.6 0 0 3.0 5.3

Year first worked in the current***,b organization

1996.4 10.6 2001.8 7.7 1997.2 10.4

Current position*,a

Lecturer 47 27.2 12 40.0 59 29.1Junior Researcher/Scientist 3 1.7 0 0 3 1.5


Variable CollaboratorsNon-

collaborators Total

Senior Researcher 36 20.8 12 40.0 48 23.6Senior lecturer 30 17.3 32 15.8 2 6.7Associate professor 13 7.5 0 0 13 6.4Professor 22 12.7 0 0 22 10.8Others—Academics 5 2.9 1 3.3 6 3.0Others—Institute scientists 16 9.2 3 10.0 19 9.4

Discipline***,a

Agriculture 11 6.4 4 12.9 15 7.4Engineering 8 4.6 6 19.4 14 6.9Life sciences 54 31.2 3 9.7 57 27.9Natural sciences 86 49.7 17 54.8 103 50.5Others 14 8.1 1 3.2 15 7.4

Field of specialization**,a

Biology 25 14.5 2 6.5 27 13.2Geology 1 0.6 1 3.2 2 1.0Chemistry 20 11.6 5 16.1 25 12.3Agriculture 11 6.4 4 12.9 15 7.4Zoology 17 9.8 0 0 17 8.3Environmental science 15 8.7 6 19.4 21 10.3Physics 23 13.3 0 0 23 11.3Mathematics 15 8.7 4 12.9 19 9.3Biochemistry 3 1.7 0 0 3 1.5Marine science 11 6.4 1 3.2 12 5.9Engineering 6 3.5 4 12.9 10 4.9

Others 20 11.4 1 3.2 21 10.3Last sector worked**,a

University 84 48.8 7 25.0 91 45.5Research institute 12 7.0 2 7.1 14 7.0Private 16 9.3 1 3.6 17 8.5Government 3 10.7 6 3.5 9 4.5NGO 3 1.7 0 0 3 1.5Other 25 14.5 4 14.3 29 14.5No previous employment 26 15.1 11 39.3 37 18.5


Total number of research projects***,b,#

6.4 8.5 2.9 2.6 5.98 8.1

Total number of projects directed***,b

4.1 6.9 1.4 1.7 3.8 6.6

Mean number of collaborative projects**,b

5.6 8.9 0.6 1.4 5.0 8.6

Mean number of international projects**,b

0.8 0.9 0.03 0.2 0.7 0.9

Notes: a Chi-square test; b independent t-test. # Maximum possible number of projects is three, as we have asked the respondents to give the details of their first three projects. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.


in collaborative and non-collaborative research, applying appropriate statistical tests.

The data can be looked at from two different perspectives: one, the percentage of men and women among the collaborators and non- collaborators; and two, the percentage of men and women within their gender categories. The second one is more precise to understand the divide. Among the collaborators, according to the first scale, there were twice as many men as women. In the second scale of counting men and women within their gender categories, both men and women were frontrunners in collaboration, but with a slight edge for men. Partly, this explains the increased number of men in the sample and the institutions where they come from (Table 6.4). About 84 per cent of the total number of women worked in partnership research but lagged behind men by about only one percentage point (85%). Irrespective of gender, collabora-tion was favoured by academics and scientists in the chosen institutions.

The interracial collaborative pattern agrees with the racial proportion of the sample. Whites were the predominant collaborators, followed by Indians and Africans. Within the racial categories (collaborators out of the total number in each ethnic category) the percentages were 89 per cent for whites, 81 per cent for Indians, and 78 per cent for Africans. On both measures, the pattern was the same while the gap is not that significant in the second measure of collaborators within each ethnic group. Collaborators were older than the non-collaborators by seven years, by five years in academic age, and by five years in insti-tutional experience. The collaborators who were born elsewhere had moved to South Africa 15 years earlier than the non-collaborators who were born elsewhere. Academic and work experience is an important factor in collaboration. The more experienced the respondents, the more collaborative they are. Collaborative researchers moved to South Africa much earlier than non-collaborators. Marriage does not cause any difference in collaboration, for both the married and the single were equally collaborative. The divorced were also inclined to collaboration but not to the same extent as the married or the single respondents. The chances of collaboration are increased if the respondents are experi-enced and have higher educational qualifications.

Two measures explain the amount of time the respondents spend out-side the country for either higher education or for residence. Evident from these two measures is that the collaborators spend more time in the developed countries and outside South Africa than their non-collaborative colleagues. The higher the position in the career, the higher the chances are for collaboration. Most of the senior respondents (senior researcher/


lecturer, all associate professors and most of the professors) were collabo-rators. A good number of lecturers were also collaborators. There is a posi-tive correspondence between the tendency to seek collaboration and the experience and age of the respondents. No particular field was singular in collaboration; all were more or less uniformly distributed. Previously, the respondents worked in a host of sectors, not only in universities but also in the private sector and in NGOs. A majority of collaborators continued to emerge from universities and research institutes (56%) and another ten per cent from the private sector. A quarter of the sample can be safely treated as beginners, for it is their first employment as scientists and aca-demics. More fresh graduates were therefore non-collaborators (40%), and they were possibly yet to start their individual research projects.

Considered together, collaborators had done appreciably well on the number of completed and directed research projects (Table 6.4). The mean values for this were 6.4 and 4.1, respectively, collaborators maintaining a lead in the projects. Collaborators had 3.3 more com-pleted research projects and 2.7 directed research projects than the non-collaborators, which is significantly different in an independent t-test. This validates and confirms the earlier finding that more research projects lead to more collaborative possibilities. Of the maximum three projects, two fell under domestic collaboration and the remaining one was left for international partnerships.

Predicting collaboration

Are background factors such as gender, age, marital status, and PhD sig-nificant in predicting collaboration? This question leads to a subsequent level of analysis of multiple regression models of collaboration on these control variables. The results of the models were however not very sig-nificant save some individual factors (Table 6.5). Three separate models using the number of all collaborative, international and domestic pro-jects as dependent variables were run. In the first model of the number of all collaborative projects, only the academic age of the respondents was significantly associated. Here, the standardized beta coefficient was significant at .05 level. The control variable of age has a negative beta coefficient, although it is not significant at any chosen levels. This sug-gests that the younger the respondent, the higher the number of col-laborative projects. Perhaps those in the early years of their career were enthusiastic and seek opportunities to work in partnership with others who were doing research in similar areas of specialization. This urge can be due to professional reasons such as to build up one’s knowledge base,


Table 6.5 Regression of collaboration on background factors

Professional factorsCollaborative

projects

International collaborative

projects

Domestic collaborative

projects

Age −0.072 −0.045 0.006Academic age 0.312** −0.003 −0.005Gender (1 = male, 0 = others) 0.017 0.010 0.011PhD (1 = PhD, 0 = others) 0.091 0.381*** 0.29***Married (1 = married, 0 = others) 0.016 0.143* −0.015R2 0.088 0.170 0.087N 174 188 188

Note: Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

to establish a standing in the discipline, to get exposure to advanced techniques, to learn new skills and expertise and, above all, the enthusi-asm to be research active. Clearly, neither gender nor marital status had any influence on the collaborative behaviour of respondents, in regard to the dependent variable of the number of collaborative projects. In several other related variables such as scientific productivity and career advancement, there is evidence in the literature that substantiates gen-der differentiation and variations according to marital status, but not in collaboration. Comparison of the mean values (and the t-test), however, showed a difference in the number of collaborative projects and inter-national collaborative projects. Admittedly, this contradicts the find-ings of some of the earlier studies. In their study of university faculty in Norway, Kyvik and Teigen (1996) reported that men collaborated more than women. Long (1990) pointed out that the opportunities for col-laboration for women were decreased by their having children. This, according to Long (1990), is because women scientists were disadvan-taged when their male mentors were reluctant to enter into close work-ing relationships for fear that colleagues or family will assume other aspects to the relationship. Egalitarian roles are necessary for collabora-tion; due to the absence of such egalitarian roles, women scientists gain less from collaborations than men (Long, 1990).

When international and domestic collaboration were run individually for the regression models, the scenario changed; this is shown in the next two models that employ the number of international and domestic projects as dependent variables (Table 6.5). In the second model (num-ber of international collaborative projects as the dependent variable) two factors were significantly associated: PhD and marriage. International


collaboration was related to the possession of PhD and marital status. Married scientists were more internationally collaborative than other marital categories.

In predicting the number of domestic collaborative projects, the third model provided some indicators (Table 6.5). In this, PhD was the only relevant variable. All these three models that regress on background fac-tors have to be treated with caution as they explain only a small percent-age of variance.

What other factors come into play in the collaborative work of scien-tists and academics? As discussed earlier in this chapter, the respondents associate with numerous other professionals such as scientists, engi-neers, and doctoral and master’s students. Using these as independent variables, a set of three regression models that are presented in Table 6.6 were run. The number of all collaborative projects is associated with the total number of research projects, projects directed and the number of doctoral students. Thus, those who had a greater number of research projects and projects to direct and doctoral candidates to supervise were likely to have more collaborative projects. When it comes to interna-tional collaborative projects, the influential variables were a doctorate and the number of doctoral students supervised. Characteristically,

Table 6.6 Regression of collaboration on professional factors

Professional factorsCollaborative

projects

International collaborative

projects

Domestic collaborative

projects

Sector (1 = academics, 0 = others) −0.041 0.059 0.083Total number of research projects 0.206** 0.006 0.028Total number of projects directed 0.184* −0.062 0.065Doctorate degree (1 = doctorate,

0 = others)−0.106 0.255** 0.131

Number of doctoral students supervised

0.503*** 0.252* 0.069

Number of master’s students supervised

−0.026 −0.068 0.234*

Number of professional scientists work closely with

−0.026 −0.138 −0.100

Number of doctoral students work closely with

0.053 −0.023 −0.090

R2 0.449 0.187 0.183N 157 159 159

Note: Sig: *p < 0.1; **p < 0.05; ***p < 0.01.


active domestic collaborators were scientists who supervised more mas-ter’s students. This is in agreement with the previous findings.

Close collaboration with doctoral and master’s students came up con-sistently in the conversation with informants. Having tested this vari-able in the regression models, its significance in collaborative research in the scientific system of South Africa became evident. This should be read with the current emphasis the Department of Higher Education and Training (DHET) is giving to the production of postgraduates and PhDs in the country. Most of the universities in the country are try-ing hard to attract postgraduates and PhDs by waiving student fees and offering bursaries.

The factors that determine the collaborative approach of scientists and academics can be looked at from different angles. Partnerships are encouraged and sustained by incentives which act as built-in support systems (Bozeman and Boardman, 2003a; Hafernik et al., 1997). These incentives are relative and vary from individual to individual (Landry et al., 1996). Some would have more incentive to collaborate than others. In research productivity, the academics in the South African uni-versity system are on a par with—or even more productive than—their counterparts in research institutes. In the prevailing system, research is as integral as teaching in universities in the country. As discussed in chapter 2, research is further supported, both financially and profession-ally, by incentive structures. The value attached to research outcomes energises academics to be productive, and collaboration often comes as a means to be vigorously productive. As in the provision of an effective incentive system to promote research productivity, structures are the key players in collaboration. If the structures are prohibitive and intimidat-ing, research and collaboration are likely to suffer. Studies (Genuth et al., 2000, for instance) had reported the importance of structural conditions in multi-institutional collaborations. A curious scientist who is enthu-siastic about research will always look forward to chances that enhance his/her own research productivity, as is clear from one of our chemical scientists.

Conclusion

Scientific research in South Africa, as seen through the cross-section in the sample, continues to hold on to its traits inherited from the colonial period. The prominent one in these traits is its accent on the collabora-tive component, bringing in partners that cut across geographical and disciplinary boundaries. Still carried through are those relations with


some select countries that it had in the colonial and apartheid past. This happens not only in collaborative ventures but also in its own scientific system that attracts scientists from overseas to work for the country in its teaching and research institutions. Currently the system is strong with young, qualified and active researchers, producing a new genera-tion of researchers. The structures that are in place are supportive of research activities, inspiring scientists to tread new paths and areas of specialization. In collaboration, both universities and research institutes apparently had a cultivating attitude and environment. The respond-ents had reported no initiatives to collaborative enterprises that had met with institutional hurdles. This is a promising environment wherein sci-entists are free to explore such possibilities without hitting the barri-cades of institutional and bureaucratic impediments. Research-intensive respondents were collaborative-intensive too. Scientists and academics that had more than one project were much more collaborative than those who had a single project. As they move up their professional lad-ders they had more than a single research project to work with, which gradually improved the likelihood of collaboration. Research institutes had emerged as the centres of domestic collaboration, while academic institutions fostered international partnerships. The data permit predic-tion of the degree of collaboration of scientists on the basis of certain known factors. Exposure to science in other countries by way of spend-ing time there for higher education or living there is more influential than one’s gender, age and marital status. Along with these, there are factors—the number of research projects, having a doctoral degree, the number of professionals and students supervised and worked with—that can be safely included in predicting collaboration of scholars in South Africa.

Having examined the research projects and the collaborative aspects of the respondents, the conditions that promote or obstruct collabo-ration need to be investigated. Collaboration is believed to enhance productivity, which is conditioned by a surrounding environment of communication and networks. How collaboration is linked to research productivity, the use of information and communication technolo-gies, the Internet and email in particular, and professional networks are discussed in the next chapter.

159

The publication of findings is the ultimate outcome and purpose of sci-entific research. As the chief end product of a scientist’s work (Price, 1965a), publication carries greater meaning and value in the professional life of scientists. Publication is widely used as a measure of research pro-ductivity (Eslami et al., 2013, for instance). The publication productivity of scientists seems to be sustained by collaboration as seen in the previ-ous chapters. Reprising the connection between productivity and col-laboration, Price very succinctly notes that ‘the most prolific man is also by far the most collaborating’ (1986: 126). The most productive authors collaborate frequently and are inclined to collaborate with highly pro-ductive authors (Katz and Martin, 1997).

The general assumption is that collaboration increases research pro-ductivity, but very few studies provide empirical evidence to support this proposition (Bozeman and Boardman, 2003a). The correlation between productivity and the degree of collaboration has been established by Beaver and Rosen (1978) and Katz and Martin (1997) in their studies, to cite but two examples. Bonaccorsi et al. (2006) reported a trade-off between academic publications and industry-oriented research and suggested that collaboration with industry leads to an initial increase in productivity. Bordens et al. (1996) found among researchers in bio-medical research (gastroenterology, cardiovascular system, and neuro-sciences) a positive correlation between the total production of authors and their international, national and local production. The analysis of Abramo et al. (2009) showed positive correlation between the degree of international collaboration and the scientific production of universities.

There is a sectoral pattern in productivity as well. In a study of Canadian universities Landry and Amara (1998) noted that the research-ers brought in more publications when they collaborated within

7Communication, Professional Networks and Productivity


research institutes than within research teams. Collaboration between researchers and industry has more impact on productivity than collabo-ration between researchers and their research peers in other institutions (Landry et al., 1996). Glänzel and de Lange (2002) suggested a strong relationship between the number of international links of countries and the number of full and fractional counts of publications. A trade-off between writing articles and engaging in collaborative research and development (R&D) is observed, and publication increases as the col-laboration network gets progressively larger (Barjak, 2006). Scientists in some European countries with larger collaboration networks have writ-ten relatively more journal articles (Barjak, 2006). In some African coun-tries, the count of publication is associated with the increasing level of international collaboration (Narváez-Berthelemot et al., 2002). They found that there was a 54 per cent increase in publications of papers co-authored with scientists from overseas.

Productivity is uneven due to the nature of disciplines and the geo-graphical locations of authors. Book publication is lower for experimen-tal physicists than social scientists, while multi-authored papers are the highest among experimental physicists (Roe, 1972). In the productivity of collaborative partners, variation is effected by geographical proximity or fields of research (Landry et al., 1996). Barjak (2006) gathered a host of factors from the literature that determine the productivity of scien-tists: research motivation, stamina, creativity, age, gender, rank and pro-fessional recognition, burden of other obligations such as teaching and administration, communication with colleagues, participation in col-laborations, training environment, size of the research group, prestige and research focus of the institution, organizational freedom, liberty to select the content of the research, scientific discipline, and the country of the scientists.

Productivity is also a function of the time that is effectively spent (or can be spent) on research. In a study of South African scholars, Jacobs and Ingwersen (2000) identified the length of time in teaching and doing research as a decisive constituent in their productivity. Senior scientists and professors, as opposed to their junior colleagues, are able to devote a larger share of their time to research that eventually trans-lates into conference presentations, workshop participations and pub-lications. Nevertheless, this is contingent upon the roles and functions attached to positions and how they are expected to be performed by the incumbents. A respondent of our study revealed that he had experi-enced a perceptible change in his productivity when he was relieved of some of his non-research functions at a university. While serving as the

Communication, Professional Networks and Productivity 161

administrative head of his division for a long period, his productivity was affected and fell off significantly. After he relinquished his headship his publication rate has picked up again.

Rank and productivity are allies. A marked difference in the productiv-ity of professors, associate professors and senior lecturers is discernible in South Africa. Professors and associate professors preferably publish in inter-national journals, rather than in domestic journals, because of their higher rank, prestige, recognition and the deliberate choice to publish in interna-tional outlets (Jacobs and Ingwersen, 2000). Publication, therefore, does not invariably follow a curvilinear fashion. Sometimes it is stagnant, and at times it flows like a stream. A researcher in inorganic chemistry said that she did not publish anything for one or two years because of the nature of her research, which took years to complete.

The productivity of scientists is swayed by certain external conditions too, for instance, access to and use of electronic communication and the Internet. Evidence of the impact of Internet tools on productivity is common. A stable relationship between Internet use and productivity has been reported (Barjak, 2006). In a study covering a large sample1 of scientists across European countries, Barjak (2006) notes that scientists who communicate more in general and via email produce more regard-less of the form of publication, namely, working papers, journal arti-cles, book chapters, monographs, conference presentations or reports. Higher productivity is thus associated with overall communication in general and email communication in particular.

Although publication is an indicator of research, research productiv-ity and publication productivity are not strictly identical (Fox, 1992). Publication of journal articles, co-authored publications in particular, is an acceptable quantitative measure of collaboration (Merton, (1938)[1970]). Some have used publications and the development of patented and unpatented products in the measurement of the productivity index (Landry et al., 1996). Although productivity can be quantita-tively measured by the number of publications, productivity really encompasses a range of variables such as ability, professional recogni-tion, endurance, communication and associative relationship (Kundra and Kretschmer, 1999). Productivity can also be measured by con-verting the publications into article equivalents and ascribing points to them (Kyvik, 1990a). In scientific circles, the count of citations is also accepted as an index to the success of productivity (Sonnenwald, 2007). A key issue in the measurement of publication productivity is the use of ‘normal’ or ‘fractional’ counts. Normal count of papers gives every author one credit; straight count assigns all the credit to first


author only; and fractional or adjusted count gives credit equal to 1/n to each of the n co-authors (Gupta and Karisiddippa, 1999). All pub-lications are weighed equally in the normal count, regardless of the number of co-authors. In fractional count, the number of publications is divided by the number of co-authors to correct for the partial con-tributions implied by the division of labour in multi-authored papers. Using curriculum vitae data on publications, Lee and Bozeman (2005) show that correlation between normal and fractional count productiv-ity is extremely high, and association between normal count produc-tivity and collaboration is even stronger than between fractional count productivity and collaboration.

To measure productivity, a series of questions were put to the respond-ents of this study. Included among these were questions about their pro-ductivity over the last five years (papers at state or national workshops, international conferences, reports whether published or not, articles in foreign journals, articles in domestic journals and chapters in books). In addition, they were asked about the count of research papers written over the past 12 months (this measure did not distinguish between single and co-authored papers). Productivity in this chapter is, therefore, dis-cussed in terms of the number of publications in these listed categories. They are further condensed into domestic productivity of publications in domestic peer-reviewed journals; foreign productivity of publications in foreign international peer-reviewed journals; and other productivity that covers books based on original research, chapters in edited volumes, published reports of studies and papers given at conferences and work-shops. Domestic journals, which are edited in the country, can also be international journals. For this reason, a classification of ‘domestic’ and ‘foreign’ is more meaningful than a division of ‘national’ and ‘interna-tional’ journals/productivity. These productivity variables are analysed first for any sectoral differences—academic or research institutes—and then for any collaborative and non-collaborative variance. The analy-sis is also meant to see how productivity varies across respondents on account of their collaboration such as domestic, international, both or none. This is followed by the examination of the predictors of produc-tivity using a set of background, professional and collaboration factors.

The Internet use covered both the email and Web use of the respond-ents, which is in agreement with the measures used in the literature (Duque et al., 2005; Sooryamoorthy et al., 2007; Sooryamoorthy and Shrum, 2007). Email usage in this analysis is operationalized using measures such as the average time spent on using emails in a normal week, the number of emails related to research, the number of emails


received and sent, and the time spent in using the Web. The Internet use is related to research (searching for information, accessing data, col-laboration via Web, collecting reference material, downloading software and publishing papers).

In this examination of the interrelationships between productivity and collaboration, the former is the dependent variable and the latter an independent variable. Two more variables, namely, communication and professional contacts, which are frequently discussed in collabora-tion studies, are brought into this analysis.

Productivity and collaboration

The productivity of South African scientists can be compared to the pre-dominant standards of the developed world. South African scientists in the sample produced an average of 3.32 papers a year (Table 7.1). This can be compared to the findings of another similar study conducted among scientists and academics from the same province. During the previous five-year period, as reported in the previous study (Sooryamoorthy and Shrum, 2007) referring to 2004–05, the respondents wrote 7.03 papers, presented 4.35 papers at national workshops and 3.63 papers at interna-tional conferences, wrote 5 research reports and published 0.85 chapters for edited volumes. As for the mean number of articles in domestic (edited within the country) and foreign (edited outside the country) journals, the respondents published a higher number in foreign journals than in domes-tic journals—5.36 and 2.08 papers respectively—in the last five years. The respondents, on average, also edited 0.2 books and wrote 0.08 original books, with a significant difference between academics and scientists.

Productivity within sectors features characteristic patterns (Table 7.1). A number of variables in this analysis emerged with significant differ-ences in the independent t-test. Academics presented more papers at international conferences than scientists in research institutes; the for-mer also published more in peer-reviewed journals, both domestic and foreign. In foreign-originated journals the presence of the academics was very conspicuous, with an increase of 5.84 papers over the respondents in research institutes. In contrast, research institutes produced more research reports than the university sector. In the measure of the total productivity that combines the number of papers in both foreign and domestic journals, chapters, edited books and books published in the previous five years, the difference between sectors was evident.

Co-publication is an outcome of collaborative alliance. With the purpose of ascertaining the extent of co-publications, co-authored papers (in both


Table 7.1 Productivity by sector

Productivity#

Academics Scientists Total


No. of papers written in the last 5 years

9.02 27.27 2.70 4.99 7.03 22.91

No. of papers published in the previous year*,a

3.65 3.55 2.58 3.52 3.32 3.57

No. of papers presented at national workshops in the last 5 years*,a

4.94 6.63 2.98 4.55 4.35 6.13

No. of papers at international conferences in the last 5 years

3.76 5.63 3.33 13.28 3.63 8.62

No. of reports in the last 5 years***,a 2.49 3.86 11.05 14.49 5.23 9.63No. of chapters written for edited

volumes in the last 5 years1.02 2.80 0.48 1.27 0.85 2.44

No. of papers published in foreign journals in the last 5 years**,a

7.10 18.65 1.26 2.66 5.36 15.90

No. of papers published in domestic journals in the last 5 years

2.37 8.71 1.46 3.22 2.08 7.42

Edited books*,a 0.21 0.49 0.19 0.52 0.20 0.50Books*,a 0.16 0.52 0.02 0.14 0.08 0.35Total publication productivity*,a 9.82 29.82 3.11 5.19 7.68 24.94

Co-publicationsCo-authored papers in foreign

journals***,a

6.90 18.6 1.04 2.23 5.16 15.86

Co-authored papers in domestic journals**,a

1.73 3.38 0.84 1.54 1.45 2.96

Co-authored books**,a 0.11 0.41 0.02 0.14 0.08 0.35Co-authored papers (foreign and

domestic combined) **,a

8.27 19.78 1.88 3.12 6.25 16.69

Total co-publication productivity (articles and books) **,a

7.87 20.22 1.6 2.47 5.85 16.95

Notes: # represents the mean value for all variables in the column. a independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

domestic and foreign journals), books and total co-publication productiv-ity were gathered. A large majority of the papers scholars had published in foreign journals during the previous five-year period were co-authored papers. The mean number of papers was 5.36 against 5.16 for co-authored publications. This was not repeated in the same way in the number of papers in domestic journals (2.08 and 1.45 respectively). Scholars in


Table 7.2 Productivity and collaboration

Productivity#

CollaboratorsNon-

collaborators Total


No. of papers written in the last 5 years***,a

8.24 24.69 0.30 0.61 7.03 22.91

No. of papers written in the previous year***,a

3.79 3.67 0.72 0.84 3.32 3.57

No. of papers presented at national workshops in the last 5 years***,a

4.97 6.41 0.64 0.91 4.35 6.13

No. of papers at international conferences in the last 5 years***,a

4.22 9.19 0.18 0.48 3.63 8.62

No. of reports in the last 5 years***,a 5.94 10.25 1.29 2.48 5.23 9.63No. of chapters written for edited

volumes in the last 5 years**,a

1.00 2.63 0.04 0.19 0.85 2.44

No. of papers published in foreign journals in the last 5 years***,a

6.21 17.01 0.22 0.58 5.36 15.90

No. of papers published in domestic journals in the last 5 years***,a

2.44 8.01 0.07 0.27 2.08 7.42

Edited books***,a 0.24 0.53 0 0 0.20 0.50Books**,a 0.13 0.48 0 0 0.08 0.35Total publication productivity***,a 9.04 26.95 0.35 0.63 7.68 24.94

Co-publicationsCo-authored papers in foreign

journals***,a

5.99 17.0 0.22 0.58 5.16 15.86

Co-authored papers in domestic journals**,a

1.70 3.14 0.04 0.19 1.45 2.96

Co-authored books***,a 0.09 0.38 0 0 0.08 0.35Co-authored papers (foreign and

domestic combined)***,a

7.33 17.93 0.26 0.59 6.25 16.69

Total co-publication productivity (articles and books)***,a

6.90 18.28 0.27 0.60 5.85 16.95

Notes: # represents the mean value for all variables in the column. a independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01 (significance between yes and no in each variable; if the mean is higher for yes, only the significance of difference is shown).

general preferred to write co-authored books than sole-authored ones. Between sectors, scientists wrote more co-authored books than academics.

Some meaningful comparison is seen in Table 7.2 with respect to col-laboration and productivity. All the productivity measures had recorded


differences between those who collaborated and those who did not. In contrast to non-collaborative researchers, collaborative researchers had written more papers in the previous year, presented more papers at national and international conferences, produced more reports and published more chapter sections and peer-reviewed articles in both domestic and foreign journals during the reference period. They wrote more books and had more edited books to their credit.

The data on the extent of co-authored publications is also seen in Table 7.2. Distinctly, collaborative scientists had significantly higher number of co-authored papers in foreign and national journals and books and in co-publication productivity. The difference between the number of papers published and the number of co-authored papers is clear. For example, out of the average of 5.36 papers published by the respondents in foreign journals in the previous five years, 5.16 were co-authored. In the case of papers in domestic journals the figures were 2.08 and 1.45 respectively. More collaborative papers appeared in foreign journals than in domestic journals. This applied to the produc-tion of books as well, wherein all the books were co-authored.

In Table 7.3, productivity is compartmentalized for domestic, inter-national, both or no collaboration. Productivity variables were grouped under four collaboration categories of ‘no collaboration’ (people who have not yet done any collaborative research project), ‘domestic collabo-ration’ (those who have research projects that involve partners from the province or in the country), ‘international collaboration’ (people having research partners from across the borders of South Africa), and ‘domestic and international collaboration combined’ (those having both types of collaboration).

The mean number of papers written in the previous year of data collec-tion was low for ‘no collaborators’ when compared with ‘domestic and international collaborators’ (Table 7.3). This discrepancy disappeared but showed a wide variation in the case of scientists who run both domes-tic and international collaborative projects. Collaborators—domestic, international and both—presented a relatively larger number of papers at national and international conferences than the non-collaborators. The same is true of the number of reports produced. Non-collaborators had published less chapter sections than the domestic collaborators and even fewer chapters when compared with other types of collaborative researchers (international and both). Importantly, collaborators had a higher productivity when it came to papers in domestic and foreign journals. It is evident in the number of papers in foreign journals that relate to the categories of international collaboration and combined

Communication, Professional Networks and Productivity 167Ta

ble

7.3

Prod

uct

ivit

y an

d t

ype

of c

olla

bora

tion

Pro

du

ctiv

ity

No

co

llab

ora

tio

nD

om

esti

c co

llab

ora

tio

nIn

tern

atio

nal

co

llab

ora

tio

n

Do

mes

tic

and

in

tern

atio

nal

co

llab

ora

tio

n

Mea

nS.

D.

Mea

nS.

D.

Mea

nS.

D.

Mea

nS.

D.

No.

of

pap

ers

wri

tten

in

th

e la

st 5

yea

rs0.

300.

618.

5126

.812

.53

33.7

8.51

25.2

6N

o. o

f p

aper

s w

ritt

en i

n t

he

pre

viou

s ye

ar0.

720.

843.

813.

824.

543.

943.

893.

71N

o. o

f p

aper

s p

rese

nte

d a

t n

atio

nal

w

orks

hop

s in

th

e la

st 5

yea

rs0.

640.

915.

386.

905.

625.

925.

066.

52

No.

of

pap

ers

at i

nte

rnat

ion

al

con

fere

nce

s in

th

e la

st 5

yea

rs0.

180.

486.

4018

.45

5.07

5.88

4.26

9.28

No.

of

rep

orts

in

th

e la

st 5

yea

rs1.

292.

486.

6310

.98

5.18

10.7

46.

1110

.44

No.

of

chap

ters

wri

tten

for

ed

ited

vo

lum

es i

n t

he

last

5 y

ears

0.04

0.19

1.10

2.84

1.57

3.53

1.05

2.68

No.

of

pap

ers

pu

blis

hed

in

for

eign

jo

urn

als

in t

he

last

5 y

ears

0.22

0.58

6.4

18.4

59.

3322

.77

6.43

17.3

5

No.

of

pap

ers

pu

blis

hed

in

dom

esti

c jo

urn

als

in t

he

last

5 y

ears

0.07

0.27

2.42

8.33

3.32

10.8

32.

488.

18

Edit

ed b

ooks

00

0.27

5.49

0.30

0.57

0.25

0.54

Boo

ks0

00.

120.

440.

130.

480.

130.

48To

tal

pu

blic

atio

n p

rod

uct

ivit

y0.

350.

639.

3429

.55

14.1

637

.69

9.37

27.6

1

Co-

publ

icat

ions

Co-

auth

ored

pap

ers

in f

orei

gn j

ourn

als

0.22

0.58

6.23

18.4

98.

9322

.65

6.15

17.3

3C

o-au

thor

ed p

aper

s in

dom

esti

c jo

urn

als

0.04

0.19

1.61

2.56

2.08

3.68

1.69

3.19

Co-

auth

ored

boo

ks0

00.

110.

410.

090.

330.

100.

39C

o-au

thor

ed p

aper

s (f

orei

gn a

nd

d

omes

tic

com

bin

ed)

0.26

0.59

7.55

19.2

610

.91

23.8

97.

4818

.29

Tota

l co

-pu

blic

atio

n p

rod

uct

ivit

y

(art

icle

s an

d b

ooks

)

0.27

0.60

7.03

19.7

910

.63

25.1

77.

0518

.69


collaboration; their figures were considerably above the average of the full sample. The trend is repeated in national journals as well.

