1

PUBLIC OPINION AND FOREIGN POLICY:

A CONSTRUCTIVIST ANALYSIS OF BRITAIN’S INVASION OF IRAQ

Dr James Strong

London School of Economics j.strong1@lse.ac.uk

DRAFT – Please do not cite or circulate without permission

Introduction

Democracies are not supposed to fight unpopular wars. But Britain did just that, in

Iraq. Democratic leaders are not supposed to court career disaster by leading their

countrymen into unwanted conflicts. But when British Prime Minister Tony Blair launched

military action against Saddam Hussein he both sacrificed long-term political capital (Hill

2007, 276, Dunne 2008, 340), and directly risked his job by submitting to a House of

Commons vote. This paper analyzes the interaction between British public opinion and the

Blair government’s decision to join the 2003 invasion of Iraq, aiming both to understand the

particular case and consider what insights it might offer for Foreign Policy Analysis (FPA)

more generally. It does three things. First, the paper asks how it was possible for

policymakers to decide on war in the face of negative opinion polls, hostile press

commentary and both the largest parliamentary rebellion and the largest street protests

against any British government policy, ever. Second, the paper considers what this specific

case reveals about the relationship between public opinion and foreign policy in general. In

particular, it asks whether what happened was really as surprising from an FPA perspective

as it may at first appear (Schuster and Maier 2006, Chan and Safran 2006). Finally, the paper

makes the case for a constructivist approach to studying public opinion as a potential

influence on foreign policy decision-making. There are clear ontological and epistemological

tensions within the specialist public opinion literature that constructivist insights can help us

resolve. Adopting a constructivist mindest also keeps this paper focused where an FPA study

properly should be. FPA is less interested in what public opinion is. Public opinion specialists

can handle that question. FPA is much more interested in what public opinion does.

2

The paper comprises three main sections. The first section makes the case for treating

public opinion as a social fact, linking celebrated critiques of opinion polling (Blumer 1948,

Bourdieu 1979) with constructivist approaches to understanding foreign policy (Doty 1993,

Weldes 1996, Althaus 2004, Houghton 2007) and proposing a novel ‘holistic’ analytical

approach. The second section presents quantitative data derived from a detailed content

analysis of press and parliamentary debate during the fifteen months before the invasion,

alongside opinion poll results. It identifies moments of particular intensity in the debate, and

maps trends in support for and opposition to war. The third section takes a more qualitative

approach, investigating in detail three major ‘spikes’ in issue salience and focusing on how

particular patterns of public and policymaker behaviour made certain policy outcomes

possible. A short conclusion ties the empirical findings together, highlighting their

implications for understanding Britain’s war in Iraq and for how FPA conceptualizes public

influence.

The case for a constructivist approach

For much of the twentieth century, most scholars interested in the relationship

between public opinion and democratic foreign policy accepted what Ole Holsti termed the

“Almond-Lippmann consensus” (Holsti 1992). Ordinary people, they concluded, were too

uninterested in and ignorant about international affairs to exercise real influence over a

decision-making process dominated by elites (Lippmann 1922, 241, Almond 1950, 53,

Rosenau 1961, 36). Public opinion was both “permissive” and driven by irrational “mood”

rather than calculation. It largely let leaders behave as they saw fit (Caspary 1970). Later

studies revised this position somewhat. They revealed that though the mass public may know

little about foreign policy (Converse 1964, 245, Holsti 1996, 215), it can still generate foreign

policy opinions (Holsti 1992, 450, Sobel 2001, 21, Isernia, Juhasz and Rattinger 2002,

Klarevas 2002, Jacobsen 2008, 351) and its opinions can still have concrete policy effects.

Ordinary citizens asked to judge foreign policy consider whether it looks successful, the

benefits it promises or the costs it seeks to avoid in order to reach an impressionistic view

(Gelpi, Feaver and Reifler 2005, Eichenberg 2005, 163, Klarevas 2006, 193, Gelpi, Reifler

and Feaver 2007, 158). Individuals’ personal ties to affected regions (Hill 2007, Ross 2013,

Koinova 2013), cultural and religious beliefs (Lacina and Lee 2013), patriotism (Nincic and

Ramos 2012) and elite leadership (Mueller 1973, Eshbaugh-Soha and Linebarger 2014)

matter too. Collective public attitudes show greater consistency and coherence than do the

3

views of individuals (Page and Shapiro 1992, 14, Nincic 1992, 31, 42, 45), though this does

not necessarily make them rational (Zaller and Feldman 1992, 579, Althaus 2004, 278).

Recent studies have shown an association between public attitudes and policy shifts (Page

and Shapiro 1992, 2, Shapiro and Jacobs 2000, 225, Everts and Isernia 2001, 17, Aldrich, et

al. 2006, 478), and between policy failures and electoral punishment (Verba, Brody, et al.

1967, 317, Hagan 1990, 4, Nincic 1992, 91-92, Bueno de Mesquita and Siverson 1995, 853,

Partell 1997, 508-509). Some even suggest that, under the right circumstances, consistent

negative poll results can sweep even the most dearly-held policy positions away (Graham

1994, 195-196). In sum, though ordinary people may not know that much about international

issues and events, they nevertheless make influential judgements that help shape foreign

policy. Their influence is often iterative (Powlick 1995, 430). It is constrained by the access

public actors enjoy to policymaking (Risse-Kappen 1991, 479-480, Risse-Kappen 1994, 238,

Cohen 1995, 53, Chan and Safran 2006, 137), and mediated by policymakers’ personal views

(Foyle 1997, 145).

