Maps and the settlement of southern Palestine, 1799–1948: an historical/GIS analysis

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Maps and the settlement of southern Palestine, 1799–1948: an historical/GISanalysis

Noam Levin*, Ruth Kark and Emir GalileeDepartment of Geography, The Hebrew University of Jerusalem, Mount Scopus, Jerusalem 91905, Israel

Abstract

Historical maps of the Negev Desert which comprises half of the total land area of Palestine can be viewed from several intersecting perspectives relatingto aspects such as their contribution to tracing patterns of settlement and agricultural history, imperialism and mapping, and legal geography of landownership and indigenous people. Here we focus mainly on the first theme, incorporate new methods and demonstrate their application to studies inhistorical geography.Since the end of the 18th century the Negev has attracted considerable attention due to its strategic location straddling three continents, its history, and itsarcheology. After the European powers recognized the geopolitical importance of this area in the mid 19th century, numerous surveys and mapping effortswere conducted. In this study we reviewed 375 historical maps covering parts or all of the Negev between 1799 and 1948. These historical maps are crucialto the understanding of colonial developments, as well as landscape and settlement processes and the sedentarization of the Bedouin population. Wescanned and rectified these maps using geographic information systems (GIS) to enable quantitative analysis of their accuracy, and to reveal new insightsinto settlement and sedentarization processes. Whereas the median error of maps that were based on explorers notes during most of the 19th centurywere at the order of several kilometers, the various Palestine Exploration Fund surveys (1872–1890) reduced these errors to the order of several hundredmeters, and later maps produced by the British during World War I and by British Mandatory Survey of Palestine obtained errors well below 100 m.Careful analysis of these maps allows us to delineate the boundary between cultivated land and the desert, to follow the establishment of new settlements,and to quantify the sedentarization process of the nomadic Bedouin population. We conclude that analyzing historical maps with GIS provides a tool todetermine their accuracy and hence potential usefulness for the study of topics such as settlement processes and legal disputes over land ownership.� 2009 Elsevier Ltd. All rights reserved.

Keywords: Palestine; Negev; Historical maps; Geographic Information Systems; Colonialism and mapping; Bedouin; Sedentarization

Introduction

Maps have been defined as a ‘symbolized image of geographicreality, representing selected features or characteristics’.1 Mostmapmaking has been undertaken by dominant groups that con-structed maps to promote their interests and worldview, leadingsome to conclude that, ‘maps are preeminently a language ofpower’.2 The era of colonialism was dependent on exploration andmapping, and led to innovations in surveying and mapmaking.The colonial powers produced a vast wealth of maps and surveys.3

According to Benedict Anderson they created three integrated

institutions of power: the census, the map and the museum –which shaped the ways in which the colonial state imagined itsdominion, the nature of the people it ruled, the geography of itsdomain, and the legitimacy of its ancestry.4 Following in this vein,Matthew Edney’s study of Imperial British India demonstratedhow mapmaking and imperialism intersected, and howmapmaking was used as a highly effective informational weapon,even as it pointed to a ‘structure of feeling’ that legitimated,justified, and defined imperialism’.5 Beyond military applications,maps were used for purposes of government and control, to settleland ownership issues, to draw boundaries, to partition territories,

* Corresponding author.E-mail address: [email protected] (N. Levin).

1 International Cartographic Association (1995), cited in Dorling and Fairbairn, Mapping: Ways of Representing the World, Harlow, 1997, 3.2 J.B. Harley, Maps, knowledge and power, in: D. Cosgrove and S. Daniels (Eds), The Iconography of Landscape, Cambridge, 1988, 277–312.3 A.J. Christopher, The British Empire at its Zenith, London, New York and Sydney, 1988.4 B. Anderson, Imagined Communities, Reflections on the Origin and Spread of Nationalism, London and New York, 1983.5 M.H. Edney, Mapping an Empire: the Geographical Construction of British India 1765–1843, Chicago and London, 1997, 1 and 340.

Contents lists available at ScienceDirect

Journal of Historical Geography

journal homepage: www.elsevier .com/locate/ jhg

0305-7488/$ – see front matter � 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.jhg.2009.04.001

Journal of Historical Geography 36 (2010) 1–18


and to rename places that made them more familiar to thecolonizers.6

In the wake of this literature, historical maps of the NegevDesert in southern Palestine might be viewed from at least threedifferent intersecting perspectives involving:

First, the application of GIS (Geographic Information Systems) tech-niques to incorporate historical maps in a spatial analysis. What wasthe accuracy of these maps, and what kind of spatial informationcan be extracted on historical settlement processes such as sed-entarization of nomadic peoples?Second, a focus on Imperialism and mapping. Considers maps as oneof the means used by colonial powers to legitimize and justifyimperialism, and settle border and land ownership issues.Third, an interest in legal geography, regarding maps as means tolegitimize ownership claims by the dominant group to indigenouslands, and as sources that may be used in present land ownershipdisputes.

Here we focus mainly on the first approach. We analyze multiplehistorical maps of the Negev using the modern tools of GIS, toanalyze their spatial accuracy, completeness and reliability asa quantitative source for the study of historical settlementprocesses at the margins of the Ottoman Empire and BritishMandatory Palestine.

The area known as Palestine was under Ottoman control for fourcenturies from 1517 until 1917. However, modern land registrationwas only introduced in this area following the Ottoman Land Codeof 1858.7 Detailed literal definitions of the geographical boundariesof each parcel of land appeared in many Ottoman surveys, and insome cases large scale maps of villages and their lands were also

drawn (at the scale of 1:10,000). However, very few of these mapssurvived and the Ottoman Empire had no systematic cadastralmapping within Palestine.8

The gradual entry of European nations into Palestine after theNapoleonic invasion of 1799, and their ultimate conquest of theseterritories in World War I, was accompanied by intensive mappingefforts. Systematic surveys of Palestine were performed in the 19thcentury by European organizations and individuals, mainly theFrench and the British, but also to a lesser extent by the Germans,Russians, Dutch, French, Americans and other nationalities.9

The first trigonometrically based topographic map of Palestinewas produced in 1799 by the French cartographer Pierre Jacotin atthe scale of 1:100,000 during Napoleon’s military campaign thatextended from Egypt to Acre. Six sheets of the Carte topographique del’Egypte were dedicated to Palestine, but they have been heavilycriticized by modern researchers such as Karmon and Gavish,because of the haste with which the survey was performed and theinclusion of fictitious information.10 During the 19th century therewere repeated calls for an organized survey of Palestine, ‘recognizingthe ineffectiveness of the haphazard collection of data by a lonetraveler working for a short time wherever he happened to be’.11

Interest in the mapping of Palestine increased due to its biblicalsignificance as well as for scientific reasons (such as determination ofthe altitudes along the Jordan valley). However, the mapping of thecountry was delayed because it was not especially important, eitherstrategically or geo-politically, to the European nations until the lastthird of the 19th century.12 The establishment of the British ResearchSociety of the Palestine Exploration Fund (PEF) in 1865 led to severalsurveys and mapping projects, the most notable of them being TheSurvey of Western Palestine that was conducted by lieutenants ClaudeReignier Conder and Horatio Herbert Kitchener in 1871–1877.13

6 D. Pinder, Mapping worlds: cartography and the politics of representation, in: A. Blunt, P. Gruffudd, J. May, M. Obgorn and D. Pinder (Eds), Cultural Geography in Practice,London, 2003, 172–187.

7 R. Kark and H. Gerber, Land registry maps in Palestine during the Ottoman period, The Cartographic Journal 21 (1984) 30–32; R. Kark, Land purchase and mapping ina mid-nineteenth-century Palestinian village, Palestine Exploration Quarterly 129 (1977) 150–161; R. Kark, Mamluk and Ottoman cadastral surveys and early mapping oflanded properties in Palestine, Agricultural History 71 (1997) 46–70; R.S. Fischel and R. Kark, Sultan Abdulhamid II and Palestine: private lands and imperial policy, NewPerspectives on Turkey 39 (2008) 129–166.

8 Kark and Gerber, Land registry maps in Palestine during the Ottoman period (note 7); D. Gavish and R. Kark, The cadastral mapping of Palestine, 1858–1928, TheGeographical Journal 159 (1993) 70–80.

9 Y. Ben-Arieh, The geographical exploration of the Holy Land, Palestine Exploration Quarterly 104 (1972) 81–92; Y. Jones, British military surveys of Palestine and Syria,1840–1841, The Cartographic Journal 10 (1973) 29–41; D. Gavish, A Survey of Palestine under the British Mandate, 1920–1948, London, 2005; B. Rosen, Mapping the coastline ofIsrael by the British navy, Cathedra 64 (1992) 59–78 (in Hebrew); H. Goren, Scientific organizations as agents of change: the Palestine Exploration Fund, the Deutsche Vereinzur Erforschung Palastinas and nineteenth-century Palestine, Journal of Historical Geography 27 (2001) 153–165; H. Goren, Sacred, but not surveyed: nineteenth-centurysurveys of Palestine, Imago Mundi 54 (2002) 87–110; M. Frumin, R. Rubin and D. Gavish, A Russian Naval Officer’s Chart of Haifa Bay (1772), Imago Mundi 54 (2002) 125–128;R. Kark, The lands of the Sultan – newly discovered Ottoman cadastral maps in Palestine, in: G. Tolias and D. Loupis (Eds), Mediterranean Cartographies, Athens: Institute forNeohellenic Research INR/NHRF 2004, 197–222.

10 As detailed in Gavish (see note 9). A detailed analysis of the geometric, thematic and toponymic errors in Jacotin’s map is given by Y. Karmon, An analysis of Jacotin’s mapof Palestine, Israel Exploration Journal 10 (1960) 155–173. The map itself was published as part of: P. Jacotin, Memoire de la Construction de la Carte de l’Egypte, Description del’Egypte XVII, Paris, 1824.

11 Goren (2002) 92 (see note 9). R. Kark and H. Goren, Pioneering British-Holy Land Exploration: the Palestine Association, The RGS and the PEF, ASTENE Association for theStudy of Travelers to Egypt and the Near East, International Conference, Manchester, UK, July 2005, mention an earlier attempt of the British Palestine Association establishedin 1804 to investigate and map Palestine which failed after a few years of activity.

12 J.J. Moscrop, Measuring Jerusalem, The Palestine Exploration Fund and British Interests in the Holy Land, London and New York, 2000; J.J. Moscrop, Strangers within the gates:the Royal Engineers and the Palestine Exploration Fund 1865–1870, Cathedra 103 (2002) 53–68 (in Hebrew); Interest in the region grew at the last quarter of the 19th centuryfollowing the Crimean War (1856), and especially with the building of the Suez Canal by France, and its opening in 1869. The British and the French were then colonial rivals,and as the French began mapping the Galilee in 1870, and with a pending war between Russia and the Ottoman Empire (which indeed erupted in 1876), the British came tounderstand that it was in their interest to have good maps of Palestine (Goren, Sacred, but not surveyed (note 9)). As in other areas around the globe, the colonial era gave riseto a vast wealth of maps and mapping activities. In fact, British imperialism was highly dependent on maps (see notes 3 and 5).