Productivity is inspired by factors that are both personal and profes-sional. The literature is abundant with antecedents such as gender, age and marital status. Kyvik and Teigen (1996) used the variables of aca-demic rank, age, research finding, research collaboration, international contact and child care as the determinants of productivity. Women pub-lish less than men in the same positions but publish more than men in lower positions; female professors publish more than male associate pro-fessors, and female associate professors publish more than male assistant professors (Kyvik, 1990a). Mauleón Bordons (2006) have reported the inter-gender differences in productivity. Reviewing productivity studies, Kyvik (1990b) concluded that productivity expands with increasing age and reaches a peak when scientists are in their late thirties and early forties and then declines; scientists who are most productive at a young age are also the most productive when they grow older; and there are large differences between disciplines in the relationship between age and productivity. Two theoretical propositions are of importance here. One, the utility maximizing theory proposes that productivity declines with increasing age because the expected utility of keeping publication activity on a high level decreases. Two, the obsolescence theory says pro-ductivity declines because older researchers have problems coping with scientific developments and become obsolete (Kyvik, 1990b). Gonzalez-Brambila and Veloso (2007) did not find age to be a determining factor in the productivity of Mexican researchers. Productivity is influenced by the number of children the researchers, particularly women researchers, have (Hunter and Leahey, 2010).

Along with these pertinent variables, other relevant variables such as the year in which the scientists migrated to the country and the time they spent outside the country for higher education and stay are checked. These two, in accord with the insights gained in chapter 3, have relevance in South Africa. In the regression models three measures of published papers in domestic journals, in foreign journals and an additive measure of these two were used to calculate the total count of papers in domestic and foreign journals. Papers in peer-reviewed jour-nals are valued higher than other outputs such as research reports and chapters for compiled volumes. In the first model (Table 7.4) of regress-ing productivity in domestic journals on background factors, only insti-tutional experience is related. Two new background control variables in these models, which are not discussed in literature, invite explanation. Academic age is the number of years the scientists have completed after


obtaining their highest degree. More often than not, scientists become productive after they successfully complete their higher degrees like the PhD, which is their primary preoccupation and concern while they are students. Scientists usually start publishing in peer-reviewed journals once they have fulfilled their career requirements such as completing a doctorate. Institutional familiarity is another variable that impacts on productivity in a more or less a similar way. It is measured by the time (in years) scientists have been with the same organization, and it is indi-rectly related to the organizational context of the institution.

Evidence (Long, 1978; Long and McGinnis, 1981) supports that organizational context, such as the prestige of the organization, affects the productivity of the people working there. Departmental prestige is reportedly related to scientific productivity (Cole and Cole, 1973), while prestigious departments select more productive scientists as their fac-ulty (Long, 1978). Academic age in the model was negatively associated with domestic productivity. Gender and academic age were clear deter-minants in productivity, as shown in the publication of peer-reviewed papers in foreign journals (model 2). What is seen in the second model is that men scientists and academics with a higher institutional experi-ence produced more publications in journals that were edited outside South Africa. Academic age was negatively correlated with publica-tion in foreign journals. In the third model that tests the regression of total productivity in peer-reviewed journals, gender and academic age were significantly correlated. Men recorded a higher productivity in

Table 7.4 Regression of productivity on background factors

Background Factors

Papers in domestic journals

Papers in foreign journals

Papers in domestic and

foreign journals

1 2 3

Gender (1 = male, 0 = others) 0.116 0.213*** 0.196**Academic age −0.071 0.410*** 0.330***Married (1 = married, 0 = others) 0.003 0.049 0.023Institutional experience 0.214* −0.133 −0.052Any degree from developed

countries (1 = yes, 0 = others)0.040 0.029 0.107

R2 0.049 0.199 0.179N 145 154 142

Notes: The dependent variables are log converted ones. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.


foreign-originated journals and in the total count of domestic and for-eign journals. Gender did not make any detectable change in domestic productivity, that is, in the count of publications in domestic journals.

Productivity is also induced by certain structural factors. Collaboration impels productivity within a specific but characteristic institutional context. In particular, the academic context, as opposed to the research sector, promotes higher productivity of the incumbents. This is further accentuated by collaboration opportunities and initiatives. This finding partly corroborates that of Duque et al. (2005) whose data refer to some African countries like Ghana and Kenya.

Inferred from these models is that the four factors that relate to the exposure of scientists to science in other countries—the time spent outside the country for higher education, degree from developed coun-tries, time spent in developed countries or the year they moved to South Africa—do not have any special effect on increasing productivity. Rather, productivity remains invariant, irrespective of exposure. Unlike the reported findings in some other studies discussed earlier, age is not a predictor of productivity in any of the regression models. Given the R2 values of the three models, these control variables do not seem to be strong predictors of domestic productivity of South African scientists.

To test the hypothesis regarding the relation between collaboration and productivity, another set of three regression models were run (Table 7.5). Collaboration variables in the analysis included the number of domestic collaborative projects, the number of international collaborative projects, and the number of total collaborative projects. In view of their perti-nence, as had emerged in chapter 4, the number of professional scien-tists and engineers and the number of doctoral students with whom the respondents worked closely with were also included. Along with these other relevant variables—sector, doctoral degree, academic age and insti-tutional experience—were incorporated.

In domestic productivity (model 1, Table 7.5), academic age, insti-tutional experience, domestic collaborative projects and total number of collaborative projects were related. Put differently, the higher the institutional experience of the scientists and the total number of col-laborative projects, the greater their domestic productivity. Academic age, as earlier, was negatively correlated. Similarly, the more domestic collaborative projects the scientists had, the more was their productivity in domestic journals. Publication in foreign journals was determined, as seen in the second regression model, by doctoral degree, academic age and total number of collaborative projects. Those who had a doctorate, a higher academic age and more collaborative research projects possessed


Table 7.5 Regression of productivity on collaboration and professional factors

Professional factors

Papers in domestic journals

Papers in foreign journals

Papers in domestic

and foreign journals

1 2 3

Sector (1 = research institute, 0 = others) 0.023 0.084 0.090Doctoral degree (1 = doctorate, 0 = others) 0.123 0.36***5 0.378***Academic age −0.199* 0.19**9 0.094Institutional experience 0.193* −0.074 −0.035No. of professional scientists and

engineers work closely with0.003 0.071 0.004

No. of doctoral students work closely with

0.073 0.082 0.092

No. of domestic collaborative projects 0.155* 0.055 0.132**No. of international collaborative

projects0.091 0.105 0.209***

Total no. of collaborative projects 0.293*** 0.197*** 0.210***R2 0.220 0.424 0.514N 152 158 152

Notes: The dependent variables are log converted ones. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

a higher level of productivity in foreign journals. Academic age, unlike in the domestic productivity, was a significant factor in the productivity of respondents in foreign-based peer-reviewed journals.

How can we predict the total output of the respondents? Institutional experience and domestic, international and total number of collabo-rative projects were the predictors in the total count of publication productivity in model 3. The more the experience in working in the present organization, the more the total productivity. Working in the same organization for a while has a settling effect in the environment in terms of practising one’s profession and doing research. In another way, it is related to the institutional support that scientists receive for their research and publication endeavours. Being familiar with the insti-tutional environment while in the institution helps scientists to make effective use of the available structures for research and publication output. Institutional familiarity therefore corresponds to the institu-tional culture that fosters research, collaborative or non-collaborative. Institutional familiarity also relates to the number of years one has been with the organization for reasons of permanency or tenure. In research universities promotion to higher ranks is another variable that


is positively associated with productivity (Long et al., 1993; Long and McGinnis, 1981). Productive scientists are promoted early to keep them from changing institutions, while less productive scientists may wait for or be denied promotion (Long et al., 1993). Related to this is recogni-tion in the profession and career-oriented motivation that contribute to production (Barjak, 2006).

Emphasized also in the third model (Table 7.5) is the account of inter-national collaboration in increasing productivity. Productivity (both domestic and foreign) was significantly associated with collaboration for the full sample and for the academic sector in particular, at varying levels of statistical significance. Collaboration according to the model was related to neither domestic nor foreign productivity of scientists in research institutes. Of relevance here is the model of Landry and Amara (1998) that predicted that researchers controlling the greater publication assets are more likely to choose to work in formal insti-tutional structures. In other words, collaborators bring in more pub-lication assets when they collaborate within research institutes than within the research teams and outside formal structures. On the con-trary, there are lower publication assets when collaboration takes place outside formal structures than within research teams (Sooryamoorthy and Shrum, 2007).

Increasing the chances of predictability, all these three models exhib-ited a relatively moderate variance level (22–51%) when compared to the former models that regressed productivity on background factors. Collaboration was associated with productivity, which reaffirms the find-ings of studies that have found the same kind of relationships (Defazio et al., 2009; Duque et al., 2005; Landry et al., 1996; Lee and Bozeman, 2005; Narváez-Berthelemot et al., 2002; Pao, 1982; Price and Beaver, 1966; Smalheiser et al., 2005; Sooryamoorthy et al., 2007; Sooryamoorthy and Shrum, 2007; Tijssen, 2007; Zuckerman, 1967). Unfailingly, total num-ber of collaborative projects had shown its consistent presence in these three models that predicted productivity of domestic, foreign and com-bined collaborations. The association between domestic collaboration and productivity—domestic and total publication output—in the two models suggests that domestic collaboration causes a change in produc-tivity (domestic and combined) or acts as a powerful factor in increasing the productivity of scientists.

The increase in productivity, apart from the factors that are influen-tial, is contingent upon institutional character such as the prevalence of a reward system. A reward system, as Cole and Cole (1967) emphasize, encourages creative scientists to be highly productive. The increase in


productivity through the thirties, and the decrease in productivity after the age of 50 are attributed to the reward system (Cole, 1979 cited in Kyvik, 1990b) through which scientists seek recognition by publishing. The cumulative advantage theory suggests that the average productiv-ity of a cohort declines with age because those who are not rewarded for their research lose their inspiration for new achievements (Kyvik, 1990b). This was confirmed when a geologist revealed his views on the rewards linked to publication in the institution where he was work-ing. At the time of data collection, the sampled institution was pay-ing ZAR18,000 per publication in approved journals, which was enough to attend one international conference. ‘If you publish more,’ as the respondent said, ‘you will have enough money to attend more interna-tional conferences that would lead to more research publications.’

To validate the views of this geologist we have two concrete cases—one a university and another a research institute—that can be cited as a testimony to an outcome of a similar reward system in South Africa. The Human Sciences Research Council (HSRC) attracted scientists offer-ing higher salaries than the existing university scales, and its productiv-ity jumped from 0.18 peer-reviewed publications per researcher in 1997 to 0.8 publications in 2004. In the case of the University of KwaZulu-Natal, a new reward system for research publications resulted in a cred-ible increase of its productivity from 448 units in 2001 to 582 in 2004 (Habib and Morrow, 2006). The incentives for the production of new knowledge are also found in the intrinsic motivation to advance knowl-edge or exchange of new knowledge for some recognition (Gläser, 2003).

Communication and collaboration

As discussed in the previous chapters, there are constraints in the con-tinuing activity of collaboration between partners scattered far and wide and who are engrossed in serious scientific endeavours. In endur-ing collaborative research that invariably demands years of committed, unremitting and technically skilled work of the collaborators, there are several intervening and influential variables that come into play. Communication on a regular, task-oriented and sustained basis is a sali-ent one among those decisive variables.

Scientific collaboration is viewed as an intellectual pursuit depending on effective access to and sharing of digital research data and the use of information tools (Olson et al., 2007). The impact of information and communication technologies (ICTs) on science has been paramount to the development community (Duque et al., 2005); ICTs provide the


elixir that could free Third-World science from its isolation and integrate it into the global scientific community (Davidson et al., 2002). A major implication of this technology is that it can open up possibilities for collaboration beyond the borders of nations and increase productivity (Lee and Bozeman, 2005). ICTs, the Internet in particular, have helped increase collaborative scientific research between countries and insti-tutions (UNDP, 2001). The Internet reigns supreme over other means of communication in centres of knowledge—research institutions and universities—where research is conducted. The power of the Internet lies in its ability to support coordination functions of collaboration by linking collaborators for real-time interaction (Teasley and Wolinsky, 2001). The Internet, by virtue of its ability to transfer information rapidly, certainly has the potential to enhance collaboration (Teasley and Wolinsky, 2001) and can give rise to new types of collaboration when scientists are not co-located (Sonnenwald, 2007). An unswerving increase in co-authored papers, a product of collaboration, in various scientific disciplines across distant locations has been made possible through this medium (Cronin et al., 2003; Moody, 2004; Wagner and Leydesdorff, 2005a). As to the costs, with the improvements in connec-tivity a sharp decline has occurred in the cost of collaboration between partners who are geographically dispersed (Adams et al., 2005). Adams et al. (2005) believed that the decline in the cost of collaboration in the US was due to the deployment of the National Science Foundation’s NSFNET in 1987 and its connection to networks in Europe and Japan. In one of the case studies conducted in the US, Bozeman and Boardman (2003a) showed how the key to the effectiveness of collaboration lay in the extensive and intensive nature of communication among project partners. At the same time, infrequent communication between collabo-rators can disconnect goals and tasks (Olson et al., 2007).

Reviewing studies on Internet use, Zhao (2006) summarized three major but conflicting views: that Internet use decreases social ties; increases social ties; and neither decreases nor increases social ties. Duque et al. (2005) examined this connection between ICTs and sci-entific collaboration in a novel study of developing countries, namely, Kenya, Ghana and the state of Kerala in India, and concluded that Kenyans were ahead of Ghanaians and Keralites in external collabora-tion. The vast majority of Kenyans (86%, and 1.71 collaborative projects on average) in the sample were engaged in collaboration as against 75 per cent of Ghanaians (1.37 collaborative projects) and 39 per cent of Keralites (0.64 collaborative projects) (Duque et al., 2005). Various studies (Kerr and Hiltz, 1982; Walsh and Bayma, 1996a, 1996b; Walsh et al., 2000)


have also produced empirical evidence to support the relation between email use and collaboration.

The growth of scientific capacity may lead to the improved ability of connectivity and can, in turn, build collaborations (Wagner and Leydesdorff, 2005b). The availability of and access to ICTs do not always stimulate national and international collaboration as in the case of sci-entists in Kerala (India), where low levels of collaboration have been identified (Sooryamoorthy et al., 2007). The adoption, application, and use of ICTs vary widely with disciplines. In some parts of the world, uni-versities appear to provide promising settings for new ICTs but have not exploited them to the extent that other organizations have made use of them (Komsky, 1991). Similarly, the capacity of the Internet in carry-ing huge quantities of information all the time depends largely on the periods (peak and lean periods) of its use. A limited relationship between Internet use and research productivity was noted by Vasileiadoua and Vliegenthart (2009). Gläser (2003) demonstrated that the Internet leaves the social order of scientific communities unchanged but affects the mode of production of some scientific communities. The point Gläser (2003) makes here is that though the Internet makes scientists’ life easier and their work faster, it does not change the content, procedures and importance of communication.

Reporting from Brazil on researchers and graduate students, Pachi et al. (2012) noted a relationship that existed between connectivity and scientific production. Their study suggested that the increase in the number of academic theses in the country was connected to the band-width that was available to them and that free access to the Internet produced an increase in their productivity. In another study from the Philippines, Ynalvez and Shrum (2011) showed that the diversity of email use related to research had a positive effect on the produc-tion of papers published in foreign journals. Butler and Butler (2011) explained that reducing communication costs can lead to collabora-tion activities of scientists and thereby their co-authorships. According to them international collaboration has a strong causal impact on the research output of scientists but the research output does not cause any impact on the future international collaboration of scientists. Kaminer and Braunstein (1998), studying the effect of Internet use on the pro-ductivity of natural scientists in a US university, found the positive effect of Internet use on productivity. Walsh et al. (2000), on the other hand, emphasized that the email use helps scientists to integrate their network, which is associated with increased collaboration and higher productivity.


Though studies that investigate the connection between the Internet, collaboration and productivity are not rare, only a few studies speak about this connection in Africa and in South Africa. Ehikhamenor’s (2003) analysis of the productivity of scientists in Nigerian universi-ties suggested that connectivity does not correlate with productivity. A similar study of South African scientists by Sooryamoorthy and Shrum (2007) concluded that there was a relationship between productivity and the Internet and email usage. This study showed that though the usage of Internet and email was positively associated with collaboration, collaboration was not associated with productivity (Sooryamoorthy and Shrum 2007). It has been confirmed in another study that an increase in the number of international collaborative research projects is associated with an increase in the publication output of South African scientists (Sooryamoorthy, 2013a).

Presuming that the centres of advanced learning, research and knowl-edge production are equipped with the latest technologies essential for faster and reliable communication, one question remains to be addressed. How far these technologies are utilized by the incumbents in these institutions for their research and advanced learning is the pertinent question. This question is important given that broadband Internet access is a reality in most of the scientific and higher educa-tion institutions in South Africa. The use of Internet technologies and its magnitude, therefore, becomes a critical component in collabora-tive research activities. One has to bear in mind the need to distinguish between local and distant collaborations as far as their communication needs, channels and usage are concerned.

Now, let us consider the African situation. African countries remain on the periphery of Internet connectivity (Sonaike, 2004). African institutions and those in the developed world are not comparable in access, capacity (bandwidth) and cost. Figures suggest that, on aver-age, an African university pays 50–100 times more than their European and North American counterparts for the same bandwidth (Jensen et al., 2007). Cross Atlantic fibre for research and education networks can be obtained for $1/mbps/month while African institutions pay over $5,000/mbps/month or more for the same bandwidth (Jensen et al., 2007). Circumventing this limitation, a few countries in Africa, includ-ing South Africa, Egypt and Morocco, have formed the National Research and Education Network (NREN) but are still faced with technical and policy barriers (Jensen et al., 2007). This being the case, South Africa lags behind much of the developed world in implementing high-speed ser-vices such as ADSL (Asymmetric Digital Subscriber Line) (Parker, 2007).


Only 3.5 per cent of the total Internet users in the world are in Africa (Internet World Statistics, 2007). In 2004, Internet access in South Africa per 1,000 people was 8.4, and the telephone density per 1,000 people was 270 (Department of Science and Technology, 2002). In 2007, 10.3 per cent of the population (5,100,000 of 49,660,502 people) in South Africa were Internet users (Internet World Statistics, 2007) and this is expected to grow by about 3 per cent to reach 800,000 broadband subscribers in South Africa. Not to be overlooked is the fact that Internet connections in South Africa are usually ‘capped,’ that is, the amount of data received and sent is monitored and billed for, limiting the use of the Internet (Parker, 2007) and discouraging new subscribers. Nevertheless, the position of the country in its Internet status compares well with that of most European countries (Sonaike, 2004). As of June 2012, according to the Internet World Statistics (2012), South Africa ranked only fifth in Africa among the number of Internet users. The percentage of Internet penetration in South Africa is only 17.4 per cent compared to 28.4 per cent for Nigeria, 35.6 per cent for Egypt, 51 per cent for Morocco, and 28 per cent for Kenya.

The following section presents the scale of usage of the Internet and its supportive or obstructive function in initiating, establishing and implementing collaboration undertakings by South African scientists. The focus is on the scale of usage of the Internet. For this purpose, email use, Web use and collaboration are the variables, and collaboration is a dependent variable. But it is difficult to establish causality because both directions of causations are possible (Gläser, 2003).

Email use for professional and collaborative research is diverse. Some of the respondents are heavy users and value the medium greatly for their research. For some others, email is seldom the ideal form to meet their communication objectives, requiring them to substitute with other modes such as the telephone and fax. Due to some of the inherent limi-tations of email—not sure of reaching the destination or disappearance into cyberspace without bouncing back—researchers do not completely trust it as an entirely reliable mode of communication. A seed physiolo-gist shared the view that most of his communication is via email though it is not the ideal one. The varying levels of access to connectivity of partners affect the speed of communication.

Instances in which emails are used along with other slower media of communication are not uncommon amongst scientists. There is often a choice of using the medium for its properties of speed and convenience and for the exchange of scientific information wherein time is a pre-eminent factor. If the users from both ends do not exploit the medium


to its full capacity, communication cannot be expected to be effective or to be used to its optimum.

Email is invaluable for communication with distant colleagues and professional partners, and is supplemented and followed by face- to-face interaction at meetings and conferences. A geographer keeps her communication channels open for her international partners with email while not missing any opportunity to meet them in person at conferences and professional meetings. A life scientist who works with national and international partners in his nature conservation project does the same.

Email not only connects people in distant locations but also those in close proximity. In their daily communication with colleagues next door, respondents combine emails with face-to-face interactions.

The analysis presented here has a number of variables that measured the email and Web use of the survey respondents. First, these measures were examined on the basis of their sectoral affiliation—academic and research institute—and then on their collaborative propensities.

Table 7.6 presents the data of email and Web use. Email was adopted widely by the respondents, and all had access to it thanks to institutional policies and support. Over 50 per cent of them spent one to five hours in a typical week, while another 32 per cent used it for more than five hours. For about 50 per cent of the full sample, more than two of their emails a day were directly related to their research. More scientists than academics were in this category (more than two emails a day). When calculated for the proportion for each type of usage, significant sectoral difference was visible in the number of hours academics and scientists allocated for their email and Web usage. In the independent t-test of this ordinal variable, significant difference between sectors was evident. Academics had a higher average of sent/received emails. Though scien-tists spent more time using the Web the difference was not significant from academics.

In the case of the number of emails that are related to research activi-ties, no significant association existed in the Chi-square test. Sixty-one per cent of the scientists in research institutes and 44 per cent of aca-demics had more than two emails for their research. In contrast, nine per cent of scientists and seven per cent of academics had less than one email a week for their research. This means scientists in research institutes had more emails that were used for their research than the academics. Not surprisingly, this explains the nature of work, exclusively research, the scientists do in their research institutes. For academics, research is only one of many responsibilities such as teaching, administration and


Table 7.6 Email and Web use by sector

Email and Web use

Academic sector

Research institute Total

N % N % N %

No. of hours a week on email***,a,!

Less than 1 hour 18 12.8 4 6.6 22 10.9Between 1 and 5 hours 68 48.2 44 72.1 112 55.4Between 5 and 10 hours 33 23.4 9 14.8 42 20.8Between 10 and 20 hours 14 9.9 2 3.3 16 7.9Over 20 hours 8 5.7 0 0 8 4.0

No. of emails related to research#

Less than 1 a week 9 6.5 5 8.5 14 7.1Between 1 and 6 in a week 43 31.2 9 15.3 52 26.4Usually 1 or 2 daily 24 17.4 9 15.3 33 16.8More than 2 daily 61 44.2 36 61.0 97 49.2

No. of hours a week using Web**,a,! Less than 1 hour 12 8.6 8 13.8 20 10.1Between 1 and 5 hours 78 55.7 29 50.0 107 54.0Between 5 and 10 hours 30 21.4 5 8.6 35 17.7Between 10 and 20 hours 12 8.6 12 20.7 24 12.1Over 20 hours 8 5.7 4 6.9 12 6.1

Ever submitted or reviewed manuscript via email***,a

113 80.1 30 49.2 143 70.8

Acquired or used data from Web 122 86.5 56 91.8 178 88.1Conducted information search 133 94.3 55 90.2 188 93.1Used an electronic journal**,a 136 96.5 54 88.5 190 94.1Collaborated on a scientific project

via Internet**,a

96 68.6 32 53.3 128 64.0

Found reference material on Web 127 90.7 52 85.2 179 89.1Accessed research reports from

Internet136 96.5 56 91.8 192 95.0

Hours spent a week in sending and receiving email#,b

2.48 (1.03) 2.08 (0.69) 2.36 (0.95)

Hours spent a week in using Web#,b 2.47 (0.97) 2.57 (1.17) 2.5 (1.03)

Notes: a Chi-square test; b independent t-test; S.D. in parentheses. Sig: *p < 0.1; **p < 0.05; ***p < 0.01; # Ordinal variable, 0—none, 1—low, 2—medium, 3—high; ! Ordinal variable, 1—first, 5—fifth.

community engagement. In the sampled institution, the 45 per cent of the academics’ time was used for teaching, 40 per cent for research, 10 per cent for community engagement and the remaining 5 per cent for administration. This distinction between academics and scientists


recurred in the usage of the Web as well, where the percentage of higher users (10–20%, and 20 plus hours) was among scientists.

Academics and scientists have had a head start in the use of these communication technologies. Some of the science departments in the academic sample were the pioneers in bringing the technology to their doorstep. For instance, the biology department of the academic sample was one of the first departments to get the email facility.

The exact time required for receiving and sending email cannot be calculated accurately in most cases. This partly depends on the right time in terms of the speed of the browser, giving us a glimpse of the ICT situation in the institution. Some were able to give us a fair idea about the number of emails they usually handled. One academic respondent counted that he received about 5–6 work-related emails a day and spent about 45 minutes writing and reading them. Another respondent used to send emails in the region of 25–50 and received almost double this number a day. This young researcher subscribes to a few newsletters that are in his branch of science.

Email/Web tops the list in the ranking of the means of communica-tion respondents used in their work. In the order of importance were email/Web, face-to-face contact, telephone, fax and postal mail. In two of these—postal mail and telephone use—academics and scientists were not alike. Scientists in research institutes relied on the postal mail more than the academics at the time to data collection. Compared to scien-tists, academics preferred the telephone (Table 7.6).

Turning to the actual usage, about two-thirds (71%) of the respond-ents used email for purposes such as submission and review of research papers. More than one-third of the respondents (36%) used the Web for more than five hours a week (the range here goes beyond 20 hours). Eighteen per cent of them can be termed as intensive users for using it more than ten hours a week. More or less the same proportion of aca-demics and scientists was in the category of those who spent more than five hours a week with the Internet. On the other side of the use were 64 per cent of scientists and 65 per cent of academics who took less than five hours a week for their Web needs.

The nature of Web usage was measured using indicators such as its use for acquiring data, conducting information search, accessing electronic journals, collaborating on scientific research projects, finding reference material and downloading research reports. In all these purpose-oriented uses, the majority were active, except in collaborative projects through the Web, which favoured only 64 per cent. Inter-sectoral contrast (academics and scientists) prevailed only in two variables—use of


electronic journal and collaboration on a scientific project via the Internet. In the first variable, more academics in higher institutions than scientists in research institutes used electronic journals. In the second instance, more academics than scientists collaborated via the Internet for their research projects.

Studies indicate a mixed relationship between the Internet, produc-tivity and collaboration. Komsky’s (1991) study investigated the differ-ence between frequent and occasional email users in matters including the preference for telephony and email, perceptions of the use of email at work and the degree to which email is used for oneself or to assist someone to send and receive. Using the data presented in Table 7.7, the relevance of these theoretical propositions and empirical findings are reviewed in this study, across both collaborators and non-collaborators.

First, let us consider email use. When grouped into two classes of ‘below’ and ‘above’ five hours of email use a week, collaborators dominated non-collaborative researchers in the higher-user class of above five hours a week (Table 7.7). Thirty-six per cent of the collaborators spent over five hours a week; for non-collaborators the figure was 16 per cent. About 14 per cent of the collaborators as against 3 per cent of non- collaborators had reported that they spent more than ten hours per week. Again, in the lowest use category of ‘less than one hour’, there were fewer collabora-tors (9%) than non-collaborators (19%). When the number of emails that were related to research was considered, there was a significant differ-ence between collaborators and non-collaborators. Within this category, 52 per cent of the collaborators sent or received more than two emails that were related to their research a week, as against 35 per cent for non-collaborators. Collaborators not only make more use of the computer for work purposes but also clock more average time receiving and sending emails. The same is the pattern in the submission and review of papers via email. More collaborators than non-collaborators made use of email for these purposes. Collaborators, in contrast to non-collaborators, had more years of experience in using email and the Web.

Sending and receiving papers by email is one of the major uses that the respondents find functional and practical in their research endeavours. The facility to send file attachments is considered as one of the principal uses of this medium. The papers they sent to journals, colleagues and peer reviewers were not always text documents but contained diagrams, figures, charts, images and pictures. A marine geophysicist appreciates this property of the email. She found the Web useful to download infor-mation including maps and diagrams that were relevant to the project she was engaged in at that time. Another geographer holds the same


Table 7.7 Email, Web use and collaboration

Email and Web use

CollaboratorsNon-

collaborators Total

N % N % N %

No. of hours a week on email !

Less than 1 hour 16 9.4 6 19.4 22 10.9Between 1 and 5 hours 93 54.4 19 61.3 112 55.4Between 5 and 10 hours 38 22.2 4 12.9 42 20.8Between 10 and 20 hours 15 8.8 1 3.2 16 7.9Over 20 hours 8 4.7 0 0 8 4.0

No. of emails related to research #,*,a

Less than 1 a week 11 6.5 3 10.3 14 7.1Between 1 and 6 in a week 43 25.6 9 31.0 52 26.4Usually 1 or 2 daily 27 16.1 6 20.7 33 16.8More than 2 daily 87 51.8 10 34.5 97 49.2

No. of hours a week using Web !

Less than 1 hour 17 10.1 3 10.7 20 10.1Between 1 and 5 hours 90 52.9 17 60.7 107 54.0Between 5 and 10 hours 31 18.2 4 14.3 35 17.7Between 10 and 20 hours 21 12.4 3 10.7 24 12.1Over 20 hours 11 6.5 1 3.6 12 6.1

Ever submitted or reviewed manuscript via email***,a

135 78.9 8 25.8 143 70.8

Acquired or used data from Web

149 87.1 29 93.5 178 88.1

Conducted information search 161 94.2 27 87.1 188 93.1Used an electronic journal 163 95.3 27 87.1 190 94.1Collaborated on a scientific

project via Internet***,a

123 72.4 5 16.7 128 64.0

Found reference material on Web*,a

155 90.6 24 80.0 179 89.1

Accessed research reports from Internet

162 94.7 30 96.8 192 95.0

Hours spent a week in sending and receiving email***,b

2.43 (0.96) 1.94 (0.77) 2.36 (0.95)

Hours spent a week in using Web

2.52 (1.05) 2.36 (0.95) 2.5 (1.03)

Emails first used year**,b 1994.8 (4.6) 1998.8 (4.9) 1995.4 (4.9)Web first used year***,b 1996.4 (4.3) 1998.3 (4.3) 1996.8 (4.4)

Notes: a Chi-square test; b independent t-test; S.D. in parenthesis. Sig: *p < 0.1; **p < 0.05; ***p < 0.01; # ordinal variable, 0—none, 1—low, 2—medium, 3—high; ! Ordinal variable, 1—first, 5—fifth. The column for total has not been used as in the previous chapters as the mean for the total would also include those without any projects.


view, not wanting to wait and waste her time to receive a paper which she has to work on. Here one gets an idea about the medium that the scientists prefer in their professional activity. Because of the speed with which submitted papers are processed for publication, the users are able to save time. The change in the broadband capacity has also contributed to this.

Apprehensions about electronic communication among users cannot be ruled out. In the transfer of files, sketches and images might be lost or get distorted in transmission. Transferring such files hits the permis-sible space available for email in the institution where the respondent is working. Added to this is the inadequate speed of the server that slows down the transfer of attachments. Scientists were also concerned about the possible incompatibility of their software programmes with those of their partners and the lack of adequate capacity for emailing files.

In some cases, email becomes very intrusive and this worries scien-tists. For them, this is not a pleasant thing despite the advantages email has to offer. Most of them do know how to tackle those emails. As a scientist reported, ‘The email has made life a lot easier in many respects but it has made it more difficult in others, because now people send an email and expect something to be done immediately.’

In regard to the preference of the means of communication, there is no great variation between those who collaborate and those who do not. Email takes up the first position in the list, followed by face-to-face contact, telephone, fax and finally the postal mail (Table 7.8). Preference does not change whether one is keen on collaborative research or not, but more to do with access to ICTs. In some cases this is a choice of con-venience. ‘Writing a letter, putting it in a brown envelope and waiting for days to get a reply is inconvenient,’ as one scientist said. Therefore, the preference of collaborators and non-collaborators is not likely to vary remarkably. The advantages of email communication over other means were quite clear for many. Let us examine this in detail in the next section.