Many of these studies follow the ontological and epistemological assumptions that

underpin the specialist public opinion literature. At an ontological level, most specialists

conceptualize ‘public opinion’ as the opinion of the public, and assume it exists ‘out there’,

awaiting observation. At an epistemological level, they employ survey methods, or ‘opinion

polls’ to measure public attitudes. Thanks to considerable methodological progress made

since George Gallup first demonstrated the power of polls prior to the 1936 US presidential

election, survey methods do offer highly reliable insights into ordinary people’s attitudes. But

their use still raises serious questions about validity. This is because public opinion

specialists generally do not distinguish adequately between survey methods and their object

of analysis. They insist that ‘public opinion’ is best defined as whatever opinion polls reveal,

and that opinion polls observe public opinion (Converse 1987, 14). This claim is tautological

(Splichal 1999, 90). If public opinion is whatever polls show, and polls measure public

opinion, there is no way of knowing if public opinion exists independently of the survey

process. Specialists argue that an public opinion must exist objectively. If it did not, “the

activity of all those who claim to explore, measure and interpret it would be a little odd”

(Osborne and Rose 1999). This claim does not follow logically. The fact that there is an

opinion polling industry does not necessarily prove that public opinion exists, nor that polling

is the best way to measure it. As Pierre Bourdieu’s celebrated argument that “public opinion

does not exist” makes clear, these shortcomings raise significant conceptual difficulties.

4

Bourdieu highlighted three “implied postulates” that underpin polling methods despite

not holding up empirically (Bourdieu 1979, 124). To begin with, Bourdieu complained that

polls operate on the assumption that every individual in a society has an opinion on every

major political issue of the day. This assumption looks unrealistic from an FPA perspective.

It is simply not rational for most ordinary people to pay attention to foreign affairs most of

the time (Page and Shapiro 1992, 14, Nincic 1992, 31). As Walter Lippmann put it, “so long

as nobody has to fight and nobody has to pay” governments can largely do what they want

(Lippmann 1922, 241). By assuming that every individual has foreign policy opinions,

polling methods wind up focusing attention on the majority of citizens who actually do not

have a position on the major foreign policy issues of the day (Cohen 1973, 4, Althaus 2004,

292). Polls do not ask whether attitudes exist, they only measure the distribution of attitudes

that are assumed to exist (Peer 1992, 231). This is a hangover from the original invention of

polls, as a mechanism for predicting voting behavior. Voting behavior exists independently

of polling. Foreign policy attitudes may not (Splichal 1999, 242). Indeed, there is good

evidence that individuals largely respond ‘on the fly’ to polls, answering questions they have

never previously considered (Mueller 1973, Zaller 1992). They draw on their own existing

beliefs to give an answer, beliefs that in the foreign policy arena in particular may not be

internally consistent (Gaubatz 1995). And they respond to the specific structure of the

question placed before them (Achen 1975). This helps explain why similar questions asked to

the same people at different times can produce radically different results (Zaller and Feldman

1992, 579-580).

Secondly, Bourdieu complained that polls treat every individual citizen as if their

opinions mattered equally. The public opinion specialist Herbert Blumer raised exactly this

point in an early critique of polling. Adopting a sociological perspective in slight contrast to

Bourdieu’s critical stance, Blumer argued that societies actually operated through social

structures; political parties, interest groups, epistemic communities and the like (Blumer

1948, 544). Treating the public as an undifferentiated mass of individuals meant failing to

appreciate the actual mechanisms through which individuals developed, refined and

articulated their views, and how in turn those views both came to the attention of and

influenced policymakers (Berinsky 1999, 1209). Within the FPA literature, this attitude is

commonly reflected in the division of ‘the’ public into distinct classes, either hierarchically

into ‘mass’, ‘attentive’ and ‘elite’ strata on the basis of knowledge, interest and influence

(Almond 1950, 138, Rosenau 1961, 33-34, Hilsman 1987, 284, Risse-Kappen 1991, 482,

Sobel 2001, 12) or functionally on the basis of behavior (Mueller 1973, 71, 116, 140,

5

Splichal 1999, 16). Distinguishing between types of public actors seems logical (Hurwitz and

Peffley 1987). We know that different individuals have different levels of interest in and

knowledge about foreign policy. It follows that they should hold different levels of influence,

too. It makes less sense then to continue to use methods designed on the assumption that

every possible respondent is equal. It is theoretically straightforward but empirically very

difficult to identify distinct ‘elite’ and ‘attentive’ actors from the wider ‘mass’ (Key 1961,

536, Mueller 1973, 164). It is generally not possible to say which particular views within a

survey sample belong to individuals in a position to exert real policy influence (Verba,

Brody, et al. 1967, 331-332). Mass opinion polls, in other words, offer surprisingly little

insight into what ‘the’ public actually looks and acts like.

Finally, Bourdieu pointed out that the act of conducting a survey imposed elite beliefs

onto the mass. On one level, polls serve a democratic purpose by bringing the views of

ordinary people into play and helping to ensure that “the wind of opinion” does not always

blow “from the better parts of town” (Verba 1996, 6, Converse 1987, 14). But their ability to

fulfil this function is bounded. Conducting opinion polls costs money. Polls only happen

when actors with the capacity to pay for them choose to do so, making polling an elite

activity by definition (Bourdieu 1979, 124). In practice, most polls are paid for by politicians

interested in winning elections and journalists interested in writing headlines (Beniger 1992,

216, Herbst 1992, 223). Neither are primarily concerned with advancing the public interest.

Both have interests of their own to promote. That raises two distinct issues. First, it raises the

prospect that elites will manipulate the design of polls themselves to produce results that

legitimize positions they intend to take regardless (Mueller 1973, 121). Second, it highlights

the considerable power elites hold to set the agenda for public opinion. Given how little

attention most ordinary people give to foreign policy, the very act of conducting a poll

inevitably imposes elite views onto a mass that may not previously have held any views on

an issue at all (Gamson and Modigliani 1989, Page and Shapiro 1992, 4-5, Cohen 1995, 53-

54, Powlick and Katz 1998, 33). This is the case even for well-designed polls that

successfully minimize question bias. For example, the Guardian newspaper published an

editorial on 11 October 2001 arguing against extending the nascent ‘war on terrorism’ to

Baghdad, and presenting an opinion poll it paid for that showed respondents largely agreed

(The Guardian 2001). In the process it began what became the Iraq debate, long before

British policymakers thought seriously about using force against Saddam Hussein, and before

rival publications even considered seeking out public views. Indeed, the piece appeared just

two days after US-led coalition forces began operations in Afghanistan. Ordinary people did