13 This series of 26 sheets at 1:63,360 (inch to the mile scale) was considered the best map of the Holy Land/Palestine for at least 50 years, serving as a basis for updated andnew maps, e.g. in the time of World War I and for archaeological surveys. Claude Reignier Conder and Horatio Herbert Kitchener, Map of Western Palestine, in 26 Sheets, fromsurveys conducted for the Committee of the Palestine Exploration Fund, London, The Palestine Exploration Fund, 1880; C.R. Conder and H.H. Kitchener, in: E.H. Palmer and W.Besant (Eds), The Survey of Western Palestine, Memoirs of the Topography, Orography, Hydrography and Archaeology, 3 Vols., London, 1881–1883 (Originaly 7 Vols., 3 Vols. in thefacsimile Edition of Kedem publishing house, 1970); Y. Shatner, The Map of the Land of Israel (Eretz-Yisrael) and its History, Israel, 1951 (in Hebrew); J. Elster, The BritishPalestine Exploration Fund map, in: J. Elster, M. Gilead, D. Amiran, N. Rosenan, M. Girdon, M. Zidon and N. Kadmon (Eds), Atlas of Israel, Jerusalem, Israel, 1956, I/6 (inHebrew); I.W.J. Hopkins, Nineteenth-century maps of Palestine: dual-purpose historical evidence, Imago Mundi 22 (1968) 30–36; R. Frankel, Some notes on the work of thesurvey of Western Palestine in Western Galilee, Palestine Exploration Quarterly 130 (1998) 99–105; Gavish (2005); Gavish, A Survey of Palestine under the British Mandate,1920–1948 (note 9); Y. Hodson, An introduction to the publication of the map and memoirs, in: D.M. Jacobson and Y. Hodson (Eds), Survey of Western Palestine, includinga Survey of Eastern Palestine: Introductory Essays, The United Kingdom, 1999, 33–71; N. Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land as a tool foranalyzing landscape changes: the coastal dunes of Israel as a case study, The Cartographic Journal 43 (2006) 45–67.

N. Levin et al. / Journal of Historical Geography 36 (2010) 1–182


World War I opened a new era in the mapping of Palestine withthe extensive use of aerial photographs for producing town mapsand updating the PEF maps.14 The establishment of the Departmentof Surveys in Tel-Aviv under the British Mandate in 1920, where themain offices of the Survey of Israel are located to the present day,led to a new triangulation and survey of Palestine, which replacedthe PEF maps.15

However, these mapping activities were concentrated inselected parts of Palestine, and only a few of them included theremote, arid southern region known as the Negev Desert.16 Thisarea was perceived as an extension of the Sinai Peninsula and wasmost often surveyed and mapped in association with that penin-sula or the southern parts of Eastern Palestine (known today asJordan).

The Negev is of great historical interest17 for three main reasons,first, because of the demarcation, from 1841 onwards, of its borders(by the Ottoman Empire/British Mandatory Palestine/Israel) withSinai (Egypt) to the west and Jordan to the east.18 Second, because itwas one of the major war fronts during World War I between theBritish-Egyptian Expeditionary Forces and the Ottoman (andGerman) forces, and an arena of military innovations in mappingfrom aerial photographs.19 And third, because of the contestbetween nomadic and sedentary cultures, the shifting boundary

between settled land and wilderness, and the land dispute betweenthe State of Israel and the Negev Bedouin since 1948.20

This study demonstrates the value of combining historical mapswith GIS-based analysis for assessing all of these issues, previouslyexamined almost entirely through the interrogation of writtensources.

In the pages that follow we analyze maps of the Negev producedbetween 1799 and 1948 using GIS to establish the spatial accuracyof the maps and the factors affecting their precision, and theinformation that can be extracted from these maps about changesin landscape and settlement patterns and the shifting boundaries ofthe Negev. We also identify the people and organizations respon-sible for producing these maps seeking to establish what motivatedtheir work.

We reviewed and collated some 375 maps produced between1799 and 1948 covering all or part of the Negev. All maps werescanned and then geo-referenced to allow their analysis within theframework of GIS.21 The planimetric accuracy of the maps wasassessed by comparing the coordinates of locations on historicalmaps with their coordinates on modern topographic maps(Fig. 1).22 To assess how the details shown on the maps are affectedby the scale and the date and context of the mapping, we focusedon three features: rivers, communication networks (including

14 Aerial photos were undertaken by both the German and the British-led Egyptian Expeditionary Force (EEF) which included Australians, New Zealanders, Egyptians,Indians, and even Italian and French companies. D. Amiran, Topographic maps of Israel from the days of World War I, Eretz-Yisrael 2 (1953) 33–40 (in Hebrew); D. Gavish andG. Biger, Innovative cartography in Palestine 1917–1918, The Cartographical Journal 22 (1985) 38–44; P. Collier, Innovative military mapping using aerial photography in theFirst World War: Sinai, Palestine and Mesopotamia 1914–1919, The Cartographical Journal 31 (1994) 100–104; B.Z. Kedar, The Changing Land: Between the Jordan and the Sea:Aerial Photographs from 1917 to the Present, Jerusalem, 1999; C.J. Mugnier, Grids and datums: the State of Israel, Photogrammetric Engineering and Remote Sensing 66 (2000)915–917.

15 A detailed history of the development of the British mapping during the Mandate is given in Gavish (2005); Gavish, A Survey of Palestine under the British Mandate,1920–1948 (note 9).

16 Definitions of the Negev varied with time. Being an extension of the Sinai Peninsula, the eastern limit of the Negev is commonly recognized as the Arava Valley (howeverits cartographic delineation and actual demarcation was one of the reasons for dispute between Israel and Jordan: E. Efrat, The Israel-Jordan boundary dispute in the AravaValley’, British Journal of Middle Eastern Studies 21 (1994) 229–239; Y. Gil-Har, The south-eastern limits of Palestine at the end of the Ottoman rule, Middle Eastern Studies 28(1992) 559-564; Y. Gil-Har, Boundaries delimitation: Palestine and Trans-Jordan, Middle Eastern Studies 36 (2000) 68–81. Whereas the Negev is delimited in the south by theGulf of Eilat/Akaba, its northern and western limits are less easy to define. The western boundary of the Negev is a political line which at present is demarcating the borderbetween Israel and Egypt (following previous border conflicts, see note 35. The northern limit of the Negev is more fuzzy and may represent the boundary between the‘desert and the sown’ land (a subject we deal with in detail within this manuscript) or a definition of where Desert and Mediterranean climate and biogeographic regionsmeet, based on a combination of climatic, pedologic and biogeographic considerations (see note 99).

17 R. Rubin, The Romanization of the Negev, Israel: geographical and cultural changes in the desert frontier in late antiquity, Journal of Historical Geography 23 (1997)267–283.

18 G. Biger, The Encyclopedia of International Boundaries, Israel, 1995.19 Gavish and Biger, Innovative cartography in Palestine 1917–1918 (note 14); Collier, Innovative military mapping using aerial photography in the First World War (note 14);

Y. Sheffy, British Military Intelligence in the Palestine Campaign 1914–1918, London, 1998.20 E. Epstein, Bedouin of the Negeb, Palestine Exploration Quarterly 71 (1939) 59–73; S. Krakover, Urban settlement program and land dispute resolution: the State of Israel

versus the Negev Bedouin, GeoJournal 47 (1999) 551–561; A. Meir and E. Marx, Land, Towns and Planning: the Negev Bedouin and the State of Israel, Geography ResearchForum 25 (2005) 43–62; H. Yahel, Land disputes between the Negev Bedouin and Israel, Israel Studies 11 (2006) 1–22.

21 All maps were scanned at a resolution of at least 300 dpi. Maps that were created prior to the British Mandate were rectified to the geographic (longitude/latitude)coordinate system, whereas the more recent maps were rectified to the Palestine 1923 Israel Cassini-Soldner coordinate system, depending on the coordinate system used onthe map. For a history of the development of the grids and datums used in Palestine/Israel, refer to Mugnier, Grids and datums (note 14). Map rectification was done usingcontrol points at the crossings of longitudes and latitudes (according to the tics of longitude and latitude drawn along the maps edges), as is the common practice in studies inwhich historical maps are used (e.g., M. Crowell, S.P. Leatherman and M.K. Buckley, Historical shoreline change: error analysis and mapping accuracy, Journal of CoastalResearch 7 (1991) 839–852; W.W. Locke and W.K. Wyckoff, A method for assessing the planimetric accuracy of historical maps: the case of the Colorado-Green river system,Professional Geographer 45 (1993) 416–424; Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land as a tool for analyzing landscape changes (note 13). Thereason for this is that the position of graticules intersection can be accurately pinpointed on a map (whereas symbols of other features may occupy an area and not a point); inaddition some of the smaller scale historical maps lack points of known coordinates (such as trigonometric points). More details of the basic mathematics of historical maprectification can be found in Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land as a tool for analyzing landscape changes (note 13) as well as in D.L.Verbyla, Practical GIS Analysis, London and New York, 2002. The order of the polynomial transformation used was determined according to the distortions in the scanned mapand its projection, following H.G. Buiten and B. van Putten, Quality assessment of remote sensing image registration – analysis and testing of control point residuals, ISPRSJournal of Photogrammetry and Remote Sensing 52 (1997) 57–73; B. Zitova and J. Flusser, Image registration methods: a survey, Image and Vision Computing 21 (2003)977–1000. The rectification of the maps and the calculation of root mean squared error (RMSE) of the control points used for the transformation were done using ENVI 3.4remote sensing software package (Research Systems, Inc. (2001). ENVI Version 3.4, 4990 Pearl East Circle, Boulder, CO 80301, USA: The Environment for Visualizing Images.).Appendix 1 lists the transformation details for some of the key maps of the Negev that were analyzed.

22 We assessed the planimetric accuracy of the historical maps by comparing the positions of 40 control points (representing villages, ruins or archeological sites; Fig. 1) inthe historical maps and in 1:50,000 topographic maps of the Survey of Israel (2000), after adjusting for possible constant shifts between maps (such as the shift in the mapsof the Survey of Western Palestine as described by Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land as a tool for analyzing landscape changes (note13)). As the scale of most of the maps is equal to or smaller than 1:200,000 most of these places are represented as points and not as polygons, facilitating the determinationof their exact location. In the larger scale maps, the position of the control point was determined either at the centre of the polygon, or according to the location ofa prominent feature within it (e.g. the central mosque). The majority of these control points were collected from the northern part of the Negev – the transition zone betweenthe desert and the Mediterranean regions, as this area is shown on maps that do not include the Negev (e.g. The Survey of Western Palestine by Conder and Kitchener, 1880)or on maps that are devoted to the Negev or to the Sinai Peninsula.

N. Levin et al. / Journal of Historical Geography 36 (2010) 1–18 3


roads, and trails), and isolated buildings/houses. All rivers androads were digitized from 15 selected maps within an area of100 km2 north of Beersheba. Isolated buildings or houses (outsidethe areas of villages, towns or cities) were digitized from fiveselected maps within an area of 600 km2 north and north-west ofBeersheba.23

Mapping the Negev

Maps by explorers and professional survey teams

Most maps produced by explorers and professional survey teamswere motivated by geopolitical interests and events such as changeof regime (from Ottoman Empire to British Mandate) and by wars;

most were published during the 20th century (Fig. 2A) generally atever larger scales (Fig. 2B).

French surveyors engaged in Napoleon’s campaign in Egypt andPalestine in 1799 worked mainly in the coastal plain where theFrench army advanced. Indeed, the area of the Negev on Jacotin’smap (Fig. 3A) is almost empty; it shows few place names, someimaginary mountain ridges, and the administrative boundarieswithin the Ottoman Empire by a dotted line in the northernNegev.24 This reconnaissance map was the least accurate of all themaps we analyzed, with a median planimetric error of more than8.8 km (Appendix 1).

Many early explorers in this area lacked the ability to determinetheir positions using astronomical measurements, and producedmaps on the basis of azimuth measurements (using a compass),and distance measurements (derived by counting their steps or

Fig. 1. The distribution of the control points that were used for assessing the accuracy of the maps.