In collaboration that draws on international partners, email com-munication is often the right option for scientists. International phone calls, which come after email in preference, are not always a viable alter-native due to their prohibitive cost in South Africa. Email has made it so much easier to be in contact with people, especially internationally. When scientists cannot afford international phone calls, and often face time zone problems, email comes handy. Email, as one scientist noted, ‘lets people to communicate internationally so much more easily than, say, a fax or telex or telephone call used to’.


Table 7.8 Professional contacts and sector

Size of networks#,a

Academic sector

Research institute Total


Total network** 4.06 3.03 3.3 1.92 3.83 2.75Network within KwaZulu-Natal* 1.41 2.0 1.89 1.49 1.56 1.87Network within South Africa*** 1.33 1.65 0.65 0.90 1.12 1.49Network outside South Africa but

within Africa0.19 0.71 0.08 0.33 0.16 0.62

International network*** 1.27 1.86 0.75 1.12 1.11 1.68Diversity of relationship locations 1.60 0.98 1.62 0.91 1.61 0.95Domestic network (province and

country)2.74 2.52 2.54 1.57 2.68 2.27

Email network** 3.70 2.88 2.92 2.0 3.46 2.66Face-to-face network** 2.18 2.47 1.56 1.70 1.99 2.28Telephone network 1.87 2.42 2.0 1.69 1.91 2.22Fax network* 0.35 1.28 0.06 0.30 0.26 1.08Postal mail network** 0.26 1.03 0.05 0.22 0.19 0.87

Notes: # represents the mean value for all variables in the column. a independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

For collaborative researchers, this is a ‘wonderful’ (to borrow one informant’s adjective) way of communication as they work on schedules and cannot afford delay when decisions are to be made. When partner-ship extends abroad, as is usually the case, respondents time their com-munication in such a way that brooks no delay while experiments are running or awaiting action.

Web use among collaborative and non-collaborative researchers did not present any deviating pattern. Both used it in a more or less simi-lar way. There were of course differences in the way collaborators and non-collaborators used the Web (Table 7.7). This is quite conspicuous in collaborative projects conducted online (a few of the non- collaborators had also previously collaborated via the Internet), where the Web was used for information search and for finding reference material. In all these variables, collaborators were way ahead of the others. Collaboration over the Internet, as in the case of a structural geologist, is ‘triggered when scientists chance upon others working in the same area of specialization’.

Scientists were quite aware of the quality of the information they down-loaded from websites. They confirmed the worth of the information they


obtained from the Web. For reasons of authenticity, they were cautious in using the information that is not refereed. A researcher in chemistry said that the information on the Web was indispensable to his research on natural production of chemical elements and also in the way it is going to affect his publications. An informant was explicit about this. She thought that if she did not have the Internet she would be able to access no more than half to two-thirds of the publications that she cur-rently could. In her view, the Internet made her a better researcher as it allowed her to access more stuff.

The query on the effects of the Internet on one’s career returned an assortment of opinions, ranging from one extreme to the other. For some, the use of the Internet has been central and ‘part of the career’; they ‘could not do without it.’ South African scientists who had to spend a considerable part of their time in libraries looking for refer-ences now find it easier to locate them by browsing the web pages of the libraries of their institutions. This aside, the immediate use of the Internet when it arrived in the country, was in collaborative research and in establishing international contacts. As a zoologist studying mammals remarked, when the World Wide Web first hit this country, in 1993–94, his first use of it was that of ‘making available informa-tion on his research group’. He compiled a list of web pages detailing his team’s research activity to develop a fairly good source of contact, mostly international contacts.

Since the advent of the Internet service for science journals, there has been an incredible expansion in their use by researchers in the coun-try. By accessing online journals and electronic versions of journals and downloading databases, the research time of scientists has been effectively enhanced. In order to economize the use, searches are well directed and narrowed down to specific sites and themes, avoiding lengthy periods of time on the Internet. For some, there is so much of information to filter through.

Evidently, email and the Internet are the prominent media of commu-nication that advance the research interests of scientists in collaborative research. One interesting enquiry at this stage is to see how the means and modes of communication culminate in the formation of contacts of a professional character. Professional contacts are indispensable for scientists to keep abreast of the developments in their own fields of interest. From the angle of collaboration too professional contacts have an entrenched position. The next section thus examines how such con-tacts affect communication and research.


Professional contacts and communication

Science is organized through networks (Etzkowitz and Leydesdorff, 2000). While research is becoming more interdisciplinary it is increas-ingly conducted in domestic and international networks (Hicks and Katz, 1996). Networks between individuals and institutions help scien-tists in many ways: getting access to laboratories, participating in con-ferences, publishing in journals (Okubo and Sjöberg, 2000), exchanging knowledge and skills, and updating themselves on developments in the discipline and on methodological breakthroughs, to name some. The professional network of scientists is therefore an integral part of the sci-entific and technical human capital of scientists (Bozeman and Corley, 2004). Growth in international collaboration is interpreted as intensi-fication of research networks transcending national boundaries (Kim, 2006) while dense networks make scientific collaboration easier (Porac et. al., 2004). Without networked communications, international col-laboration would be highly limited (Wagner and Leydesdorff, 2005b).

Callon (1994) believed that scientific activities produce heterogene-ous networks. Networks and scientific activity operate both ways; net-works result in scientific activities and vice versa. One nurtures the other. Collaboration has effects on the ability to develop network ties and contacts that are critical to the careers of scientists (Bozeman and Corley, 2004). A greater network increases the likelihood of future col-laborations, forging new research alliances and relationships (Goldfinch et al., 2003). From one set of network ties, collaborative research extends further by building up new sets of networks, bringing in the potential opportunities for more collaborative ventures. Networks, more often than not, multiply in geometrical progression.

Agrawal and Goldfarb (2005) estimated that the Internet connectivity in connected universities increased collaboration by 85 per cent (cited from Butler and Butler, 2011). The Internet, by way of reducing com-munication costs, could help transform the collaboration activities of scientists, women in particular, and could improve their co-authorships (Butler and Butler, 2011). International collaboration is positively related to future research output more so than to domestic collaboration. This study is interesting as the authors examine reverse causality between international collaboration and research output.

In the Chilean academic and research system, as Duque et al. (2012) argued, the Internet was instrumental in promoting collaboration. Among women academics in political science, as the study by Butler and Butler (2011) demonstrated, the use of the Internet has reduced the


cost of communication and helped establish collaborative opportuni-ties that have culminated in co-authored publications. The advantage of Internet communication has made it possible for women academics to increase their co-publication activities at a faster rate than their male counterparts in the field (Butler and Butler, 2011). There was not only an increase in the productivity of women academics, but also the Internet and its implied lower costs of communication enabled women to take up positions in smaller departments where there were fewer women. The Internet and the allied communication channels, therefore, served facil-itators in collaboration, affecting the outcome of co-productivity (Butler and Butler, 2011). Among Canadian university scholars, as Landry et al. (1996) observed, scientific collaboration had a contributing effect on the productivity of academics but the effect of collaboration on productivity varied according to the geographical closeness and the areas of research of the partners.

Concerning the effect of Internet use on the productivity of natural scientists in a US university, a study by Kaminer and Braunstein (1998) in the early years of the Internet pointed out that Internet use had a positive effect on productivity. Email use by scientists is helping them to integrate their professional network, and it is associated with increased collaboration and higher productivity (Walsh et al., 2000). Although communication via the medium of computer is not the sole and pri-mary reason for collaboration, it nevertheless assists collaboration, par-ticularly between partners in remote locations (Haythornthwaite, 2005; Walsh et al., 2000). Collaboration is also strongly associated with the email measures they have used in this study, as does the amount of co-authorship and email use. It is further clear from the study that the productivity of the respondents—as measured in terms of the number of referred papers published in the previous two years of the study—is positively associated with the measures of email use.

Ehikhamenor (2003) held a different view on the connection between Internet use and productivity. Based on a survey of scientists in Nigerian universities, who were mostly at an early stage in their careers, the study found that those who use the Internet did not think their productivity was facilitated by Internet use and that the association between Internet use and productivity was only coincidental.

The media of communication are powerful players in the crea-tion of professional networks. Muliplexed social interaction is accom-plished through a variety of media that are available for interaction (Haythornthwaite, 2001). Technologically oriented individuals do not confine themselves to a single technology but adopt a cluster of


technologies that enrich their communication repertoire. The emer-gence of a new medium may, as Dimmick et al. (2000) argued, lead to the exclusion, replacement or displacement of the old media, and the patterns of interpersonal use of media depend on geographical location (Baym et al., 2004). A recent analysis shows that ICTs are not associated with the relational structure but are significantly associated with the locational structure of social networks (Sooryamoorthy et al., 2007). For most local relationships, telephones are used, while emails are preferred for long-distance relationships (Chen et al., 2002). As modern means of communication, the Internet and email are capable of fashioning net-works with colleagues at work, both local and global. To form meaning-ful new relationships and to extend the existing social networks, the Internet is used (Baym et al., 2004; Thurlow and McKay, 2003). In a pro-fession like science and scientific research, professional networks lead to the movement of scientists from one location to another or become the basis of future collaborative research. Email makes those contacts faster and easier. A researcher in chemistry shared how he got his present job using his professional contacts maintained via email.

In this study, professional contacts (locational) and means of contact were measured by asking the respondents first to list all the people with whom they had professional contacts for seeking advice or consulting for professional reasons. They should be individuals who have similar work interests to those of the respondents. Such networks are to be located in the province, country, region (Africa) or outside the country, but not within the department or institution. From these variables their domestic and international networks were calculated. All the major means of communication—face-to-face, telephone, fax, postal mail and email—were taken into account. The same dichotomy of academics/ scientists and collaborators/non-collaborators is followed in the analysis.

The network size tended to be significantly larger for academics than for scientists (Table 7.8). This is a combined measure encompassing all the networks in the province and country as well as internationally. In network locations, academics differed from scientists only in the average size of the networks they had with individuals outside Africa and in domestic locations such as within the province and the coun-try. Academics reportedly maintained larger contacts with professionals within the country, within the continent, and outside the continent, but for the scientists in research institutes, domestic networks were more important within the province of KwaZulu-Natal. In domestic networks that comprise networks both within the province and within the coun-try, the academics were ahead of the scientists with networks double the


size of that of scientists. Scientists were more provincially connected, than nationally or internationally. The size of the network outside the country was significantly higher for academics (1.27) than for scientists in the research institutes (0.75). Both sectors were alike when it came to the diversity of locations of professional networks. Diversity covers networks in locations such as the province, the country, within the con-tinent but outside the country, and outside the country.

The means of communication employed to maintain these profes-sional contacts varied in all modes of email, face-to-face, telephone, fax and postal mail. Academics relied more on networks through all these media except telephone networks than scientists. The difference was quite significant in the t-test. Scientists used telephone more frequently than the academics to establish and maintain their contacts. Although email and telephones were available in all the institutions where the study was conducted, respondents, to a certain extent, used fax and postal mail for their professional communications. Scientists were least interested in using fax and postal mail for their professional contacts.

In some of the measures, the dichotomy of collaborators and non-collaborators of professional contacts is obvious. Collaborators reported a larger network (total) size than non-collaborators (Table 7.9). In the same manner, they had larger networks in the country and outside the country than their non-collaborative colleagues. Within the province

Table 7.9 Professional contacts and collaboration

Size and type of networks#,a

Collaborators Non-collaborators

Mean S.D. Mean S.D.

Total network*** 4.03 2.87 2.71 1.62Network within KwaZulu-Natal 1.51 1.93 1.81 1.49Network within South Africa*** 1.24 1.55 0.42 0.81Network outside South Africa but

within Africa0.14 0.64 0.23 0.50

Diversity of relationship locations 1.61 0.94 1.58 1.03Domestic network 2.76 2.37 2.23 1.56International network*** 1.22 1.78 0.48 0.77Email network*** 3.71 2.74 2.10 1.58Face-to-face network** 2.09 2.37 1.39 1.54Telephone network*** 2.06 2.32 1.10 1.22Fax network* 0.29 1.16 0.10 0.20Postal mail network** 0.22 0.94 0.03 0.18

Notes: # represents the mean value for all variables in the column; a Independent t-test. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.


of KwaZulu-Natal, the difference was insignificant. In international net-works, the mean size was higher for the collaborators by about three times that of the non-collaborators. Diversity of locations that showed where the professional relationships were based is similar for both col-laborators and non-collaborators.

The email network of collaborators hints that they make greater use of emails than non-collaborators to contact their professional associ-ates scattered in other locations. Email topped the list of all the means of communication used, with the largest size of networks. The least used means for both collaborators and non-collaborators were fax and postal mail, whose relevance as media of communication is declining. In the case of all other means of communication, the difference was significant between collaborators and non-collaborators. Collaborators built and maintained more networks, as clear from the size, than non- collaborators through email, telephone, fax, postal mail and face-to-face contact. Collaborations in other words necessitate more frequent and regular contacts.

Collaborators form, create and nurture their contacts in several forms. Some contacts begin at a conference or on the Internet. These contacts might give rise to meaningful research enterprises. One young scientist engaged in forestry studies met people working in his area and it led to an international network, the Plant Tissue Culture Network. This online network binds its members together and serves as a vehicle for exchange of ideas, clarification of queries and finding of solutions to scientific problems. The contacts that were revealed in the interviews were far reaching, across nations and continents. A physiologist’s networks, as reported in the interview, were predominantly national but were care-fully cultivated with regular contacts. For a senior physical chemist, the net was cast widely across the world from Africa to Europe and to North America with his collaborating partners.

Respondents had extensive networks and linkages all over the world, particularly for those running collaborative projects with others. The contacts they had were not restricted to a country or a region, or just around a single collaborative project. Many of them had more than one project at the time of the interviews and each one brought in a new set of network ties.

Communication, collaboration and productivity

Returning to productivity, let us now consider the relationship it has with communication. While examining the connection between electronic


communication, collaboration and productivity, it is also important to see the directions they follow: whether email leads to collaboration or vice versa or whether email use promotes productivity.

Not all respondents were sure whether the Internet and electronic communication had made any beneficial impact on their productiv-ity, that is, whether their productivity was a function of their Internet use. To some respondents, the use of the Internet had widened their knowledge base as they could find the details and specific information that they needed in their research. Widened knowledge, they agree, was manifest in the production of research papers, as was also their quality.

The relationship between collaboration and higher productivity has been reaffirmed in numerous studies (Abramo et al., 2011; Butler and Butler, 2011; Duque et al., 2005, 2012; Katz and Martin, 1997; Landry et al.,1996; Lee and Bozeman, 2005; Mouton, 2000; Ponomariova and Boardman, 2010). Abramo et al. (2009) sought to find out whether external collaboration intensity is correlated with research performance in the Italian university system. Adopting a bibliometric approach, they recognized that international collaboration had produced a significant effect on the publication quality of the academics with sectoral differ-entiation but had no similar effect on domestic collaboration. In a sub-sequent study, they corroborated the strong link between the intensity of international collaboration and the productivity of Italian university researchers in hard science. The increase in the number of publications has a corresponding increase in the number of cross-national publica-tions as well (Abramo et al., 2011).

In Table 7.10, the productivity of the respondents across sectors is given. This variable was measured taking into account the normal pro-fessional activities academics and scientists are expected to do. They participate and present papers at workshops/conferences, write reports, publish papers and books. The data in this table refer to the period of five years prior to the year of data collection. In all the measures of this publication productivity, clear statistical difference was found between academics and scientists. Except in the production of research reports, academics scored better than their counterparts in research institutes. In relation to scientists, academics presented significantly more papers at national workshops, published more papers in foreign journals, edited and wrote more books. When productivity was measured in terms of combined publication productivity of papers in journals and total pub-lication productivity, academics perform well.

Further to this, the co-publication productivity of respondents was examined. This gives an idea about the collaboration they engage in


Table 7.10 Productivity

Productivity

Academics Scientists All


Papers at national workshops* 4.94 6.63 2.98 4.55 4.35 6.13Papers at international

conferences3.76 5.63 3.33 13.28 3.63 8.62

Reports*** 2.49 3.86 11.05 14.49 5.23 9.63Papers in foreign journals** 7.10 18.65 1.26 2.66 5.36 15.90Papers in national journals 2.37 8.71 1.46 3.22 2.08 7.42Chapters in books 1.02 2.80 0.48 1.27 0.85 2.44Edited books* 0.21 0.49 0.19 0.52 0.20 0.50Books* 0.16 0.52 0.02 0.14 0.08 0.35Productivity of papers

(national and foreign)*9.02 27.27 2.70 4.99 7.03 22.91

Total publication productivity* 9.82 29.82 3.11 5.19 7.68 24.94

Notes: Total productivity includes papers in foreign and national journals, chapters, books and edited books. Sig: *p < 0.1; **p < 0.05; ***p < 0.01.

and the outcome of these alliances. Five variables had been used to measure this co-productivity of the scholars over the past five years (Table 7.11). Obviously, as the data show, academics were the leaders in co-authoring. They produced more co-authored papers in both national and foreign journals and co-authored more books. They had a clear edge over the researchers in research institutes in the productivity of co-authored papers and books combined. In some of the measures, the difference was quite big. If co-productivity is considered an indicator of collaboration, then the researchers in institutes cannot be treated as active collaborators.

What kind of relationship could be expected between professional networks, productivity and collaboration? One way to grapple with the interactive behaviour of these variables is to run a regression analysis. The results are given in Table 7.12. The total professional network size is the dependent variable in the first model that employed seven predictors such as sector, total number (log) of papers in domestic and foreign jour-nals, total productivity, total co-productivity and number of collabora-tive, domestic and international projects. This model (R2 = 0.173) showed significant relationships with two factors: number of international col-laborative projects and number of domestic collaborative projects. The size of the total professional network can be predicted by the number of domestic and international collaborative projects the respondents have.


Table 7.11 Co-publication productivity

Co-productivity

Academics Scientists All


Co-authored papers in foreign journals***

6.90 18.60 1.04 2.23 5.16 15.86

Co-authored papers in national journals**

1.73 3.38 0.84 1.54 1.45 2.96

Co-authored books** 0.11 0.41 0.02 0.14 0.08 0.35Co-productivity of papers

(national and foreign)***8.27 19.78 1.88 3.12 6.25 16.69

Combined co-productivity of papers and books)***

7.87 20.22 1.60 2.47 5.85 16.95

Note: Sig: **p < 0.05; ***p < 0.01.

Table 7.12 Regression of network structure on collaboration, productivity and email network

Collaboration and productivity variables

Total network

size

Domestic network

size

Interna-tional

network size

1 2 3

Sector (1 = academics, 0 = others) −0.095 −0.086 −0.047No. of papers in domestic and foreign

journals −0.200 −0.108 0

Total productivity 0.125 0.070 −0.036Total co-productivity 0.134 −0.044 0.202No. of collaborative projects −0.107 −0.093 −0.054No. of international collaborative projects 0.306*** −0.021 0.452***No. of domestic collaborative projects 0.245*** 0.327*** 0.010R2 0.173 0.110 0.271N 159 159 159

Note: Sig: ***p < 0.01.

The second model in the same table of regression refers to the domes-tic professional network, using the same independent variables as in the previous model. Here, only the domestic projects were related to domes-tic network. Similarly, for the third model of international network, only the number of international collaborative projects was positively asso-ciated. The number of domestic collaborative projects was negatively


related, though it was insignificant. An increase in the number of inter-national collaborative projects obviously results in an increase in the size of international network of the scientists. Conversely, domestic col-laborations do not affect international networks.

Collaboration and networks are therefore closely related. However, one is not sure which comes first, similar to the proverbial chicken-and-egg situation. Networks might be a sequel to collaborative research requiring scientists to maintain their professional network ties. It might also be the other way round. Contacts and networks might have spurred collaborative alliances between scientists. Whichever direction (collabo-ration–network–collaboration) it takes, collaboration and networks are indeed interrelated. What flows from this analysis is that international networks have a positive relationship with international collabora-tive projects. This goes with the earlier finding. Productivity does not increase the network size of the scientists, nor do networks trigger pro-ductivity. Network size across all categories of total, domestic and inter-national is emphatically determined by the email communication of scientists. Email has become the prominent means of communication for our scientists, regardless of their sectoral affiliation. Domestic col-laboration does not cause any positive change in the domestic network size, nor does an increase in domestic networks culminate in domestic collaborative projects. On the other hand, international collaborative research leads to a corresponding increase in international networks, and an increase in international networks leads to more collaborative research.

Conclusion

Collaboration has thus shown its strong connection to productivity and communication in the data that is drawn from a sample of academics and scientists in South Africa. Productivity varies according to the col-laborative and non-collaborative propensities of scientists, the former turning out to be the more productive in the number of publications in peer-reviewed journals. This is confirmed again when productivity is grouped on the basis of ‘no collaboration’, ‘domestic collaboration’, ‘international collaboration’ and ‘combined collaboration’ (domestic and international), where productivity is clearly associated with collabo-ration. Productivity in journals that are published outside South Africa depends on the gender (male) and the number of years the scientists have lived outside the country. The gender factor is not significant in the productivity of scientists in domestically edited journals. Academic


age and institutional familiarity alter the productivity of respondents favourably, particularly those who manage international collabora-tive research projects. Productivity increases correspondingly with an increase in the number of collaborative projects. Surprisingly, this is not the case with domestic collaboration, which is not associated with pro-ductivity. In short, international collaboration influences productivity in a positive manner.

When communication was introduced as a factor, a new dimension was added to the analysis. Collaborative researchers allocate more of their time for email use and have more email and Internet experience than their non-collaborative counterparts. In network structure and location, collaborators maintain larger networks. Network size, collabo-rative projects and email networks of respondents are positively asso-ciated. Directly related to this is the earlier finding that collaboration of scientists in research institutes is more domestic than international. International networks are related to international collaborative pro-jects. Networks do not increase productivity or vice versa.

Collaboration thus has a decisive effect on the productivity of South African scientists. Although it has no direct effect, broadband connectiv-ity and its incessant usage provide a conducive environment for collabo-ration. Among the types of collaboration, the international rather than the domestic variety has more potential for the productivity of South African scientists. As discussed earlier in this chapter, this finding coin-cides with that of many previous studies carried out in other countries.

It is interesting to examine a prolific researcher’s experience of col-laboration, productivity, communication and networks using historical, bibliometric and primary data. The next chapter presents a long conver-sation with a renowned researcher to gain some useful insights into her research.

197

Patricia Berjak is an eminent seed scientist, highly esteemed in Africa and outside the continent. Her admission to the Order of Mapungubwe is but one of her many accolades. In 2001, she was assessed as an A-rated scientist by the National Research Foundation of South Africa to become one among the very few A-rated scientists in the country. She serves the body of scientists, nationally and internationally, in numerous capacities. In 2004, the Department of Science and Technology in South Africa selected her for the award of Distinguished Woman Scientist. In 2005, she was the president-elect of the International Society for Seed Science, becoming the president for the 2008–11 triennium. She is also a Fellow of the Third World Academy of Sciences and a vice-president of the Academy of Science of South Africa (ASSAf). Over several years of intensive research, Patricia (affectionately called Pat by her colleagues) has made several groundbreaking discoveries in cryopreservation of the genetic resources of species producing unstorable seeds, among which several are of importance in traditional medicine.

This is the transcript of the interview she gave us on camera on 20 February 2007. Geoff Waters, a sociologist and writer, interviewed her while I handled the camera. In this interview she opens up about her long-standing career, research interests, collaboration, and her passions. She passed away in late January 2015.

Geoff Waters: Could you begin by telling us a little about your personal background—where you were born, where you grew up, about your fam-ily, schools that you went to?

Patricia Berjak: I was born in Johannesburg and educated there right through to honours level. We lived on one of the mines and the philosophy—my father’s philosophy—was that there was no point in travelling right across Johannesburg for a couple of hours each

8Collaboration Experience: Portrait of an Eminent Scientist


side of the day to go to school. The local school was good enough, and it was a rough school and it taught me a hell of a lot about working on my own and a lot about life and the others who went there, but there was no polish, no finesse. But there I learnt a lot about life and how to survive and, then, after not a particularly distinguished matric, I went to Wits [University of Witwatersrand] to study there, as far as honours.

GW: For the benefit of lay people, could you broadly explain your field of academic interest?

PB: I can try. All right, let me tell you how I got into it. I came to the university with the misguided idea of doing chemistry and physics . . . I took one look at physics, particularly, and decided then that maths was not particularly attractive to me and decided to do something else. And I went to a chemistry/botany combination for which I had to do zoology as well, and I suddenly found out about biology. I’d never done it in school, and found it extremely intriguing. So I changed to biochemistry and so my first degree was in biochemistry. I then came down here to Durban deciding that Jo’burg was not really the place that I wanted to stay and got a position down here, in fact, in Physiology at the Medical School doing physiological and biochemical research involving a heart condition with experimental rats, which led to me earning the M.Sc. degree. But I had a mind shift and decided that there was more to biology, more to life, more than solutions of blood, or whatever, in test tubes. So, I . . . actually, the person who started this department in fact, Professor Villiers, was just about to start this department. He was a very eminent electromicroscopist. So I talked to him and he said fine, come and do a PhD with me, but it will have to be on seeds. This was interesting because I had never even thought of that. Being a student I had not taken to botany that much, and it is quite ironical that I have gone purely into the plant line, into seeds. That is how it all started. It was really the particular expertise of the person who supervised my PhD. It was never planned that way.

GW: And your personal experiences? Maybe then I will go on to ask, could you explain with regard to your important research on seeds, the back-ground to it and your role in it?

PB: Sure, all right. When I said that I got into a PhD topic on seeds, it was not the kind of seeds we work on presently. We worked on maize seeds, maize being important in Africa, and the deteriora-tion of those seeds when they are stored. Because seeds, no matter

Collaboration Experience 199

how dry they are, if they are stored badly, and by badly one means when the relative humidity is too high and the temperature is too high, they deteriorate badly and they grow fungus and become useless for planting, and actually, very bad for eating. So, the research of my PhD was focused on what causes the deterioration of seeds, maize seeds, in storage. Now, maize seeds are typical of a category of orthodox seeds. It is a horrible name, ‘orthodox seeds’, implying that they will do what you want them to. In other words, you can dry them and you can store them. If you store them cor-rectly, you can keep them for a long, long time. How do we get to where we are now?

Well, in 1980, I was one of the small group of people invited to a meeting in Reading in England to talk about seeds that are quite different—recalcitrant seeds. I unashamedly now admit that I did not know what the recalcitrant seeds were when I went to that meeting. I did not say a word in the meeting, just listened. And again it was the story of how we can store the seeds. Such seeds are amazingly short-lived, some of them last a few days or a few weeks only, depending on the species. And the big question was: What can we do? There are a lot of important seeds of this category such as natural rubber, cocoa, tea and so forth and the focus was how we could store these because this seed storage is extremely impor-tant both for planting, genetic resources, food and feed and so forth. I came away from that meeting with just one idea—nobody understands these seeds, nobody knows why they behave the way they do. What did come to me quite strongly there was that most of the plant species that have been characterized as recalcitrant which means ‘disobedient’—another terrible word to describe seeds—were tropical. But most of the technology in 1984 for doing sophisticated work was in the northern hemisphere. So, there was a geographical gap between the origin of the seeds in the tropics and occurrence of the seeds where the work could be done. I thought in Durban we are in a wonderful position because we are tropical or at least, sub-tropical, we have access to species produc-ing seeds like this and our technology is not at all bad. So, really, I went to a meeting and I was at the right time and in the right place and I had the right background. We never looked back since we started working on recalcitrant seeds.

GW: OK, much of your work has been done in collaboration with others?PB: Oh, yes.GW: Would you talk a little bit about how this collaboration operates?


PB: Sure.GW: And also, what you consider the positives and negatives?PB: Yeah, all right. In the first place, in science and in the practical

sciences, you cannot do a one-man, a one-woman, show anymore. Research is terribly time consuming, terribly time demanding and, in fact, really the way to make good progress is to build up a research team, which, of course, comes mainly from your research students acting on your ideas. So, in science it is unlike the way that humanities research is done, and publication is done by a team, with the senior author being the one who did most of the work. So, we started off by starting to build a team. I was very lucky in having extremely able postgraduate students at the time and particularly Norman Pammenter, who is also my husband. He is a plant physiologist interested in water and has become my major collaborator through all the years and we built up a team which largely consisted of people who were our students and who are now professionals in their own right. We have actually built up a really strong team. You cannot operate in the practical sciences by yourself, you just cannot.

GW: And positives and negatives of that collaborative research?PB: I . . . I can’t think of any negatives. There’s really only a positive

side. To my mind the students presently are in a very fortunate position because Norman and I—sometimes with others too—supervise them as a team. Not all of us may be involved with any one student but there will always be two or three people involved per student. I think that it is exceedingly positive; it is positive for students because the student gets other points of view . . . It is posi-tive for us because in discussions all sorts of ideas come out that one may not have thought of oneself. I really cannot think of any negatives, not in our particular situation. We’ve been fortunate in terms of personalities and just the way everything has gone. I think that collaboration is part of the strength of our success. We’ve so far had generally harmonious relationships with our stu-dents. We have had the odd disagreements but nothing major and the supervisory team approach is very good.

GW: Yeah, yeah. Going back to collaboration. From what you’ve said, it sounds as if it’s more of a hierarchy . . . or how do you see it? I mean, is it a senior person who pulls together the materials?

PB: Ok, look. I reckon that I do drive research. I’m innovative and I actually supervise with quite an outwards-looking attitude. So, I think while some of the ideas and the driving force come from me,


everybody in the team contributes. I am talking about at the super-visory level. I and my colleagues are equal participants once the ideas are born. I mean I might have an idea which somebody will say no, no and what about this aspect and what about that aspect. So, we normally get onto a common ground. But I will say that I have a sort of an intuitive flair that makes me think of the ideas in the first place generally . . . but not exclusively.

GW: If we look at the practical consequences of your work, what do you see as being major practical consequences for South Africa, for developing countries generally, and for the world globally?

PB: Alright. I need to explain a little bit more about the recalcitrant seeds. They are unlike the orthodox seeds that we were talking about earlier. Recalcitrant seeds are wet and if you try to dry them, you will kill them. The reason that this happens, that you can kill them, is they are what we call metabolically active. If you imagine yourself, and we now dehydrate you, you are going to die pretty rapidly because you are an actively metabolizing system. Similarly, the seeds are actively metabolizing systems. That actually was one of our discoveries—to realize that this was the basic reason they could not be dried. So, as a consequence of not being able to dry them, you certainly cannot store them in a conventional sense. So, in terms of science, we wanted to understand all the phenom-ena that underlie the characteristics of seed recalcitrance. In terms of practical value, we must learn how to best conserve the resources of the African recalcitrant seeds. Now, this is as true for planting materials as for orthodox forms, consumption . . . for genetics resources, for conservation. It is a huge global concern these days that people, mankind, humankind is being so careless; species go extinct as you blink. There is a huge concern now for conserving plant genetic resources. So, how are we going to conserve the resources of species producing recalcitrant seeds? I’ve mentioned a couple of species. I mean, avocado pears, litchis, mangoes . . . these are common examples but there are many forest trees many, many, many, many species that haven’t been explored by biologists and are still unknown. So, we are looking as broadly as possible.

Are the seeds responsive to drying? Are they not? How could we best conserve them? So, there are two aspects to this. One is what do you do in the short to medium term. You wish to gather some seeds now and plant them in six months’ time. Can we do this? Here we are working on a system called hydrated storage that does not let them dry out. The biggest problem is the fungus that


proliferates in the seeds. There are no seeds produced in the tropics and sub-tropics that are free of fungus, and that goes for orthodox seeds too, for example, mealies. But, in hydrated storage of recalci-trant seeds, the fungus is as active as the seed is. Think about a loaf of bread in a plastic bag in Durban and how in a shorter time at household temperatures, you see fungus. Same problem. So, we are working on this. One of our big drivers is getting rid of these fungi that are common in recalcitrant seeds. The other drive is storing the seeds, as germinating occurs in hydrated storage, which I will explain. These seeds are shed from the parent tree with water content sufficient to germinate, that is, they need no additional water to germinate. And, in fact, because they are actively meta-bolic, sooner or later depending on how developed they are when they are shed, they will germinate. As a consequence, in your nice warm hydrated storage bucket they will start to germinate, some of them quickly, some of them less quickly. You might say, why don’t you cool them down, because everything slows at a cooler temperature. But there are seeds that are sensitive to cooling, like cocoa, which must be kept at 15 degrees centigrade or so. We have all these problems to contend with. So, on a species basis, we try to optimize, get rid of the fungus, find the best temperature, and the best storage conditions. So that gives us a window of weeks, months if we are lucky but not more . . .