6

not push for a stand against war with Iraq in October 2001. The Guardian did that itself,

recruiting ordinary people via a poll to support its stance. In a similar vein, the Mirror, which

also opposed the invasion, found 71% of respondents rejected the use of force against Iraq by

specifically asking about a war without any UN support, an option explicitly ruled out by the

government at the time (Seymour and Blackman 2002). As Blair himself pointed out, no-one

asked about military action in Afghanistan before 9/11, therefore there were no polls showing

how unpopular such a proposal would have been (Blair 2002f). He made the reasonable point

that decisions about national security could not always wait for the public to appreciate a

threat, given the damage an attack might do. Poll results accordingly often tell us at least as

much about elite views than they do about mass attitudes (Jervis 1976, Hill 1981, 59). These

two issues are collectively problematic because a feedback loop exists between what public

actors think and what polls tell us. Polls influence public opinion (Herbst 1993, 41). Elites

commission and publish favorable opinion polls to legitimize their own established views,

and to strengthen their positions with regard to other participants in elite debate (Brookes,

Lewis and Wahl-Jorgensen 2004, 63, Solomon 2009, 269).#

A constructivist mindset

Adopting a constructivist mindset allows us to compensate for the ontological and

epistemological issues raised by opinion polling, by bringing in a wider set of sources and

adopting a more discursive definition of ‘public opinion’. Public opinion clearly does exist,

as even Bourdieu admitted. But it exists as a social rather than a material fact (Adler 1997,

327). It exists to the extent society, and in particular elite actors within a society, think it

exists. The conduct, publication and use of polls forms just part of a broader series of

discursive interactions that collectively constitute ‘public opinion’ (D. Campbell 1992, 4,

Hansen 2006, 10, Althaus 2004, 297). The concept of public opinion, after all, predates

polling. Before Gallup, however, it was “usually equated with riots, strikes, demonstrations,

and boycotts” (Ginsberg 1986, 48). Why should it not be? As Charles Tilly notes,

“if we push back into the strange terrain of western Europe and North America before the middle of the nineteenth century, we soon discover another world. In that world, most people did not vote, petition, or take positions on national affairs in anything like the contemporary meanings of those terms. Yet they did act together on their interests, broadcasting their demands, complaints, and aspirations in no uncertain terms” (Tilly 1983, 462).

7

In more contemporary times this ‘active’ public opinion, what Lippmann called “Public

Opinion with capital letters” (Lippmann 1922, 29), appears in the opinion pages of the press,

in speeches by parliamentarians, in the statements and actions of pressure groups and in street

protests like the ‘Stop the War’ marches that paralyzed London and other major cities in

February 2003. It is constituted by and through public debate whether elites seek to measure

it or not. It derives its ‘publicness’ from its “publicity” (Habermas 1962, 247), from the fact it

exists in public rather than because it reflects the views of the public (Noelle-Neumann 1979,

152, Splichal 1999, 30). Christopher Hill warned against “the tendency to confuse a part with

the whole...to see the will of the majority in a stormy parliamentary debate or a persistent

pressure-group lobby” (Hill 1981, 57). But Hill missed the point. Noisy minorities may not

necessarily represent the population as a whole. But they often are influential nonetheless

(Key 1961, 17-18, 92, Risse-Kappen 1991, 510, Everts 2000, 187, Hill 2007). They cause

problems for governments, and there is no effective way of knowing whether the silent

majority even has an opinion on a topic (Verba, Brody, et al. 1967, 328). Polling, as we have

seen, imposes elite views onto the mass rather than letting it speak for itself. No-one solicits

active public opinion. It may not represent the mass. But it does at least represent itself, and

policymakers listen.

There are thus good conceptual grounds for approaching the study of public opinion

on constructivist terms, treating polls as one powerful element in a discursive process through

which elites constitute ‘public opinion’ as a social fact, along with media commentary,

legislative debate, pressure group activities of various sorts and street protests. From an FPA

perspective, however, there are practical grounds for adopting such an approach, too. It

reflects how policymakers actually behave. Public opinion, as we have seen, cannot influence

foreign policy directly. It can influence policymakers to the degree they consider it important

(Foyle 1997). As David Patrick Houghton pointed out, there is considerable scope for linking

the cognitive school of FPA research to constructivist efforts to understand political

phenomena (Houghton 2007).

Policymakers definitely do pay attention to opinion polls. The Blair government

employed Philip Gould to provide regular reports on the public mood derived from a

combination of quantitative and qualitative approaches (Campbell and Stott 2007, 610). As

Zaller pointed out, however, “when a politician believes that he knows better than the public

what means will lead to desired ends, he will ignore the polls and follow his own beliefs”

(Zaller 2003, 313). Jonathan Powell served as Chief of Staff to Tony Blair throughout Blair’s

8

ten years in office. He wrote that “a wise leader should use quantitative polling as a tool but

not as a substitute for his own political instincts” (Powell 2010, 136-137). Polls, after all,

provide only a historical snapshot, not a guide to future trends (Key 1960, 56). Blair himself

agreed. He embodied Elizabeth Noelle-Neumann’s argument that decision-makers construct

their own individual images of what the “climate of opinion” might be, and act accordingly

(Noelle-Neumann 1979, 147-148, 154, Blair 2010, 298). They weigh poll results in the

context of exchanges they have with non-governmental actors, including legislators,

journalists, academics, business leaders and pressure groups as well as ordinary citizens

(Cohen 1973, 78, 107, 125, Herbst 1992, 228, Powlick 1995, 433-435, Entman 2004, 21). In

other words, they construct an image of public opinion for themselves in light of exchanges

with other actors. For foreign policy decision-makers, public opinion is an intersubjective

construct. They are “practical-intuitive” (Vertzberger 1986) constructivists.

A holistic approach

The following empirical analysis looks to understand the relationship between British

public opinion and the Blair government’s decision to go to war in Iraq in 2003. It does this

in two stages. First, the next section presents a quantitative overview of the balance between

pro and anti-war arguments expressed in published opinion polls, parliamentary debates and

press commentary. Second, a further section presents qualitative accounts reconstructing

three key moments in the pre-invasion debate, moments highlighted as potentially significant

by the quantitative stage. Both stages rely on three main types of source material, though they

adopt complementary epistemologies, one more positivist and the other more in keeping with

the interpretive tradition of research into British foreign policy (Beech 2011, Bratberg 2011,

Daddow and Gaskarth 2011, Bevir, Daddow and Hall 2013). Published opinion poll results

come primarily from ICM surveys conducted on behalf of the Guardian newspaper. A

handful of additional polls are referenced where different approaches to question wording

skewed the results in interesting ways, but focusing on a single company for the most part

allows us to exclude this sort of variation. ICM asked respondents “Would you approve or

disapprove of Britain backing American military action against Iraq?” at regular intervals

from the Summer of 2002 until the start of the war in early 2003. This allows us to track

changes over time.