23 Only five historical maps had detailed enough depiction of isolated buildings to allow this analysis. The five maps used were the following: Conder and Kitchener, 1880;Survey of Egypt, 1919; Ottoman, 1919; Survey of Palestine, 1936–1939 and 1945/6.

24 Shatner, The Map of the Land of Israel (Eretz-Yisrael) and its History (note 13); Karmon, An analysis of Jacotin’s map of Palestine (note 10); Gavish (2005); Gavish, A Survey ofPalestine under the British Mandate, 1920–1948 (note 9).



travel time). Few were able to cover the whole country, but theirnotes characteristically provided detailed descriptions of ruins,villages, towns and the landscapes through which they traveled.Some of these explorers (such as John Lewis Burckhardt, HenryBaker Tristram, Edward Robinson and Rev. Eli Smith) handed theirdata to professional cartographers (such as Heinrich Berghaus orHeinrich Kiepert) to compile their maps (Table 1 and Appendices).The maps produced by Kiepert (1840, 1841; Fig. 3B), based on the1838 travels of Robinson and Smith, are much more detailed thanthose compiled by Jacotin.25 The routes of the explorers are shown,and many place names and names of Bedouin tribes and majortrails are provided. Large areas beyond the explorers’ traversesremain blank, however. Other explorers compiled their own maps;among these the maps by Victor Guerin and Charles W. MeredithVan de Velde show only the northern part of the Negev, although

this is represented more accurately than on the Jacotin map(Appendix 1).26

The establishment of the British Palestine Exploration Fund(PEF) in 1865 revolutionized the mapping of Palestine, as itssurvey parties included British Royal Engineers officers equippedwith professional surveying equipment. The first PEF mapcovering the Negev was the ‘Route Map of the Negeb of SouthCountry and part of the Desert of Et Tih’ by Edward Henry Palmerand Charles F. Tyrwhitt Drake (1871) based on a continuation ofthe Sinai survey of 1869–70 (Fig. 3C).27 The survey of WesternPalestine, led by Conder and Kitchener between the years 1871–1877, was one of the greatest achievements of the PEF.28 Theeponymous 1880 map they produced on a scale of 1:63,360 wasthe largest scale and most comprehensive map of Palestine to thatdate (Fig. 2B). It was also considerably more accurate than its

Fig. 2. A: Frequency of Negev maps along the years; B: Improvement of scale in Negev maps along the years.

25 E. Robinson, Biblical Researches in Palestine, Mount Sinai and Arabia Petraea, 3 Vols., Southampton, 1841.26 V. Guerin, Description Geographique, Historique et Archeologique de la Palestine, Paris, 1869 (‘Judee’), 1874 (‘Samarie’), 1880 (‘Galilee’), 7 Vols; C.W.M. Van de Velde, Memoir

to Accompany the Map of the Holy Land, Gotha, 1852; C.W.M. Van de Velde, Narrative of a Journey through Syria and Palestine in 1851 and 1852, 2 Vols., London, 1854.27 E.H. Palmer, The Desert of the Exodus, 2 Vols., Cambridge, 1871.28 Conder and Kitchner, Map of Western Palestine (note 13); Conder and Kitchner, The Survey of Western Palestine, Memoirs of the Topography, Orography, Hydrography and

Archaeology (note 13); For more details on the accuracy of the Survey of Western Palestine, see Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land asa tool for analyzing landscape changes (note 13).



Fig. 3. Examples of some Negev maps. A: Jacotin (1826); B: Kiepert (1841); C: Maps covering the Negev by the Palestine Exploration Fund: Palmer and Drake (1871), Conder andKitchener (1880), Schumacher (1886) and Kitchener, Armstrong and Hull (1890); D: Musil (1902); E: TSGS (1906, 1907); F: Fischer (1910); G: War Office (1915); H: Survey ofPalestine (1936–1939). This map is overlaid by the current boundaries between Israel and its neighboring countries (in black) for orientation.



predecessors; the median planimetric error in the northern Negevwas less than 300 m (Appendix 1).29 This was the first topographicmap to cover the biblical extent of the Land of Israel. It showed allruined and permanent settlements as well as isolated housesoutside villages and agricultural plantation areas, and providedmore than seventy names of Bedouin tribes in Palestine, althoughit extended only to Wady es Seba (Beersheba river-bed) in thesouth, and left most of the Negev unmapped (Fig. 3C). Later,smaller surveys for the PEF by Gottlieb Schumacher (1886) in thecoastal plain south of Gaza (which followed the Survey of WesternPalestine in style but was at a smaller scale of 1:100,000), and byKitchener, Armstrong and Hull (1890) from which the Map of theWady el Arabah was produced (Fig. 3C) extended careful mappingto the south, but many areas of the Negev remained un-surveyed.30

Early in the 20th century a remarkable map based on extensivesurveys by the Czech scholar Alois Musil in the Negev and the areanow known as southern Jordan provided invaluable detail aboutvarious types of archeological sites, permanent settlements, semi-permanent villages, and plantations although its planimetricaccuracy was less than that of the Survey of Western Palestineundertaken in the years 1871–1877 (Appendix 1).31

Atlas maps

Atlas maps were commonly compiled by professional cartogra-phers, and their accuracy was based on their ability to selectreliable sources. These maps showed the whole of Palestine ina single sheet (e.g. the maps compiled by Hans Fischer and pub-lished in 1890, 1910, 1911), or various sections of the territory onlarger scales (e.g. the map made by Carl Zimmerman in 1850).Fischer’s maps are quite accurate, with median planimetric loca-tion errors of less than 1 mm on the map at this scale (Fig. 3F,

Appendix 1).32 They are also unique in showing the boundariesbetween permanently settled areas and the wilderness. Workingfor the Deutschen Palastina-Vereins (DPV, the German Society ofPalestine) Fischer also published a detailed description of thedevelopment of the cartography of Palestine and of the sources heused in his maps.33

British maps 1900–1915

A dispute over the boundary between the Ottoman Empire andBritish-governed Egypt brought British-Egyptian and Ottoman forcesto the brink of war in January 1906.34 A Turkish-Egyptian boundaryagreement was signed in October 1st 1906, and that boundary(slightly modified in 1915) forms the basis for the border betweenIsrael and Egypt today.35 Following surveys by Jennings Bramley inthe summer of 1902 and the boundary demarcation in the summer of1906 by E.B.H. Wade (on the British side), the British TopographicalSection General Staff (TSGS,1906,1907) issued several maps showingthe boundary but lacking in other detail (Fig. 3E).

In the years immediately following, the British began to surveythe Sinai Peninsula and the Negev more systematically, to producemaps suited to the logistical needs of the army (although twoarcheologists, Charles Leonard Woolley and Thomas Edward Law-rence from the British Museum, were attached to the survey toconceal its purpose).36 The whole of Sinai (then under Britishcontrol) north of lat. 28�300 was triangulated by the Survey of Egyptand mapped topographically by parties of officers from theGeographical Section of the General Staff, under the direction ofColonel Stewart Francis Newcombe from the Royal Engineers,between 1909 and 1914.37 On the 29th of October, 1913 theOttoman Empire granted permission for a survey of southernPalestine. Five survey parties under Newcombe’s directioncompleted work in the Negev in the first four months of 1914.38 The

29 This is slightly higher than the median error of 154 m of the whole map as reported by Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Land as a toolfor analyzing landscape changes (note 13).

30 G. Schumacher, Researches in southern Palestine, Palestine Exploration Fund Quarterly Statement for 1886 (1886) 171–197.31 A. Musil, Arabia Petraea, Wien, 1907–1908. The map was drawn in 1902, the book was published in 1908, and the Survey was taken in the years 1895–1902.32 I.e. that the distance between the position as shown on the historical map and that shown on modern topographic maps translates to less than 1 mm on the map at the

scale of the historical map. Note that in historical maps errors of at least 0.81 mm are expected: K. Thapa and J. Bossler, Accuracy of spatial data used in geographicinformation systems, Photogrammetric Engineering and Remote Sensing 58 (1992) 835–841; see also Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Landas a tool for analyzing landscape changes (note 13).

33 H. Fischer, Begleitworte zur karte des Syrisch-Agyptischen Grenzgebiets, Zeitschrift des Deutschen Palastina-Vereins 33 (1910) 188–221; H. Fishcer, Geschichte der kar-tographie von Palastina, Zeitschrift des Deutschen Palastina-Vereins 62 (1939) 169–189; H. Fishcer, Geschichte der kartographie von Palastina, Zeitschrift des Deutschen Pal-astina-Vereins 63 (1940), 1–111.

34 Following the British occupation of Egypt in 1882, the British aimed to have full control of the Sinai Peninsula as a buffer zone to protect the strategic Suez Canal. Since1841 the official eastern boundary of Egypt in the Sinai Peninsula was along a line that started near Khan Junis (south of Gaza) and ended in Suez – the southern end of theSuez Canal (G. Biger, The rediscovery of the first geographic-political map of Sinai, in: A. Degani (Ed.) Sinai, Vol. B, Tel-Aviv, 1987, 907–911 (in Hebrew); see also Biger, TheEncyclopedia of International Boundaries (note 18). The British Major W.E. Jennings Bramley was appointed in 1902 to investigate the water sources in the Sinai Peninsula andin the Negev and to determine the optimal route through which to pass the border between the Ottoman Empire in Palestine and British controlled Egypt (see note 37). Hisobservations on the land and people were published in thirty-three installments over nearly a decade in the Quarterly Statements of the Palestine Exploration Fund(commonly known as Palestine Exploration Quarterly).

35 E.B.H. Wade, Report on the Delimitation of the Turko-Egyptian Boundary in the Sinai Peninsula, June–September, 1906, Egypt, 1908; M. Brawer, Geographic factors in thedemarcation of the Israel-Egypt boundary, Researches in the Geography of the Eretz-Yisrael (Israel) 7 (1970) 125–137 (in Hebrew); G.R. Warburg, The Sinai Peninsula Borders,1906–1947, Journal of Contemporary History 14 (1979) 677–692; N. Kadmon, A cartographic survey of the mapping background to the demarcation of the internationalboundary near Taba in the peace agreement between Israel and Egypt, Researches in the Geography of the Eretz-Yisrael (Israel) 14 (1993) 50–70 (in Hebrew); A. Kemp andU. Ben-Eliezer, Dramatizing sovereignty: the construction of territorial dispute in the Israeli–Egyptian border at Taba, Political Geography 19 (2000) 315–344; N. Kliot,The evolution of the Egypt–Israel boundary: from colonial foundations to peaceful borders, in: C. Schofield (Ed.), Boundary and Territory Briefing, Vol 1(8), (1995) 1–22.International Boundaries Research Unit.

36 J. Wilson, Lawrence of Arabia – the Authorised Biography, London, 1989; C.L. Woolley and T.E. Lawrence, The Wilderness of Zin, London, 1915; B.A. Saidel and G.L. Chris-topherson, Four days at Khalasa: using aerial photography and GIS analysis to reappraise Woolley and Lawrence’s survey of Byzantine Elusa in the Western Negev Desert,Palestine Exploration Quarterly 137 (2005) 53–63.