Then we get the other side of the coin. What do we do with the long term? And this is where we come to this aspect called cryos-torage. The long-term storage of these resources is best done at liquid nitrogen temperature or certainly at very low sub-zero tem-peratures. Conventionally, the technique as used is to immerse in liquid nitrogen at minus 196 degrees centigrade. In order to do this, there is a lot of technology we’ve got to develop. Because, number one, the seeds are generally big, and they are wet. Now, picture yourself putting a lettuce leaf in the freezer; when you take the lettuce leaf out and you get this little rag. Because in a wet tis-sue ice forms and ice shreds the tissues to pieces. You’ve got to get around that. How do you get around that? How are you going to prevent it? I said earlier you can’t dry them. But one of the discov-eries we made along the line was that the faster you could dry them, although it sounds ridiculous, the more water loss. Relatively speaking, they would tolerate. However, recalcitrant seeds are large and cannot be sufficiently dried to avoid damage. But if we cut out of the seeds just what’s called the embryonic axis, we have a small


enough structure . . . If we cut the axis free of its food supply, then it is a tiny piece of tissue with the potential to produce a new plant, we can dry axes very quickly, but they will stay alive only in the short term. If an axis has been dried for minutes maybe we can plunge it into liquid nitrogen: it’s dry enough for cryostorage . . . I make it sound very easy but we’ve got to, number one, coming back to the fungus, get rid of the fungus, number two, get the axis to dry just enough to freeze and not so much that it is damaged, and three, determine the parameters for each species, including the rate of introduction into the liquid nitrogen and the rate of thawing, and how to handle these recovered axes.

All those things give us a huge amount of work because pres-ently just looking across species we don’t have uniformity. Every species tends to behave a bit differently. Under those circum-stances, we really are looking at things on a per species basis . . . with the objective in the end to be able to generalize though, and ultimately to produce a manual, possibly online that people world-wide can actually follow these principles. But you know I made it sound easy. It is quite complicated actually. There are numerous stumbling blocks along the way. For example, if you get seeds of species that are related to temperate regions, Western Cape, or Europe or wherever. The axes of those seeds are much easier to manipulate successfully than those of tropical and sub-tropical species. So, there is a difference. We need to understand the differ-ences. Are there differences in membranes or cells? There must be a difference and one of our difficulties in our tropical, sub-tropical species is after all these manipulations, even after just cutting the axis out we do damage the shoot tip such that the seed does not establish a normal shoot. We’ve got to overcome this. So, one of our big drives is to stem the damage, but to do this, one must understand what is causing the damage, and we are making a lot of progress with this. So, as I say, there are a lot of problems, but that makes the project, that’s really exciting.

GW: Let me ask you, on average how much time do you spend working on . . .?PB: Not enough. At the moment, not enough.GW: What would you say would be the ideal per day, per week, if you do

not mind?PB: I really don’t think I can answer that; it would depend, really

depend, on where we are in our investigation and at what stage. I mean I would actually enjoy microscopy. It is occupational therapy for me.


GW: Do you use it every day?PB: No. At the moment, I still have got an enormous amount of out-

side responsibility in terms of teaching undergrads, honours and then even when I stop teaching undergrads, which hopefully will be next year, except for one course, a seed biology course, there will be honours students and postgrads. I don’t actually have the pleasure of enough time. But in fact you see, again I could have a student looking at something and saying just come and see this; then, I’m also learning, and thinking of new avenues of investigation.

GW: Well my next question is somewhat different, but in a sense related to what we’ve been talking about. If you were a South African government minister today what policies would you advocate for current agricultural practices and also for conservation which often needs to be at odds?

PB: Yeah. Alright. One has to try and maximize the quality and quan-tity of whatever is being produced without actually keeping on degrading more and more the natural landscape for planting. So, you have to go in for breeding high quality stuff that actually grows faster and produces more. I think this is the way I would go because, in fact, degrading the landscape in the short term to feed a population that is too big is not sound procedure. If you want to address the root cause it is population. Population is too large and every encouragement should be actually given to curb population growth. Unfortunately it is a political hot potato. There are too many people but also just to degrade the landscape in order to feed a burgeoning population is going to get humankind nowhere in the long run. We cannot live in a totally degraded landscape. For example, let’s move to the South American continent. The destruc-tion of rain forests for short-term aims of growing whatever they might be growing there is totally deleterious because the oxygen in the atmosphere, in our atmosphere, comes from plants. If you get rid of large, huge tracts of natural vegetation you are actually degrading the whole atmosphere, not only the landscape.

GW: Well, to change . . . initially, I don’t think that you want to be asked about it. But the natural sciences have been considered a male world. As a woman do you think that this is the case, is it changing, is it changing at all?

PB: Alright.GW: Would you recommend anything?PB: I would say, Geoff, that for the last 30 years at least there have

been no barriers. If you are the kind of person that wants to get on,


then you get on. Well, when I came into the job, I did not get my first job that I applied for and it was given to a male with lesser qualifications. When I ultimately asked, they said, oh, yes, the par-ticular selection committee at the time was not keen on appoint-ing women. This, actually, was the best shot in the arm for me: my reaction was all right, I’ll show them. And that was actually before I got my PhD and I think if I’d got a lectureship at that stage I would never have got to where I have. So the bias was actually very good for me. Now it’s wide open. There is every opportunity. In fact, there are opportunities for women, these days are good. I do not believe that somebody should get a job because she is a woman: rather, appointment must be on merit.

GW: Ok, fair enough. Enough said. If you look at science education and par-ticularly education in the biological sciences in South Africa today both at secondary and tertiary levels, do you have any thoughts on this? Is it going in the right direction? Is it where it should be? Does it need to change?

PB: Ja. You know I’m really not in touch with the secondary level any more.

GW: OK.PB: At one stage, I’ve found the school syllabi were terribly con-

strained, terribly prescribed. And every time something new heaved into view it was included in the secondary syllabus with nothing falling out; it actually became a jam packed syllabus that the kids could not really get enthusiastic about because they didn’t have the time. So, I can’t actually comment now because I’m really not in touch much with the school situation. The university is much freer. On the whole, your mode of teaching within the sub-ject is dictated by your own sort of view of the subject. If you are teaching first year it is a little bit different because particularly now with merging universities here, we now have common modules between biology here and in Pietermaritzburg and so forth. So, those syllabi and that teaching, to me, have become more con-strained . . . Once I teach students in the senior years I actually have no constraints whatsoever. So, I think that it is more at the junior levels where you’ve got more students that things are becoming constrained; once you’ve specialized, your teaching is your own ‘baby’.

GW: And in terms of throughput from school, do you feel that you are get-ting sufficient numbers of . . .?

PB: We are getting more and more numbers, actually. We have, over the last maybe ten years, had too many for subjects that are


laboratory-based, and I feel that the quality of the work and what we can offer in the laboratory may well suffer as a consequence.

GW: What advice would you give to young biological scientists in South Africa today?

PB: Alright. I think that they must follow their noses. They must get into what they really enjoy doing. So in biology we have got many students who arrived, particularly in old days, wanted to be a marine biologist because they have seen nice stuff on the televi-sion–adventures. As things go along, suddenly they found branches of biology which they never knew existed. They’ve got enthusiastic and have done really well. So, I think to my mind, don’t undertake something if you don’t enjoy it. Obviously at undergraduate level you will have courses that you do not enjoy but ultimately a student can pick what he or she really enjoys, and ultimately engage in research that is exceedingly exciting.

GW: Ja.PB: You know every day it is a different story. It is great. You also have

got to become philosophical though. In research you can spend two-thirds of your time apparently not going anywhere because you are developing methods and trying this and that, modifying that, but ultimately it’s worth it. So, my advice is unless you have the mindset to say alright, two-thirds of the time I am going to be quite disappointed because it’s drudgery, but the last third of the time it’s fantastic—if you are going to be a researcher, that is what’s important. I don’t think that training as a biologist constrains peo-ple. I think it’s a training for people to question, to ask questions, to seek answers themselves, to think laterally and, as a trainee, I think that makes a person really good to undertake almost any-thing. I mean I always said to students, if you do not want to do research, if you do not want to become an academic teacher with your biology you can attach yourself to something . . . public rela-tions or as a journalist or whatever; take your science that way, and a lot of people have done this. You know some people, many peo-ple, practise as biologists. Some don’t. They are going to totally different areas and do very well.

GW: You spent most of your career at the University of Natal, now the University of KwaZulu-Natal. Could you tell us why?

PB: Well, I came here from Wits and not directly here; I did little bit school teaching, came back to the medical school where I did my master’s degree, and then came to the biological sciences at the then University of Natal, initially in MTB [Memorial Tower


Building]. I found the University of Natal a very friendly warm place in contrast to Wits. Maybe because I was not a postgrad there, but I found it a very nice place to be, so I did my PhD here. We then went overseas for four years. We were in the UK in Leeds. The way we worked is . . . Norman worked here when I was doing my PhD and then I worked over there while he was doing his PhD and we were very keen to come back and when the opportunity came up for both of us, and remember this is the difficulty when you’ve both got a PhD in the same game. We were not exactly in the same specialty within the plant biology but we were in the same game. We were just really lucky that we both got opportuni-ties to come here at the end. I never really wanted to go anywhere else but sometimes I get a little bit impatient . . . but what can you do? That’s life. Probably won’t be better anywhere else. Ja.

GW: Well, in your time particularly at this University, you’ve done a lot of teaching and research.

PB: Ja.GW: And a lot of administration?PB: Indeed, yes.GW: Which of those can you get you in the morning, and why?PB: Well, research is still the more exciting and I quite enjoy teaching.

But sometimes you get there in the face of students dribbling in at one per minute. I get quite irritated. But I do enjoy teaching but I think that research is it. I administer well but I don’t enjoy it.

GW: Talking about something which you discussed further with Moorthy at an earlier stage, about your hobbies?

PB: Hum Hum.GW: Light aircraft, fast cars, and ball room dancing?PB: Not at all. Not at all. We started. If you want to put it in order fast

cars came first when we were postgrad. As students we had a Mini Cooper S. In fact, we had one of them then we had another one. We didn’t eat very well because the car cost money, and the couple of dogs that we had cost money but we’ve been quite keen on per-formance cars. We acquired the Scimitar when we were in England actually. It was one of those strange things. We were there not hav-ing a car at all, at first and then having an old Anglia, and then it was one of those interesting things. I was on the staff at Leeds and they forced us to take a subsidiary pension on which I was not in the least keen. I said: What happens when I leave? They said it goes with you and if you don’t have a place which has the same scheme, it becomes a private insurance policy. So I had no choice


and then when we were about to leave they said would you like to be paid out in cash? Indeed, yes!! This is how we got the Scimitar. And it is much about how do you choose your cars. I fancied in those days, misguidedly, that Mercedes coupe with a slightly con-cave roof. Norman fancied more the Rover. They were both way out of our range. When we saw the Scimitar, we both liked it, and the fact that it has a rust-proof fibreglass body, and that’s how the Scimitar came into our lives. We brought it back from England with us and that was our only car for a long time and then we’ve got a subsidiary car and that became more the weekend car and we went through a couple of other makes, the first, because it was cheap and then the Alfa Romeo, because Norman always wanted one . . . and then another Mini Cooper S!

GW: Yes.PB: When they reappeared in the market, we just simply couldn’t

resist it.GW: Ah.PB: So, that’s how the present Cooper S came to our lives. Dancing

came to our lives also a long time ago, we both danced quite well. But we were not tutored and we decided one particular year we would actually take it up a bit more seriously and we enjoy it thor-oughly. I would say it’s the kind of hobby that keeps one sane—it’s a total release, something different totally different challenge. We play bridge occasionally too and as far as flying was concerned, it was quite a challenge. We don’t fly anymore because it became prohibitively expensive and we didn’t have time to do it. Flying is one of those things, if you don’t keep flying, you become unsafe.

GW: Yes.PB: So, we gave that side of it up.GW: OK, if I asked you to give us five words that describe yourself, what

would those five words be?PB: Oh my, that is a very difficult question. Alright, I know, I know

I’m not simply driven or determined. I’m intuitive, I’m enthusias-tic and when I get enthusiastic about something, then I’m driven with determination along that track yes. That’s only three words. Eh, beyond that . . .

GW: One last question. When you do retire finally, what would you like to do?PB (laughing): Look, let me put it this way. As long as we feel we are

really having an impact with the research and we are really doing well . . . we’ve got the funding, the students are still attracted to stay, we’ll stay with it you know. Beyond that I really don’t know.


It’s very difficult to say what would one like to do. Norman face-tiously mentions being driven to and from boozy lunches, ha ha ha! I certainly wouldn’t find that fulfilling (and neither would he, I suspect)!

GW: Well, that’s it. Thank you very much indeed.PB: Ok.

Collaboration and publication productivity

The interview with Pat provides a closer view of her research life. It high-lights a number of issues which relate to the key themes this book has dealt with in the previous chapters. Pat’s long journey in the world of science was not undertaken alone, but with a team of scientists and stu-dents drawn from both within and beyond South Africa’s borders. In her view, teamwork is paramount in science. She acknowledges that science is no more a one-person show and that it is essential to build a compe-tent research team in order to advance in scientific undertakings. This teamwork has led to considerable progress in her research, career, and in the production of publications. Pat’s remarkable accomplishments in the field of seed science are firmly rooted in her collaborative research and her success in building a strong research team which is willing to carry forward the original ideas and work through to their fruition. Pat’s experience as a researcher shows that a research team works together in the production of publications. It is led by senior members of the team whose ideas are taken forward by other team members. Postgraduate students are part of this collective enterprise. In this partnership, even life partners are involved. Everybody makes a contribution to it. This is unlike in the social sciences where a senior member of the team often does most of the work. Postgraduate students in Pat’s team have an added advantage. They are not supervised only by one member. Rather, colleagues in the team are available for further assistance, which actu-ally helps the students. For her, teamwork is the recipe for success in research at a higher level since teamwork has had only positive effects on her research activities. Are these effects reflected in her publication productivity?

Pat’s publication profile shows how productive she has been in her career. In line with the data used in chapters 4 and 5, the database of the Web of Science is utilized. The Web of Science stores a total of 142 of her publications (papers and reviews). These extend over a 42-year period, from 1972 to 2014. While this might not be the complete count of all her publications, it is a good sample for purposes of analysis.


In the previous chapters, the relationship between collaboration and the productivity of scientists was examined using both bibliometric and empirical data. Is there any connection between productivity and col-laboration evident in Pat’s scholarly career?

During the period 1972–2014, she produced an average of 3.38 publications a year. True to her conviction that science is a group activ-ity, with the exception of four (2.8% of the total), all of her papers were jointly written. This reveals that she practices what she preaches in sci-entific research. In a single year (1992), her productivity rose to a total of 12 publications. In 1992, 1997, 2000 and 2004 she had more than 10 publications each. Amongst the total 142 publications stored in the Web of Science database, Pat worked with an average of 2.32 (S.D. = 1.14) authors per publication. Most of her publications are with two authors (mode), while some 13 per cent of these 142 papers involved more than four authors, with a maximum of seven authors. For the production of research publications Pat associated with 324 partners (some partners or authors worked with Pat more than once) in her career. In comparison to South African scholars (referred to in chapter 5), the average number of Pat’s co-authors was lower by about two authors (4.27 versus 2.32).

The nature of Pat’s research collaboration suggests that she preferred association with colleagues and students in her department and in her institution, and those in other institutions within the country. This type of collaboration—domestic—exceeds her international research collaborations. Pat wrote 99 scientific papers (74%) with colleagues in the country while 34 papers (26%) were written with colleagues from overseas. (This excludes sole-authored papers and incomplete publica-tion records.) Two-thirds of her papers involved domestic collaboration as against one-third of papers that were produced with international collaboration. Pat had the opportunity to work with 275 scholars from within the country and 49 scholars from abroad. These figures include authors who had worked with her more than once. The publications (in her case 3.38 papers a year) Pat produced is above the norm for aca-demics at her level (full professor), demonstrating that she has been very productive. Among her publications, 87.8 per cent were co-authored. A relationship between collaboration and publication productivity can be deduced from this data.

As regards the patterns of collaboration, Pat’s international partners came from the US, England, Scotland, Canada, Italy, France, Ukraine, China, Nigeria, Benin, and Ghana. She conducted and produced more research with local scholars than with international scholars (85% and 15%, respectively). Thus the indisputable international reputation she


has gained in the field of seed science is not entirely due to her collabora-tion with partners from overseas. This contradicts the view held by some scholars that local collaboration is not as significant as international collaboration. The research produced in collaboration with domestic partners is as important, valuable and visible as those produced in inter-national alliances. Locally produced knowledge, as Pat’s research profile confirms, can be of international quality and provide international rec-ognition. This is analysed in greater detail in the following paragraphs using her citation statistics.

As captured in the Web of Science as of 24 September 2014, Pat received 3,005 citations for her 142 publications with an average of 21.16 citations per paper. Pat’s h-index for these publications stands at 30. The highest number of citations received by her publication is 246 (including citations covered by other databases such as BIOSIS, Chinese Science Citation Database and SciELO Citation Index), for a paper pro-duced in 1999 jointly with another author, in this instance her husband. The paper, ‘A review of recalcitrant seed physiology in relation to desiccation-tolerance mechanisms’, was published in the journal Seed Science Research.

The literature reviewed in the preceding chapters suggests that cita-tion is often dependent upon the number of authors who produced the paper. In other words, more citations can be expected from co-authored publications than from sole-authored publications. There are also vari-ations in the count of citations depending on the nature of the col-laboration (domestic/international) involved in the production of the publications. These two theories (increased number of citations for co-authored and for internationally collaborated papers) do not fit well with Pat’s publication record. Some of her sole-authored publications were highly cited, in fact, more so than many of her co-authored pub-lications. For instance, one of her single-authored publications had a total citation count of 59 (paper entitled ‘Unifying perspectives of some mechanisms basic to desiccation tolerance across life forms’, published in Seed Science Research). Conversely, there were some joint publications that have not been cited so far. Taking the analysis one step further to identify the relationship between citations and international col-laboration (read production co-authored with international partners), it emerges that two of her highly cited papers were written in collabora-tion with local partners and not with international partners. This cor-roborates the earlier conclusion that locally produced knowledge is as visible and important as that produced in association with international partners.


The responses Pat gives in her interview provide some insights into the way collaborative research teams function in the life sciences. Often the internal understanding and functioning within a research team is not fully known. In some collaborative enterprises, scientific research proceeds in a structured fashion. Many of the issues dealt with in the earlier chapter on conceptualization apply to teamwork. A hierarchy of roles and ranks might appear in teamwork. In some other cases, as Pat conducts her own research, the leader proposes a new idea which is developed with the contribution of other members of the team. But this is not done in a hierarchically structured or determined manner. In the end they all come to a ‘common ground’ on what is to be done and how it is to be done. This agreement is obvious in the research Pat is doing. In successful collaborations, members of the team recognize the strengths of other members and these are valued for the collective benefit of the team. Collaboration becomes an enriching experience for partners within the team and every contribution of the partners adds to the quality of research and publication outcomes. This is relevant when scientists are working on leading-edge subjects and themes, such as in genetics, which require intensive, lengthy and committed participation. This is what Pat and her team have been doing—engaging in the chal-lenging task of discovering innovative ways to protect and store recal-citrant seeds. The use of advanced technology is part of experiments of this kind.

As elaborated in Chapter 3, there are a number of factors which moti-vate scientists to work together, of which the sharing of resources such as equipment and technology is one. The cryostorage technology that Pat and her team employ for scientific experiments in genetics is an expensive one. Not everyone in the field, not even every institution, can afford to have a cryostorage facility. Researchers, therefore, look for its availability and establish contacts with those who have access to it to conduct their research.

Another prominent feature in Pat’s research career is that she worked with a team of researchers, many of whom worked repeatedly with her and took part in the production of several of her publications. This reveals that partners continued to associate with Pat on subsequent pro-jects. As members of the team, they enjoyed being part of the team, receiving the advantages of collaboration. Partners will not continue to associate in collaborative research undertakings unless they benefit professionally. Without professional benefit, they would terminate their participation after the first project and leave the team. As discussed in chapter 3, there can be a host of benefits and incentives to participate in


collaboration, including the sharing of resources, learning from partners (relevant for student partners also), developing and honing skills and contributing to publications. Since the partners were not interviewed, what kept them in collaborative research projects is not clear, but pro-ductivity could be one reason that encouraged Pat’s partners to associate on a continuous basis. Pat has been very productive, publishing regu-larly and in high impact journals, and many of her research partners would have found this beneficial to their career advancement.

The continued association of partners, obvious in Pat’s research, also relates to trust. Science is conducted by human beings, with machines which are controlled by human beings. The literature suggests that trust is the foundation on which partners build their team for the success-ful completion of research. In her interview, Pat reiterates that she has formed a strong research team which is the basis of the success of her research endeavours. This would not have been possible without ensur-ing the trust of her team members. Once trust is established, commit-ment to the objectives of the project can be guaranteed.

Opportunities for breakthrough research is yet another factor that motivates scientists to collaborate. There were bright prospects for a breakthrough in Pat’s field of research: drying of seeds in artificial conditions and using advanced technology. Scientists are particularly attracted to potential possibilities where they can be part of a ground-breaking research endeavour. Perhaps this could be the reason for the successful building of an efficient and committed research team such as Pat has developed over the years. She worked continuously with other researchers and produced more than one publication in tandem with many of them.

The next chapter incorporates the findings of this chapter and those of the analysis chapters preceding it.

215

Scientific collaboration

Science is no longer a centralized activity located in a single place, but is dispersed far and wide. Hicks and Katz (1996) predicted that, in future, a piece of knowledge will be produced by more people in more loca-tions. Increasingly it is carried out collaboratively in teams of individu-als from different institutions and countries. The growing number of multiple-author publications (Hafernik et al., 1997) is an index of col-laboration. As Whyte (1957, cited in Hagstrom, 1964) correctly said, the scientist is becoming an ‘organization man’, accustomed to work-ing in groups, taking decisions in committees and conforming to the rules of the organization. Collaboration enhances competency, skills, and knowledge of the partners while ensuring the quality of research. It strengthens scientific activities and capabilities, and failure in collabora-tion weakens science and technical enterprise (Bozeman and Boardman, 2003b). The substantial fall in the costs of air travel and communication and advances in effective information and communication technologies (ICTs) have accelerated collaborative activities (Katz and Martin, 1997). In collaborative enterprises, the location of partners is becoming more dispersed (Adams et al., 2005). Distance is no longer a matter of serious concern in collaboration, although progress in ICTs has not taken place uniformly in all parts of the world.

International consortia of sciences are taking shape (Hicks and Katz, 1996). After the Second World War, Japan put in place a number of systems to promote collaborative research and development amongst researchers in universities, industries, and government laboratories (Wen and Kobayashi, 2001). Research associations have been estab-lished nation-wide and have a wide reach in Japanese society (Wen and

9Science and a Model for Scientific Collaboration


Kobayashi, 2001). The International scientific unions in the US are the backbone of international scientific collaboration (Bok, 1955). The US has developed an institutional mechanism to conduct regular monthly meetings to discuss opportunities for new collaborations, review and assess the existing collaborations, and recommend future collaborations (Bozeman and Boardman, 2003a). The European Research Area within the European Union is another configuration for scientific and techno-logical cooperation (Georghiou, 2001).

The production of scientific knowledge has ceased to be the monopoly of universities and academic institutions. Non-academic institutions are equally diligent and are engaged in groundbreaking research. Alliances are imperative. Closer university–industry alliance is backed by both parties for mutually beneficial research outcomes. Industries profit from the intellectual pool of academics in universities for inventions that lead to new industrial products and applications. In return, universities get funding and equipment. Despite the differences in the two cultures, there needs to be a meeting point in university–industry collaboration on their priorities and the benefits that accrue to the partners.

From the traditional forms of scientific collaboration have emerged new phases, characterized by complexity in the division of work, an interdisciplinary approach, adoption of procedures, formation of assist-ing structures and utilization of communication technologies. Modern forms of collaboration entail more dependence on external authorities and structures and greater centralization of authority in research organi-zations (Hagstrom, 1964). Collaboration, in essence, is between individ-uals and not institutions (Katz and Martin, 1997). Individual scientists are the real players in research tie-ups, while institutions play second fiddle. In institution-initiated alliances too, individual scientists are the key actors while the institution provides the required support. Again, institutions do not have much to do with the maintenance of relations between associates. It is ultimately the responsibility of the scientists.

Collaboration is a working relationship between equipment and lab-oratories, and between human beings. Often individual scientists are the initiators, banking on informal contacts and acquaintances. When alliances stem from informal contacts responsibilities are unclear, and when commitment is uncertain collaboration can become stressful (Fox and Faver, 1984). Since not much is known about the cultural and attitu-dinal dimension of academic behaviour (Lee, 1996), understanding the personal components in collaboration is not always easy. But knowledge about the cultural and attitudinal dimensions of academic activity can shed light on the human side of collaboration. Inadequate consideration

Science and a Model for Scientific Collaboration 217

of this human component will undermine the spirit of collaboration, resulting in the break-up or premature closure of the collaboration.

To benefit fully from collaboration, the parties (individuals, institu-tions or countries) need to reach a certain level of scientific absorptive capacity, including the infrastructure of support, communication and research (Wagner et al., 2001 cited in Olson et al., 2007). They ought to have a fair idea about the costs and benefits (Sonnenwald, 2007). Meticulous cost–benefit analysis works in multiple ways. It gives the partners the opportunity to assess the worth of their involvement, maintains their interest and commitment throughout the duration of the project and gives them the strength to endure times of adversity. Costs and benefits vary according to the size—number of institutions or scientists—of the collaboration. Costs are incurred in terms of adminis-tration, coordination, travel, communication, computer programmes, occasional face-to-face meetings, and the real work of the partners. Amongst the benefits are access to equipment, knowledge, skills, exper-tise, interaction, publication and citation.

Transaction cost theory stipulates that collaborations veer towards avenues that minimize costs. When institutional structures cost too much for collaborative ventures to take off there is the option of seek-ing informal structures. This is perhaps the reason why researchers rely on their personal and professional contacts to initiate research alliances rather than using formal institutional channels. For scientists this is an easier route, saving a great deal of their time and energies. They are able to avoid the hardships that would accompany the bureaucratic, administrative and legal rules in rigid formal institutional structures. Institutions do not usually object to these forms of partnership as indi-vidual involvement is fundamental to collaborations. In multi-institu-tional collaborations involving huge funds, expensive equipment and a large number of participants, informal structures cannot work well.

Collaboration has structural and personal elements. If personal com-ponents are not aligned properly with the structural ones, collaborations would fall away. The balance between these complementary elements contributes to the maintenance of collaborative efforts. Structural ele-ments are to get access to the infrastructure and the skills required in a research project while the personal ones are the individual reasons to (or not to) collaborate with someone who might be a friend, com-panion, long-standing associate or a professional contact. Intellectual companionship and freedom from academic isolation motivate sci-entists to seek productive, fulfilling research alliances. Personal, face- to-face and informal contacts can spur people to associate on a research


project. It is the intrinsic social nature that makes collaboration a per-sonal choice—whether to collaborate or not, when and where, and for how long. Focussed scientific purpose or a joint experiment which has the potential for a new discovery might not serve to bind the partners for the entire span of collaboration if the idiosyncrasies of the participat-ing scientists are not taken into account. Collaborators leave midway for personal reasons. As collaboration is a socio-cognitive process, the suc-cess of the researchers in a collaboration depends on their success both as social individuals and as scientists (Melin, 2000).

Reiterated in the literature is the importance of prior relations, work-ing or personal, in successful alliances. Partners—individuals and insti-tutions—who have known each other or maintained contacts are likely to work well in collaborations. This gives a feeling of solidarity in part-nerships, underlining the target of collective goals. The choice of a part-ner, if not the country or institution, is invariably a personal decision, taking into account the constituent components such as informal rela-tionships, acquaintance, previous knowledge, prior working experience and trust. Research partnerships have spontaneous origins as well, such as informal conversations in scientific meetings or conferences.

Managing collaboration, even if it involves a small number of collabo-rators, entails a great deal of planning, coordination and administration (Peterson, 1993). Collaboration, as Bozeman and Corley (2004) rightly put it, is much more than just getting the work out the door. It creates a sense of responsibility and commitment amongst the partners (Fox and Faver, 1984). In large-scale collaborations more formal organizational and managerial structures are required (Bozeman and Boardman, 2003a). Corresponding and appropriate institutional changes (Sonnenwald, 2007) have to be made, and the required structures and mechanisms need to be in place to avoid administrative and technical hitches. The administrative component is as central as its academic counterpart. When the responsibility of administration and management of the pro-ject falls on the shoulders of the senior scientist/s or the originator, it turns out to be a burden (Peterson, 1993). Distributing these hard tasks among partners and thereby minimizing the load on the senior(s) is a way to handle the situation. This is not a feasible formula when the originator does not like to part with project information, funding in particular, and prefers to hold onto the strings of power and authority.

Collaborative efforts become efficacious if there is agreement on the division of labour, shared decision making and collective responsibility. Maintenance of leadership and participation of partners in a team are critical (Bush, 1957). Leadership in collaboration is not authoritarian


nor is it similar to that found in universities or research institutes. It has to be on the basis of equality and understanding (Bush, 1957). As a respondent of Hagstrom (1964: 242) said, ‘There you have a strong norm. Telling someone what to do is a taboo. The greatest man in sci-ence cannot tell the lowest what to do.’ The notion of equality comes into play at different stages (Bickel and Hattrup, 1995). Disparity of part-ners in their professional standing and experience should not under-mine the merit of egalitarianism and mutual recognition in collective labour. Mutual respect and recognition irrespective of skills, knowledge or location, regular communication on project matters without regard for seniority, and more or less equally distributed benefits are contribut-ing conditions in collaboration. The relationship between partners is symbiotic, characterized by dissimilarity, pursuit of mutual self-interests and attainment of self-interest (Bickel and Hattrup, 1995). A sound col-laboration relies on a ‘give and take’ attitude (Bickel and Hattrup, 1995: 36). No individual’s point of view dominates and the authority for deci-sions and actions resides with the group, and not with one or two part-ners (Minnis et al., 1994 cited in John-Steiner et al., 1998). Predisposition of the partners to domination and subjugation can be tolerated only at the cost of coalition.

Lack of transparency in matters of objectives, process, allocation of work, benefits, obligations, coordination and funds can cause ruptures in scientific alliances. Partners might not consider this as a serious mat-ter as long as there is an intermittent flow of funds and the work is carried out on a predetermined schedule. Conditions agreed to in the form of a formal contract might allay many of the anxieties regarding transparency.

Collaboration is to be balanced, not skewed in favour of one party. The popular perception is that in such alliances the developing coun-tries realize a larger return in equipment, materials and training (Arunachalam and Doss, 2000). That collaboration is a way to get access to the developed scientific knowledge and their technology (Kim, 2006) is a deliberately propagated view. This might be true in cases where purely technological collaborations take place, but not necessarily in scientific collaborations. In part this view originates from the divide between the developed and the developing countries, and the notion that the developing countries always need the support of the developed countries. Neglected in this discourse is the value and worth of the sci-entific data and knowledge that the developed countries acquire from the developing countries through such alliances. In successful instances of collaboration, the balance in the collaboration configuration is quite


evident as we find when Africans provide access to local communities and the non-Africans provide needed equipment and training (Cohen, 2000). Collaborative activities function and get on well when the part-ners recognize the reality that the source of funds is equivalent to the source of resources; none is superior to the other. For scientists in many developing countries (except for a few prestigious institutions within them such as the Institutes of Technology in India), in contrast to those in developed countries, the cost of collaboration is at a premium. Basic essentials for conducting research, namely, phone calls, postage, the Internet, email, stationery, printing and copying, library search, data-bases, assistance and local travel—not to mention equipment and labo-ratory material—are not always at the disposal of the scientists in poorer countries. These are structural hurdles in materializing alliances, at least in the conceiving phase before funds are actually released.