A Lexis Nexis search for the period from early 2002 to March 2003 using the

keywords “Iraq” and “Blair” (to keep the focus on policy-oriented coverage) produced a

9

corpus of newspaper stories drawn from the leading UK newspapers of the day, the Daily

Mail, Daily Telegraph, Guardian, Independent, Mirror, Times, Sun and their respective

Sunday sister papers. Not only does this list cover the leading morning daily and Sunday

publications in the UK’s crowded newspaper market, it also encompasses the full spectrum of

mainstream political views. Two coders independently assessed the resulting corpus. A

primary coder manually coded all 4,450 articles, distinguishing between 2,335 news stories

and 2,115 commentary pieces, and identifying the particular stance each of the latter type

took towards the prospect of military action. A secondary coder coded a randomly-selected

sample of texts drawn from the same corpus using a codebook produced by the primary

coder, in order to test the reliability of the coding frame. The coders agreed 93% of the time

on the type of article an individual text represented, and 92% of the time on whether

comment pieces expressed pro-war, anti-war or neutral views. A test of inter-coder reliability

using Krippendorff’s alpha (Krippendorff 2004) showed scores of 0.849 for publication type

and 0.893 for article stance. A content analysis frame is generally considered very highly

reliable at agreement levels of 0.90 and above, highly reliable at levels of 0.80 and above and

acceptable at levels of 0.66 and above (Bauer 2000, 144). These results suggest the coding

framework used was highly, bordering on very highly, reliable.

Newspaper coverage should serve as a reasonable proxy for media debate more

generally. At the time most UK news consumers still gained their information either through

print or broadcast media. The mass switch to digital sources had begun but had not yet shifted

the focus decisively online. It seems unlikely that any major arguments either for or against

the war in Iraq surfaced on television without appearing also in print, especially given the

greater space newspapers provide for commentators to express their views in detail.

Alongside newspapers, a third set of source material derived from Hansard, the formal record

of debates in the House of Commons, the UK’s primary legislative chamber. A search of the

Hansard index identified thirty individual discussions touching upon the prospect of war in

Iraq between early 2002 and March 2003, and a total of 333 individual speeches. A similar

content analysis process to that discussed above generated data showing the distribution of

MPs’ views.

Taken both individually and collectively these source materials offer an imperfect

insight into the social construction of public opinion by British elite actors during the pre-

invasion period. They also offer little direct insight into decision-makers’ behaviour. The

final stage of the analysis reported below used participant accounts, in the form of

contemporary diaries and internal government documents as well as later recollections, to

10

triangulate observations drawn from these different publicly-available sources. This helped

validate observations as well as relating the ‘outsider’ view to what Blair government actors

actually thought at in 2002 and early 2003.

Quantitative analysis

Figure 1 describes the distribution of source materials identified across the time

period under analysis. It is immediately apparent that both press and parliamentary interest in

Iraq waxed and waned considerably from one month to the next.

FIGURE 1: Volume of press and parliamentary sources included in analysis, by month.

Figure 1 reveals three major spikes in public attentiveness to the prospect of British

involvement in Iraq. The first, in March and April 2002, surrounded Prime Minister Blair’s

visit to President Bush’s ranch at Crawford in Texas. The second, in September 2002, both

triggered and reflected the release of the government’s now-notorious ‘dossier’ on Iraq’s

Weapons of Mass Destruction. The third and most substantial, in February and March 2003,

accompanied the final drive to secure a diplomatic consensus, the mass street protests of 15

February and the climactic parliamentary debate of 18 March. The qualitative discussion

below focuses on understanding these periods.

The results of the quantitative stage show that British public actors broadly opposed

the invasion of Iraq. The data presented in Table 1 shows net poll, parliamentary and press

approval of the prospect of war during the fifteen months immediately preceding the

11 34

136 90

31 14

72 107

302

97

49 61

200

444 467

0

50

100

150

200

250

300

350

400

450

500Press

Parliament

11

commencement of hostilities. The poll figures reflect the average results generated by ICM

and YouGov polls during the period. The press and parliamentary figures derive from

assigning scores of 1 to pieces coded as pro-war, 0 to those coded as neutral and -1 to those

coded as anti-war, and then calculating the average stance taken by all of the texts generated

in a given month.

TABLE 1: Raw and salience-adjusted net public approval of the prospect of war with Iraq. Month Polls Parliament Press

Raw Adj. Raw Adj. Raw Adj.

Jan-021 -0.110 -0.103 -0.667 -0.090 -0.613 -0.033 Feb-02 - - -0.333 -0.090 0.050 0.013 Mar-02 -0.160 -0.150 -0.217 -0.225 -0.201 -0.199 Apr-02 - - -0.097 -0.135 -0.303 -0.167 May-02 - - - - -0.020 -0.004 Jun-02 - - 0.000 0.000 -0.213 -0.017 Jul-02 - - 0.000 0.000 0.029 0.017

Aug-02 -0.170 -0.159 - - -0.262 -0.232 Sep-02 -0.080 -0.225 -0.093 -0.225 -0.105 -0.256 Oct-02 -0.053 -0.150 -0.500 -0.360 -0.079 -0.047 Nov-02 -0.090 -0.084 -0.103 -0.135 0.053 0.020 Dec-02 -0.080 -0.075 0.000 0.000 -0.215 -0.088 Jan-03 -0.170 -0.159 -0.205 -0.405 -0.167 -0.246 Feb-03 -0.230 -0.216 -0.075 -0.135 -0.149 -0.425

Mar-032 0.053 0.150 0.025 0.090 0.129 0.417 Average -0.073 -0.078 -0.151 -0.114 -0.138 -0.083

The figures reported in Table 1 have undergone two adjustments in order to meet Blumer’s

injunction that scholars of public opinion should be “realistic” (Blumer 1948, 543). The first

adjustment recognized substantial differences between the popularity of different UK

newspapers. The Independent, for example, sold an average of 224,077 copies of each issue

during this period, while the Sun sold 3,529,968, sixteen times as many3. Rather than treating

such different publications as equal contributors to public debate, it seemed more realistic to

weight their contributions according to circulation figures. The effect of this adjustment,

interestingly, is to make the results look more pro-war than they otherwise might. The

average pro-war comment piece achieved a circulation of 1,808,137 during this period

compared to the 1,195,660 achieved by the average anti-war piece. This pattern reflects the

1 Poll figure for October 2001. 2 To 20 March 2003 only 3 All data from the Audit Bureau of Circulations.

12

fact that popular papers declared pro-war views rather than suggesting that pro-war views

proved more popular (though there is some evidence that the Mirror’s aggressively anti-war

line cost it market share).