37 G.W. Murray, The land of Sinai, The Geographical Journal 119 (1953) 140–153.38 The end of the Balkan wars in 1912/1913 eased the tensions in the area, and the time seemed fit to launch a survey mission to southern Palestine (which was then under

Ottoman control). As with the Survey of Western Palestine, the PEF asked permission for the mapping, to ‘record upon the map the position of all ruins of ancient towns, andof other archeological remains. The work will be exploration, not excavation, and. increase our understanding of the history of the Patriarchs and the Israelites’ (C.M.Watson, The desert of the wanderings, Palestine Exploration Fund Quarterly Statement for 1914 (1914) 18–23.



resulting map carried the triangulation work of the Egyptian Surveyeast from the Sinai Peninsula and discovered that the longitudinalinformation in Conder and Kitchener’s Survey of Western Palestinewas about half a mile in error.39

Newcombe’s maps were published by the British War Office in1915, at two scales: 1:125,000 and 1:250,000 (Appendices 1 and 2,and Fig. 3G). They completed and replaced maps derived from theearlier Survey of Western Palestine. The military nature of thesemaps is evident. They contain information on the roads (whetherthey are passable by light wheeled traffic or by fully laden camels),on sand dune areas (where favorable crossings are noted), on thelocation of water supplies (including their quality and quantity anddepiction of temporary or permanent water holes), and on relief(contours at approximately 100 feet intervals). In addition, eachsheet includes written details (‘General Note’) about the nature ofthe country, roads, fuel, supplies, etc.

German and Ottoman maps in World War I

Newcombe’s maps of the Negev were kept from the Turkish-German forces at the outbreak of World War I. The Germanstherefore ordered Major Von Ramsay to prepare a 1:250,000 mapcovering the Negev and the northern Sinai Peninsula. He did this bycompiling existing maps rather than by surveys and his work wasbarely half as accurate as Newcombe’s map (Appendix 1). As thewar proceeded, both sides realized that their maps were insuffi-ciently detailed for tactical purposes, especially when the British-led Egyptian Expeditionary Forces reached the populated and well-defended areas in southern Palestine, along the frontline betweenGaza and Beersheba.40 The German Palestine World War I maps at1:50,000 (prepared in 1917–1918) did not include the Negev, andthose undertaken at a 1:200,000 scale by Ottoman surveyorsduring 1917 were not published until 1918.41

Maps and aerial photography

In 1915 the British began to develop new methods of mappingusing aerial photographs, and the fact that the northern Negev wasthe scene of many battles led to a dramatic increase in the prepa-ration of maps (Fig. 2A).42 In March 1917 the War Office expanded

the Topographical Section of the Egyptian Expeditionary Force(EEF) to become the 7th Field Survey Company, Royal Engineers,which included some 190 men by the end of 1917.43 Altogether theEEF mapped an extensive area of 1280 km2 between Gaza andBeersheba using aerial photographs.44 The rate of the photoreconnaissance increased during 1917, and each month hundreds ofphotographs were taken, from which thousands of prints wereprepared.45 Supplementary control was provided by theodoliteobservations wherever possible.46 From these photographsdetailed town maps and frontline maps were prepared at scales of1:10,000 and 1:15,000. A series of contour maps at 1:20,000 and1:40,000 were also produced, mostly based on aerial photographs.The Seventh Field Survey Company used aerial photography almostexclusively for mapping areas behind Turkish lines. As these areascame under British control, the air survey maps were replaced bymaps based on conventional plane table survey techniques.47

While these maps were very detailed (they presented houses,plantations and detail about the quality of roads) and much moreaccurate than previous maps of the area (median planimetric errorswere well below 100 m; Appendix 1), they lacked grid coordinatesand some of the contours were inaccurate.48

Maps by the Survey of Palestine 1920–1947

Article 11 of the Council of the League of Nations to the Mandate ofPalestine states that the Mandatory administration ‘. shall intro-duce a land system appropriate to the needs of the country, havingregard, among other things, to the desirability of promoting theclose settlement and intensive cultivation of the land’.49 To meetthis mandate, which they assumed on April 15th, 1920, the Britishneeded civilian maps for administrative and especially cadastralpurposes to settle issues related to land ownership.50 The Survey ofPalestine was founded in the summer of 1920, to conduct a cadas-tral survey.51 It began near Gaza, and the hill of Ali el Muntar, whichoverlooks Gaza from the south, received the coordinates of 100E–100N (km). The triangulation network was largely completedbetween May 1921 and 1923, but most of the Negev was excluded.By 1930, the Survey of Palestine had established a geodetic networkof 105 major and 20,000 minor control points, based on the Cassini-Soldner projection, and claimed that Palestine was ‘the only

39 S.F. Newcombe, Annual meeting, Palestine Exploration Fund Quarterly Statement for 1914 (1914) 121–124; S.F. Newcombe, The Survey of Sinai and south Palestine, PalestineExploration Fund Quarterly Statement for 1914 (1914) 128–133. Compare with Hopkins, Nineteenth-century maps of Palestine (note 13); Levin, The Palestine Exploration Fundmap (1871–1877) of the Holy Land as a tool for analyzing landscape changes (note 13).

40 See additional details in Gavish (2005); Gavish, A Survey of Palestine under the British Mandate, 1920–1948 (note 9).41 D. Gavish, The Ottoman topographical mapping in Palestine, 1917–1918, in: E. Pimentel and E. Schiller (Eds), The First World War in Palestine: the Day After, The Battles

Around Beer Sheva in November 1917, Sde-Boker, 2006, 13–17. These maps had several versions (in Turkish or later on also in French), were first published in 1918 and coveredthe Negev only north of a line extending from Gaza to Beer Sheba and to the southern end of the Dead Sea. These maps had a median error of about 600 m as in the maps ofVon Ramsay, however they presented the locations of some isolated houses and buildings outside of villages.

42 H.H. Thomas, Geographical reconnaissance by aeroplane photography, with special reference to the work done on the Palestine front, The Geographical Journal 55 (1920)349–370; G. Salmond, C. Hedley, H.S.L. Winterbotham, A.R. Hinks and E.M. Jack, Geographical reconnaissance by aeroplane photography, with special reference to the workdone on the Palestine front: discussion, The Geographical Journal 55 (1920) 370–376; Gavish and Biger, Innovative cartography in Palestine 1917–1918 (note 14); Collier,Innovative military mapping using aerial photography in the First World War (note 14); Sheffy, British Military Intelligence in the Palestine Campaign 1914–1918 (note 19);Kedar, The Changing Land: Between the Jordan and the Sea (note 14); P. Collier and R.J. Inkpen, Mapping Palestine and Mesopotamia in the First World War, The CartographicJournal 38 (2001) 143–154.

43 Sheffy, British Military Intelligence in the Palestine Campaign 1914–1918 (note 19).44 See additional details in Gavish (2005); Gavish, A Survey of Palestine under the British Mandate, 1920–1948, (note 9).45 Sheffy, British Military Intelligence in the Palestine Campaign 1914–1918 (note 19).46 P. Collier, The impact on topographic mapping of developments in land and air survey: 1900–1939, Cartography and Geographic Information Science 29 (2002) 155–174.47 Collier and Inkpen, Mapping Palestine and Mesopotamia in the First World War (note 42).48 Amiran, Topographic maps of Israel from the days of World War I (note 14); Sheffy, British Military Intelligence in the Palestine Campaign 1914–1918 (note 19).49 http://www.yale.edu/lawweb/avalon/mideast/palmanda.htm (accessed 2007 January 1st).50 ‘Cadastral maps were used by state governments in the Old and New Worlds to organize, control and record the settlement of ‘empty’ lands’: R.J.P. Kain and E. Baigent,

The Cadastral Map in the Service of the State, A History of Property Mapping, Chicago and London, 1992. See additional details regarding the cadastral mapping of Palestine inGavish (2005); Gavish, A Survey of Palestine under the British Mandate, 1920–1948 (note 9).

51 See note 14.



country in the world which has succeeded in completing its entireframework in advance of the detail survey’.52

Despite the experience gained in aerial mapping during WorldWar I, air surveys were considered inadequate for the cadastralsurvey because they were insufficiently exact for a 1:2,500 scale,because they would be more expensive, and because field surveyswere required for naming and verification.53 The basic topo-cadastral survey began in 1928 and ended in November 1933.However, in southern Palestine only the southern coastal plainareas were surveyed and mapped. Even at the end of the BritishMandate, the 1:20,000 map sheets covered only part of thenorthern Negev. The southernmost sheets (each sheet covering anarea of 10�10 km) did not go beyond Beersheba, and to thenortheast of Beersheba the southernmost sheet was that coveringthe village of Dhahiriya, prepared in 1945.

In 1933 the Survey of Palestine began preparing a topographicmap series at the scale of 1:100,000, under the direction of Lieu-tenant-Colonel Frederick John Salmon. Salmon believed that ‘eachdifferent type of country has a different system of mapping suitedto it, while the principal uses to which the map is to be put willaffect the style’.54 For Palestine, he intended that the topographicmap would ‘distinguish the desert from the sown’. Had ‘time andmoney allowed of mapping on the 1/50,000 scale,’ he would haveproceeded thus, because ‘on the 1/100,000 scale, which it has beennecessary to adopt, the inclusion of so many names, some of themrather long, has been difficult’. Salmon also ‘preferred to coverPalestine slowly, since the major task involved in the cadastre in thecountry was not the mapping but the work of the land settlementofficer in the field’.55 Production of the first version of the1:100,000 topographic maps was accelerated and completedduring the Arab Revolt of 1936–39 in Palestine, but they coveredonly the northern Negev (Fig. 3H).

In addition to topographic mapping the British issued maps atsmaller scales. Maps at a scale of 1:250,000 were used duringWorld War I, in both a small scale version of Newcombe’s 1915maps and in maps issued by the War Office (1917) – later updatedby the Ordnance Survey (e.g. in 1925). In the late 1930s, the Surveyof Palestine began to issue a new series of maps at this scale: ThePalestine Index to Villages and Settlements. These maps weremainly devoted to marking all villages and settlements in Palestine,showing district and village boundaries and distinguishingbetween Arab and Jewish villages. The northern Negev (about12 km south of Beersheba) was mapped at 1:250,000, and the restof the Negev was usually shown in an inset map at 1:1,000,000.Various versions of the Index maps were produced, and some ofthem included state domains/lands, forest reserves or Jewish-owned lands. One of the most interesting versions of this map isone titled ‘Distribution of the Nomad Population of the BeershebaSub-district’.56 This map will be analyzed in detail below.

Aerial photography over Palestine resumed only in World War II,when trained units photographed broad tracts of the territory aspart of their wide-ranging aerial photography missions throughoutthe Middle East. Still, only towards the end of the war, in December1944, when political considerations reinforced military arguments,did the British agree to conduct complete, systematic, and orderlyaerial photography’.57 This created the Palestine Survey (PS) series,photographed over a five-month period, forming the standardbaseline for comparison of photographs from various periods in thisarea.58

Jewish mapping

Working on behalf of the Jewish Agency, the Jewish surveyor andcartographer, Zalman Lif (Lifschitz) and Yehoshua Hankin issueda reduced version of the 1930s topo-cadastral maps in Hebrew ata scale of 1:40,000, showing land in Jewish possession. However,these maps barely reached the northern Negev. Lifschitz also issuedphysical maps of Palestine in Hebrew at the scale of 1:250,000,reaching as far as about 22 km south of Beersheba. As the topo-graphic mapping from the British Mandate period ended at a linesouth of Beersheba, in 1948–1949 the newly established Survey ofIsrael produced new maps at a scale of 1:100,000 covering all of theNegev down to Eilat by using old cartographic sources, includingNewcombe’s pre-World War I survey.59 This was done in order tofacilitate military action in the Negev area.