Transparency is vital in collaboration. Hamel (1991) noted some inherent, ex ante determinants of transparency: penetrability of the social context that surrounds the partner; attitudes towards outsiders; the extent to which the partner’s distinctive skills are discrete and the partner’s pace of skill-building. Transparency is important in sharing the data produced in alliances. The ownership of data in scientific collabo-ration deserves due consideration before it becomes a matter of conflict and legal battle. Dangel and Mishkin (1996) in a communiqué narrates a dispute, the Maryland Whistleblower case, of this sort in which he was a trial counsel.

Transparency and trust go together. A spark of mistrust can engulf the whole activity if it is not dealt with promptly. Trust is cardinal in risky activities and when the participants do not have the ability to see each other (Jarvenpaa and Leidner, 1999). Face-to-face interactions are irre-placeable in building and maintaining trust (Nohria and Eccles, 1992). It is highly unlikely that collaboration would develop without a measure of trust (Tschannen-Moran, 2001).

The challenges are inherent and imminent. If handled properly and timeously they can make collaboration stronger (Sonnenwald and Pierce, 2000). If not, they can cause serious trouble that would shake the foun-dation of collaboration. Walsh and Maloney (2003) remind us that size, geographic distance, interdependency and competitiveness can exacer-bate the challenges. But these can be overcome with the appropriate use of communication technologies that can bring the partners closer.

To deal with conflicts, general rules and principles should be laid down (Mishkin, 1996). Informal collaborations, involving a group of a few intimate colleagues, might not require rules to succeed. Formal


rules for such a close team of partners are a constraint rather than a precondition. Given the nature of human relations, partners cannot be assured of the same set of relationships throughout the entire period of collaboration. Agreed rules, principles and codes of conduct are useful on occasions when personal relationships are transformed into profes-sional relationships. In the light of experience, some have argued for guidelines that can be used as a reference point in these circumstances (Smalheiser et al., 2005). What is advisable, as suggested by Bozeman and Boardman (2003a), is the adoption of a framework for collaboration if that has worked out well previously, but definitely with amendments. This proposition is helpful to draw up the structures and norms for a new collaboration venture. The universal application of the framework is doubtful though, when the partners, discipline, field, topic, objectives and methodologies differ.

Some institutions work within the rules they have drawn up. These might identify their focus areas, preferred partner countries, sectors, institutions and laboratories, immediacy of problems that need to be researched, costs and benefits. For instance, institutions might have a rule for the interdisciplinary nature of collaboration, or a rule that it will work within the framework of the existing institutions rather than seek-ing to develop new institutions or another one regarding the sectoral composition of collaborations (Bozeman and Boardman, 2003a). These rules often are based on previous experiences or stem from the assess-ment of the resources, preferences and needs of individual institutions.

Collaboration, for its initiation, execution and successful conclusion, requires a great many prerequisites. In the African context, as Keay (1976) suggests, governments, without political considerations, should be willing to recognize scientific merit for the benefit of their countries.

The South African model

The model of scientific collaboration in South Africa as has emerged from this research presented in the previous chapters and other studies (Sooryamoorthy, 2009a, 2010a, 2010b, 2010c, 2013b, 2014) is unique in several respects and can perhaps offer some lessons for other countries as well. In South Africa, scientific collaboration is an accepted, recog-nized and encouraged practice among scientists, supported by the gov-ernment, higher education institutions, research institutes, industry, the private sector and individual scientists. It is considered part of the sci-entific enterprise and knowledge production. The structural support for collaboration is seen in the positive policies taken at the governmental


level and implemented in the centres of research and academic activ-ity. The reward system that is in place for the promotion of publication productivity encourages scientists to clinch tie-ups with international partners. Scientists are establishing links with their colleagues around the world.

Historically, the flow of scientists in and out of the country, the exchange of scientific personnel and collaborative scientific activities, happened unhindered. This began in the latter part of the 17th century when South Africa became a colony, first under the Dutch and then under the British. The colonial government(s) lured scientists and aca-demics to the land and supported their scientific activities. Many of them became instrumental in building scientific disciplines across the coun-try, rendering able leadership and direction to scientific research in gov-ernment departments, universities and research institutes. The colonial legacy, the focus on specific branches of science and their consequent growth in South Africa are distinctive, if not unparalleled, in the history of science. Professional associations, representing various branches of science, are proactive in their efforts towards scientific growth and col-laboration. Collaboration continued during the apartheid era as well, when the country, due to its racial segregation policies, was subject to isolation and boycott by the international community. Collaboration grew into an accepted and recognized practice and norm, which was not the case with all other African or Asian countries. Associating with a foreign partner did not invite contempt or suspicion of the scientific community in a politically stable ambience.

Distance does not matter in collaboration. South Africans partner sci-entists from countries all over the world, notably with those in faraway locations. Scientific collaboration is sustained with countries that South Africa has had a historical tradition of contacts and collaboration with. International collaboration is growing over the years. Like geographical limits, disciplinary boundaries are not restrictions for South African sci-entists. Some branches of science such as medicine, plant and animal sci-ences are more collaborative and productive than some other branches. Collaboration has helped the country to keep its advantageous position in many branches of science, bringing greater international visibility to South African science.

Scientists who are research-intensive are collaboration-intensive too. An active researcher engaged in a number of research projects is likely to be an active collaborator. Collaborators are more likely to have research projects than non-collaborative scientists. Certain key variables— gender, race, age, native place of birth, migrant status, marital position, highest


degree earned, time spent outside South Africa for higher education, time spent in developed countries, stage of the career, current position, fields of specialization, sector last worked, number of research projects, directed projects, collaborative projects and international projects, domestic col-laboration and international collaboration—are all relevant factors in the attitude of scientists to collaboration/non-collaboration. Collaborators are older in terms of age, academic age and institutional experience. Both domestic and international collaboration happen in an encouraging sci-entific environment. All major sectors—university, research institute, industry and government—are part of it. Sectorally, research institutes favour domestic collaboration involving partners within the country. Universities incline more towards international collaboration. Research institutes are thus the centres of domestic collaboration while the uni-versities foster international alliances.

There is a difference between subjects and disciplines in the extent of international collaboration. Some subjects are known for establishing international tie-ups than others. International collaborative research, not domestic collaborative research, brings more people to the research team. In other words, the size of the internationally collaborated pub-lications is significantly larger than that of domestically collaborated papers. Collaborated South African publications, in accordance with the international trends, yield more citations than sole-authored papers. The increase in the count of citations depends on the number of authors of the publications. Internationally collaborated papers, as against domes-tically collaborated papers, yield higher number of citations. Certain subjects attract increased number of citations than others.

Traditionally, the productivity of scientists has received encourage-ment in South Africa. This is being continued through appropriate institutional structures. Collaborations have a beneficial effect on the productivity of partners. Collaboration is a means to enhance produc-tivity. Productivity is a function of the institutional familiarity of scien-tists. As the number of international collaborative projects increases, the publications output of South African scientists also increases. Domestic collaboration does not have as much effect as international collabora-tion on the productivity of scientists.

Communication, specifically via email and the Internet, is a power-ful factor in the collaboration endeavours of South African scientists. Collaborators, as opposed to non-collaborative scientists, are more familiar with email and the Internet and spend more time with these. Compared to other modes of communication—postal mail, telephony, fax and face-to-face interaction—email is the most popular medium for


research communication aiding the collaborative research interests of scientists. The preference for this medium is irrespective of sectoral affil-iation. Adequate infrastructure, access to modern ICTs and their effec-tive usage serve scientists in their collaborative research. In contrast to non-collaborative researchers, collaborators maintain diverse and larger networks. In certain circumstances, contacts and networks can lead to new collaborative alliances. Network size is determined by the extent of email communication. Collaboration and networks are interrelated. International networks positively affect international collaborations, and international projects have larger international networks.

Productivity of researchers often determines their collaborative/non-collaborative propensities. The variables that determine the productivity of scientists in the country include academic age, institutional experi-ence and international collaborative research projects. Networks do not influence productivity.

The scientific system

Science is a priority for any country. For a host of reasons science is on the decline in the African continent, the cradle of humankind. Since the 1990s, after active institutional development in the 1970s and 1980s (Arvanitis et al., 2000), the crisis has been brewing. The reasons for this decline range from a lack of resources (men and machines) to inade-quate working environments. Besides, there are causes such as diminish-ing resources in general and for science in particular, tyrannical rule in many countries, deterioration in teaching and research and the infec-tious demoralization of scientists (Zeleza, 2002). As the former deputy director general of UNESCO, aptly said, ‘Political independence without scientific knowledge and competence is as contradictory as the concept of a vegetarian tiger’ (cited in Odhiambo, 1967: 881).

Science in Africa is now centred in the northern and southern extrem-ities of the continent (Arvanitis et al., 2000). Within Africa, there is a clear division in scientific productivity between the northern African states and the sub-Saharan states. Egypt, Algeria, Mauritania, Libya, Morocco, and Tunisia are prominent among the northern African coun-tries, growing in scientific output. In spite of everything, many African states realize the importance of science and are charting ways to take the path of development and progress.

The location of a nation on the map of scientific research is deter-mined not solely by its scientific activity but also by factors that con-tribute to that very activity. Studies, as we have seen in the previous


chapters, have examined conditions that control the production of sci-entific knowledge. Research done by a nation is strongly influenced by its wealth, which is a prerequisite for scientific growth (Cole and Phelan, 1999). The gross national product (GNP) and the proportion spent on research and development (R&D) are some straightforward indicators of scientific activity. But there is more to it than these, and not all are obvi-ous or manifest. Intrinsically, scientific activity is an interaction between scientists in their socio-technological environment. Processes such as collaboration are part of this interaction, influencing the production of knowledge and the scientific wealth of nations. Collaboration, domestic or international, accelerates scientific growth and advancement.

South Africa has long recognized science and its indispensabil-ity to progress. This recognition has permeated to the stakeholders. Collaboration has been considered as part and parcel of science, which is now being promoted with added vigour. Over the years the scien-tific capabilities of South Africa have grown impressively, placing it on par with partners in collaborative ventures. Having the proven ability to produce world class science, the opportunities for collaboration are increasing in South Africa (Vaughan et al., 2007). More and more, the need for collaboration is seen at various levels, not just among scientists. For instance, the chief executive officer of Clothing, Textiles, Footwear and Leather publicly called for the need to establish international col-laborations with leading overseas universities that are in the forefront of cutting-edge research and training in these industries (The Mercury, 2007). One other reason for collaboration is the shortage South Africa is facing in skilled personnel such as technologists, artisans, managerial professionals and engineers.

Given the position of South African science in Africa, South Africa has a leading role to play in collaboration with other African countries. The country is seeking ways and means to enhance its R&D performance, which has been acknowledged as a key component in its strategic eco-nomic growth (Kahn, 2007). The proposed target of the national R&D strategy is in tandem with this. By 2012, it aims at increasing research and technology enablers (matriculates with university exemptions in mathematics and science) from the current level of 3.4 to 7.5 per cent; the proportion of science, engineering and technology tertiary students from 20 to 30 per cent; the number of science, engineering and technol-ogy practitioners from 7 to 11 per 10,000 workforce; the number of US patents of South African origin from 100 to about 200 and government R&D expenditure from 0.36 to 0.66 per cent of the GDP (Department of Science and Technology, 2002). Funding is also expected to grow.


In 2008–09 the government spent 2.9 per cent of the total national gov-ernment budget on science and technology activities, which increased to three per cent in 2012–13 (Department of Science and Technology, 2013).

In the research and development strategy document (Department of Science and Technology, 2002), South Africa explicitly recognizes that science is global (South Africa has a share of 0.5% of the global science) and its scientists have to be well connected with the world body, not only developing collaborations across the African continent but also tapping international resources. At the same time, more resources and time have to be invested in basic research which in 2001–02 was about 27 per cent of the total R&D.

Revitalizing the regional and national research institutes and fine-tuning higher education institutions in South Africa will boost scien-tific cooperation. This will also help the country to take advantage of its reserve of rare natural resources and data mines that are essential for cru-cial scientific discoveries in which many developed countries are inter-ested. A set of policies coupled with fitting structures can give support to collaborative initiatives. There are already several policy initiatives in this regard. The emphasis on collaborative efforts within and outside the country is quite clear in a number of policy documents. The Innovation Fund of the Department of Arts, Culture, Science and Technology has as its objective the advancement of transdisciplinary collaboration across sectors in South Africa (Letseka, 2005). Centres of excellence, research teams, research centres and work groups have been formed to facilitate and support collaborative efforts between disciplines, universities and industries, and with other institutions, regions and countries. As argued in a well-thought out policy document (Ellis, 1994), South Africa has to open its doors wide for a strong and steady inflow of ideas and scientists. Also heard is the call for policies and programmes to promote unim-peded movement of scientific and technical information to the national and international systems and encourage South African scientists to par-ticipate in national, regional and international collaborative ventures (Habib and Morrow, 2006; IDRC, 1993). Being a regional and a continen-tal power in matters of scientific research, South Africa stands to benefit from prospective alliances. Identification of weak points in this matter is also important for further improvement. The existing networks with Southern African Development Community (SADC) and New Partnership for African Development (NEPAD) have turned out to be unsustainable for want of resources (Department of Science and Technology, 2002), pre-venting effective collaboration with member countries.


More investment in R&D, international collaboration, development of skilled persons and further opening up of the South African knowl-edge economy are necessary (Kahn, 2007). The aging and shrinking sci-entific population is skewed by gender and race (Habib and Morrow, 2006; Kaplan, 2004). As seen in the share of the output of researchers who have crossed the age of 50, the scientific system warrants immedi-ate redress, preventing the decline of the country’s scientific potential (Habib and Morrow, 2006). In 2002, half of the scientific output of the country was from scientists over the age of 50 as against 18 per cent in 1990 (Department of Science and Technology, 2002).

South Africa has points of strength in astronomy, ecology, environ-mental science, natural sciences and plant and animal sciences. Rightly identified in the National Research and Development Strategy (NRDS; Department of Science and Technology, 2002), South Africa holds key areas that are geographically and technically advantageous. Access to clear southern skies and the technology to build telescopes, excellent sites for further excavations in human palaeontology, the presence of Cape Floral Kingdom (one of the seven most diverse floral kingdoms in the world), being the only African country with a presence on the Antarctic continent, the expanse of mines and avant-garde mining technologies, the strides achieved in microsatellite engineering, the suc-cessful feats in encryption technology, the competence in fluorine tech-nology in the uranium enrichment programme and the developments in HIV/AIDS research can work in favour of South Africa, nationally and internationally, inviting prospective partners.

The working environment in the higher education sector, a major contributor to scientific research, has been deteriorating over the years. Comparing the situation in 1995 and 2001, Webster and Mosoetsa (2002) warned of the decline on many fronts including the autonomy of higher education institutions. In the last two decades academic salaries in South Africa have dropped considerably and are way below that of many other countries (Kubler and Roberts, 2005 cited in Habib and Morrow, 2006), which is aggravated by the onerous and stressful working condi-tions (Habib and Morrow, 2006). The academics in South Africa have to teach, mark and administer more, leading to demoralization, greater stress and less productivity (Webster and Mosoetsa, 2002). The emerg-ing consensus is that academics and researchers have to be adequately remunerated and that there must be supportive institutional and work-ing environments (Habib and Morrow, 2006) for the entire scientific community. The conference under the auspices of the Department of Science and Technology in association with the Africa Institute and


the Human Sciences Research Council, which brought together gov-ernment officials, higher education managers, managers of the science councils, representatives from the private sector and civil society along with researchers and experts, has formulated a plan of action. This docu-ment, amongst others, strongly advocates the recruitment and reten-tion of high-level scientific and technological personnel in the country; promotion of partnerships between universities, research councils and the industry; engagement with scientific globalization such that South Africa can become a hub of research activities in appropriate areas to attract talented researchers and promotion of institutional collabora-tions within and across national boundaries (Habib and Morrow, 2006).

A major challenge for the country is to retain its researchers and scien-tists in their professions. For every thousand members in the workforce, South Africa has less than one researcher (Department of Science and Technology, 2002). As indicated in some recent figures, the attrition rate for researchers from government laboratories in the country is 11 per cent per annum while it is 15 per cent from the universities, causing concerns about the stability and growth of the nation’s scientific system.

The call by the NRDS of South Africa to double the skills and improve the system of the existing science and technology in the country (Kaplan, 2004) needs to be heeded. In the current scenario of science in the country, this can be accomplished without much difficulty. Identifying science-oriented students at schools, universities and from amongst those who have not had any formal education is the first step towards this target. Next comes the task of finding mentors who can groom the young scientific minds and direct them to the channels of advanced education and training. They also need to be introduced to proper insti-tutional settings wherein their talents and skills are nourished and devel-oped. For mentoring, the existing pool of researchers such as those rated by and associated with the National Research Foundation and those in the private sector can be used. Established scientists would find it pro-fessionally fulfilling to train the new generation of scientists in their own fields of specialization, as no one would wish to see their branch of specialization dying out for lack of personnel to carry the mantle. For those experienced and senior scientists it is a matter of commitment to their science.

In South Africa, for reasons of common focus and need and in view of the disappearing scientific skills, there is ample room for pooling together the locally and regionally dispersed skills and expertise in sci-ence and technology. Retaining the new and young talent in the cho-sen fields of interest is another challenge. Affiliated institutions of the


new breed of scientists have to provide a functional research environ-ment. Provision of funds for research and travel to conferences without bureaucratic hurdles is not a precondition but should be matched with a human touch for personal encouragement and support. Combined with this is an incentive mechanism that takes care of the research free-dom and autonomy, career advancement, encouragement and adequate remuneration. We have already seen the effect of a similar system in the country.

The crucial part that the scientists and researchers play for human-ity should be known beyond the confining walls of scientific institu-tions, for their role to be recognized by society. This makes scientists and researchers feel that the hard work they put in and the life they live in laboratories and libraries is appreciated by the outside world. The growth of science and its advancement in society, as evidence in the his-tory of science shows, rests largely on its acceptance as a social value and the encouragement it derives from the society. A better public image of scientists and researchers will inspire newcomers to join the band and strengthen the scientific system of South Africa. As Merton (1938 [1970]) noted, those societies that ascribe a high value to scientific activity will have a higher proportion of individuals pursuing a career in science.

231

1 Introduction

1 For a recent analysis of this, see Schroeder and Swedberg (2002).2 In 2004, Chinese scientists were ranked fifth in the world, accounting for

five per cent of the global publications in three major international indices: Science Citation Index (SCI), Engineering Index (EI), and Index to Science and Technology Proceedings (ISTP) (Guan and Ma, 2007).

3 The percentage shares of scientific output and GDP on R&D (in parenthesis) for other countries in May’s (1997) analysis are: UK, 8 (2.2); Japan, 7.3 (2.9); Germany, 7 (2.3); France, 5.2 (2.4); Canada, 4.5 (1.6); Italy, 2.7 (1.2); India, 2.4 (0.7); Australia, 2.1 (16); Netherlands, 2 (1.9); Sweden, 1.7 (3.3); Switzerland, 1.4 (2.7); China, 0.9 (0.5); Denmark, 0.8 (1.8); and Finland 0.7 (2.4).

4 In Mode 2, more knowledge, often transdisciplinary, is produced in the con-text of application, and the outcomes of research are influenced by social accountability (Gibbons et al., 1994).

5 More specifically, the study pertains to the strategies researchers follow in the collaboration choices that remain within the control of the individual researchers.

6 This finding is based on the analysis of co-authored publications during 1990–2000, in which multiple co-authorships are found to be centred in regional hubs.

2 Science in Africa and in South Africa: A Historical Review

1 The CCTA originated as an international agreement in London between the governments of Belgium, Rhodesia (now Zimbabwe), Nyasaland (now Malawi), France, Ghana, Guinea, Liberia, Portugal, the UK and South Africa.

2 EBED was later renamed as the Inter-African Bureau of Animal Health (IBAH) in 1960.

3 South Africa had an active participation in the functions of CCTA and SARCCUS.

4 Arvanitis et al. (2000), analyzing the PASCAL database for 1991 and 1996, have found that more papers than in other subjects had been published in medical biology, clinical medicine, biology and in agricultural sciences. A surprising increase has been noted in engineering science mainly because of the expansion of engineering sciences in North Africa.

5 This study referred to the period 1991–97 and examined the 15 most pro-ductive African countries of South Africa, Egypt, Nigeria, Tunisia, Morocco, Kenya, Algeria, Cameroon, Zimbabwe, Senegal, Ivory Coast, Tanzania, Ethiopia, Ghana and Sudan.

Notes

232 Notes

6 Tijssen (2007: 305) cited this figure from the United Nations’ Institute of Statistics’ Bulletin on Science and Technology Statistics.

7 These were among those that were heralded as the beacons of progress in the continent (Hassan, 2001).

8 The period from 1652 had been eventful for this region. Between 1659 and 1803, the European settlers conquered the inhabitants—San (Bushmen) and Khoi (Hottentots)—and wars ensued between the settlers and the original inhabitants. The Cape then fell into the hands of the British and became its colony from 1806 to1910, after a period of Dutch rule from 1652 to 1795.

9 The museum, opened by Lord Charles Somerset, has its antecedents in a small museum established by Willem Adriaan Van der Stel in Cape Town in the 18th century (Naudé and Brown, 1977).

10 In the field of agriculture, for instance, there were associations including the South African Veterinary Association (1919), the Entomological Society of Southern Africa (1937), the Soil Science Society of Southern Africa (1953), the South African Society of Animal Production (1961), the South African Society of Plant Pathology and Microbiology (1962), the South African Institute of Agricultural Engineers (1964), the South African Institute for Agricultural Extension (1966), the South African Society of Dairy Technology (1966), the Grassland Society of Southern Africa (1966), the South African Institute of Forestry (1968) and the South African Society of Crop Production (1971) (Joubert, 1977).

11 Included in the themes for presentations were atmospheric electricity, binary star systems, meteorology, heredity, the bacterial system for sewage purifica-tion, stock diseases, ground water resources, forestry and language (Plug, 2003).

12 Modelled on the British Association, its membership continued to soar year after year, from 268 in 1902 to 765 in 1903. However, Rich (1990) recorded that the Association was founded in 1903.

13 The University of the Orange Free State (originating from a college started in 1855), Potchefstroom University for Higher Education (first a college in 1869), the University of Pretoria (originally the Transvaal University College), Victoria College (which became the University of Stellenbosch) (1918), the University of Port Elizabeth (1966), the Rand Afrikaans University, Rhodes University (initially a college formed in 1904), the University of Natal (origi-nally a college started in 1910) and the University of Witwatersrand (originat-ing from a college in 1922), and the University of South Africa (originating in 1873, and which later, in 1946, became a correspondence and examining institution taking only external students) were among those that catered to either Afrikaans-speaking people and/or the English-speaking population (Naudé and Brown, 1977; Nordkvelle, 1990).

14 South Africa became the Union of South Africa in 1910 as an autonomous dominion.

15 The National Botanic Garden at Kirstenbosch in Cape Town came into being in 1913. Around this time, South Africa started a National Standards Testing and Investigational Bureau to further its capacity in the testing and the cali-bration of precision instruments and scientific apparatus (Science, 1944).

16 The first fossils were discovered in 1830 (Cluver and Barry, 1977).17 Included are the South African Quarterly Journal (1830), the Cape Monthly

Magazine (1857), Transactions, the Annual Report of the Transvaal Geological

Notes 233

Survey (1903), the South African Architect and Builder (1903), the Transvaal Agricultural Journal (1903), the Journal of the Institute of Land Surveyors of the Transvaal (1905), the Journal of the South African Ornithologists’ Union (1905), the Transvaal Medical Journal (1905), the Onderstepoort Journal, the Annals of the Natal Museum (1906), the Annals of the South African Museum, the Records of the Albany Museum (1903), the Agricultural Journal of the Cape of Good Hope, the Annals of the Cape Observatory, the Transactions of the South African Philosophical Society, the South African Journal of Science, Agricultural Journal, the Journal of the Department of Agriculture, Farming in South Africa, the South African Medical Journal (1903), the South African Medical Record and the South African Journal of Agricultural Sciences (1958). Some of these journals have been either renamed or merged.

18 In several erstwhile British colonies including India, Australia and Canada, these types of institutions are quite common and perform an active role in the country’s research programme (Department of Science and Technology, 2002).

19 The report, ‘The Effects of Apartheid on Education, Sciences, Culture and Information’ is a working document prepared by the UNESCO for the UN Special Committee on Apartheid.

20 The point the authors make here is that a certain amount of submissions to international journals might have been rejected for policy reasons, during the apartheid era.

21 See more on this from the lecture of John Pratt (South African Journal of Science, 1977).

22 The universities are: Cape Peninsula University of Technology, Central University of Technology, Free State, Durban Institute of Technology, Mangosuthu University of Technology, Nelson Mandela Metropolitan University, North West University, Rhodes University, Tshwane University of Technology, University of Cape Town, University of Fort Hare, University of the Free State, University of Johannesburg, University of KwaZulu-Natal, University of Limpopo, University of Pretoria, University of South Africa, University of Stellenbosch, University of Venda, University of Western Cape, Witwatersrand University, University of Zululand, Vaal University of Technology and Walter Sisulu University of Science and Technology, Three more universities, one at Mpumalanga, another at the Northern Cape and a medical in Gauteng, have been established.

23 The National Research Foundation (NRF) in its database lists 95 profes-sional organizations, which also include non-scientific organizations such as Alzheimer’s South Africa, the Nursing Organization and the Nutritional society. These have been excluded in our counting, and 68 scientific organi-zations that are truly scientific in their activities were identified.

24 For more details on the research of Sydney Brenner that won him the cov-eted prize, see Manchester (2003).

25 King’s study (2004) of 31 countries including G8 countries and European Union countries showed that they account for more than 98 per cent of the world’s highly cited papers while the remaining 162 contributed less than two per cent. The countries covered in this study are Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland, France, Germany, Greece, India, Iran, Ireland, Israel, Italy, Japan, Luxembourg, the Netherlands, Poland,

234 Notes

Portugal, Russia, Singapore, Spain, South Africa, South Korea, Sweden, Switzerland, Taiwan, the UK and the US.

26 King’s study on this finding in particular is based on the number of publica-tions indexed by the ISI and referring to the period of 1993–2002.

27 India’s economic growth, as in China years ago, is set to increase. India has developed its science base very rapidly, and this has effectively contributed to its economic growth (King, 2004). India’s major science institutes (IITs, IIMs and IISc, for instance) have high international standards and have pro-duced high-quality graduates (King, 2004). A considerable brain drain from these national high quality institutes to developed countries, which is still occurring, has negatively affected the scientific productivity of India. A large number of established scientists who presently work in countries like the US are of Indian origin. NSF (2004) reported that 30 per cent of the PhD stock in the US is foreign-born.

28 http://www.dst.gov.za/index.php/internatprog, accessed 3 May 2014.

3 Scientific Collaboration: Towards Conceptual Clarity

1 Another aspect of this view is that collaboration can improve collective crea-tivity but may inhibit individual creativity and inventiveness in research (Fox and Faver, 1984).

2 In the UK, for instance, research transcending the traditional boundaries of disciplines in the 1980s has increased more rapidly than disciplinary research (Hicks and Katz, 1996).

3 Nonaka (1994) notes that the advantage of interdisciplinary collaborations is in its ability to enhance the interplay between tacit and explicit knowledge.

4 Many have documented the interlinked concepts of collaboration and sci-ence. See Ma and Guan (2005) for a review of some of those studies.

5 Collaboration, partnership, alliance, cooperative research, coalition and tie-ups are used interchangeably in this book to avoid the repetitive use of the word collaboration but without losing its meaning.

6 An invisible college is a community of scientists, active in the same field but in different institutions (Kretschmer, 1999).

7 Collaboratory is a computer-supported system which allows scientists to work with each other using facilities and databases without travelling to distant locations. It is a centre without walls wherein scientists interact with their fellow scientists and access instruments, information and data. Collaboratories allow scientists to use remote libraries, collaborate with colleagues in remote locations and interact with remote instruments and analyse data (Wulf, 1993). The term was first discussed by Lederberg and Uncapher in 1989 (Lederberg and Uncapher, 1989, cited in Finholt and Olson, 1997).

8 Scientific and technical human capital, according to Bozeman et al. (2001), is the sum of scientific, technical and social knowledge, and skills and resources embodied that a particular individual brings to the collaborative effort.

9 Hicks et al. (1996) find in their study that companies that specialize in physi-cal or biomedical science prefer collaboration. See also Bush (1957).

Notes 235

10 Godin’s inference is built upon the huge number of papers and patents pro-duced by multinationals.

11 The rise in university–industry collaboration is a matter of contention because of their differing goals, motivations of research and the levels of research freedom. See Yang (1986) for more such types of collaborations.

12 During 1991–97, the support extended to universities by industry rose to 20 per cent (USD 1.05 million), 6.5 per cent of all the basic research expendi-ture. At the same time, federal funding to R&D at universities continued (National Science Board, 1998, cited in Bozeman, 2000). On the publica-tion front, 6 per cent of all academic publications in 1995 were produced by industry scientists (Bozeman, 2000).

13 This inference is based on the analysis of co-authored papers in the fields of earth and space, mathematics, physics, biomedicine, biology, chemistry, engineering and technology, and clinical medicine. The first three fields in this list have recorded the most collaboration (Luukkonen et al., 1992). Frame and Carpenter (1979) have also come up with a similar finding in their study of international co-authorship patterns in 11 countries.

14 Region-wise, the US’ alliances with Asia have increased rapidly, followed by Europe and other regions. The foreign share of institutional collaborations in the 1980s and 1990s rose to 5.11 and 7.41 per cent per annum, respectively, accumulating a growth of 45 per cent (Adams et al., 2005).

15 Apart from this and collaboration between European scientists, there are organizations such as the European Southern Observatory (ESO) that build stronger ties between nations (Katz and Martin, 1997).

16 This refers to the study of 34 major European and Japanese companies in the pharmaceutical and electronic sectors.

17 In this regard, the EU’s share of collaborative papers with countries has grown between 1985 and 1995: with Australia, from 0.45 per cent to 0.88 per cent; with Canada, from 0.91 per cent to 1.61 per cent; with Japan, from 0.52 per cent to 1.33 per cent; with the US, from 5.89 per cent to 9.52 per cent; and with Korea, from 0.02 per cent to 0.23 per cent.

18 The study of select European and Japanese firms in the pharmaceutical and electronic sectors shows that the number of collaborative papers during 1980–89 has increased far more than that of non-collaborative papers (Hicks et al., 1996).

19 Japan is actively collaborating with the US and European countries chiefly in the fields of natural sciences. Known for the warmth and appealing man-nerisms, the Japanese offer amenities and necessities to host international collaborations (Taubes, 1994).

20 The size of the country’s scientific efforts supposedly has an influence on the country’s international collaboration (Frame and Carpenter, 1979).

21 Hafernik et al. (1997), calculating the number of papers in the six volumes (1997) of the journal Science, report that 97 per cent of the articles were co-authored. The absolute number of co-authored papers has also risen consid-erably in the UK, during 1981–91 (Hicks and Katz, 1996).

22 The number of international air passengers carried from the UK increased by 87 per cent between 1981–83 and 1989–91, international telephone calls by 138 per cent (Hicks and Katz, 1996).