The second adjustment reflected the extent to which public interest in the Iraq issue

varied over the course of this period, as revealed by Figure 1. It involved weighting the ‘raw’

monthly figures reported in Table 1 according to the proportion of the total texts covered by

this study produced in each month, in order to produce the adjusted data contained in the

same table. The ‘raw’ figures, in other words, report the balance between pro and anti-war

views in a given month, while the ‘adj’ figures also reflect the intensity of those views. For

example, the ‘raw’ column indicates that a very high proportion of the comment pieces

published on Iraq in January 2002 expressed anti-war views. The figure for January 2003 is

much closer to zero, suggesting lower levels of opposition. But while the January 2002 figure

summarizes views expressed in just 11 comment pieces, the January 2003 figure represents

200 pieces. Although the average view expressed in January 2002 was more anti-war than

that expressed in January 2003, there were in absolute terms far more anti-war comment

pieces in circulation by then than there had been a year before. The adjusted figures reflect

this variation in intensity. The raw figures do not.

FIGURE 2: Salience-adjusted net public approval of the prospect of war with Iraq.

-0.500

-0.400

-0.300

-0.200

-0.100

0.000

0.100

0.200

0.300

0.400

0.500

Jan-

02

Feb-

02

Mar

-02

Apr

-02

May

-02

Jun-

02

Jul-

02

Aug

-02

Sep-

02

Oct

-02

Nov

-02

Dec

-02

Jan-

03

Feb-

03

Mar

-03

Polls

Parliament

Press

13

Figure 2 presents the adjusted data in graphical form. This allows us to say that the

spikes in interest in March 2002 and in September 2002 depicted in Figure 1 largely reflected

surges in anti-war opinion, while that in February and (especially) March 2003 reflected a

dramatic shift from anti-war to pro-war attitudes. However we look at this data, one

overwhelming point appears clear. All three of the indicators of public attitudes discussed by

this study showed predominant public opposition to the prospect of war with Iraq throughout

the period leading up to the Blair government’s decision to go to war. The majority of poll

respondents, press commentators and elected parliamentarians rejected the idea repeatedly

and consistently during 2002 and early 2003, though there was a clear “rally effect” in March

2003 as the prospect of war became imminent (Mueller 1970, 1973, Baker and Oneal 2001,

Lai and Reiter 2005).

A simple covariance test shows that the figures for press and poll opinion are strongly

related (R=0.86). This is to be expected given the role of the press in commissioning polls,

and specifically the polls discussed here, which were mostly commissioned by the Guardian.

The link between press and parliamentary opinion is less direct (R=0.51). But it is still

substantial. The difference appears to be explained by the fact that shifts in press opinion

generally preceded shifts in parliamentary attitudes. This could be a function of MPs relying

on press commentary to help shape their views. There is some evidence of this happening, for

example during September 2002, when press hysteria about an imminent attack on Iraq drove

MPs to demand their own debate. It likely also reflects the different structural constraints the

two groups faced. Journalists could write whenever they liked. MPs could speak in

parliament only when parliament was sitting. This was particularly important during August

and December 2002, when press opposition spiked while MPs were away from Westminster

for the summer and Christmas parliamentary breaks. They were unable to respond to press

complaints in August, for example, until they forced the government to recall parliament

early for a special debate on Iraq in late September.

Qualitative analysis

The Blair government knew it faced considerable domestic opposition to the prospect

of war in Iraq, yet it proceeded regardless. This section looks in more detail at the three

moments of particularly intense public attentiveness to the Iraq question identified by the

previous section, in March 2002, September 2002 and February-March 2003. It seeks to

understand how ministers and other elite actors constructed public opposition to the prospect

14

of war in Iraq, and how their constructions made proceeding without express public support

both thinkable and politically possible.

March 2002

Although the Guardian attempted to fire the starting gun in October 2001, the Iraq

deabte did not really begin until early 2002. Several journalists speculated, once the Taliban

fell, about what ‘phase two’ of the ‘war on terrorism’ might involve. But it was not until the

end of January 2002 that President Bush, making his first State of the Union address, offered

an answer. Naming Iran, North Korea and Iraq, he warned of an “axis of evil” between rogue

states “and their terrorist allies” (Bush 2002a). He planned, several British observers inferred,

to turn his attention to Iraq. Robin Cook, the former Foreign Secretary who later resigned

from the Cabinet over Iraq, felt the resulting opposition was unrepresentative, having been

“generated by the newspapers themselves”. Afraid of missing out on a scoop, too many

journalists proved too eager to interpret a rhetorical device as evidence of a shift in US

strategy, and a meeting as a firm UK commitment to follow (Cook 2003, 113). As one

commentator put it, although the speech had been primarily rhetorical, “rhetoric on occasion

does change reality. And the axis of evil is for real” (Cornwell 2002).

Vice-President Cheney’s visit to London on 11 March and the announcement that

Tony Blair would visit President Bush at the latter’s ranch in Crawford, Texas at the start of

April further fuelled press speculation. Anticipating the arrival in London of a known

administration ‘hawk’, several journalists assumed Cheney’s primary purpose was to plot

military action. Anti-war commentators launched a pre-emptive strike, warning “it would be

lunacy to attack Iraq at the moment, madness for Britain to back an American assault, and

complete insanity to make any large-scale commitment of our troops” (Mirror 2002). Their

pro-war counterparts retaliated, insisting that Cheney was in London to “listen and not

lecture” (The Times 2002a). This was, they claimed, a genuine opportunity for Anglo-

American co-operation. A joint press conference between Vice President and Prime Minister

passed off well. Both insisted that the proliferation of WMD posed a potential threat, while

maintaining that no decisions had been made over what to do about Iraq’s alleged weapons

development. Cheney, however, caused some concern by echoing the ‘Axis of Evil’ speech,

warning of a “potential marriage” between al-Qaeda and Iraq (Blair and Cheney 2002).