In summary, mapping of the Negev intensified and maps wereproduced at larger scales (>1:100,000) over time. The introductionof modern mapping techniques and the transition from maps madeby lone explorers, to work by specific survey missions and orga-nized mapping by governments increased the planimetric accuracyof maps (Figs 2 and 4). Statistically, the correlation between theyear of a map and median error as estimated from the Negev mapswas very high when measured in real world units (m; R2¼ 0.80,n¼ 33). This is due to improved surveying methods as well as to thelarger scale of the maps which has increased with time (R2¼ 0.56,n¼ 33, for the relationship between year of publication and scalefor the maps whose accuracy was reported above).

Maps and settlement patterns in the northern Negev

The rural and nomad sectors

Towards the end of the Ottoman period, and during the Mandateperiod (late 19th century, and mainly in the first half of the 20thcentury) semi-nomadic Bedouin inhabitants of the northern Negevbegan to adopt a more settled pattern of existence associated withthe construction and use of a few permanent buildings. Whereassemi nomads (sheep and goat Bedouins) wander over smaller

52 R. Home, Scientific survey and land settlement in British colonialism, with particular reference to land tenure reform in the Middle East 1920–50, Planning Perspectives 21(2006) 1–22. Mugnier, Grids and datums (note 14).

53 H.S.L. Winterbotham, The economic limits of aeroplane photography for mapping, and its applicability to cadastral plans, The Geographical Journal 56 (1920) 481–483; D.Gavish, An account of an unrealized aerial cadastral survey in Palestine under the British Mandate, The Geographical Journal 153 (1987) 93–98; P. Collier, R.J. Inkpen,Photogrammetry in the Ordnance Survey from Close to Macleod, Photogrammetric Record 18 (2003) 224–243.

54 F.J. Salmon, Some notes on conventional signs for topographical maps, The Geographical Journal 89 (1937) 50–53.55 As quoted in Gavish, An account of an unrealized aerial cadastral survey in Palestine under the British Mandate (note 53), 97.56 This map was compiled from information collected in the census of 1946 and from aerial photographs taken January–June 1945. Some background information on the

compilation of this map can be found in Appendices 3 and 4 of the following publication: United Nations, General Assembly, ‘Ad Hoc Committee on the Palestinian Question,Report of Sub-Committee 2’ (1947), Document A/AC.14/32.

57 Quote from the website of D. Gavish: http://www.photogis.huji.ac.il/aero/public/ap-history.htm (accessed on January 3rd, 2007); see more in Gavish (2005); Gavish, ASurvey of Palestine under the British Mandate, 1920–1948 (note 9).

58 An example of a study utilizing these historical aerial photographs is that of N. Levin and E. Ben-Dor, Monitoring sand dune stabilization along the coastal dunes ofAshdod-Nizanim, Israel, 1945–1999, Journal of Arid Environments 58 (2004) 335–355.

59 D. Gavish, Foreign intelligence maps: offshoots of the 1:100,000 topographic map of Israel, Imago Mundi 48 (1996) 174–184.



Fig. 4. The density of the roads and rivers networks on maps from different periods and different scales. Each row relates to a different map, and each column to a different feature.Column I: The historical maps; Column II: The roads (including trails, dirt roads and railways); Column III: The rivers (including wadis). Row A: Conder and Kitchener (1880); Row B:Musil (1902); Row C: Survey of Egypt (1919); Row D: Survey of Palestine (1936–1939); Row E: Survey of Palestine (1945–1946); Row F: Survey of Israel (2000).



territories nearer the steppe and the cultivated regions, full nomads(camel Bedouins) are characterized by seasonal migration atgreater distance in search of suitable grazing areas. Semi nomadismmay lead to the initiation of sedentarization characterized bybuilding of storage structures, and houses.60 Written documenta-tion of this process is scarce and one of the few sources that can beused for its systematic reconstruction over time is the analysis ofmaps. In an effort to chart this shift we first examined represen-tations of two different networks on maps of the region: wadis (dryriver-beds), that are unlikely to undergo change over a century or soand trails and roads that are far more susceptible to change overtime.61 The level of detail pertaining to both networks differedgreatly among maps (Fig. 4, Appendix 1), but whereas the density ofboth networks was dependent on the scale of the maps (R2¼ 0.66and 0.69 for the rivers and roads, respectively), only the density ofthe road network increased with time (R2¼ 0.75), an indicator ofthe rapid development that took place in the area especially sinceWorld War I.

The northern Negev forms a transition zone between the desertand the Mediterranean climate regions. The Negev itself rangesbetween semi-arid to extremely arid in its southern end, while itsnorthern limit is usually defined by 200–300 mm/year isohyets.62

The native vegetation is transitional between steppe and desertassociations, and consists mainly of dwarf shrub communities. Inthe more arid areas of the southern and eastern Negev, shrubs andtrees are confined to wadis (river beds).63 The number and extent ofsettlements in the Negev has varied throughout history as a func-tion of climatic conditions, the available technology and thestrength of the central government to control the nomadic pop-ulation of the Negev.64

In such areas, on the margins of the desert, attacks by Bedouinsled the local population to concentrate in large villages that couldwithstand such raids.65 This led to larger villages at a lowerdensity in the southern Hebron Hills. According to the Survey ofWestern Palestine and Shucmacher’s 1886 map (Fig. 5A), only twoinland villages in the southern Hebron Hills were located where

yearly rainfall is lower than 350 mm: Edd-Dhahariyeh andes-Semua. The following description demonstrating the unstablesituation of settlements in southern Palestine relates to the Arabvillage of Edd-Dhahariyeh (the last village before the desert66) wasgiven by Kitchener (Fig. 5): ‘I found out here that the next villageI was going to, Dhoheriyeh, was entirely deserted. Owing to thebad year, the inhabitants were not able to pay taxes, and foundit better to desert their homes’.67 ‘This village,’ he continued,‘contained some 300–400 persons in 1874; but in 1877 it wasdeserted, in consequence of the encroachment of the Arabs[i.e. Bedouin] into the country of the fellahin’.68 Based on theanalysis of the maps we can see that along the southern coastalplain, in the 1870–80s there were several villages south of theisohyet of 300 mm/year (Fig. 5A). In these villages the local pop-ulation developed and operated a ‘Mawassi’ agricultural systemwhich relies on the high level of coastal underground water ininter-dune areas.69

As part of a policy to block or settle the Bedouins in southernand eastern Palestine, Transjordan, and the Syrian Desert, modernurban administrative centers were established on the fringe of thesettled land. In 1899, Beersheba was established where theBedouins had previously pitched their tents, in order to enforceOttoman control over the Bedouins in the Negev. Two Arab engi-neers and a Swiss and a German architect formulated the gridpattern of the town plan. From 1902 to 1911, the number of towndwellers in Beersheba increased from 300 to over 800. A few yearslater another new town was planned in Auja el-Hafir, in southernPalestine near the new (1906) border with Egypt.70

A few new villages were also founded in the late 19th century, asa means to prevent the threat of raids by wandering Bedouin intothe settled regions.71 As we can deduce mainly from detailedOttoman maps from the end of the 19th century, the OttomanSultan Abdulhamid II initiated the building of three modern newvillages: Kaufakha, El-Muharraqa, and El-Jaladia on his privatelands in the northern Negev. There are very few archival documentsthat give us information on each village, and only through the

60 Patterns of nomadism were observed and researched by many scholars, E. Epshtein, The Bedouins – Life and Habits, Tel-Aviv, 1932, 15–56 (in Hebrew); G.M. Kressel, LetShepherding Endure, New York, 2003, 1–3; S. Bar Zvi and J. Ben David, The Negev Bedouin in the thirties and the forties of the 20th century as semi Nomadic Society,Researches in Geography of Erets Israel 10 (1978) 107–136 (in Hebrew); N. Lewis, Nomads and Settlers in Syria and Jordan, 1800–1980, Cambridge, 1987, 3–9; A.M. Khazanov,Nomads and the Outside World, Cambridge, 1983, 36–39; A. Musil, The manners and customs of the Rwala Bedouins, American Geographical Society: Oriental Explorations andStudies 6 (1928) 44; R. Patai, Nomadism: Middle Eastern and central Asian, Southwestern Journal of Anrthropology 7 (1951) 401–414; M. Awad, Settlement of nomadic andsemi-nomadic tribal groups in the Middle East, International Labour Review 79 (1959) 25–56.

61 A. Yair, On the adequacy of the 1:20,000 topographic maps of Israel for a quantitative analysis of hydrological networks, Researches in the Geography of the Eretz-Yisrael(Israel) 7 (1970) 13–24 (in Hebrew).

62 M. Zohary, Geobotanical Foundations of the Middle East, Stuttgart, 1973; O. Potchter and H. Saaroni, An examination of the map of climatic regions of Israel, according to theKoppen classification, Studies in the Geography of Israel 15 (1998) 179–194 (in Hebrew).

63 D. Hillel and N. Tadmor, Water regime and vegetation in the central Negev highlands of Israel, Ecology 43 (1962) 33–41; A. Danin, Man and the natural environment, in: T.E. Levy (Ed.), The Archaeology of Society in the Holy Land, London, 1995, 24–39.

64 G. Avni, Early mosques in the Negev highlands: new archaeological evidence on Islamic penetration of southern Palestine, Bulletin of the American Schools of OrientalResearch, 294 (1994) 83–100; I. Finkelstein, Arabian trade and socio-political conditions in the Negev in the twelfth–eleventh centuries b. c. e., Journal of Near Eastern Studies47 (1988) 241–252; I. Finkelstein, Living on the Fringe: the Archaeology and History of the Negev, Sinai and Neighboring Regions in the Bronze and Iron Ages, Sheffield, 1995. Seealso in Fischel and Kark, Sultan Abdulhamid II and Palestine (note 7).

65 D. Kalner [Later Amiran], Dura – characteristics of the villages on the margins of the Bedouins, Bulletin of the Jewish Palestine Exploration Society XIV (1948–1949) 30–37.66 C.R. Conder, Tent Work in Palestine – a Record of Discovery and Adventure, London, 1878.67 Quote from pp. 11–12 in H.H. Kitchener, Lieut. Kitchener’s reports. VIII. Camp at Jerusalem, October 2nd, 1877, Palestine Exploration Fund Quarterly Statement for 1878.

January 1878 (1878) 11–14.68 Conder and Kitchener, The Survey of Western Palestine, Memoirs of the Topography, Orography, Hydrography and Archaeology (note 13), Vol. 3, 402.69 In the ‘Mawassi’ agricultural system the high level of coastal underground water was used for growing grapes, palm trees and other crops in the inter-dune areas. See

more details on this in the articles by H. Tsoar and Y. Zohar, Desert dune sand and its potential for modern agricultural development, in: Y. Gradus (Ed.), Desert Development,1985, 184–200; Levin and Ben-Dor, Monitoring sand dune stabilization along the coastal dunes of Ashdod-Nizanim, Israel, 1945–1999 (note 58).

70 M. Berman, The Evolution of Beersheba as an Urban Center, Annals of the Association of American Geographers 55 (1965) 308–326; Y. Avci, The application of the tanzimatin the Desert: Ottoman Central Government and the Bedouins of Southern Palestine, A paper presented at: The International Conference on The Application of the tanzimatReforms in Various Regions of the Ottoman Empire, Haifa University, Haifa, Israel, June 2007.