236 Notes

23 The project formation and composition dimension refers to the origin and constitution of collaboration in which one sector is dominant. Magnitude is the size of the collaboration. In other words, it takes into account the num-ber of institutions, partners, subcontractors and graduate students involved in the collaboration. Interdependence is seen in data sharing, autonomy of organizational teams and in the analysis of joint data. To carry on with the communication needs, a centre is used, mainly for a more or less informal type of communication, or a public relations officer will be in charge of the communication. Bureaucratic structure is conceived of in a number of ways. There can be a lead centre with designated scientific and adminis-trative leaders and a clear division of authority. Participation concerns par-ticipants such as graduate students, principal investigators and institutions. Technological devices of varying dependence and form are used in collabora-tion (Chompalov and Shrum, 1999).

24 Adams et al. (2005) are referring to their study of 110 top research universi-ties in the US, for the period 1981–99.

25 This classification by Wagner (2005) entails collaboration in sciences such as biomedics, genetics, demography, computer, epidemiology and virology as data driven; oceanography, geology, seismology, zoology, social science and anthropology as resource-driven; high energy physics, astronomy, energy, avionics, polymers and manufacturing as equipment-driven; and mathemat-ics, economics, sociology, anthropology, science studies and philosophy as idea/theory-driven.

26 Beaver and Rosen (1979) record that collaboration was started as a response to the professionalization of science in the 18th and 19th centuries.

27 The variables are: the length of time that the respondent has known the per-son; whether the collaboration was requested by someone in administration; whether the collaboration was to help junior colleagues; whether the collabo-ration was to help graduate students; the strength of the collaborator’s science reputation; the complementary skills of respondent and collaborator; qual-ity of previous collaboration; how far collaboration is perceived as fun and entertaining; collaborator’s fluency in respondent’s language; whether the respondent and collaborator have the same nationality; the strength of the collaborator’s work ethics; the collaborator’s ability to stick to the sched-ule; and the collaborator’s ability to assign credit (Bozeman and Corley, 2004).

28 This finding comes out of the empirical study of 86 biopharmaceutical prod-uct development projects.

29 For more on this dimension see Hamel (1991).30 Their classification is again sub-grouped into the factors that are internal and

those that are external to science.31 Hagstrom (1964) speaks of two mutually exclusive traditional forms of scien-

tific teamwork, free collaboration of peers and the professor–student group.32 The multi-disciplinary nature and the need for various resources make bio-

technology a collaborative discipline (Oliver, 2004).33 Blumenthal et al. (1986) show in a study of 1,200 faculty members in 40

major universities in the US that their respondents who enjoy the support of industry not only publish more but also participate in more administra-tive and professional activities and earn more than their colleagues without support.

Notes 237

34 Hafernik et al. (1997: 33) narrate their personal experience of collaboration thus: ‘[w]e three have found that collaborating on research and writing pro-jects has enriched our personal and professional lives and has helped us to make contributions to our institution, profession and community.’

35 It is more applicable in industries but very relevant in any scientific collaboration.

36 Presser’s study relies on submissions to a journal Sociometry (now Social Psychology Quarterly) for a select period. The percentage of initial rejections of single-author papers was 66.9 as against 52.6 for multi-author papers with an acceptance rate of 18.3 and 22.6 percentages respectively.

37 This refers to the findings of a study of nine government-owned research institutes in New Zealand by Goldfinch et al. (2003), who also note that the citation rates are higher for articles with higher levels of collaboration and greater geographical and institutional spread of co-authors.

38 This is based on the calculation of: C = Nm/(Nm + Ns), where C is the degree of collaboration, Nm is the number of multi-authored papers and Ns is the number of single-authored papers.

39 Partners are counselled for successful collaborations at regular meetings (Hafernik et al., 1997).

4 Research Publications of South African Scientists, 1945–2010

1 See Bozeman and Corley, 2004; Egghe and Rousseau, 1990; Frame and Carpenter, 1979; Fox and Faver, 1984; Hicks and Katz, 1996; Katz and Martin, 1997; Luukkonen et al., 1992; Ma and Guan, 2005; May, 1997; Moed and Hesselink, 1996; Newman, 2001; Schott, 1993; and Tijssen, 2007.

2 Erdös was an itinerant Hungarian mathematician who apparently spent a large portion of his life living out of a suitcase and writing papers with those who were willing to offer him room and board. He became the mathemati-cian who published more than any other mathematician, except Leonhard Euler (Newman, 2001).

3 See Adams et al., 2005; Anuradha and Urs, 2007; Arvanitis et al., 2000; Basu and Aggarwal, 2001; Ben-David, 1960; Frame, 1979; Frame and Carpenter, 1979; Glänzel et al., 1999; Godin, 1996; Goldfinch et al., 2003; Gómez et al., 1999; Guan and Ma, 2007; Gupta and Dhawan, 2003; Hicks and Katz, 1996; Hicks et al., 1996; Katz and Hicks, 1997; Katz and Martin, 1997; Kim, 2006; Larivière et al., 2006; Luukkonen et al., 1992; Lundberg et al., 2006; Laudel, 2001, 2002; Ma and Guan, 2005; May, 1997; Melin, 2000; Moed and Hesselink, 1996; Newman, 2001, 2004; Narváez-Berthelemot et al., 2002; Pouris, 2003; Rabkin et al., 1979; Smith, 1958; Sooryamoorthy, 2009a, 2009b; Stokes and Hartley, 1989; Wagner and Leydesdorff, 2005b; Okubo and Sjöberg, 2000; Price, 1963; Price and Beaver, 1966; Porac et al., 2004; Wagner-Döbler, 2001; Wagner, 2005; Wilson and Markusova, 2004; Yoshikane and Kageura, 2004; and Zitt et al., 2000.

4 It is now called the Department of Higher Education and Training.5 For instance, see Adams et al., 2005; Anuradha and Urs, 2007; Basu and

Aggarwal, 2001; Ben-David, 1960; Cole and Phelan, 1999; Arenas et al., 1999;

238 Notes

Frame et al., 1977; Glänzel and de Lange, 2002; Glänzel et al., 1999; Goldfinch et al., 2003; Gómez et al., 1999; Gupta and Dhawan, 2003; Hicks et al., 1996; Ingwersen and Jacobs, 2004; Jacobs, 2006; Jacobs and Ingwersen, 2000; Katz and Hicks, 1997; Katz and Martin, 1997; Kim, 2006; King, 2004; Larivière et al., 2006; Laudel, 2001, 2002; Ma and Guan, 2005; Melin, 2000; Mêgnigbêto, 2013; Moed and Hesselink, 1996; Negishi et al., 2004; Newman, 2004; Okubo and Sjöberg, 2000; Ovens, 1995; Porac et al., 2004; Pouris, 2003, 2006a; Stokes and Hartley, 1989; Torricella-Morales et al., 2000; Tijssen, 2007; Wagner and Leydesdorff, 2005b; Wilson and Markusova, 2004.

6 The degree of clustering is the probability that two of a scientist’s collabora-tors have themselves collaborated (Newman, 2001).

7 Godin et al. (1995) examined four sectors of universities, hospitals, industry and government.

8 The disciplines included are agriculture, astronomy, biochemistry, biology, biotechnology, chemistry, computer science, materials science, mathemat-ics, medicine, neuroscience, oncology, paediatrics, pharmacology, physics, plant sciences, psychiatry, surgery, veterinary science and zoology. This clas-sification, in principle, has been adopted with amendments for our analysis in this chapter.

9 The categories are: all documents, article, abstract of published item, art exhibit review, bibliography, biographical item, book review, chronology, correction, correction–addition, dance performance review, database review, discussion, editorial material, excerpt, fiction, creative prose, film review, hardware review, item about an individual, letter, meeting abstract, meeting summary, music performance review, music score, music score review, news item, note, poetry, record review, reprint, review, script, software review, TV review, radio review and theatre review.

10 Some bibliometricians (Kim, 2006, for instance) also favour articles, reviews and notes in their analyses. Reviews refer to research.

11 ‘Scholar’ is a generic term used here to denote scientists and researchers in universities, technikons, research institutes, government, industry and hos-pitals. This generic labelling is necessary as we do not specifically examine the affiliation details of the authors of all papers as they are primarily incon-sequential in this analysis.

12 This refers to the doctorates from the University of Cape Town.

7 Communication, Professional Networks and Productivity

1 The sample size is 1,400 scientists in the chosen disciplines in select European countries of Denmark, Germany, Ireland, Italy, the Netherlands, Switzerland and the UK.

239

Abramo, G., D’Angelo, C. A., & Costa, F. D. (2009). Research collaboration and productivity: is there correlation? Higher Education, 57(2), 155–171.

Abramo, G., D’Angelo, C. A., & Solazzi, M. (2011). The relationship between sci-entists’ research performance and the degree of internationalization of their research. Scientometrics, 86(3), 629–643.

Acosta, M., Coronado, D., Ferrándiz, E., & León, M. D. (2011). Factors affecting inter-regional academic scientific collaboration within Europe: the role of eco-nomic distance. Scientometrics, 87(1), 63–74.

Adams, J. D., Black, G. C., Clemmons, J. R., & Stephen, P. E. (2005). Scientific teams and institutional collaborations: evidence from U.S. universities, 1981–1999. Research Policy, 34(3), 259–285.

Alberts, B., & Shine, K. (1994). Scientists and the integrity of research. Science (New Series), 266(5191), 1660.

Alter, C., & Hage, J. (1993). Organizations Working Together: Coordination in Interorganizational Networks. Newbury Park, CA: Sage Publications.

Anuradha, K. T., & Urs, S. R. (2007). Bibliometric indicators of Indian research col-laboration patterns: a correspondence analysis. Scientometrics, 71(2), 179–189.

Arenas, J. L. D., Valles, J., & Arenas, M. (1999). Profile of the Mexican health sci-ences elite: a bibliometric analysis of research performance. Scientometrics, 46(3), 539–547.

Arrow, K. (1974). The Limits of Organization. New York: Norton.Arunachalam, S., & Doss, M. J. (2000). Mapping international collaboration in

science in Asia through coauthorship analysis. Current Science, 79(5), 621–628.Arvanitis, R., Waast, R., & Gaillard, J. (2000). Science in Africa: a bibliometric

panorama using PASCAL database. Scientometrics, 47(3), 457–473.Atkinson, P., Batchelor, C., & Parsons, E. (1998). Trajectories of collaboration and

competition in medical discovery. Science, Technology and Human Values, 23(3), 259–284.

Balakrishnan, S., & Koza, M. (1993). Information asymmetry, adverse selection and joint ventures. Journal of Economic Behaviour and Organization, 20(1), 99–117.

Barabasi, A. L., Jeong, H., Neda, Z., Ravasz, E., Schubert, A., & Vicsek, T. (2002). Evolution of the social network of scientific collaborations. Physica A: Statistical Mechanics and its Applications, 311(3–4), 590–614.

Barjak, F. (2006). Research productivity in the internet era. Scientometrics, 68(3), 343–360.

Barnard, H., Cowan, R., & Müller, M. (2012). Global excellence at the expense of local diffusion, or a bridge between two worlds? Research in science and tech-nology in the developing world. Research Policy, 41(4), 756–769.

Bartneck, C., & Hu, J. (2010). The fruits of collaboration in a multidisciplinary field. Scientometrics, 85(1), 41–52.

Basu, A., & Aggarwal, R. (2001). International collaboration in science in India and its impact on institutional performance. Scientometrics, 52(3), 379–394.

References

240 References

Baym, N. K., Zhang, Y. B., & Lin, M.-C. (2004). Social interactions across media: interpersonal communication on the internet, telephone and face-to-face. New Media and Society, 6(3), 299–318.

Beaver, D. D. (2001). Reflections on scientific collaboration (and its study): past, present and future. Scientometrics, 52(3), 365–377.

Beaver, D. D., & Rosen, R. (1978). Studies in scientific collaboration, Part I: the professional origins of scientific co-authorship. Scientometrics, 1(1), 65–84.

Beaver, D. D., & Rosen, R. (1979). Studies in scientific collaboration, Part II: scien-tific coauthorship, research productivity and visibility in the French scientific elite, 1799–1830. Scientometrics, 1(2), 133–149.

Belkhodia, O., & Landry, R. (2007). The triple-helix collaboration: why do researchers collaborate with industry and the government? what are the factors that influence the perceived barriers? Scientometrics, 70(2), 301–332.

Benavent-Pérez, M., Gorraiz, J., Gumpenberger, C., & Moya-Anego, F.d. (2012). The different flavors of research collaboration: a case study of their influence on university excellence in four world regions. Scientometrics, 93(1), 41–58.

Ben-David, J. (1960). Scientific productivity and academic organization in nine-teenth century medicine. American Sociological Review, 25(6), 828–843.

Berjak, P., Cowan, D., Cowling, R., Pammenter, N., & Scholes, R. (2010). Biological sciences. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 133–163). Pretoria: ASSAf.

Bickel, W. E., & Hattrup, R. A. (1995). Teachers and researchers in collaboration: reflections on the process. American Educational Research Journal, 32(1), 35–62.

Blankley, W., & Kahn, M. (2004). South African research and development: pre-liminary results and indicators from the latest survey. South African Journal of Science, 100(1 & 2), 9–11.

Blumenthal, D., Gluck, M., Louis, K. S., Stoto, M. A., & Wise, D. (1986). University-industry research relationships in biotechnology: implications for the univer-sity. Science (New Series), 232(4756), 1361–1366.

Bok, B. J. (1955). Science in international cooperation. Science (New Series), 121(3155), 843–847.

Bonaccorsi, A., Daraio, C., & Simar, L. (2006). Advanced indicators of productiv-ity of universities: an application of robust nonparametric methods to Italian data. Scientometrics, 66(2), 389–410.

Bordons, M., Gómez, I., Fernández, M. T., Zulueta, M. A., & Méndez, A. (1996). Local, domestic, and international scientific collaboration in biomedical research. Scientometrics, 37(2), 279–295.

Boshoff, N. (2009). South-South research collaboration of countries in the Southern African Development Community (SADC). Scientometrics, 84(2), 481–503.

Bozeman, B. (2000). Technology transfer and public policy: a review of research and theory. Research Policy, 29(4–5), 627–655.

Bozeman, B., & Boardman, C. (2003a). Research and technology collaboration and linkages: implications from two U.S. case studies. Georgia Institute of Technology, http://www.csta-cest.ca/files/usa.pdf. Accessed 24 April 2007.

Bozeman, B., & Boardman, C. (2003b). Managing the New Multipurpose, Multidiscipline University Research Centers: Institutional Innovation in the Academic Community. Washington, DC: IBM Center for the Business of Government.

Bozeman, B., & Corley, E. (2004). Scientists’ collaboration strategies: implications for scientific and technical human capital. Research Policy, 33(4), 599–616.

References 241

Bozeman, B., Dietz, J., & Gaughan, M. (2001). Scientific and technical human capital: an alternative model for research evaluation. International Journal of Technology Management, 22(7/8), 636–655.

Brock, J. F. (1977). Some highlights of medical research in South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 192–213). Cape: Royal Society of South Africa.

Brown, K. (2005). Tropical medicine and animal diseases: Onderstepoort and the development of veterinary science in South Africa 1908–1950. Journal of South African Studies, 31(3), 513–529.

Browne, C. A. (1936). Academies of science and the cooperative spirit in scientific research. Science (New Series), 84(2166), 1–7.

Bruce, D. (1905a). The advance of our knowledge of the causation and methods of prevention of stock diseases in South Africa during the last ten years. Science (New Series), 22(558), 289–299.

Bruce, D. (1905b). The advance of our knowledge of the causation and methods of prevention of stock diseases in South Africa during the last ten years. Science (New Series), 22(559), 327–333.

Bunt, E. A. (1977). Some highlights of engineering research in South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 427–453). Cape: Royal Society of South Africa.

Burns, M., Audouin, M., & Weaver, A. (2006). Advancing sustainability science in South Africa. South African Journal of Science, 102(9/10), 379–384.

Bush, G. P., & Hattery, L. H. (1956). Teamwork and creativity in research. Administrative Science Quarterly, 1(3), 361–372.

Bush, V. (1957). Professional collaboration. Science, 125(3237), 49–54.Butler, D. M., & Butler, R. J. (2011). The internet’s effect on women’s coauthoring

rates and academic job market decisions: the case of political science. Economics of Education Review, 30(4), 665–672.

Callon, M. (1994). Is science a public good? Science, Technology and Human Values, 19(4), 395–424.

Chen, W., Boase, J., & Wellman, B. (2002). The global villagers: comparing inter-net users and users around the world. In B. Wellman & C. Haythornthwaite (Eds.), Internet in Everyday Life (pp. 74–113). Malden, MA: Blackwell.

Cherry, M. (2010). South African science: black, white and grey. Nature, 463, 726–728.

Chin, G., Myers, J., & Hoyt, D. (2002). Social networks in the virtual science labo-ratory. Communications of the ACM, 45(8), 87–92.

Chompalov, I., Genuth, J., & Shrum, W. (2002). The organization of scientific collaborations. Research Policy, 31(5), 749–767.

Chompalov, I., & Shrum, W. (1999). Institutional collaboration in science: a typology of technological practice. Science, Technology and Human Values, 24(3), 338–372.

Clark, C., Herter, R. J., & Moss, P. A. (1998). Continuing the dialogue on collabo-ration. American Educational Research Journal, 35(4), 785–791.

Cluver, M. A., & Barry, T. H. (1977). Advances in South African vertebrate paleon-tology. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 318–338). Cape: Royal Society of South Africa.

Cohen, J. (2000). Balancing the collaboration equation. Science, 288(5474), 2155–2159.

242 References

Cole, J. R., & Cole, S. (1973). Social Stratification in Science. Chicago: University of Chicago Press.

Cole, S. (1992). Making Science: Between Nature and Society. Cambridge, MA: Harvard University Press.

Cole, S., & Cole, J. R. (1967). Scientific output and recognition: a study of the operation of the reward system in science. American Sociological Review, 32(3), 377–390.

Cole, S., & Phelan, T. J. (1999). The scientific productivity of nations. Minerva, 37(1), 1–23.

Corley, E. A., Boardman, P. C., & Bozeman, B. (2006). Design and the manage-ment of multi-institutional research collaborations: theoretical implications from two case studies. Research Policy, 35(7), 975–993.

Cronin, B., Shaw, D., & LaBarre, K. (2003). A cast of thousands: coauthorship and subauthorship collaboration in the 20th century as manifested in the scholarly journal Literature of Psychology and Philosophy. Journal of the American Society for Information Science and Technology, 55(2), 160–168.

Cummings, L. L., & Bromiley, P. (1996). The organizational trust inventory (OTI): development and validation. In R. M. Kramer & T. R. Tyler (Eds.), Trust in Organizations: Frontiers of Theory and Research (pp. 302–330). Thousand Oaks, CA: Sage.

Dangel, E. T., & Mishkin, B. (1996). Collaboration and data sharing: continued. Science (New Series), 271(5246), 129–130.

Davidson, T., Sooryamoorthy, R., & Shrum, W. (2002). Kerala connections: will the internet affect science in developing areas? In B. Wellman & C. Haythornthwaite (Eds.), The Internet in Everyday Life (pp. 496–519). Oxford: Blackwell.

Day, J. H. (1977). Marine biology in South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 86–108). Cape: Royal Society of South Africa.

Defazio, D., Lockett, A., & Wright, M. (2009). Funding incentives, collaborative dynamics and scientific productivity: evidence from the EU framework pro-gram Research Policy, 38(2), 293–305.

Department of Science and Technology. (2002). South Africa’s National Research and Development Strategy. Pretoria: Department of Science and Technology, http://www.dst.gov.za/publications/reports.php. Accessed 31 August 2007.

Department of Science and Technology. (2006). National Survey of Research and Experimental Development (R&D), 2004–5. Pretoria: Department of Science and Technology.

Department of Science and Technology. (2013). 2012/13 Report on Public Funding for Scientific and Technological Activities. Pretoria: Department of Science and Technology, Republic of South Africa.

Department of Science and Technology & HSRC. (2013). National Survey of Research and Experimental Development High-level Key Results (2009/10 Fiscal Year). Pretoria: Department of Science and Technology and HSRC.

Diab, R. (2010). Conclusion. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 277–284). Pretoria: ASSAf.

Dienel, H.-L. (1999). Bilateral scientific and technical collaboration between hos-tile countries in Europe: France and Germany 1860–1950. Innovation, 12(4), 517–524.

References 243

Dillon, W. S. (1966). Africa science board, national academy of sciences, USA. The Journal of Modern African Studies, 4(1), 98–101.

Dimmick, J., Kline, S., & Stafford, L. (2000). The gratification niches of personal e-mail and the telephone: competition, displacement, and complementarity. Communication Research, 27(2), 227–248.

Dodgson, M. (1991). Technological learning, technology strategy and competi-tive pressures. British Journal of Management, 2(3), 133–149.

Dubow, S. (1995). Scientific Racism in Modern South Africa. New York: Cambridge University Press.

Duque, R. B., Sooryamoorthy, R., Ynalvez, M., Mbatia, P., Dzorgbo, D.-B., & Shrum, W. (2005). Collaboration paradox: scientific productivity, the internet, and problems of research in developing areas. Social Studies of Science, 35(5), 755–785.

Duque, R., Miller, P., Omar Barriga, J., Shrum, W., & Henríquez, G. (2012). Is inter-net use associated with reporting fewer problems in collaboration? Evidence from the scientific community in Chile. Science Communication, 34(5), 642–678.

Edge, D. O., & Mulkay, M. J. (1976). Astronomy Transformed: The Emergence of Radio Astronomy in Britain. New York: Wiley-Interscience.

Egghe, L., & Rousseau, R. (1990). Introduction to Informatics. Amsterdam: Elsevier.Ehikhamenor, F. A. (2003). Internet resources and productivity in scientific

research in Nigerian universities. Journal of Information Science, 29(107), 107–116.

Eisemon, T. O., & Davis, C. H. (1992). Universities and scientific research capac-ity. Journal of Asian and African Studies, 27(1–2), 68–93.

Ellis, G. F. R. (1994). Science Research Policy in South Africa. Rondesbosch: University of Cape Town.

Ervin, E., & Fox, D. L. (1994). Collaboration as a political action. JAC: A Journal of Advanced Composition, 14(1), 53–71.

Eslami, H., Ebadi, A., & Schiffauerova, A. (2013). Effect of collaboration network structure on knowledge creation and technological performance: the case of biotechnology in Canada. Scientometrics, 97(1), 99–119.

Etzkowitz, H., & Leydesdorff, L. (1998). The endless transition: a “triple helix” of university-industry-government relations. Minerva, 36(3), 203–208.

Etzkowitz, H., & Leydesdorff, L. (2000). The dynamic of innovation: from national systems and “Mode 2” to a triple helix of university-industry-government relations. Research Policy, 29(2), 109–123.

Evans, D. S., van Zinderen Bakker Sr., E. M., & Greenwald, L. (1972). The question of South Africa. Science (New Series), 175(4023), 705.

Farahat, H. (2002). Authorship patterns in agricultural sciences in Egypt. Scientometrics, 55(2), 157–170.

Farrell, M. P. (2001). Collaborative Circles: Friendship Dynamics and Creative Work. Chicago: University of Chicago Press.

Finholt, T. A., & Olson, G. M. (1997). From laboratories to collaboratories: a new organizational form for scientific collaboration. Psychological Science, 8(1), 28–36.

Forbes, V. S. (1977). Some scientific matters in early writings on the Cape. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 33–59). Cape: Royal Society of South Africa.

244 References

Fox, M. F. (1992). Research, teaching, and publication productivity: mutuality versus competition in academia. Sociology of Education, 65(4), 293–305.

Fox, M. F., & Faver, C. A. (1984). Independence and cooperation in research: the motivations and costs of collaboration. The Journal of Higher Education, 55(3), 347–359.

Frame, D. J., Narin, F., & Carpenter, M. P. (1977). The distribution of world sci-ence. Social Studies of Science, 7(4), 501–516.

Frame, J. D. (1979). National economic resources and the production of research in lesser developed countries. Social Studies of Science, 9(2), 233–246.

Frame, J. D., & Carpenter, M. P. (1979). International research collaboration. Social Studies of Science, 9(4), 481–497.

Fujimura, J. (1988). The molecular bandwagon in cancer research: where social worlds meet. Social Problems, 35(3), 261–283.

Gaillard, J. (1992). Science policies and cooperation in Africa: trends in the production and utilization of knowledge. Science Communication, 14(2), 212–233.

Gazni, A., & Didegah, F. (2011). Investigating different types of research collabo-ration and citation impact: a case study of Harvard University’s publications. Scientometrics, 87(2), 251–265.

Genuth, J., Chompalov, I., & Shrum, W. (2000). How experiments begin: the formation of scientific collaborations. Minerva, 38(3), 311–348.

Georghiou, L. (1998). Global cooperation in research. Research Policy, 27(6), 611–626.

Georghiou, L. (2001). Evolving frameworks for European collaboration in research and technology. Research Policy, 30(6), 891–903.

Gevers, W. (2001). The Academy of Science of South Africa is now a statutory organization. South African Journal of Science, 97(11 & 12), 447–448.

Gevers, W. (2003). The Academy of Science of South Africa and the South African Journal of Science. South African Journal of Science, 99(3 & 4), 99.

Gevers, W., Mati, X., Mouton, J., Page-Shipp, R., Hammes, M., & Pouris, A. (2006). Conclusions and recommendations for a strategically enhanced role of research publishing in South Africa. Report on a Strategic Approach to Research Publishing in South African South Africa (pp. 107–120). Pretoria: The Academy of Science of South Africa.

Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P., & Trow, M. (1994). The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies. London: Sage Publications.

Gill, D. (1905). Astronomy and geology in South Africa. In W. Flint & J. D. F. Gilchrist (Eds.), Science in South Africa: A Handbook and Review (pp. 61–78). Cape Town, Pretoria and Bulawayo: T. Maskew Miller.

Gillette, R. (1975). Uranium enrichment: with help, South Africa is progressing. Science (New Series), 188(4193), 1090, 1092.

Glänzel, W. (2002). Co-authorship patterns and trends in the sciences (1980–1998): a bibliometric study with implications for database indexing and search strategies. Library Trends, 50(3), 461–471.

Glänzel, W., & de Lange, C. (2002). A distributional approach to multinational-ity measures of international scientific collaboration. Scientometrics, 54(1), 75–89.

References 245

Glänzel, W., Schubert, A., & Czerwon, H. J. (1999). A bibliometric analysis of international scientific cooperation of the European Union (1985–1995). Scientometrics, 45(2), 185–202.

Gläser, J. (2003). What internet use does and does not change in scientific com-munities. Science Studies, 16(1), 38–51.

Glover, S. W., & Bowen, S. L. (2004). Bibliometric analysis of research published in tropical medicine and international health 1996–2003. Tropical Medicine and International Health, 9(12), 1327–1330.

Godin, B. (1996). Research and the practice of publication in industries. Research Policy, 25(4), 587–606.

Godin, B., Barker, S., & Landry, M. (1995). Besides academic publications: which sectors compete or are then competitors? Scientometrics, 33(1), 3–12.

Goldfinch, S., Dale, T., & Jr., K. D. (2003). Science from the periphery: collabora-tion, networks and ‘periphery effects’ in the citation of New Zealand Crown Research Institutes articles, 1995–2000. Scientometrics, 37(3), 321–337.

Goldstein, A., & Friedhoff, A. J. (1988). Collaboration and responsibility. Science, 242(4886), 1623.

Gómez, I., Fernández, M. T., & Sebastián, J. (1999). Analysis of the structure of international scientific cooperation networks through bibliometric indicators. Scientometrics, 44(3), 441–457.

Gonzalez-Brambila, C., & Veloso, F. M. (2007). The determinants of research output and impact: a study of Mexican researchers. Research Policy, 36(7), 1035–1051.

Good, C. M. (1987). Ethnomedical Systems in Africa. New York: The Guilford Press.Gould, C., & Folb, P. (2002). The role of professionals in the South African chemi-

cal and biological warfare programme. Minerva, 40(1), 77–91.Government of South Africa. (2006). Science and Technology: http://www.info.gov.

za/aboutsa/science.htm#mine_saf_res. Accessed 30 April 2007.Greenberg, D. S. (1970a). South Africa: (I): Booming nation’s research and indus-

try benefit from close ties with the United States. Science (New Series), 169(3941), 157–163.

Greenberg, D. S. (1970b). South Africa: (II): University system follows apartheid pattern; government enforces limits on academic dissenters. Science (New Series), 169(3942), 260–264, 266–267.

Guan, J., & Ma, N. (2007). A bibliometric study of China’s semiconductor litera-ture compared with other major Asian countries. Scientometrics, 70(1), 107–124.

Gulati, R. (1998). Alliances and networks. Strategic Management Journal, 19(4), 293–317.

Gultig, J. (2000). The university in post-apartheid South Africa: new ethos and new divisions. South African Journal of Higher Education, 14(1), 37–52.

Gupta, B. M., & Dhawan, S. M. (2003). India’s collaboration with People’s Republic of China and science and technology: a scientometric analysis of coauthored papers during 1994–1999. Scientometrics, 57(1), 59–74.

Gupta, B. M., & Karisiddippa, C. R. (1999). Collaboration and author productiv-ity: a study with a new variable in Lotka’s law. Scientometrics, 44(1), 129–134.

Habib, A., & Morrow, S. (2006). Research, research productivity and the state in South Africa. Transformation, 62, 9–29.

Hackett, E. J. (2005). Introduction. Social Studies of Science, 35(5), 667–671.

246 References

Hafernik, J. J., Messerschmitt, D. S., & Vandrick, S. (1997). Collaborative research: why and how? Educational Researcher, 26(9), 31–35.

Hagedoorn, J. (1993). Understanding the rationale of strategic technology part-nering: interorganizational modes of cooperation and sectoral differences. Strategic Management Journal, 14(5), 371–385.

Hagedoorn, J., Link, A. N., & Vonortas, N. S. (2000). Research partnerships. Research Policy, 29(4–5), 567–586.

Hagstrom, W. O. (1964). Traditional and modern forms of scientific teamwork. Administrative Science Quarterly, 9(3), 241–263.

Hagstrom, W. O. (1965). The Scientific Community. New York: Basic Books.Hagstrom, W. O. (1974). Competition in science. American Sociological Review,

39(1), 1–8.Hall, A. V. (1977). A history of the Royal Society of South Africa. In A. C. Brown

(Ed.), A History of Scientific Endeavour in South Africa (pp. 474–487). Cape: Royal Society of South Africa.

Hamel, G. (1991). Competition for competence and inter-partner learning within international strategic alliances. Strategic Management Journal, 12(Special Issue), 83–103.

Hardwig, J. (1991). The role of trust in knowledge. The Journal of Philosophy, 88(12), 693–708.

Hargens, L. (1978). Relations between work habits, research technologies, and eminence in science. Sociology of Work and Occupations, 5(1), 97–112.

Harris, V., Hatang, S., & Liberman, P. (2004). Unveiling South Africa’s nuclear past. Journal of Southern African Studies, 30(3), 457–475.

Harsanyi, M. A. (1993). Multiple authors, multiple problems – bibliometrics and the study of scholarly collaboration: a literature review. Library and Information Science Research, 15(4), 325–354.

Hassan, M. H. A. (2001). Can science save Africa? Science, 292(5522), 1609.Haythornthwaite, C. (2001). Exploring multiplexity: social network structures in

a computer-supported distance learning class. The Information Society, 17(2), 211–226.

Haythornthwaite, C. (2005). Social networks and internet connectivity effects. Information, Communication and Society, 8(2), 125–147.

Hicks, D. M., Isard, P. A., & Martin, B. R. (1996). A morphology of Japanese and European corporate research networks. Research Policy, 25(3), 359–378.

Hicks, D. M., & Katz, J. S. (1996). Where is science going? Science, Technology and Human Values, 21(4), 379–406.

Hirsch, J. E. (2005). An index to quantify an individual’s scientific research out-put. Proceedings of the National Academy of Science of the United States, 102(46), 16569–16572.

Hoffman, J., & Cox, T. J. (1971). Relations with South Africa. Science (New Series), 173(4000), 868.

Hofmeyr, J. H. (1929a). Africa and science. Science (New Series), 70(1812), 269–274.