Cheney’s words comforted pro-war commentators, themselves convinced that “rogue

15

groups” were working closely with “rogue states” (The Times 2002b). Critics fulminated that

there was no evidence of any such link (Freedland 2002).

Poll figures released in March 2002 showed little change in respondents’ attitudes

from those published five months earlier. ICM identified a one percentage point drop in

support for further military action, and a four point increase in opposition. Neither was

statistically significant. Nothing had changed, in other words. Yet policymakers thought it

had. Blair wrote to Jonathan Powell on 17 March, noting his increasing concern that “public

opinion is fragile” (Blair 2002a). Cabinet ministers agreed. Foreign Secretary Jack Straw

observed that the Labour Party would not necessarily back Blair over a further Middle

Eastern war (Straw 2002). Straw could point to clear grounds for concern. Over one hundred

Labour MPs signed an Early Day Motion, a non-binding but public statement of

parliamentary opinion, condemning the presumed US intention to invade Iraq (House of

Commons 2002). While polls showed a definite lack of public enthusiasm for war in Iraq, the

real problem lay with parliamentary and journalistic elites. The latter were, after all, the ones

commissioning and reporting the polls in the first place.

What is particularly interesting about this first spike in opposition, however, is the

way the Blair government responded. It did not in any way change its substantive stance.

Blair defined the problem as one of “persuasion” rather than policy (Blair 2002a). He used

the Crawford visit to emphasize the link between terrorism, rogue states and WMD. He

provided journalists with ample copy to fill their column inches, and offered MPs a House of

Commons debate immediately after his return. He and his ministers granted a greater number

of media interviews in the following months. Blair personally sat for a three-part special

Newsnight feature in mid-May, including an entire section dedicated to Iraq. He held the first

of what became a regular series of monthly prime ministerial press conferences on 20 June.

On 16 July he appeared before the House of Commons Liaison Committee, comprised of

Select Committee chairs, to answer questions across a range of subjects. He was the first

Prime Minister to do so in more than fifty years. The strategy worked. Public opposition

declined in April and dropped to negligible levels in May. The government’s efforts

successfully satiated the appetites of both the media and parliamentarians, while the lack of

any concrete policy developments made Iraq look less important. This first wave of public

opposition largely stemmed from frustration about the disconnect between what the Bush

administration was saying and the Blair government’s initial insistence that “no decisions

have been taken” on Iraq (Blair and Cheney 2002, Blair 2002b). Once ministers began to talk

more openly about the prospect of further military action, the sense among elites that they

16

were being shut out of the process receded, and so their levels of frustration fell. A timely

communication response proved adequate, at least at first.

August – September 2002

Westminster traditionally empties in August, and it did so in 2002. Parliament was in

the midst of its long summer recess. Even the Prime Minister went on holiday. But politics

continued as usual in Washington, and in the pages of the British press. That conjunction

proved problematic for the Blair government. Once again feeling starved of information as

ministers closed up shop in late July, British journalists looked to the increasingly vociferous

US public debate over Iraq, aided by a handful of disgruntled government MPs looking to

score points at ministers’ expense. The result was a second wave of speculation, this time

centred around the belief that an attack would take place before parliament returned. This

belief was premature. But the government did conceal the true extent of its war planning. On

23 July 2002 Blair decided that meetings with the US over the summer should proceed “on

the assumption that the UK would take part” in any military action (Rycroft 2002a).

Important decisions obviously had been made. But Blair never disclosed this. Two days after

this meeting he told journalists “we are not at the point of decision yet” (Blair 2002c). The

veteran anti-war MP Tam Dalyell tried to get parliament recalled to debate the prospect of

war almost immediately after the start of the recess. He failed. But during August the media

focused relentlessly and mostly critically on the Iraq question, as the quantitative data makes

clear. This stirred up MPs. In early September the Labour MP Graham Allen gained

sufficient support from colleagues to propose an ‘unofficial’ House of Commons sitting. He

argued that the apparent imminence of an attack demanded an immediate parliamentary

response, and his colleagues responded. Allen secured a former House of Commons speaker

to chair his debate, the BBC agreed to broadcast it live and several major newspapers

declared their support (Guardian 2002, Independent 2002, Daily Mail 2002). Faced with the

prospect of losing control over parliament, let alone the Iraq debate, the government caved.

Blair promised to release a ‘dossier’ of information on Iraq’s WMD, and to recall MPs to

debate it during September. It was an embarrassing reversal.

Again the government responded to pressure from elites as if it reflected wider

popular views. Again, poll figures showed no significant changes in mass attitudes from

earlier surveys. Blair’s chief communications adviser Alastair Campbell wrote in his diary on

1 September that “it was clear that public opinion had moved against us during August”

17

(Campbell and Stott 2007, 632-633). Blair told his monthly press conference “it is clear that

the debate has moved on” from his previous public statements in July (Blair 2002e). It was

not the mass public, represented by poll results, that had moved, however. It was the press,

and parliamentary opinion itself influenced by press commentary. On this occasion the

government’s response involved both communicative and more substantive elements. On the

substantive side, ministers urged the Bush administration to confront Iraq through the United

Nations. President Bush did exactly that with an address to the General Assembly on 12

September. On the communicative side, Blair accelerated the release of the government’s

WMD ‘dossier’. This was apparently a snap decision during his 3 September press

conference. He requested a briefing on Iraq ahead of the session, but did not mention the

dossier (Blair 2002d). His speaking notes suggested only that the dossier would emerge

“when the time is right” (Rycroft 2002b). Both John Scarlett, then Chairman of the Joint

Intelligence Committee, and Alastair Campbell, who between them would drive the

production of the dossier, indicated at the Chilcot Inquiry that the process began in earnest

after the Prime Minister’s statements on 3 September (Scarlett 2009, 56, A. Campbell 2010a,

66). This point is significant. It underlines the reactive nature of the dossier, and the role

media speculation played in forcing the government to publish it ahead of schedule. This

proved important in the longer term, given the dossier’s various weaknesses.