71 Y. Braslavsky, Did You Know the Land, Part B: the Negev Land, Tel Aviv, 1946 (in Hebrew). Robinson passed in the southern coastal plain of Palestine in May 1838, and leftthe following notes (Robinson, Biblical Researches in Palestine, Mount Sinai and Arabia Petraea (note 25), Vol. 2, 44 and 49): ‘We heard much of the village of Huj, as havingbeen recently founded by order of the government in the territory hitherto occupied by the Bedawin.we reached es-Sukkariyeh. Like Huj, it had recently been built up bythe governor of Gaza, Sheikh Sa’id, upon former foundations, and was considered as his property,’ the same was the case for el-Kubeibeh (The locations of Huj andel-Kubeibeh can be seen in Fig. 5).



analysis of rare 1:5,000 Ottoman maps we can map the exactlocation and plan of these villages.72

Mapping the boundary between the ‘desert and the sown’ andsettlement expansion

The idea of depicting the boundary between the desert and thesown can be traced back to the map of Herodotus showing theOikumene, i.e. the inhabited world. In the Near East this wasa dynamically changing conflict area that is described very gener-ally in a few written historical sources, and usually not shown onmaps. The desert moved back and forth, depending on the nomadicroutes and periodic living spaces. This competition or contrastbetween the ‘desert and the sown’ was a standard theme ofOttoman Palestine and continued into the early 20th century.73 Forexample the northern Negev flourished and was densely settledwith urban and rural settlements during the Nabataean, Roman andByzantine periods. After the Muslim conquest it reverted towilderness. Whereas some scholars claim that the changes of theboundary between the desert and the sown were a function of the

power relations between the nomads and the central regime,others claim that climatic changes were the main factor controllingthe distribution of human settlements in the Negev.74 We turn tohistorical maps as our main source for learning about the boundarybetween nomadic and settled areas in the Negev and for recon-structing the process of settlement expansion from the end of the19th century onwards.

One of the first attempts at depicting nomad grazing areas inPalestine is that shown on Kiepert’s map of Palestine from 1891,where the winter grazing lands of the nomads are given in yellowcolors (‘Wustengebiete, d.h. nur von Nomaden bewohntes Winter-weidelenad’), in contrast with alluvial soils of permanent agricul-ture that are shown in green colors (Fig. 5C). Based on Kiepert’smap and additional sources,75 the German cartographer HansFischer uniquely mapped what he called the ‘current limit ofpermanent settlement’ (Jetzige Grenze seBhaften Wohnens). OnKiepert’s map the northern limit of the nomads’ winter grazinglands may signify that in those yellow areas the climatic conditionsare more extreme, and that these areas were less favored even bynomadic peoples. On the maps that Fischer published at the turn of

Fig. 5. Maps showing permanent settlements indicating the border between the desert and the inhabited land. A: Villages and towns depicted on the PEF maps by Conder andKitchener (1880) and by Shumacher (1886). Overlaying the map are equal rainfall isohyets of 50 mm/year. B: The current limit of permanent settlements as drawn on the map byFischer and Guthe (1911). C: Alluvial plains (in green) and winter grazing lands of the nomads (in yellow) on Kiepert’s map from 1891. D: Villages and towns depicted on the Surveyof Palestine (1947) 1:250,000 map. Fischer’s (1911) limit of the desert/sown is shown on all maps for comparison (For interpretation of the references to colour in this figure legend,the reader is referred to the web version of this article).

72 R. Kark, Consequences of the Ottoman Land Law: Agrarian and Privatization processes in Palestine, 1858–1918, in: D. Kushner (Ed.), The Application of the Tanzimat Reformsin Various Regions of the Ottoman Empire (in press); Maps of Lands Arrangement of the çiftliks of Muharraqa, Kaufakha and Jaladiyya, which are attached to Kaza Gaza, Sancakof Jerusalem, 1:5,000, 1309 [1893], Kark Map Collection, Jerusalem (in Turkish); Fischel and Kark, Sultan Abdulhamid II and Palestine (note 7).

73 See, for instance, Lewis, Nomads and Settlers in Syria and Jordan (note 60).74 The argument is still far from its end: M. Sharon, Processes of Destruction and Nomadization in Eretz Israel under the Islamic Regime (633–1517) in Idem (Ed.) Topics in

the History of Eretz Israel under the Islamic Rule, Jerusalem, 1976, 21–24 (in Hebrew); R. Rubin, The Negev as a Settled Land, Urbanization and Settlement in the Desert in theByzantine Period, Jerusalem, 1990, 182–189 (in Hebrew); A.S. Issar and M. Zohar, Climate Change – Environment and Civilization in the Middle East, Berlin, 2004; A. Reifenberg,The Battle between the Sown and the Desert – Transformations in Agricultural Civilization in Israel and its Neighbours, Jerusalem, 1950 (in Hebrew).

75 H. Kiepert, Neue Handkarte von Palaestina [cartographic material]/bearbeitet von H. Kiepert.Berichtigt von Richard Kiepert. Lith.Anst.v. L.Kraatz, Berlin, ImprintBerlin, 1891.



the twentieth century, a thick red line separates permanentlyinhabited areas from the areas where the population is mostlynomadic. This red line is located in the Negev to the north ofKiepert’s yellow areas. When permanent settlements are locateddeep within nomadic territories, they are encircled by this red line(Fig. 5B). The first map of this type for the Negev area is the one byFischer and Guthe (1890), and the last one is again by Fischer andGuthe (1911; Fig. 5B). We are not certain about Fischer’s motivationto map this boundary. Although these maps use firmly delineatedlines, this frontier usually consisted of a large overlapping area inwhich two different ways of life encountered each other, the onepredominantly pastoral-nomadic, the other mainly sedentary-agrarian.76 This spatial gradient between permanently settled areasinto nomadic populated areas can be better appreciated from the1946 British map showing the distribution of Bedouin tents inthe Negev (Fig. 6).77 From this map it can be appreciated that theBedouin tent distribution was not merely a function of rainfall,but also of settlement patterns. The Bedouin tent distributiontowards the north ceases relatively abruptly (mostly restrictedwithin the Beersheba District: Fig. 5D), whereas towards the southBedouin tents distribution dwindles gradually (Fig. 6). This isprobably due to the exclusion of Bedouin tents from agriculturaland settled areas within village boundaries in the north. The peakconcentration of Bedouin tents was between rainfall isohyets of200–250 mm, and above 350 mm there was a sharp decrease inBedouin presence (Fig. 7).

In both Fischer’s 1890 and 1911 maps the village of Edd-Dha-hariye was depicted outside the permanently inhabited land. Beingon the margins of the settled area may have been the cause fortemporary desertion of the village about which we learn fromliterary sources. Only from Fischer’s map we can learn about twomore new settlements, apart from the new town of Beersheba.These two settlements were completely encircled by this red line in1911 (and not in 1890): Tell el Buredsch (probably a neighborhoodof Beersheba) and El-Kusefe to the east. Musil had marked bothBeersheba and Kseife (El Kusefe) on his 1902 map as semi settledplaces (Zur Hafte bewohnter Ort).

An example for the role of the Ottoman Sultan and the centralgovernment in expanding the permanently inhabited areas, is thecase of two additional villages, Kaufakha and El-Muharraqa. Thesevillages were founded in the north-western Negev to the south ofthe southernmost permanent village of Huj (Fig. 7C), apparently inthe early 1890s on jiftlik (private) lands of the Sultan Abdulhamid II:(Fig. 5D).78 These were planned villages with a grid plan of straight

intersecting and parallel roads (there are Ottoman maps showingthe plan of these villages from 1893; Ruth Kark’s Archive).

Following the improvement of security conditions after WorldWar I and during the British Mandate period other small sub-villages were founded, mostly in the foothills of the Hebron Hillsand to the north of Edd-Dahariye.79 These villages are depicted onthe Survey of Palestine 1:20,000 and 1:100,000 maps (1930s–1940s). While most of the Arab population in the area remainednomadic, Jewish settlements were founded in the northern Negev,demonstrating the power of motivated groups to settle ‘frontier’areas. The first Jewish settlement in the area (Ruhama or Jamame)was founded in 1913. Over 20 Jewish settlements were founded inthe northern Negev in the 1940s, on lands bought by the JewishNational Fund (Fig. 5D).80 Only by a detailed spatial analysis ofrectified maps, it is possible to follow the spatial distribution andthe extent of this new settlement phenomenon, what theyreplaced, and their impact on the landscape.

The Arab District Administrative Officer of Beersheba in the1930s, ’Aref el-’Aref, stated that apart from the inhabitants ofBeersheba, all the population of the District of Beersheba was thencomprised of nomadic or semi-nomadic Bedouins (Figs. 5D and6).81 This situation was also noted by Joseph Braslavsky whoreported that in the mid 1940s to the south of Huj, Jammama,Kaufakha and El-Muharraqa there were no other permanentvillages.82 El-’Aref notes that under the British Mandate, in additionto the hospital in Beersheba, there were three outpatient clinicswithin the District of Beersheba (El ’Auja, El ’Imara, and Jammama),and nine police stations, located in Beersheba, Bir ’Asluj, El ’Auja,Kurnub, Ras ez-Zuweira, El-Ghamar, Um-Rashash, El-’Imara andJammama. These stations are noted on some of the editions of the1:250,000 maps of the Survey of Palestine as well as in El-’Aref’s1934 map in Arabic which includes the location of the Negev sub-tribes, cisterns, and other details).83

Sedentarization of the Bedouin

Although most of the Bedouin population was nomadic, initiationof the sedentarization process of the Bedouin may have alreadybegun at the end of the 19th century. As we have very few writtencontemporary sources relevant to this process, it is mostly from thecomparison of old maps that we can learn about sedentarization ina more systematic mode. In the northern Negev, this processcommenced with the construction of buildings by the Bedouin.Made of adobe or from secondary use of building materials from

76 J. Gommans, The silent frontier of South Asia, c. A.D. 1100–1800, Journal of World History 9 (1998) 1–23.77 On this map (see note 56) Bedouin tent concentration is depicted in red and each dot represents one tent (within those red areas). To map all these tents, we applied

a supervised classification method (maximum likelihood, using the remote sensing software ENVI 4.4; ITT Visual Information Solutions Inc.) on this historical map, extractingall the red areas as a new layer. Individual tents were digitized within an area of 600 km2 (within the limits of the area analyzed in Fig. 8), to estimate the accuracy of theclassified map of Bedouin tent areas. The correlation between classified tent area and tent numbers was high (r¼ 0.73), and allowed us to then estimate the number of tentswithin each area, arriving at a sum of 8722 tents with the census area, the same number reported by the British Census (see note 56). It is important to remember that thelocation and numbers of tents have varied geographically, were changing according to the climatic conditions and that Bedouins at that time lived in small numbers alsoalong the shores of the Mediterranean sea, the plain of Esderalon and the Judean hills (see note 58).

78 Braslavsky, Did You Know the Land, Part B (note 71); D. Gasit, Settlement processes in the Besor region in the days of the Sultan Abdulhamid II, in: Y. Schwartz, Z. Omer andI. Tsipar (Eds), Jerusalem and Eretz-Yisrael, Jerusalem, 2000, 183–186 (in Hebrew); R. Fischel, Sultan Abdulhamid II (1876–1909) and Palestine: Imperial Policy and Private Lands.Unpublished MA thesis, Jerusalem, 2006, in Hebrew; Fischel and Kark, 2008, see note 7.

79 Kalner, Dura (note 65); D. Grossman, The bunched settlement pattern: Western Samaria and the Hebron Mountains, Transactions of the Institute of British Geographers,New Series 6 (1981) 491–505.