Hofmeyr, J. H. (1929b). Africa and science II. Science (New Series), 70(1813), 294–299.Hsu, J-w., & Huang, D-w. (2011). Correlation between impact and collaboration.

Scientometrics, 86(2), 317–324.Hunter, L. A., & Leahey, E. (2010). Parenting and research productivity: new evi-

dence and methods. Social Studies of Science, 40(3), 433–451.

References 247

IDRC. (1993). Towards a Science and Technology Policy for a Democratic South Africa: Mission Report. Johannesburg: The International Development Research Centre.

Ingwersen, P., & Jacobs, D. (2004). South African research in selected scientific areas: status 1981–2000. Scientometrics, 59(3), 405–423.

Internet World Statistics. (2007). Internet usage statistics for Africa. Retrieved 1 November 2007, from Internet World Statistics: http://www.internetworldstats.com/stats1.htm

Internet World Statistics. (2012). Usage and population stats: Internet users in Africa. Retrieved 22 November 2012, from http://www.Internetworldstats.com/stats1.htm

Irikefe, V., Vaidyanathan, G., Nordling, L., Twahirwa, A., Nakkazi, E., & Monastersky, R. (2011). Science in Africa: view from the front line. Nature, 474(7353), 556–559.

Irvine, J., & Martin, B. (1985). Basic research in the east and west: a comparison of the scientific performance of high-energy physics accelerators. Social Studies of Science, 15(2), 293–343.

Jacobs, D. (2006). Analysis of scientific research in selected institutions in South Africa: a bibliometric study. South African Journal of Library and Information Science, 72(1), 72–77.

Jacobs, D. (2008). An informetríc analysis of publication and research collabora-tion patterns in natural and applied sciences in South Africa. South African Journal of Libraries and Information Science, 74(1), 41–48.

Jacobs, D., & Ingwersen, P. (2000). A bibliometric study of the publication pat-terns in the sciences of South African scholars 1981–96. Scientometrics, 47(1), 75–93.

Jansen, B. C. (1977). The growth of veterinary research in South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 164–191). Cape: Royal Society of South Africa.

Jarvenpaa, S. L., & Leidner, D. E. (1999). Communication and trust in global vir-tual teams. Organization Science, 10(6), 791–815.

Jeenah, M., & Pouris, A. (2008). South African research in the context of Africa and globally. South African Journal of Science, 104(9/10), 351–354.

Jeffrey, P. (2003). Smoothing the waters: observations on the process of cross-disciplinary research collaboration. Social Studies of Science, 33(4), 539–562.

Jensen, M., Belcher, M., & Ngimwa, P. (2007). Information access and issues of bandwidth. In W. Shrum, K. Benson, W. Bijker, & K. Brunnstein (Eds.), Past, Present & Future of Research in the Information Society (pp. 189–192). New York: Springer.

Jewkes, J., Sawers, D., & Stillerman, R. (1959). The Sources of Invention. New York: St. Martin’s Press.

John-Steiner, V., Weber, R. J., & Minnis, M. (1998). The challenge of studying col-laboration. American Educational Research Journal, 35(4), 773–783.

Joubert, D. M. (1977). Agricultural research in South Africa: an historical over-view. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 265–282). Cape: Royal Society of South Africa.

Juritz, C. F. (1915). Science and progress in South Africa. Journal of the Royal African Society, 15(57), 17–23.

Juritz, C. F. (1916a). Science and progress in South Africa (continued). Journal of the Royal African Society, 15(58), 184–186.

248 References

Juritz, C. F. (1916b). Science and progress in South Africa (continued). Journal of the Royal African Society, 15(59), 256–260.

Juritz, C. F. (1916c). Science and progress in South Africa. Journal of the Royal African Society, 16(61), 30–35.



Kahn, M. (2007). Internationalization of R&D: where does South Africa stand? South African Journal of Science, 103(1 & 2), 7–12.

Kaminer, N., & Braunstein, Y. M. (1998). Bibliometric analysis of the impact of internet use on scholarly productivity. Journal of the American Society for Information Science, 49(8), 720–730.

Kaplan, D. (2004). South Africa’s national research and development strategy: a review. Science, Technology and Society, 9(2), 273–294.

Katila, R., & Mang, P. Y. (2003). Exploiting technological opportunities: the tim-ing of collaborations. Research Policy, 32(2), 317–332.

Katz, J. S., & Hicks, D. (1997). How much a collaboration worth? a calibrated bibliometric model. Scientometrics, 40(3), 541–554.

Katz, J. S., & Martin, B. R. (1997). What is research collaboration? Research Policy, 26(1), 1–18.

Keay, R. (1976). Scientific cooperation in Africa. African Affairs, 75(298), 86–97.Kerr, E. B., & Hiltz, S. R. (1982). Computer-Mediated Communication Systems.

New York: Academic.Khumalo, G. (2001). An Investigation and Identification of Indigenous Science

Understandings among Zulu Community Elders and the Impact of these Understandings on the Zulu Secondary School Learners. Durban: University of Durban-Westville.

Kim, K.-W. (2006). Measuring international research collaboration of peripheral countries: taking the context into consideration. Scientometrics, 66(2), 231–240.

King, D. A. (2004). The scientific impact of nations: what different countries get for their research spending. Nature, 430, 311–316.

Klevorick, A., Levin, R. C., Nelson, R. R., & Winter, S. G. (1995). On the sources and significance of inter-industry differences in technological opportunities. Research Policy, 24(2), 185–205.

Komsky, S. H. (1991). A profile of users of electronic mail in a university: frequent versus occasional users. Management Communication Quarterly, 4(3), 310–340.

Kramer, R., & Tyler, T. (1995). Trust in Organizations. Thousand Oaks, CA: Sage Publications.

Kretschmer, H. (1999). A new model of scientific collaboration. Part 1: Theoretical Approach. Scientometrics, 46(3), 501–518.

Krige, E. J. (1974). The Social Systems of the Zulus (6th Impression ed.). Pietermaritzburg: Shutter and Shooter.

Krige, S. (1997). Segregation, science and commissions of enquiry: the contesta-tion over native education policy in South Africa, 1930–36. Journal of South African Studies, 23(3), 491–506.

Kundra, R., & Kretschmer, H. (1999). A new model of scientific collaboration: Part 2, collaboration patterns in Indian medicine. Scientometrics, 46(3), 519–528.

References 249

Kyvik, S. (1990a). Motherhood and scientific productivity. Social Studies of Science, 20(1), 149–160.

Kyvik, S. (1990b). Age and scientific productivity, differences between fields of learning. Higher Education, 19(1), 37–55.

Kyvik, S., & Teigen, M. (1996). Child care, research collaboration, and gender dif-ferences in scientific productivity. Science, Technology and Human Values, 21(1), 54–71.

Landry, R., & Amara, N. (1998). The impact of transaction costs on the institu-tional structuration of collaborative academic research. Research Policy, 27(9), 901–913.

Landry, R., Traore, N., & Godin, B. (1996). An econometric analysis of the effect of collaboration on academic research productivity. Higher Education, 32(3), 283–301.

Langer, E. (1967). Science in South Africa: the effects of apartheid. Science (New Series), 155(3768), 1387–1389.

Larivière, V., Gingras, Y., & Archambault, E. (2006). Canadian collaboration net-works: a comparative analysis of the natural sciences, social sciences and the humanities. Scientometrics, 68(3), 519–533.

Laudel, G. (2001). Collaboration, creativity and rewards: why and how scientists collaborate. International Journal of Technology Management, 22(7/8), 762–781.

Laudel, G. (2002). Collaboration and reward: what do we measure by co-author-ships? Research Evaluation, 11(1), 3–15.

Lawani, S. (1986). Some bibliometric correlates of quality in scientific research. Scientometrics, 9(1&2), 13–25.

Lee, S., & Bozeman, B. (2005). The impact of research collaboration on scientific productivity. Social Studies of Science, 35(5), 673–702.

Lee, Y. S. (1996). ‘Technology transfer’ and the research university: a search for boundaries of university-industry collaboration. Research Policy, 25(6), 843–863.

Letseka, M. (2005). Government incentivization of partnerships in South Africa: an audit of Thrip and the Innovation Fund. Industry and Higher Education, 9(2), 161–168.

Levine, J. M., & Moreland, R. L. (2004). Collaboration: the social context of theory development. Personality and Social Psychology Review, 8(2), 164–172.

Lewis, J. D., & Weigert, A. (1985). Trust as a social reality. Social Forces, 63(4), 967–985.

Link, A. N., & Bauer, L. L. (1989). Cooperative Research in U.S. Manufacturing. Lexington: Lexington Books.

Lister, S. (1929). Medical research: an account of the South African Institute for Medical Research. In H. J. Crocker & J. McCrae (Eds.), South Africa and Science: A Handbook (pp. 288–301). Johannesburg: Hortors Limited.

Lomas, J., Molengraaff, G. N., Penck, A., Hobson, B., Beck, P., Anderson, W., et al. (1905). Permian glaciation in South Africa. Science, 22(563), 475.

Long, J. S. (1978). Productivity and academic position in the scientific career. American Sociological Review, 43(6), 889–908.

Long, J. S. (1990). The origins of sex differences in science. Social Forces, 68(4), 1297–1315.

Long, J. S., Allison, P. D., & McGinnis, R. (1993). Rank advancement in academic careers: sex differences and the effects of productivity. American Sociological Review, 58(5), 703–722.

250 References

Long, J. S., & McGinnis, R. (1981). Organizational context and scientific produc-tivity. American Sociological Review, 46(4), 422–442.

Lundberg, J., Tomson, G., Lundkvist, I., Skar, J., & Brommels, M. (2006). Collaboration uncovered: exploring the adequacy of measuring university-industry collaboration through co-authorship and funding. Scientometrics, 69(3), 575–589.

Luukkonen, T., Persson, O., & Sivertsen, G. (1992). Understanding patterns of international scientific collaboration. Science, Technology and Human Values, 17(1), 101–126.

Ma, N., & Guan, J. (2005). An exploratory study on collaboration profiles of Chinese publications in molecular biology. Scientometrics, 65(1), 343–355.

MacRoberts, M. H., & MacRoberts, B. R. (2010). Problems of citation analysis: a study of uncited and seldom-cited influences. Journal of the American Society for Information Science and Technology, 61(1), 1–13.

Maienschein, J. (1993). Why collaborate? Journal of History of Biology, 26, 167–183.

Majozi, T., Mbuli, N., Ferguson, K., & Vaal, P. D. (2010). Engineering. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 263–275). Pretoria: ASSAf.

Manchester, K. (2003). Nobel laureate at last. South African Journal of Science, 99(1 & 2), 10.

Marais, H. C. (2000). Perspectives on Science Policy in South Africa. Pretoria: Network Publishers.

Marshall, E. (1988). South Africa blocks AAAS visit. Science (New Series), 240(4852), 595.

Mauleón, E., & Bordons, M. (2006). Productivity, impact and publication habits by gender in the area of materials science. Scientometrics, 66(1), 199–218.

May, R. M. (1997). The scientific wealth of nations. Science, 275(5301), 793–796.Meadows, A. J. (1974). Communication in Science. London: Butterworths.Mêgnigbêto, E. (2013). International collaboration in scientific publishing: the

case of West Africa (2001–2010). Scientometrics, 96(3), 761–783.Melin, G. (2000). Pragmatism and self-organization: research collaboration on

the individual level. Research Policy, 29(1), 31–40.Melin, G., & Persson, O. (1996). Studying research collaboration using co-author-

ship. Scientometrics, 36(3), 363–377.Mercury, T. (2007, 6 June 2007). Training authorities in big push to develop essen-

tial skills. The Mercury, p. 6.Merton, R. K. (1938[1970]). Science, Technology and Society in Seventeenth Century

England. New York: Howard Fertig Press.Meyer-Krahmer, F., & Schmoch, U. (1998). Science-based technologies: university-

industry interactions in four fields. Research Policy, 27(8), 835–851.Milliken, F. J., & Martins, L. (1996). Searching for common threads: understand-

ing the multiple effects of diversity in organizational groups. Academy of Management Review, 21(2), 402–431.

Mishkin, B. (1996). Collaboration and data sharing: continued. Science (New Series), 271(5246), 130.

Moed, H. F., & Hesselink, F. T. (1996). The publication output and impact of aca-demic chemistry research in the Netherlands during the 1980s: bibliometric analyses and policy implications. Research Policy, 25(5), 819–836.

References 251

Moody, J. (2004). The structure of a social science collaboration network: discipli-nary cohesion from 1963 to 1999. American Sociological Review, 69(2), 213–238.

Moraal, H. (2010). Physics and astronomy. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 83–102). Pretoria: ASSAf.

Morris, A. G. (2010). Archaeology and palaeoanthropology. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 241–260). Pretoria: ASSAf.

Mouton, J. (2000). Patterns of research collaboration in South Africa. South African Journal of Science, 96(9 &10), 458–462.

Mouton, J., Boshoff, N., & Tijssen, R. (2006). A comprehensive analysis of South African research journals. A Report on a Strategic Approach to Research Publishing in South Africa (pp. 29–59). Pretoria: Academy of Science of South Africa (ASSAf).

Mouton, J., & Gevers, W. (2010). Introduction. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 39–67). Pretoria: ASSAf.

Murphy, M. (2011). Historical highlights of the rise and demise of South Africa’s nuclear capability. Africa Insight, 40(4), 121–135.

Narváez-Berthelemot, N., Russel, J. M., Arvanitis, R., Waast, R., & Gaillard, J. (2002). Science in Africa: an overview of mainstream scientific output. Scientometrics, 54(2), 229–241.

Nature. (2002). Don’t boycott Israel’s scientists. Nature, 417, 1.Naudé, S. M., & Brown, A. C. (1977). The growth of scientific institutions in

South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 60–85). Cape: Royal Society of South Africa.

Negishi, M., Sun, Y., & Shigi, K. (2004). Citation database for Japanese papers: a new bibliometric tool for Japanese academic society. Scientometrics, 60(3), 333–351.

Newman, M. E. J. (2001). The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences of the United States of America, 98(2), 404–409.

Newman, M. E. J. (2004). Coauthorship networks and patterns of scientific col-laboration. Proceedings of the National Academy of Sciences of the United States of America, 101(Suppl. 1), 5200–5205.

Niu, F., & Qiu, J. (2014). Network structure, distribution and the growth of Chinese international research collaboration. Scientometrics, 98(2), 1221–1233.

Nohria, N., & Eccles, R. G. (1992). Face-to-face: making network organizations work. In N. Nohria & R. G. Eccles (Eds.), Network and Organizations (pp. 288–308). Boston, MA: Harvard Business School Press.

Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 14–37.

Nordkvelle, Y. (1990). The academic boycott of South Africa debate: science and social practice. Studies in Higher Education, 15(3), 253–275.

NSF. (2004). Science and Engineering Indicators. Washington, DC: National Science Foundation.

Odhiambo, T. R. (1967). East Africa: science for development. Science (New Series), 158(3803), 876–881.

Okubo, Y., & Sjöberg, C. (2000). The changing pattern of industrial scientific research collaboration in Sweden. Research Policy, 29(1), 81–98.

Oliver, A. L. (2004). Biotechnology entrepreneurial scientists and their collabora-tions. Research Policy, 33(4), 583–597.

252 References

Olson, G. M., David, P. A., Eksteen, J., Sonnenwald, D. H., Uhlir, P. F., Tseng, S.-F., et al. (2007). International collaborations through the internet. In W. Shrum, K. Benson, W. Bijker, & K. Brunnstein (Eds.), Past, Present & Future of Research in the Information Society (pp. 97–114). New York: Springer.

Olson, G. M., & Olson, J. S. (2000). Distance matters. Human-Computer Interaction, 15(2–3), 139–178.

Onyancha, O. B., & Jacobs, D. (2009). Capacitating national research: a review of South African natural sciences research projects, theses and dissertations, 1986–2006. South African Journal of Libraries and Information Science, 75(2), 117–130.

Ovens, C. S. H. (1995). Citations patterns of University of the Orange Free State scientists. South African Journal of Library and Information Science, 63(2), 47–53.

Pachi, C. G. d. F., Yamamoto, J. F., Costa, A. P. A. d., & Lopez, L. F. (2012). Relationship between connectivity and academic productivity. Scientometrics, 93(2), 265–278.

Pao, M. L. (1982). Collaboration in computations musicology. Journal of the American Society for Information Sciences, 33(1), 38–43.

Parker, A. (2007). Fast, trusty internet connection is at hand. The Mercury, p. 8.Pelz, D., & Andrews, F. (1966). Scientists in Organizations: Productive Climates of

Research and Development. New York: John Wiley.Pérez, A. I., Blanco, N., Ogalla, M., & Rossi, F. (1998). The flexible role of the

researcher within the changing context of practice: forms of collaboration. Educational Action Research, 6(2), 241–255.

Perez, A. M., Ahmed, N., & London, L. (2012). Racial discrimination: experiences of black medical school alumni at the University of Cape Town, 1945–1994. South African Medical Journal, 102(6), 574–577.

Persson, O. (2010). Are highly cited papers more international? Scientometrics, 83(2), 397–401.

Peterson, B. J. (1993). The costs and benefits of collaborative research. Estuaries, 16(4), 913–918.

Pick, W. (2010). Health and Medical Sciences. In R. Diab & W. Gevers (Eds.), The State of Science in South Africa (pp. 187–209). Pretoria: ASSAf.

Plug, C. (2001). Southern African science in the year 1901–100n. South African Journal of Science, 97(1 & 2), 41–42.







Ponomariova, B. L., & Boardman, P. C. (2010). Influencing scientists’ collabora-tion and productivity patterns through new institutions: university research centers and scientific and technical human capital. Research Policy, 39(5), 613–624.

References 253

Porac, J. F., Wade, J. B., Fischer, H. M., Brown, J., Kanfer, A., & Bowker, G. (2004). Human capital heterogeneity, collaborative relationships and publication pat-terns in a multidisciplinary scientific alliance: a comparative case study of two scientific teams. Research Policy, 33(4), 661–678.

Porter, M. E. (1980). The Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York: The Free Press.

Porter, M. E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance. New York: The Free Press.

Pouris, A. (2003). South Africa’s research publication record: the last ten years. South African Journal of Science, 99(9 & 10), 425–428.

Pouris, A. (2005). An assessment of the impact and visibility of South African journals. Scientometrics, 62(2), 213–222.

Pouris, A. (2006a). Assessing scientific strengths of academic institutions: the example of the University of Pretoria. South African Journal of Science, 102(1 & 2), 23–28.

Pouris, A. (2006b). A bibliometric assessment of South African research publica-tions included in the internationally indexed database of Thomson ISI. Report on a Strategic Approach to Research Publishing in South Africa in South Africa (pp. 9–29). Pretoria: Academy of Science of South Africa.

Pouris, A. (2010). A scientometric assessment of the Southern Africa Development Community: science in the tip of Africa. Scientometrics, 85(1), 145–154.

Pouris, A. (2012a). Science in South Africa: the dawn of a renaissance? South African Journal of Science, 108(7/8), 1–6.

Pouris, A. (2012b). Scientometric research in South Africa and successful policy instruments. Scientometrics, 91(2), 317–325.

Pouris, A., & Ho, Y.-S. (2014). Research emphasis and collaboration in Africa. Scientometrics, 98(3), 2169–2184.

Pouris, A., & Pouris, A. (2009). The state of science and technology in Africa (2000–2004): a scientometric assessment. Scientometrics, 79(1), 297–309.

Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational col-laboration and the locus of innovation: networks of learning in biotechnology. Administrative Science Quarterly, 41(1), 116–145.

Presser, S. (1980). Collaboration and the quality of research. Social Studies of Science, 10(1), 95–101.

Price, D. (1965). Is technology historically independent of science? A study in statistical historiography. Technology and Culture, 6(4), 553–568.

Price, D. J. d. S. (1963). Little Science, Big Science. New York: Columbia University Press.

Price, D. J. d. S. (1965b). The science of science. Bulletin of the Atomic Scientists, 21(8), 2–8.

Price, D. J. d. S. (1986). Little Science, Big Science . . . and Beyond. New York: Columbia University Press.

Price, D. J. d. S., & Beaver, D. (1966). Collaboration in an invisible college. American Psychologist, 21(11), 1011–1018.

Puuska, H.-M., Muhonen, R., & Leino, Y. (2014). International and domestic co-publishing and their citation impact in different disciplines. Scientometrics, 98(2), 823–839.

R. G. (1974). South Africa’s process may not be so new after all. Science (New Series), 183(4131), 1271.

254 References

Rabkin, Y. M., Eisemon, T. O., Lafitte-Houssat, J.-J., & Rathgeber, E. M. (1979). Citation visibility of Africa’s science. Social Studies of Science, 9(4), 499–506.

Rich, P. (1990). Race, science, and the legitimization of white supremacy in South Africa, 1902–1940. The International Journal of African Historical Studies, 23(4), 665–686.

Roe, A. (1972). Patterns in productivity of scientists. Science (New Series), 176(4037), 940–941.

Rogers, A. W. (1929). Geology. In H. J. Crocker & J. McCrae (Eds.), South Africa and Science: A Handbook (pp. 88–110). Johannesburg: Hortors Limited.

Rosenberg, N., & Mowery, D. C. (1990). Why do firms do basic research (with their own money)? Research Policy, 19(2), 165–174.

Rosenfeld, S. A. (1996). Does cooperation enhance competitiveness? Assessing the impacts of inter-firm collaboration. Research Policy, 25(2), 247–263.

Schaffer, W. (1977). A history of physics in South Africa. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 372–387). Cape: Royal Society of South Africa.

Schapera, I. (Ed.). (1966). Bantu-Speaking Tribes of South Africa (8th Impression ed.). Cape Town: Maskew Miller Limited.

Schapera, I., & Farrington, B. (1933). The Early Cape Hottentots. Cape Town: The Van Riebeeck Society.

Scholes, R. J., Anderson, F., Kenyon, C., Napier, J., Ngoepe, P., Wilgen, B. v., et al. (2008). Science councils in South Africa. South African Journal of Science, 104(11–12), 435–438.

Schott, T. (1993). World science: globalization of institutions and participation. Science, Technology and Human Values, 18(2), 196–208.

Schroeder, R., & Swedberg, R. (2002). Weberian perspectives on science, technol-ogy and the economy. British Journal of Sociology, 53(3), 383–401.

Science. (1885). The Stone Age in Africa. Science, 5(119), 405–406.Science. (1905a). Meeting of the British Association in South Africa. Science (New

Series), 21(533), 435–437.Science. (1905b). Meeting of the British Association in South Africa. Science (New

Series), 22(551), 93–94.Science. (1925). The Yale Astronomical Station in South Africa. Science (New Series),

61(1567), 36.Science. (1926). The Harvard Expedition to Africa. Science (New Series), 63(1638),

517–518.Science. (1929a). Science in Africa. Science (New Series), 70(1806), 143–144.Science. (1929b). Astronomy in South Africa. Science (New Series), 70(1825), 600.Science. (1944). A government testing laboratory in South Africa. Science (New

Series), 99(2574), 338.Science. (1960). Scientific activity in Africa growing; U.S. education aid rising

slowly. Science (New Series), 131(3396), 287–289.Science Watch. (1994). Crowd control? Multiauthor papers appear to level off in

recent years. Science Watch, 15(4), 1–2.Seaborg, G. T. (1970). A scientific safari to Africa. Science (New Series), 169(3945),

554–561.Shepherd, N. (2003). State of the discipline: science, culture and identity in South

African Archaeology, 1870–2003. Journal of South African Studies, 29(4), 823–844.

References 255

Shrum, W., Jr., Chompalov, I., & Genuth, J. (2001). Trust, conflict and perfor-mance in scientific collaboration. Social Studies of Science, 31(5), 681–730.

Sindiga, I., Nyaigotti-Chacha, C., & Kanunah, M. P. (Eds.). (1995). Traditional Medicine in Africa. Nairobi: East African Educational Publishers Ltd.

Smalheiser, N. R., Perkins, G. A., & Jones, S. (2005). Guidelines for negotiating scientific collaboration. PLoS Biology, 3(6), e217–e218.

Smart, J. C., & Bayer, A. E. (1986). Author collaboration and impact: a note on citation rates of single and multiple authored articles. Scientometrics, 10(5–6), 297–305.

Smith, M. (1958). The trend toward multiple authorship in psychology. American Psychologist, 13, 596–599.

Smith, D. (1967). Scientific research centres in Africa. African Studies Bulletin, 10(3), 20–47.

Sonaike, S. A. (2004). The internet and the dilemma of Africa’s development. Gazette: The International Journal for Communication Studies, 66(1), 41–61.

Sonnenwald, D. H. (2003). Managing cognitive and affective trust in the concep-tual R & D organization. In M. Huotari & M. Livonen (Eds.), Trust in Knowledge Management and Systems in Organizations (pp. 82–106). Hershey, PA: Idea Publishing.

Sonnenwald, D. H. (2007). Scientific collaboration: a synthesis of challenges and strategies. In B. Cronin (Ed.), Annual Review of Information Science and Technology (Vol. 4, pp. 643–683). Medford, NJ: Information Today Inc.

Sonnenwald, D. H., & Pierce, L. (2000). Information behaviour in dynamic group work contexts: interwoven situational awareness, dense social networks, and contested collaboration in command and control. Information Processing and Management, 36(3), 461–479.

Sooryamoorthy, R. (2009a). Collaboration and publication: how collaborative are scientists in South Africa? Scientometrics, 80(2), 419–439.

Sooryamoorthy, R. (2009b). Do types of collaboration change citation? Collaboration and citation patterns of South African science publications. Scientometrics, 81(1), 171–193.

Sooryamoorthy, R. (2010a). Medical research in South Africa: a scientometric analysis of trends, patterns, productivity and partnership. Scientometrics, 84(3), 863–885.

Sooryamoorthy, R. (2010b). Science and scientific collaboration: apartheid and after. Scientometrics, 84(2), 373–390.

Sooryamoorthy, R. (2010c). Scientific publications of engineers in South Africa, 1975–2005. Scientometrics, 86(1), 211–226.

Sooryamoorthy, R. (2013a). Scientific collaboration in South Africa. South African Journal of Science, 109(5/6), 1–5.

Sooryamoorthy, R. (2013b). Scientific research in the natural sciences in South Africa: a scientometric study. South African Journal of Science, 109(7/8), 1–11.

Sooryamoorthy, R. (2014). Publication productivity and collaboration of researchers in South Africa: new empirical evidence. Scientometrics, 98, 531–545.

Sooryamoorthy, R., Duque, R. B., Ynalvez, M. A., & Shrum, W. (2007). Scientific collaboration and the Kerala Model: does the internet make a difference? Journal of International Development, 19(7), 982–996.

256 References

Sooryamoorthy, R., & Shrum, W. (2007). Does the internet promote collaboration and productivity? Evidence from the scientific community in South Africa. Journal of Computer-Mediated Communication, 12(2), 733–751.

South African Journal of Science. (1977). Uneven support for science. South African Journal of Science, 73(1), 2–3.

South African Journal of Science. (2005). A first for tuberculosis research in South Africa. South African Journal of Science, 101(9 & 10), 393–394.

Stephen, P., & Levin, S. (1987). Demographic and Economic Determinants of Scientific Productivity. Atlanta: Georgia State University Press.

Steyn, A. G. W. (2002). State funding of universities and technikons, 1993 to 2001. South African Journal of Education, 22(4), 253–269.

Stichweh, R. (1996). Science in the system of world society. Social Science Information, 35(2), 327–340.

Stokes, T. D., & Hartley, J. A. (1989). Coauthorship, social structure, and influence with specialties. Social Studies of Science, 19(1), 101–125.

Subramanyan, K. (1983). Bibliometric studies of research collaboration: a review. Journal of Information Science, 6(1), 33–38.

Talbot, W. J. (1977). Pathfinders and pioneers, explorers and scientists 1487–1976. In A. C. Brown (Ed.), A History of Scientific Endeavour in South Africa (pp. 1–32). Cape: Royal Society of South Africa.

Taubes, G. (1994). Japan learns to accommodate its global research partners. Science (New Series), 266(5188), 1181–1183.

Teasley, S., & Wolinsky, S. (2001). Scientific collaborations at a distance. Science, 292(5525), 2254–2255.

Thurlow, C., & McKay, S. (2003). Profiling “new” communication technologies in adolescence. Journal of Language and Social Psychology, 22(1), 94–103.

Tijssen, R. J. W. (2007). Africa’s contribution to the worldwide research literature: new analytical perspectives, trends, and performance indicators. Scientometrics, 71(2), 303–327.

Tilley, H. (2011). Africa as a Living Laboratory: Empire, Development and the Problem of Scientific Knowledge, 1870–1950. Chicago: University of Chicago Press.

Tobias, P. V. (1965). Early man in Africa. Science (New Series), 149(3679), 22–33.Toivanen, H., & Ponomariov, B. (2011). African regional innovation systems: bib-

liometric analysis of research collaboration patterns 2005–2009. Scientometrics, 88(2), 471–493.

Torricella-Morales, R. G., Hooydonk, G. V., & Araujo-Ruiz, J. A. (2000). Citation analysis of Cuban research. Part I. A case study: the Cuban Journal of Agricultural Science. Scientometrics, 47(2), 413–426.

Traoré, N., & Landry, R. (1997). On the determinants of scientists’ collaboration. Science Communication, 19(2), 124–140.

Tschannen-Moran, M. (2001). Collaboration and the need for trust. Journal of Educational Administration, 39(4), 308–331.

UNDP. (2001). Human Development Report 2001. New York: Oxford.Vasileiadoua, E., & Vliegenthart, R. (2009). Research productivity in the era of the

internet revisited. Research Policy, 38(8), 1260–1268.Vaughan, C. L., Reddy, B. D., Noakes, T. D., & Moran, V. C. (2007). A commentary

on the intellectual health of the nation. South African Journal of Science, 103(1 & 2), 22–26.

References 257

Velho, L., & Krige, J. (1984). Publication and citation practices of Brazilian agri-cultural scientists. Social Studies of Science, 14(1), 45–62.

Vieira, E. S., & Gomes, J. A. N. F. (2010). Citations to scientific articles: its distri-bution and dependence on the article features. Journal of Informetrics, 4(1), 1–13.

Wagner, C. S. (2005). Six case studies of international collaboration in science. Scientometrics, 62(1), 3–26.

Wagner, C. S., & Leydesdorff, L. (2005a). Mapping the network of global science: comparing international co-authorships from 1990 to 2000. International Journal of Technology and Globalization, 1(2), 185–208.

Wagner, C. S., & Leydesdorff, L. (2005b). Network structure, self-organization, and the growth of international collaboration in science. Research Policy, 34(10), 1608–1618.

Wagner- Döbler, R. (2001). Continuity and discontinuity of collaboration behav-iour since 1800: from a bibliometric point of view. Scientometrics, 52(3), 503–517.

Walsh, J. P., & Bayma, T. (1996a). The virtual college: computer-mediated com-munication and scientific work. Information Society, 12(4), 343–363.

Walsh, J. P., & Bayma, T. (1996b). Computer networks and scientific work. Social Studies of Science, 26(3), 661–703.

Walsh, J. P., Kucker, S., Maloney, N. G., & Gabbay, S. (2000). Connecting minds: computer-mediated communication and scientific work. Journal of the American Society for Information Science and Technology, 51(14), 1295–1305.

Walsh, J. P., & Maloney, N. G. (2002). Computer network use, collaboration struc-tures and productivity. In P. Hinds & S. Kiesler (Eds.), Distributed Work (pp. 433–458). Cambridge, MA: MIT Press.

Walsh, J. P., & Maloney, N. G. (2003). Problems in scientific collaboration: does email hinder or help? http://tigger.uic.edu/~jwalsh/. Accessed 12 December 2006.

Walsh, J. P., & Maloney, N. G. (2007). Collaboration structure, communication media, and problems in scientific work teams. Journal of Computer-Mediated Communication, 12(2), http://jcmc.indiana.edu/vol12/issue12/walsh.html. Accessed 17 May 2007.