Whereas the government’s communications efforts largely succeeded in dampening

down speculation and opposition amongst elite public commentators in early 2002, by

September attitudes were hardening. Parliament’s disquiet lasted beyond the special debate

on 24 September and carried on well into October; indeed, the greater availability of

parliamentary time after the end of the regular recess ensured MPs grew more rather than less

vocal even as press interest again began to wane.

February – March 2003

By early 2003 most observers recognized that the US was going to war in Iraq,

probably with Britain in tow. At first, this caused resentment, and public opposition spiked

for a third time. On 15 February the largest public protests in British history shook London

and several other major cities. Around one million people marched against the prospective

invasion in the capital alone. Media coverage emphasized that most of those marching were

not the “usual suspects” who protested against any use of force abroad (Daily Mail 2003,

Cook 2003, 298). They were “middle-class, middle-aged, politely-mannered and jolly angry”

18

(Ellam 2003). Journalists and MPs both participated in and responded to the protests, while

within Downing Street Alastair Campbell noted most officials had family members who

joined the march (A. Campbell 2010b, 54). Blair, in Campbell’s words, “knew he was in a

tight spot” (Campbell and Stott 2007, 667). Again the government responded with a

communicative salvo. Blair was speaking at the Labour Party’s spring conference in Glasgow

on the day of the march (this was no accident, the marchers targeted his speech). Apparently

undeterred, he declared that “the moral case against war has a moral answer” (Blair 2003b).

Having believed since early in the Iraq debate that removing Saddam Hussein from power

was simply the right thing to do (cf. Blair 2002a), he began for the first time to make the

point in public. Shortly afterwards he sought to harness the xenophobia of the right-wing

tabloid press, by far the largest component of the UK’s fractious media market, by blaming

France for the failure of the UN inspection process.

Both strategies worked to some extent. Blaming France for the UN failure rallied

right-wing press commentators, attracting widespread favourable coverage (Evans 2003,

Rees-Mogg 2003, The Sun 2003, The Times 2003). It influenced a number of anti-war MPs

(Grice 2003). And it impacted poll results. An ICM poll published on 16 March jointly by the

pro-war News of the World and Sunday Times (both part of the News International stable)

showed 51% of respondents thought President Chirac “has poisoned the diplomatic process”

while 61% said “France has put its own interests above the rest of the world by opposing

war” (News of the World 2003, Smith and Speed 2003). Pro-war elites finally possessed poll

results worth talking about. What was particularly interesting about this episode was the

shaky foundations for the claim. President Chirac did cause consternation when he

announced on 9 March that “whatever the circumstances, France will vote no” (Chirac 2003).

Jack Straw told the Chilcot Inquiry that this intervention made securing agreement at the

Security Council impossible. Wavering states saw no benefit from speaking out once France

promised to veto regardless (Straw 2010, 19). But Jeremy Greenstock, then Britain’s

Ambassador to the UN, admitted he never secured the nine votes required to pass a

resolution, veto or not (Greenstock 2009, 72). Not for nothing did former Cabinet Minister

Clare Short label the effort to blame France “one of the big deceits” of the entire pre-war

period (Short 2010, 104). The point, however, is that the tactic worked. Elite participants in

the public debate constructed an image of a broader public satisfied with the government’s

characterization of its UN failure as a product of French perfidy.

To a significant extent, the moral arguments and the anti-French rhetoric salved

consciences rather than changing minds. Blair’s ‘moral turn’ helped those on the political left

19

resolve the stark choice they faced between backing Blair and saving Saddam. Philip Gould’s

focus groups, meanwhile, showed “an instinctive understanding that no Prime Minister would

do anything as difficult or unpopular as this for the hell of it” (Campbell and Stott 2007, 669).

Media commentators acknowledged that Blair was not simply following President Bush’s

orders, undermining the argument that the Prime Minister was acting like a ‘poodle’

(Rawnsley 2003, Orr 2003). Blair had long believed that public opinion would swing behind

the prospect of war once people faced the question of whether or not to do something about

Saddam Hussein head on (cf. Rycroft 2002a). It appeared he might be right. The same sense

of inevitability that triggered opposition in February undermined it in March, as resentment

gave way to resignation. Journalists concluded that “the time has passed when war could be

averted, so any vote against war cannot be effective or persuasive” (Daily Telegraph 2003).

Several MPs acknowledged that “we cannot prevent war. The choice is either to go in

alongside the Americans to topple Saddam Hussein, or to let the Americans go in on their

own” (Hansard 2003, cc. 867, 869). Although far from solid, as indeed it ultimately proved,

this fatalistic attitude helps explain the ‘rally effect’ apparent in the data for March 2003.

Conclusions

British public opinion largely opposed the 2003 invasion of Iraq. Britain’s

government, led by Prime Minister Tony Blair, recognized this fact. But Blair, in particular,

held his ground. David Patrick Houghton has rightly argued that foreign policy analysts

should investigate the relationship between constructed social facts and individual leaders’

cognitive beliefs (Houghton 2007). From this perspective, the broad consensus amongst

British media and political elites between early 2002 and March 2003 that ‘public opinion’

opposed the idea of invading Iraq collided with Blair’s underlying faith in his own persuasive

capabilities. In Foyle’s terms, Blair believed public support was desirable, but not necessary

(Foyle 1997). He further believed strongly in his own power to win opponents over. Though

at one stage he said he ‘did not claim to have a monopoly of wisdom in these issues’ (Blair

2003c), he acted quite differently. He told audiences ‘I do what I think is the right thing,

whether on foreign or domestic policy’ (Blair 2002c), and ‘polls or no polls, my job in a

situation like this is sometimes to say the things that people don’t want to hear’ (Blair 2003a).