80 R. Kark, The History of the Pioneer Jewish Settlement in the Negev until 1948, Jerusalem, 2002 (in Hebrew).81 A. El ’Aref, The History of Beer Sheba and its Tribes, Jerusalem, 1934 (in Arabic), and Tel Aviv, 1937 (in Hebrew, translated from Arabic).82 Braslavsky, Did You Know the Land, Part B (note 71).83 El ’Aref, The History of Beer Sheba and its Tribes (note 81). As noted by Avci, The application of the tanzimat in the Desert (note 70); Gil-Har, The south-eastern limits of

Palestine at the end of the Ottoman rule (note 16), some years before World War I the Ottoman Government, in order to control the Bedu in the extreme South and South-eastof Palestine, built a small town at El-Auja with residences for officials, a hospital, barracks and store houses for troops and houses and shops for merchants. Wells were dug,a water supply with engines and pumps installed and pipes laid through the main streets. The town was joined to Beersheba by a metal road and railway line. A Kaimakamand staff were appointed and a Police Post was formed. At the outbreak of war El-Auja was rapidly developing into a thriving administrative and commercial centre. Howeverthe town of El-Auja was evacuated during the war and fell again into ruins afterwards. For further details on Ottoman policies see: Avci, The application of the tanzimat in theDesert (note 70).



archeological sites, they were intended to be used for storage andbarns. The Bedouin term for these buildings was ‘Baika’ and theywere mostly built in the ‘heart’ of a family’s grazing area, not alwayswith direct connection to the family’s temporary dwelling. The‘bajkat’ on Musil’s map or ‘baikat’ on the British maps that werefound and mapped demonstrate the extent and distribution of thisstage of sedentarization.84 These buildings were later used as

shelters in winter when some of the Bedouins were engaged infarming; however there is no evidence for new permanent Bedouinsettlements.85

One example for this phenomenon is found in Musil’s map andtext. On August 20th, 1902 Musil noted that ‘. from the Bawajekabu Susen [located 25 km south of Gaza, and 25 km west ofBeersheba on Musil’s map] there develops a village whose building

Fig. 6. Distribution of the nomad population of the Beersheba sub-district, based on the Survey of Palestine map of 1947. The thick black line represents the Bedouin census limits.Overlaying the map are equal rainfall isohyets of 50 mm/year (shown in blue) (For interpretation of the references to colour in this figure legend, the reader is referred to the webversion of this article.).

84 Musil, Arabia Petraea (note 31); Braslavsky, Did You Know the Land, Part B (note 71); Bar Zvi and Ben David, The Negev Bedouin in the thirties and the forties of the 20thcentury as semi Nomadic Society (note 60).

85 E. Marx, Land and work: Negev Bedouin struggle with Israeli bureaucracies, Nomadic Peoples 4 (2000) 106–121; G. Kressel, J. Ban-David and K. Abu Rabia, Changes in landusage by the Negev Bedouin since the mid-19th century: the intra-tribe perspective, Nomadic Peoples 28 (1991) 28–55.



materials are taken from the field of ruins.’, and predicts that ifthe conditions prevail, that area will be populated by severalvillages.86 British Mandate maps indeed subsequently show anincrease in the number of houses in this and other areas in thenorthern Negev.

A comparison of the maps using GIS methods can reconstructthis transition process, and help us deduce whether it was part ofthe Bedouin permanent settlement, or reflected only partialengagement in agriculture without settling. This is a very importantissue in the debate over the legal status of the lands on which theBedouin grazed and built their baikat, between the successiveregimes (Ottoman Empire, British Mandate, and Israeli Govern-ment) and the Bedouin, from 1858 until the present. Indeed, insome areas the concentration of baikat formed the nucleus forfuture settlement. A careful analysis of five historical maps onwhich such isolated buildings are depicted, allowed us for the firsttime to monitor this process and to identify, for the first time, thespatial distribution of this phenomenon (Fig. 8, Table 2). There wereno isolated buildings in the 600 km2 area north-west of Beershebaon the Survey of Western Palestine map of 1880. However, in thePEF Memoirs the surveyors note that in Kh. el Kauwukah there were‘four or five modern houses’; Fig. 8A.87 Maps from World War Iyears show many more houses in the northern Negev (for the samearea of 600 km2): 42 houses in the Ottoman 1:200,000 maps from1919 (Fig. 8B) and 153 houses in the Survey of Egypt (1919)1:40,000 maps (Fig. 8C). Twenty years on, The Survey of Palestinetopographic 1:100,000 maps from 1936 to 1939 show 159 houses(Fig. 8D), and 566 houses are depicted in the 1:20,000 maps ofSurvey of Palestine from 1945 to 1946 in the 600 km2 area of study(i.e. almost 1 house/km2; these numbers relate to houses andbuildings outside of permanent settlements; Fig. 8E).

A comparison of several World War II aerial photographs withthe 1:20,000 maps, showed full correspondence between thenumbers of houses in both sources, indicating the completeness ofthe topo-cadastral maps. Based on the PEF maps and memoirs, wehave shown that there were only 4–5 isolated houses in the 1870s.The observed density of houses on the maps gives contrastingvalues for 1919 (0.09 and 0.26 houses/km2 based on the Ottoman

1:200,000 maps and on the Survey of Egypt 1:40,000 maps,respectively; Table 2), and suggests stagnation between 1919 to thelater 1930s (0.26 and 0.27 houses/km2, respectively). However, itshould be remembered that each map was drawn to a differentscale, and therefore its details are different. First we calculated thatthe yearly increase in the number of houses in the studied areawould be about around eight houses per year (or 0.01 houses/km2;assuming a linear increase between the time of Survey of WesternPalestine and World War II). We were then able to calculate theexpected density of houses for the years 1919 and 1936: 0.60 and0.82 houses/km2, respectively. We can then calculate the ratiobetween the observed and the expected densities, and plot themwith respect to the scale of the map. As we found a perfect linear fitbetween the scale of the maps (from 1919 and 1936) and the ratiobetween the observed and the expected densities for maps from theyears 1919 and 1936 (R2¼ 0.999, n¼ 3), we can infer that the numberof houses in the area increased linearly with time from the 1870s tillWorld War II. A linear increase may signify that the sedentarizationprocess of the Bedouin was not accelerated due to British Mandatepolicies (as the rate did not change during the British Mandate), andhad started in the late 19th century in part as a response to Ottomanpolicies. According to the British Mandate map of Bedouin tents, in1945/6 there were 1720 tents in this area (Fig. 8F), i.e. more thanthree times the number of houses (566) shown on 1:20,000 maps ofthe same area; thus, the majority of the Bedouin population wassemi-nomadic and not settled in the mid 1940s.

The archeologist Dan Gasit describes several other villages to thesouth of Kaufakha and Muharraqa supposedly planned by theOttoman authorities.88 However, he quotes no solid sources, andnone of those additional villages appear in any of the Negev mapsas a formal village. In contrast with Kaufakha and Muharraqa whichwere located on State (Sultan) Lands (as shown in the corre-sponding Survey of Palestine 1:250,000 map), the settlementsmentioned by Gasit were not located on State Lands. Based on themap materials we collected, it seems that those settlementsreferred to by Gasit represent locations where the density of baikatand houses was relatively large, however they were not recognizedas official villages on the British Mandate maps.

0%

5%

10%

15%

20%

25%

30%

35%

40%

0-50

50-100

100-150

150-200

200-250

250-300

300-350

350-400

400-450

Percen

tag

e w

ith

in

1946 cen

su

s area

Annual rainfall (mm)

Tents

Area

Fig. 7. The distribution of Bedouin tents as a function of rainfall in the Negev, in comparison with area distribution.

86 Musil, Arabia Petraea (note 31), Vol. 2, 62.87 Conder and Kitchener, The Survey of Western Palestine, Memoirs of the Topography, Orography, Hydrography and Archaeology (note 13).88 Gasit, Settlement processes in the Besor region in the days of the Sultan Abdulhamid II (note 78).



Fig. 8. Distribution of houses outside of villages and towns in five maps. A: Conder and Kitchener (1880); B: Ottoman (1919); C: Survey of Egypt (1919); D: Survey of Palestine(1936–1939); E: Survey of Palestine (1945–1946); F: Survey of Palestine (1947) – distribution of Bedouin tents; on this map the red areas represent concentrations of Bedouin tents(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).



Increasing the area of cultivated land was one of the mainobjectives of the 1858 Ottoman Land Code. It sought to bring asmuch mahlul (land uncultivated for over 3 years), and mewat Statelands under cultivation as possible.89 The Tabu Law of 1858 and the1867 protocol that permitted foreign citizens to acquire urban andrural land led to changes in ownership of village lands, and toprivatization and land reclamation in uninhabited regions.90

Although the British Mandate’s formal policy was not to introducelarge changes in the existing law of the land, such changesoccurred, and not only as a result of the interpretation of Ottomanlaws by British colonial officials in Palestine.91 The British admin-istration used existing laws to define indigenous land rights asnarrowly as possible and to vest as large an area as possible in thestate.92 Later Israeli policies similarly kept the earlier legal status ofmost of the Negev lands as mewat State Land.93 Kedar and Meirassert that the State of Israel conceived the Negev as vacuumdomicilium (i.e. terra nullius, or uninhabited land), but in fact theIsraeli law preserved the Ottoman and Mandatory legal status ofthe Negev lands as mewat, based on the lack of inhabited places andpermanent villages or towns in the Negev prior to 1900.94 In thisrespect, the use of historical maps and GIS showing the location anddistribution of permanent villages is a crucial testimony in legalland claims for both the Bedouin claimants and the State of Israel.This approach may also be interpreted such that the state’s policy isintended to objectify the denial of indigenous claims of landownership, and the legal system is used by the state to maintain its

rights to the land.95 A similar policy was also adopted by otherMiddle Eastern and North African countries, which were underOttoman and Madatory rule. In a few countries however, indige-nous land rights have been recognized recently, as in the judgmentby the Australian High Court in the landmark case of Mabo vs.Queensland.96 This is not likely to happen in Israel in the nearfuture, as long as the State policy is based on legality, due to thepolitical conflict between Jews and Arabs over land. However therecent report (Dec. 2008) of the Commission to Propose a Policy forRegulating Bedouin Settlement in the Negev, headed by Justice A.Goldberg, points to a new direction for finding a solution forBedouin settlement. This includes recognition to a certain extent ofland ownership, ‘taking into account the historic ties of theBedouins to the land’, and formal recognition of presently existingunauthorized villages. The fact that land settlement was notundertaken in the Negev under the Ottoman or the British regimesmay have contributed to current disputes over land ownershipbetween the Israeli government and the Bedouins.97

Conclusions

Historical maps have often been used by historical geographers tostudy changes in land cover, land use and the distribution ofsettlements.98 However the accuracy of those maps was often notassessed. The accuracy in the position of the features on historicalmaps and the level of detail shown may serve as a means to

89 M. Bunton, Demarcating the British colonial state: land settlement in the Palestinian Jiftlik villages of Sajad and Qazaza, in: R. Owen (Ed.), New Perspectives on Property andLand in the Middle East, Cambridge MA and London, 2000, 121–158. Most uninhabited and uncultivated land was defined as ‘mewat’ land (the term being derived from theverb ‘dead’ in Arabic). According to Article 6 of the Ottoman Land Code, ‘mewat’ land was land that was located ‘at such a distance from a village or town from which a loudhuman voice cannot make itself heard at the nearest point where there are inhabited places, that is a mile and a half, or about half an hour’s distance from such.’ Likewise,Article 103 defined mewat as ‘dead land . [meaning] vacant (khali) land, such as mountains, rocky places, stony fields, pernallik and grazing ground which is not in thepossession of anyone by title-deed or assigned ab antique to the use of inhabitants of a town or village, and lies at such a distance from towns and villages from whicha human voice cannot be heard at the nearest inhabited place’. Taken from A. Kedar, The legal transformation of ethnic geography: Israeli law and the Palestinian landholder1948–1967, New York University Journal of International Law and Politics Quarterly 33 (2001) 923–1000. This translation of the Ottoman Land Code is based on F.M. Goadby andM.J. Doukhan, The Land Law of Palestine, Tel-Aviv, 1935 (reprinted 1998 by Gaunt, Holmes Beach, Florida).