Watts, L. A., & Monk, A. F. (1998). Reasoning about tasks, activities and technol-ogy to support collaboration. Ergonomics, 41(11), 1583–1606.

Webster, E., & Mosoetsa, S. (2002). At the chalk face: managerialism and the changing academic workplace 1995–2001. Transformation, 48, 59–82.

Wen, J., & Kobayashi, S. (2001). Exploring collaborative R&D network: some new evidence in Japan. Research Policy, 30(8), 1309–1319.

Wilson, C. S., & Markusova, V. A. (2004). Changes in the scientific output of Russia from 1980 to 2000, as reflected in the Science Citation Index, in relation to national politico-economic changes. Scientometrics, 59(3), 345–389.

Wohlin, C. (2009). A new index for the citation curve of researchers. Scientometrics, 81(2), 521–533.

Woods, D. R. (2005). Science after a century at Rhodes University. South African Journal of Science, 100(11 & 12), 518.

Wulf, W. M. (1993). The collaboratory opportunity. Science (New Series), 261(5123), 854–855.

258 References

Yang, J. T. Y. (1986). Collaboration between nonprofit universities and commer-cial enterprises: the rationale for exempting nonprofit universities from federal income taxation. The Yale Law Journal, 95(8), 1857–1881.

Ynalvez, M., & Shrum, W. (2011). Professional networks, scientific collaboration, and publication productivity in resource-constrained research institutions in a developing country. Research Policy, 40(2), 204–216.

Yoshikane, F., & Kageura, K. (2004). Comparative analysis of coauthorship net-works of different domains: the growth and change of networks. Scientometrics, 60(3), 433–444.

Yoshikane, F., Nozawa, T., Shibul, S., & Suzuki, T. (2009). An analysis of the con-nection between researchers’ productivity and their co-authors’ past attribu-tions, including the importance in collaboration networks. Scientometrics, 79(2), 435–449.

Zachariah, M., & Sooryamoorthy, R. (1994). Science in Participatory Development. London: Zed Books.

Zeleza, P. T. (2002). The politics of historical and social science research in Africa. Journal of South African Studies, 28(1), 9–23.

Zhao, S. (2006). Do internet users have more social ties? A call for differentiated analyses of internet use. Journal of Computer-Mediated Communication, 11(3), 844–863.

Ziman, J. (1994). Prometheus Bound: Science in a Dynamic Steady State. Cambridge: Cambridge University Press.

Zitt, K., Bassecoulard, E., & Okubo, Y. (2000). Shadows of the past in international cooperation: collaboration profiles of the top five producers of science. Scientometrics, 47(3), 627–657.

Zuckerman, H. (1967). Nobel laureates in science: patterns of productivity, col-laboration and authorship. American Sociological Review, 32(3), 391–403.

Zuckerman, H. A. (1968). Patterns of name-ordering among authors of scientific papers: a study of social symbolism and its ambiguity. American Journal of Sociology, 74(3), 276–291.

259

academic boycott, 40, 42, 43. See also academic isolation; closed-off period

academic isolation, 41, 119. See also academic boycott; closed-off period

Academy of Science of South Africa (ASSAf), 51

Africa, 17, 118African Academy of Sciences, 14Association of African Geological

Survey, 25collaboration in, 19, 224colonial research, 13Commission for Technical

Cooperation in Africa South of Sahara (CCTA), 231

connectivity, 176Internet users, 177PhD holders, 17population, 15production of knowledge, 19publications, 18research in, 18research personnel, 15science in, 15, 17, 18. See also sciencescientific cooperation, 16scientists, 15share of publications, 17, 19universities, 15, 18

African Academy of Sciences, 14agreements, 54. See also collaborationAgricultural Research Council (ARC), 36agriculture, 32Animal and Dairy Science Institute, 35apartheid, 34, 36–9, 42–5, 62, 92, 107,

108, 111, 122, 222race, 38racist policies, 41science, 44students, 38

Archaeological Survey, 31archaeology, 31, 37

Association of African Geological Survey, 25

astronomy, 26Australia, 11–13, 113

Backlund, Johan (1846–1916), 24Barnard, Christian Neethling, 42Berjak, Patricia, 197–209bibliometrics, 85, 86, 89, 91,

102, 127Biko, Steve, 41Bohle, Hermann (1876–1943), 28brain drain, 92, 94Brenner, Sydney, 46BRICS (Association of Brazil,

Russia, India, China and South Africa), 53, 55, 111

British Association for the Advancement of Science (BAAS), 24

Broom, Robert, 21Burchell, John (1781–1863), 20Bureau of Archaeology, 31

Canada, 108, 110, 112, 139, 210, 231, 233, 235

Cape of Good Hope Veterinary Medical Society, 22

Cape Society of Engineers, 22Carmichael, Dugald (1772–1827), 21cattle disease, 23, 26, 53Centre for Research in

Agroforestry, 14Chemical Metallurgical

and Mining Society, 22China, 1, 49, 90, 94, 110,

130, 231citations, 17, 19, 76, 86–8, 116, 120,

127, 128, 223analysis, 119journals, 86, 88, 90, 111, 120, 121,

124, 125, 128, 211, 217publications, 111, 116, 124–5

Index

260 Index

closed-off period, 36, 42, 118. See also academic boycott; academic isolation

co-authorships, 87–8, 102–3, 143, 175, 186, 231. See also publications, co-authored

collaboration, 2–3, 6, 12, 13, 16, 17, 24–6, 28, 34, 40, 73

administrative, 72, 218advantages, 59Africa, 13, 14, 16, 17, 19, 224apartheid, 40, 41, 222approaches, 7associations, 55. See also

professional associationsasymmetrical, 69authority, 218benefits, 59, 73, 74categories, 69challenges, 80, 220citations, 127classification, 70, 106co-authored publications, 102–3,

163, 174. See also publicationscollaboratories, 58, 63, 234colonial South Africa, 108, 156communication, 54–5, 78,

79, 173, 174, 185, 190–1, 195, 220, 223

competition, 65, 81. Compare cooperation

competitive force approach, 7components, 4, 64, 57, 79, 218concept, 64–5, 67conflict, 78, 81, 82connectivity, 175contacts, 12context, 5cost-benefit, 6, 60, 217costs, 66, 220with countries, 55, 61, 62, 68,

96–100, 108, 112, 113culture, 72, 79, 80, 171definition, 57degree of, 76, 104, 107–8, 159determinants, 5, 67–8for developed world, 83for developing world, 83development, 6

dictatorships in, 81. Compare collaboration, trust

dimensions, 64, 83, 122disagreements, 78, 82. See also

collaboration, conflictdistance, 68, 129, 215, 222domestic, 6, 61, 69, 106, 107, 125,

129, 147, 148, 154, 157, 166, 172, 186, 191, 194, 223

duration, 65, 145–8, 217effectiveness, 5, 60, 65, 80, 174effects, 68, 73, 76egalitarianism, 219elements, 64–5, 217, 236email, 181, 182, 223Europe, 13experience, 67external-institutional, 106–7,

124, 129facets, 4, 82, 135, 144factors, 3, 4, 7, 60, 66–8, 80, 149,

153–6, 223features, 145–6foreign institutions, 13, 54, 55, 75forms, 68–70, 124, 147funding, 6, 17, 63, 66, 70, 73,

113, 208gender, 152, 154government, 24, 30, 33, 34, 54,

55, 59history, 58, 93, 157ICTs, 4, 65, 82, 173–5, 183, 215, 224impact, 5, 75, 160incentives, 67, 73–4, 156index, 63, 76, 104, 129, 215indicators, 86–7, 192industry-university, 5informal, 5, 67, 217, 220informal contacts, 67, 216–18institutional, 25, 59–60, 69, 82institutional ambience, 72institutional structures, 4–6,

72, 80institutional support, 107, 171, 178institutions, 51, 62, 63, 75, 80, 103,

106, 124, 216inter-country, 13, 14inter-institutional, 6, 60, 61, 63,

69, 70

Index 261

collaboration – continued inter-university, 61internal-institutional, 106–7, 124, 129international, 6, 13, 17, 25, 26,

46, 54, 61–3, 66, 69, 77, 79, 103, 106, 107, 124, 125, 128, 129, 147–9, 154, 157, 160, 166, 175, 186, 191, 194, 195, 211, 222–4, 227

Internet in, 174–7, 186, 223intra-institutional, 69invisible college, 58knowledge production, 221leadership in, 71, 218–19local, 54–5, 69, 176, 211management, 4, 64, 70, 72marital status, 154meaning, 64–5, 67, 216, 234measures, 3, 65, 81, 88, 103, 106,

143, 152models, 153, 156, 221motives, 59, 66–7multi-institutional, 68, 73, 78, 82national, 69, 149networks, 59, 62, 67, 73,

157, 160, 186, 194, 224non-local, 107number of, 76, 107, 125organizational requirements, 72outcomes, 76, 103ownership of data, 220partners, 68, 73, 81, 82, 93, 108,

111, 112, 115, 124, 125, 129, 160, 211, 219, 221, 223

partnerships, 218, 228. See also research partnerships; collaboration

patterns, 86, 152personal elements, 7, 64, 67, 81,

216–18policies, 4, 16, 61, 63post-apartheid, 108power, 218prediction, 153, 155private sector, 61, 221process, 212, 218 productivity, 5, 66, 73, 76, 77, 101,

122, 166, 170, 192, 193, 196, 209, 210, 223

projects, 148, 172, 222properties, 83publications, 66, 73, 83, 87,

107, 111, 211rate of, 124regional, 69, 111, 149relationships, 82relevance, 57, 61research effectiveness, 80research-industry, 7rewards, 73, 75science, 58, 125,

180, 200scientific capability, 65sectors, 223significance, 57size, 78, 104, 126, 127,

130–4South Africa, 40, 43, 44, 46,

48, 49, 51–4, 74, 93, 101, 104, 124, 153, 157, 225, 226

spirit of, 81, 82structural elements, 83, 217structural support, 107structure, 6, 72, 156subjects, 130–4symmetrical, 69taxonomy, 70technical, 62technologies, 79theory, 5, 7, 217time factor, 81trade-off, 81transaction cost, 74, 217transparency, 219, 220Triple-Helix model, 30, 116trust, 77–8, 220types, 89, 104university-industry, 60, 116, 216,

235university-industry- government,

116uses, 50. See also collaboration,

benefitsvariables, 65, 80, 149, 153, 155,

222–3, 236co-publications, 124, 175, 193.

See also co-authorshipscollaboratories, 58, 63, 234

262 Index

Commelin, Caspar (1667–1731), 21Commission for Technical

Cooperation in Africa South of Sahara (CCTA), 13, 40, 231

communication, 78, 79, 163, 177, 178, 184, 186, 195

costs, 186, 187ICTs, 4, 65, 82, 173–5, 183, 215, 224means of, 183, 185, 188–90, 194media, 187problems, 183productivity, 161, 187, 190–1repertoire, 188science, 79South Africa, 176technologies, 180

Consultative Group of International Agricultural Research (CGIAR), 16

contacts, 12, 25cooperation, 11, 14, 16–18, 53–5, 106,

112, 127, 216, 226. See also collaboration

Corstorphine, George. S, 22Council for Mineral Technology

(MINTEK), 36Council for Science in Africa, 40councils, 36, 47, 51, 55, 136, 228Scientific Council for Africa South of

Sahara (CSA), 14, 25, 40, 54Council for Scientific and Industrial

Research (CSIR), 33, 36, 44, 116mission, 51

Daniell, George (1864–1937), 27Dart, Raymond, 31Davy, Joseph Burtt (1870–1940), 28development, 1, 6, 11–12, 32, 36,

39, 42, 47, 69, 72, 92, 108, 128, 144, 224

advancement, 53economic, 51economy, 144GDP, 225, 231growth, 43, 122National System of Innovation

(NST), 51–2productivity, 50programmes, 52R&D, 46, 50, 160

science, 55, 71scientific growth, 52–3South Africa, 48

de Zuid-Afrikaanse Akademie voor Taal, Letteren en Kunst, 51

Department of Agriculture, 22, 27, 33disciplines, 2, 5, 31, 63, 69, 71,

83, 92, 123, 125–7, 138, 144, 151, 154

discrimination, 39training, 39education, 39

domestic cooperation, 106Drury, James (1875–1962), 27du Toit, Petrus Johann, 31Duckit, William (1768–1825), 21

Economic Commission for Africa, 40economic development, 51economy, 144education, 27, 29, 37

apartheid, 38discrimination, 38race, 38, 39

Egypt, 17, 19, 49, 79, 124, 177, 224email use, 162, 175–80, 188, 190

collaboration, 181, 182, 187measures, 178variables, 178

engineering sciences, 50, 116, 119, 120, 231

England, 93, 94, 96, 110, 119, 112Evans, Illtyd Pole (1879–1968), 22

framework autonomy, 36France, 12, 13, 40, 54, 62, 94,

96, 110, 112, 113, 231Foundation for Research

Development (FRD), 36funding, 35, 37, 50, 52, 66, 75, 113,

225–6

GDP, 2, 43, 50, 225, 231geology, 31, 44, 48Germany, 22, 40, 52, 62, 93, 94, 96,

110, 112, 113Ghana Academy of Science and

Learning, 13Gill, David (1843–1914), 27

Index 263

Groote Schuur Hospital, 42–3Gross expenditure on R&D (GERD), 50

Hatch, Frederick. H, 22heart transplant, 42higher education, 46, 135, 176

funding, 35, 75Holloway formula, 35incentive system, 75Internet, 181race in, 37, 38South Africa, 227South African Post Secondary

Education (SAPSE) formula, 35–6

van Wyk de Vries formula, 35h-index, 75, 119, 211Historically Backward Universities, 39history

South Africa, 232Hodgson, Arthur, 28Human Sciences Research

Council (HSRC), 36

Information and Communication Technologies (ICTs), 4, 17, 65, 79, 82, 173–5, 183, 215, 224. See under collaboration

Immunology Biotechnology Laboratories, 17

impact factor, 74, 77. See also journals; publications

incentive mechanismSouth Africa, 229

incentive system, 75India, 1, 7, 49, 54, 94, 96, 110, 111,

174, 175, 231, 233, 234Ingle, Herbert, 28Institute of Tropical Meteorology, 13Institute of Veterinary Research at

Onderstepoort, 30. See also Onderstepoort Veterinary Research Institute

International Dairying Federation and the International Wine Office, 26

International Institute of Insect Physiology and Ecology, 14

International Institute of Tropical Agriculture (IITA), 16

International Laboratory for Research on Animal Diseases (ILARD), 16

International Livestock Centre for Africa, 16

International Office of Epizootic Diseases, 25–6

International Seed Testing Association, 26

internationalization, 62, 118Internet, 185, 186, 188, 191

Africa, 177collaboration, 175higher education, 181productivity, 176South Africa, 176, 177use, 161–3, 174, 176, 187

Japan Society for the Promotion of Science (JSPS), 61

Japan, 49, 50, 61, 94, 96, 110, 113, 215, 231, 235

Journal Citation Report (JCR), 89, 90journals, 32, 33, 42, 48, 49, 119,

129, 173citations, 76. See also citations,

journals; publicationsh-index, 119publications, 93South Africa, 88–9, 232

Kanthack, Francis (1872–1961), 27Kenya, 13, 14, 16, 17, 49, 96, 170,

174, 177Klug, Aron, 46knowledge

mode of, 2Mode 2, 2production, 2, 44, 63, 144, 176, 221

Koch, Robert, 21, 26Kolb, Peter, 20, 26KwaZulu-Natal, 8, 135, 147

Lehfeldt, Robert (1868–1927), 27Lichtenstein, Martin (1780–1857), 27

Mackrill, Joseph (1762–1820), 21Mandela, Nelson (1918–2013), 45

264 Index

mathematics, 15, 63, 71, 86, 88, 93, 118, 134, 225

McCrae, John (1875–1960), 24medical and health sciences, 50Medical Research Council (MRC),

36, 116medicine, 17, 32, 42, 46, 48, 71, 86,

92, 108, 118, 126, 222mineral resources, 43mining, 23, 30, 40, 45, 48, 227

National Aeronautics and Space Administration (NASA), 43

National Development Plan, 50National Institute of Health and the

Atomic Energy Commission, 43National Research and Development

Strategy (NRDS), 227National Research Foundation (NRF),

63, 228, 233National System of Innovation (NSI),

51–2networks, 58, 59, 67, 73, 157, 160,

175, 186, 188, 190, 192, 193, 195collaboration, 62, 194, 224domestic, 194international, 194science, 52size, 189type, 189

New Zealand, 48, 52, 96, 113, 237Nigeria, 14, 16, 46, 49, 79, 94,

96, 177Nobel laureates, 67Nobel Prize, 46, 58North America, 109, 112, 190

Observatories, 20, 21, 24, 26Onderstepoort Veterinary Research

Institute, 25, 27, 31, 40. See also Institute of Veterinary Research at Onderstepoort

Order of Mapungubwe, 197Organization for Economic

Co-operation and Development (OECD), 54

paleoanthropology, 32paleontology, 21, 31

partnerscollaboration, 211, 219, 221, 223

partnerships, 111, 113, 144, 149, 156, 228. See also research partner-ships; collaboration

patents, 49, 76, 225Persoon, Christiaan Hendrik

(1762–1836), 22physics, 15, 17, 24, 54, 58, 63, 66, 70,

71, 86, 93, 118, 120, 126, 127Policy, 54, 55, 61

South Africa, 226post-apartheid, 107, 141Potter, Pieter, 20productivity, 4, 5, 7, 17, 30, 66, 73,

75–7, 91–2, 102, 128, 156, 160, 164, 195

academic age, 169, 194–5activities, 192administration, 161age, 168, 173children, 168classification, 162collaboration, 124, 157, 159, 163,

166, 172, 174, 187, 192, 196, 209, 210, 223

communication, 194co-publications, 191determinants, 169domestic, 170education, 169factors, 160, 168, 170features, 167, 193gender, 194incentives, 173Internet, 161, 175, 187levels of, 171meaning, 162measures, 161–2, 191men, 168models, 168, 170–2, 192, 193patterns, 159, 163prediction, 171–3publications, 50, 80, 101, 102, 124,

159, 161rank, 161reward system, 172, 222salaries, 173scientific, 7

Index 265

sectors, 191South Africa, 124, 169variables, 160, 168, 170, 172,

193, 224women, 168

professional activities, 141–3, 183, 203professional associations, 22–5, 29, 36,

37, 44, 222, 233professional contacts, 163, 165,

184–6, 189professional networks, 187–9. See also

networksprofessional relationships, 190Project Coast, 35. See under nuclearprojects, 153

collaboration, 144, 148, 194, 222publications, 18, 19, 24, 27, 35, 36,

42, 48, 62–3, 65, 66, 73, 80, 83, 85–7, 89, 91, 92, 94, 101, 107, 119, 112–16, 118, 122, 125–6, 128, 129, 160–2, 172, 173, 183, 185, 212–13, 233

citations, 76, 86, 111, 211co-authored, 88, 102–3, 124, 126,

161, 163, 166, 187co-publications, 166honorary coauthors, 103impact factor, 74, 77Journal Citation Report (JCR),

89, 90and productivity, 17Relative Citation Impact (RCI), 119South Africa, 63, 120trends, 90visibility, 87–8world, 94, 96–100

race, 37, 38discrimination, 38, 39South Africa, 141

Relative Citation Impact (RCI), 119Rendell, Fermour, 28Research and Development (R&D), 2,

39, 43, 46–8, 50, 51, 55, 62, 160, 225, 231, 235

research effectivenesscollaboration, 80

research institutes, 115, 141, 156, 157, 178

research outputs, 127research partnerships, 67, 74,

80, 82, 93, 107, 218. See also collaboration; partnerships

research programme, 86researchers, 44, 47, 51reward system, 75, 172,

173, 222Rinderpest, 27Robertson, William, (1872–1918), 25Rogers, Arthur. W, 22Rousselet, Charles, 27Royal Observatory (Cape), 21, 27Royal Society of Northern

Antiquaries, 23Royal Society of South Africa, 22, 52Russia, 62

Sclater, William (1863–1944), 27science, 1, 2, 11–12, 15, 20, 24–31,

34, 229academic boycott, 40, 42, 43academic isolation, 41advancement, 45, 53, 63Africa, 12, 17, 22Agriculture Research Council

(ARC), 36Alzheimer’s disease, 58apartheid, 34–45archaeology, 31, 37astronomy, 26businesses in, 30capabilities, 42challenges, 228closed-off period, 118collaboration, 5, 58colonial South Africa, 25, 27, 28,

30–2, 34, 40, 53communication, 79, 180communism, 41contacts, 25Council for Mineral Technology

(MINTEK), 36Council for Science in Africa

(CSA), 40Council for Scientific and Industrial

Research (CSIR), 30, 33, 36, 44, 51, 116

266 Index

science – continued councils, 36, 47, 51, 55, 136, 228demographics, 52development, 36, 42, 108disciplines, 116, 144education, 39, 205, 228engineering sciences, 50, 116, 119,

120, 231explorations, 23gender, 140geology, 31, 44, 48government, 24, 31, 53heart transplant, 42higher education, 46Human Sciences Research Council

(HSRC), 36ICTs, 4, 65, 82, 173–5, 183, 215, 224immigration, 42incentive mechanism, 229India, 1, 7, 49, 54, 94, 96, 110, 111,

174, 175, 231, 233, 234institutions, 44internationalization, 32, 34, 62,

118, 129, 141isolation of, 42journals, 33, 42, 48, 49, 232Kenya, 16, 17mathematics, 15, 63, 71, 86, 88, 93,

118, 134, 225medical and health sciences, 50Medical Research Council (MRC),

36, 116medicine, 17, 32, 42, 46, 48, 71, 86,

92, 108, 118, 126, 222migration, 140National Research Foundation

(NRF), 233networks, 52, 160, 186, 192Nigeria, 14, 16, 46, 49, 79, 94,

96, 177Nobel laureates, 46, 67Nobel Prize, 46, 58nuclear, 35paleoanthropology, 32paleontology, 31patents, 225physics, 15, 17, 24, 54, 58, 63, 66, 70,

71, 86, 93, 118, 120, 126, 127policy, 54, 55, 61, 226

post-apartheid, 45, 47–9, 51–3professional activities, 141, 142professional associations, 22–5, 29,

36, 37, 222, 233Project Coast, 35. See under nuclearprojects, 48, 60, 135, 144, 145, 153promotion of, 16, 18publications, 27, 36, 42, 48, 62, 63,

65, 83, 85–7, 89–94, 101, 102, 118

R&D, 39, 43, 46, 47, 50, 51, 55, 62race, 38, 152research institutes, 44research partnerships, 80, 82researchers, 44, 47science citation rank order, 49science councils, 36, 47, 51, 55,

136, 228scientific societies, 44. See also

professional associationsscientific system, 52scientists, 23, 32, 43, 137–8,

140, 222sectors, 115, 125, 136, 140social sciences and humanities, 50sociology of, 2South Africa, 19, 20, 22, 23, 25,

27, 29, 32, 48, 51, 54, 55, 86, 108, 112, 116, 120, 122, 125, 127, 129, 144, 154, 156, 206, 225, 226

South African Association for the Advancement of Science (SAAAS), 23, 24, 32

South African Museum, 21, 27South Africanization, 32, 35strengths of, 122, 227structures of, 34subjects, 116toxicology, 31training, 42transformation, 37, 52tuberculosis, 48veld, study of, 22veterinary science, 22, 30

science and technology, 1, 52science citation rank order, 49science observatories, 26science policy, 54, 55, 61

Index 267

scientific cooperation, 13, 14, 16. See also collaboration

Scientific Council for Africa South of Sahara (CSA), 13

scientific growth, 48scientific institutions, 44scientific output, 119, 129Scientific Revealed Comparative

Advantage (SRCA), 48scientists, 23, 32, 43, 46, 54, 62,

142, 148brain drain, 92emigration, 92, 139

segregationracial, 41

Simpson, C. B., 28Smith, Andrew, 21Smith, Frank (1864–1950), 27social sciences and humanities, 50Somerset, Charles, 21Somerville, William (1771–1860), 21South Africa, 3, 4, 7–8, 19, 22, 24

academic boycott, 40, 42, 43academic isolation, 41, 119Academy of Science of South Africa

(ASSAf), 51agreements, 54. See also

collaborationAgriculture Research Council

(ARC), 36agriculture, 32Animal and Dairy Science

Institute, 35apartheid, 34, 36–9, 41–5, 62, 92,

107, 111, 222Archaeological Survey, 31archaeology, 31, 37astronomy, 26Biko, Steve, 41Bureau of Archaeology, 31Cape fauna, 20Cape of Good Hope Veterinary

Medical Society, 22. See also professional associations

Cape Society of Engineers, 22. See also professional associations

cattle disease, 23, 26, 53. See also Onderstepoort Veterinary Research Institute

challenges, 228Chemical Metallurgical and Mining

Society, 22. See also professional associations

citations, 116closed-off period, 36, 42, 118collaboration, 24, 28, 34, 44, 46, 62,

93, 101, 153, 157, 225, 226colonial era, 23, 27, 30, 31–3, 40,

53, 108Commission for Technical

Cooperation in Africa South of Sahara (CCTA), 40

contribution to science, 112Council for Mineral Technology

(MINTEK), 36Council for Science in Africa

(CSA), 40Council for Scientific and Industrial

Research (CSIR), 30, 33, 36, 44, 51, 116

councils, 36, 47, 51, 55, 136, 228de Zuid-Afrikaanse Akademie voor

Taal, Letteren en Kunst, 51Department of Agriculture, 27, 33development, 32, 48, 50, 51, 92discrimination, 39economy, 140education, 27, 29, 37, 38educational reforms, 28–9email use, 178emigration, 139engineering sciences, 50foreigners, 138Foundation for Research

Development (FRD), 36framework autonomy, 36funding, 35, 37, 50, 52, 75GDP, 43, 50gender, 141geology, 31, 44, 48gold production, 43Groote Schuur Hospital, 42Gross expenditure on

R&D (GERD), 50heart transplant, 42higher education, 46, 113, 135, 227Historically Backward Universities,

38, 39

268 Index

South Africa – continuedhistory, 232Holloway formula, 35Human Sciences Research Council

(HSRC), 36imperial contact, 23incentive system, 75, 229internationalization, 32, 34, 118,

129, 141Internet, 176Internet users, 177isolation of, 40, 42journals, 32, 33, 42, 48, 49, 88–9, 232KwaZulu-Natal, 8, 135, 147marine life, 20medical and health sciences, 30, 50Medical Research Council (MRC),

36, 116mineral resources, 43mining, 23, 30, 40, 45, 48, 227National Development Plan, 50National Research and Development

Strategy (NRDS), 227National Research Foundation

(NRF), 228, 233National System of Innovation

(NSI), 51–2Nobel laureates, 46Nobel Prize, 46nuclear bombs, 35nuclear reactor, 35observatories, 20, 21, 24, 26Onderstepoort Veterinary Research

Institute, 25, 27, 31, 40Order of Mapungubwe, 197paleoanthropology, 32paleontology, 21, 31patents, 49, 76, 225policy, 43, 55, 61, 226post-apartheid, 45, 52, 53, 107, 108productivity, 30, 50, 91–2professional associations, 22, 23, 29,

36, 37, 222, 233programmes, 52Project Coast, 35publications, 19, 35, 36, 42, 48, 89,

91, 92, 94, 101, 102, 112, 118, 122, 160, 169, 170–1

R&D, 39, 43, 46, 48, 49, 51, 55

race, 37, 38, 141, 152racial segregation, 41racist policies, 41research institutes, 44, 141research partnerships, 93researchers, 44, 47, 51reward system, 172, 173, 222Royal Observatory at the

Cape of Good Hope, 27Royal Society of Northern

Antiquaries, 23Royal Society of

South Africa, 22, 52science, 19, 20, 22, 23, 25, 27, 29,

47–9, 51, 54, 55, 86, 116, 127, 129, 156, 206, 225, 226

science and technology, 52science citation rank order, 49science councils, 51science education, 228science education, 228science in post-apartheid, 51science policies, 54, 55scientific associations, 46. See also

professional associationsscientific growth, 48scientific institutions, 44scientific societies, 44scientific system, 52scientists, 20, 32, 42, 54, 137–8,

140, 141, 222skills shortage, 225social sciences and

humanities, 50South African Academy for

Science and Art, 51South African Association for the

Advancement of Science (SAAAS), 23, 24, 32

South African Association of Analytical Chemists, 23, 24

South African Bureau of Standards (SABS), 36

South African College, 28, 29South African Institute for Medical

Research, 30South African Literary and

Philosophical Society, 22South African Literary Society, 22

Index 269

South Africa – continuedSouth African Museum, 21, 27South African National Committee

on Oceanographical Research (SANCOR), 35

South African Post Secondary Education (SAPSE), 35–6, 75

South African School of Forestry, 22South African Wool Textile

Research Institute, 35South Africanization of science,

32, 35Square Kilometre Array (SKA), 52strengths in science, 35, 227subjects, 116technikons, 36, 114training, 27, 42transformation, 37, 45, 46, 52, 135,

140–1, 152Transvaal Department of

Agriculture, 28Transvaal Observatory, 24universities, 22, 37–9, 41, 48, 55, 74,

75, 115, 135, 153, 173, 227, 232, 233

University of Cape of Good Hope, 29

University of Cape Town, 28University of Durban-Westville, 39University of Fort Hare, 38University of Natal, 39, 41University of South Africa, 29University of Western Cape, 38University of Witwatersrand, 28,

31, 48University of Zululand, 38van Wyk de Vries formula, 35Victoria College of Scientific

Society, 22world science, 49world share of publications, 86

South African Academy for Science and Art, 51

South African Association for the Advancement of Science (SAAAS), 24, 32

origin, 23South African Association of

Analytical Chemists, 23, 24

South African Bureau of Standards (SABS), 36

South African College, 28, 29South African Institute for Medical

Research, 30South African Literary and

Philosophical Society, 22South African Literary Society, 22South African National Committee on

Oceanographical Research (SANCOR), 35

South African Post Secondary Education (SAPSE), 35–6, 75

South African Wool Textile Research Institute, 35

South Korea, 63Southern African Development

Community (SADC), 19, 20, 53, 55

Southern African Regional Commission for the Conservation and Utilization of Soil (SARCCUS), 14

Square Kilometre Array (SKA), 52Stockman, Steward (1869–1926), 28subjects, 116Sweden, 61–2, 94, 96, 110, 139, 231

Tachard, Father, 21technikons, 36, 114Theiler, Arnold (1867–1936), 25, 27Thunberg, Carl (1743–1828), 21toxicology, 31training, 27, 42transformation, 37, 45, 46, 52, 135,

140–1, 152transparency, 219, 220Transvaal Department of

Agriculture, 28Transvaal Observatory, 24Triple-Helix model, 30, 116trust, 220Truter, Petrus (1775–1867), 21

Union of South Africa, 29universities, 15, 18, 28, 29, 31,

37, 38, 39, 41, 48, 55, 74, 75, 115, 135, 153, 173, 187, 227, 232, 233

270 Index

University of California at Berkeley, 55

University of Cape of Good Hope, 29

University of Cape Town, 28University of Durban-Westville, 39University of Fort Hare, 38University of KwaZulu-Natal, 48, 173University of Minnesota, 55University of Natal, 39, 41University of Pretoria, 22University of South Africa, 29University of Western Cape, 38University of Witwatersrand, 28,

31, 48University of Zululand, 38US, 67, 94, 112, 174Uvarov, Boris, 13

van der Stel, Willem Adriaan, 21van Wyk de Vries formula, 35veterinary science, 30Victoria College of Scientific

Society, 22

Warren, Ernest, 28Waters, Geoff, 197–209wealth, 2web use, 178–80, 182, 184WHO, 40w-index, 119Wilman, Maria, (1867–1957), 22, 27world science, 2, 17, 20, 25–6, 48,

49, 129

Young, Andrew (1873–1937), 28Young, Robert (1874–1949), 28

Date post:	11-Sep-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Transforming Science in South Africa

Documents