90 H. Islamoglu, Property as a contested domain: a reevaluation of the Ottoman land code of 1858, in: R. Owen (Ed.), New Perspectives on Property and Land in the Middle East,Cambridge MA and London, 2000, 3–61. Kark, Consequences of the Ottoman Land Law (note 72).

91 M. Bunton, Inventing the status quo: Ottoman land-law during the Palestine Mandate, 1917–1936, The International History Review 21 (1999) 28–56.92 Bunton, Demarcating the British colonial state (note 89), 128.93 The British Mandate legislated the mewat Land Ordinance (1921) which ruled that persons cultivating mewat lands could register the land in their own names, during

a limited period of time. As before, the Bedouin did not register the lands. According to the Ordinance, after that period, ‘Any person who, without obtaining the consent ofthe government breaks up or cultivates any waste land, shall obtain no right to a title deed for such land, and further will be liable to be prosecuted for trespass.’ (Yahel, Landdisputes between the Negev Bedouin and Israel (note 20)). This was the last formal opportunity to obtain legal titles for mewat lands. Later on, a 1969 Israeli law abolished themewat category and stipulated that all such lands would be registered as state property unless a formal legal title could be produced by a claimant. (R. Shamir, Suspended inspace: Bedouins under the law of Israel, Law & Society Review 30 (1996) 231–258). See also a recent book by H. Sandberg, The Lands of Israel: Zionism and Post-Zionism,Jerusalem, 2007, 143–148.

94 From the 1858 Ottoman Land Law, and in the legal systems of the following regimes, isolated buildings were not considered as a settlement (more on this in Kedar, Thelegal transformation of ethnic geography (note 89). Meir suggested another interpretation of the term mewat: A. Meir, Alternative examination of the sources of the conflictbetween the government and the Bedouin: a geo-legal outlook, Karka 63 (2007) 19–51 (in Hebrew).

95 Kedar, The legal transformation of ethnic geography (note 89); Meir, Alternative examination of the sources of the conflict between the government and the Bedouin(note 94); this situation is not unique to the relationships between Israel and the Bedouin population. Denying that indigenous peoples possessed recognizable societies, law,property rights or sovereignty, provided the rationale for declaring newly explored territories as terra nullius by past colonial powers (as in Australia: B. Buchan and M. Heath,Savagery and civilization, from terra nullius to the tide of history, Ethnicities 6 (2000) 5–26).

96 In a landmark judgment handed down on 3 June 1992, by a vote of 6–1, the full bench of the High Court of Australia overturned legal precedent and determined thatunder certain circumstances, indigenous land rights (or ‘native title’) had not been extinguished by white European settlement and indeed, under common law, still existed insome parts of what is now called Australia in the latter years of the 20th century. In other words, at the time of the first European settlement in the late 18th century, Australiawas not as was previously claimed – terra nullius (an empty land belonging to no one) but was home to indigenous peoples who could claim long-standing prior ownershipand usufructuary rights. (D. Mercer, Aboriginal self-determination and indigenous land title in post-Mabo Australia, Political Geography 16 (1997) 189–212. See also: R. Karkand S. Franzman, Empire, state, and the Bedouin of the Middle East: a long term comparative study of land and settlement politics, Unpublished manuscript, presented in the31st International Congress, Tunis, Tunisia, August 2008.

97 As in other areas in Israel (see in Gavish and Kark, The cadastral mapping of Palestine, 1858–1928 (note 8)). See also Report of the Commission to propose a Policy forArranging Bedouin Settlement in the Negev, headed by Justice A. Goldberg, State of Israel, Jerusalem, December 2008.

98 D.J. Stanley and A.G. Warne, Nile delta: recent geological evolution and human impact, Science 260 (1993) 628–634; R.J. Fensham and R.J. Fairfax, The use of the landsurvey record to reconstruct pre-European vegetation patterns in the Darling-Downs, Queensland, Australia, Journal of Biogeogrpahy 24 (1997) 827–836; S.A.O. Cousins,Analysis of land-cover transitions based on 17th and 18th century cadastral maps and aerial photographs, Landscape Ecology 16 (2001) 41–54; C.C. Petit and E.F. Lambin,Impact of data integration on historical land-use/land-cover change: comparing historical maps with remote sensing data in the Belgian Ardennes, Landscape Ecology 17(2002) 117–132; A. Symington, M.E. Charlton and C.F. Brunsdon, Using bidimensional regression to explore map lineage, Computers, Environment and Urban Systems 26 (2002)201–218; D.K. Hall, K.J. Bayr, W. Schoner, R.A. Bindschadler and J.Y.L. Chien, Consideration of the errors inherent in mapping historical glacier positions in Austria from theground and space (1893–2001), Remote Sensing of Environment 86 (2003) 566–577; O. Bender, H.J. Boehmer, D. Jens and K.P. Schumacher, Using GIS to analyze long-termcultural landscape change in Southern Germany’. Landscape and Urban Planning 70 (2005) 111–125; Levin, The Palestine Exploration Fund map (1871–1877) of the Holy Landas a tool for analyzing landscape changes (note 13).



critically and quantitatively evaluate their reliability for historicalreconstruction and to enable comparison with contemporary maps.This is especially true in regions such as the Negev during periodsfor which we lack written sources and maps must serve as the mainsource.

Considering the process of evaluating the accuracy of historicalmaps, two stages should be distinguished: (1) determining coor-dinate grid accuracy, and (2) determining feature accuracy.99

Overall, the general trend that we found was that both the scale andthe accuracy of the maps covering the Negev improved with time,but there were exceptions which ‘preceded their time’ such as thesuperior 1870s PEF maps. Mapping efforts were not uniform acrosstime: There were several bursts of activity, mostly related in theNegev to international geopolitical events (e.g., in 1906, and beforeand during World War I and II) or to administrative needs (e.g. theOttoman land and settlement policy change, and the Mandatoryintended Land Settlement).

The maps preceding the PEF Survey of Western Palestine gavevery little information about topics such as landscape features,agriculture and distribution of Bedouin tribes. This situationchanged with the initiation of scientific and military surveys. Thusfrom the 1880s to 1948 we could reliably reconstruct from thesurprisingly large number of Negev maps, and the scarce writtensources, the distribution of villages and even isolated buildings.Focusing on five maps (at scales of at least 1:200,000) in whichisolated buildings (in addition to the villages) were depicted, aswell as a unique map that enables us to determine the exactdistribution of Bedouin tents in the Negev in 1946, we are able todemonstrate the spatial distribution of nomadism and semi-nomadism in the Negev. This also illuminates the slow southwardmovement of permanent settlements from the end of the 19thcentury, including the newer phenomenon of widespread Jewishagricultural settlement in the northern Negev, mainly from 1939 to1948, all serving to push the boundary between the desert and thesown to the south.

The density of permanent dwellings increased gradually, mostlyin the northern Negev between Beersheba and Gaza. However, noestablishment of formal Bedouin villages (recognized by BritishMandatory authorities) resulted from this process. Whereas in thelate 19th century, there were no permanent villages south of theisohyet of 300 mm, one of the recent definitions of the northernboundary of the Negev (in the Atlas of the Negev, 1986) places itmuch more to the south at the isohyet of 200 mm, which isapproximately where Beersheba, the new city founded by theOttomans in 1900, is located.100 This southwards shifting of theborder between the sown and the desert continues to the present.

This shift was a result of the Ottoman central governmentalpolicy which aimed to increase personal and property security andestablish new villages and towns in an effort to gain better controlof peripheral areas bordering with the nomadic Bedouin tribes indifferent parts of the Middle East and Palestine, including theNegev, as well as to derive economic benefit from these marginal

areas. During the British Mandate period the central governmentshowed less interest in developing the Negev and the Bedouinpopulation there.101

Only by using GIS can we combine a series of old maps, extractand compare the location of Bedouin tribes, the spatial distributionof their tents and isolated buildings, the boundary between thedesert and the sown, and the settlement expansion between 1799and 1948, on a modern map. After doing this we were able todeduce that from the end of the 19th century the boundarybetween the desert and the sown was moving southward. Howeverthe main agents determining this development were the effortsinvested by the Ottoman government and the Sultan before 1917(in founding new villages, towns and police stations in the Negev),and subsequently the Mandatory government which developedsome infrastructure, and the Jews who purchased land and estab-lished new agricultural settlements in the northern Negev in thefirst half of the 20th century.

We conclude that the study of historical maps analyzed andverified by modern GIS technology in the context of historical–geographical studies, provides a valuable tool to assess their accu-racy and consequently their potential usefulness for the study oftopics such as the reconstruction of the process of sedentarizationof nomadic populations, changes in the landscape and patterns ofsettlement, assessing the impact of historical governmental policieson these characteristics of marginal areas, and for current uses suchas planning policy, regularization of the status of unrecognizedsettlement status, and legal disputes over land ownership bynomadic and indigenous populations.

Acknowledgements

We would like to thank Rachel Kengisser and Lea Engel from theMap Library of the Department of Geography at the Mt. ScopusCampus of the Hebrew University of Jerusalem, Ayelet Rubin, EstherGugenheim and Tomer Koren from the Laor Maps collection at TheJewish and National University Library, Yelena Rubinstein from theNational Maps Archives of the Survey of Israel, Ram Almog from theAerial Photos Archive of the Department of Geography at the Mt.Scopus Campus of the Hebrew University of Jerusalem, ZvikaMednik from the Map Library of the Department of Geography andthe Human Environment of Tel Aviv University, and Prof. HaimGoren from Tel Hai Academic College, for their help in our searchfor maps covering the Negev area. We also thank Dr. Naftali Thal-man for his translations from German of various articles and maplegends. We appreciate the remarks made by the anonymousreviewers and thank them for their contribution to the improve-ment of this manuscript.

Supplementary material

Supplementary data associated with this article can be found, in theonline version, at doi:10.1016/j.jhg.2009.04.001

99 Coordinate grid accuracy. This stage involves registering the map to its respective coordinate system, based on the grid lines depicted on it. In this stage, unless the paperof the map underwent non-systematic bending, stretching, shrinking or local deformation, and excluding gross errors made by the cartographer or by the user, theregistration root mean squared error (RMSE) should be less than 1 mm, as the total error introduced in secondary methods of data collection, is expected to be equal to 0.81 mm in the worst case scenario (Thapa and Bossler, Accuracy of spatial data used in geographic information systems (note 34)). Feature accuracy. Estimating the accuracy inthe positions of features on the map can be done by comparison to modern topographic maps, orthophotos or field measurement using the global positioning system (GPS)measurements. The RMSE obtained at this stage reflects location errors produced by the mapping methods as well as by the scale of the map and the type of featureexamined.100 E. Stern, Y. Gradus, A. Meir, S. Krakover and H. Tsoar, Atlas haNegev (Atlas of the Negev), Beer Sheva, 1986 (in Hebrew).101 R. Kark and S.J. Frantzman, The Negev: land, settlement, the Bedouin and Ottoman and British policy 1871–1948, submitted for publication.


Date post:	18-Nov-2023
Category:	Documents
Upload:	bgu
View:	0 times
Download:	0 times

Maps and the settlement of southern Palestine, 1799–1948: an historical/GIS analysis

Documents