Jonathan Powell labeled this self-belief Blair’s “Messiah complex” (Powell 2010, 56, Blair

2010, 117). He acknowledged it helped at times, especially during complex negotiations. It

both insulated him from opposition and generated admiration. President Bush highlighted

20

both benefits when he told reporters in Crawford that “the thing I admire about this Prime

Minister is that he doesn’t need a poll or a focus group to convince him of the difference

between right and wrong” (Bush 2002b). It also bordered on hubris. One anonymous

Downing Street aide later remarked that Blair “would think that his own judgement was at

least as good as that of the Archbishop of Canterbury, of the Cardinal of Westminster, and of

the Pope combined” (Seldon, Snowdon and Collings 2007, 153-154). Archbishop Rowan

Williams, Cardinal Cormac Murphy-O’Connor and Pope John Paul II all spoke out against

the invasion of Iraq. Blair remained unmoved.

Blair’s self-belief broke the causal link between public opinion as a social fact and

British policy towards Iraq. Though he largely accepted the elite consensus that most

ordinary people thought the war the wrong thing to do, he justified rejecting it by recalling

earlier occasions on which his judgement about foreign affairs apparently proved superior to

his critics’ (Daddow 2009), and by insisting the public mood would ultimately shift towards

his point of view. Faced with spikes in opposition in March 2002, August 2002 and February

2003, Blair responded with additional arguments rather than substantive policy changes. On

the first two occasions, the mere fact the government said something helped neutralize a

significant part of the opposition, which often derived from journalists’ (apparently

justifiable) fear that they were being kept out of the decision-making loop. The ‘dossier’ of

September 2002, for example, changed few minds. But it answered critics’ demands for some

sort of concession and gave journalists something to report. Satisfied, they moved on to other

topics for a while, even as MPs grumbled for a few weeks more. In February and March

2003, meanwhile, successful government framing efforts around the moral case for action

and French conduct at the UN combined with opposition fatigue to stymie critics’ efforts to

‘stop the war’.

Adopting a constructivist definition of ‘public opinion’ and attempting to reconstruct

it through a holistic combination of press, parliamentary and poll data apparently works.

Though the polls largely held constant throughout the pre-invasion period, at least until its

final week, press and parliamentary attention and opposition to the prospect of war waxed

and waned. Journalists responded to developments in the United States and to political

vacuums created by government communication failures (Strong 2016). Media commentary

encouraged a response from MPs, and together they prompted government efforts that met

the procedural demand for information if not substantive calls for a change in policy course.

Bringing these additional sources into consideration helps us understand both when and how

public attitudes shifted, as well as the nature of the government response. From an FPA

21

perspective the fact policymakers responded to press and parliamentary opinion ‘as if’ it

represented the public as a whole underlines the practical benefits of the holistic approach. In

more general theoretical terms, triangulating among a range of different data sources helps

compensate for the epistemological difficulties raised by opinion polls, as well as reflecting

the foreign policy analyst’s greater interest in the political question of what public opinion

does than in the methodological question of what it says. It offers a more empirically valid

and nuanced alternative strategy for studying public opinion as an influence over foreign

policy when compared to approaches based solely on opinion polls.

22

Bibliography Achen, Christopher. “Mass political attitudes and the survey response.” American Political

Science Review 69, no. 4 (1975): 1218-1231.

Adler, Emanuel. “Seizing the middle ground: Constructivism in world politics.” European Journal of International Relations 3, no. 3 (1997): 319-363.

Aldrich, John, Christopher Gelpi, Peter Feaver, Jason Reifler, and Kristin Thomson Sharp. “Foreign Policy and the Electoral Connection.” Annual Review of Political Science 9 (2006): 477-502.

Almond, Gabriel. The American People and Foreign Policy. New York: Harcourt, 1950.

Althaus, Scott. Collective Preferences in Democratic Politics: Opinion Surveys and the Will of the People. Cambridge: Cambridge University Press, 2004.

Baker, William, and John Oneal. “Patriotism or opinion leadership? The nature and origins of the 'rally round the flag' effect.” Journal of Conflict Resolution 45, no. 5 (2001): 661-687.

Bauer, Martin. “Classical content analysis.” In Qualitative Researching with Text, Image and Sound, by Martin Bauer and George Gaskell. London: Sage, 2000.

Beech, Matt. “British conservatism and foreign policy: Traditions and ideas shaping Cameron's global view.” British Journal of Politics and International Relations 13, no. 3 (2011): 348-363.

Beniger, James. “The Impact of Polling on Public Opinion: Recalling Foucault, Habermas, and Bourdieu.” International Journal of Public Opinion Research 4, no. 3 (1992): 204-219.

Berinsky, Adam. “The two faces of public opinion .” American Journal of Political Science 43, no. 4 (1999): 1209-1230.

Bevir, Mark, Oliver Daddow, and Ian Hall. “Interpreting British Foreign Policy.” British Journal of Politics and International Relations 15, no. 2 (2013): 163-174.

Blair, Tony. “Memo to Jonathan Powell.” March 17, 2002a. http://www.iraqinquiry.org.uk/media/50751/Blair-to-Powell-17March2002-minute.pdf (accessed September 22, 2015).

Blair, Tony. “Speech at the George HW Bush Presidential Library.” April 7, 2002b. http://www.theguardian.com/politics/2002/apr/08/foreignpolicy.iraq (accessed September 22, 2015).

23

Blair, Tony. “Prime Minister's Monthly Press Conference.” July 25, 2002c. http://webarchive.nationalarchives.gov.uk/20061004051823/http://number10.gov.uk/page3000 (accessed September 22, 2015).

Blair, Tony. “Memo to David Manning.” September 1, 2002d. http://www.iraqinquiry.org.uk/media/51610/1September2002-PrimeMinistertoManningnote.pdf (accessed February 15, 2016).

Blair, Tony. “Prime Minister's Monthly Press Conference.” September 3, 2002e. http://webarchive.nationalarchives.gov.uk/20061004051823/http://number10.gov.uk/page3001 (accessed September 22, 2015).

Blair, Tony. “Speech to the TUC.” September 10, 2002f. http://webarchive.nationalarchives.gov.uk/+/http://www.number10.gov.uk/Page1725 (accessed May 27, 2010).

Blair, Tony. “Prime Minister's Monthly Press Conference.” January 13, 2003a. http://webarchive.nationalarchives.gov.uk/20061004051823/http://number10.gov.uk/output/Page3005.asp (accessed October 12, 2015).

Blair, Tony. “Speech to the Labour Party Spring Conference.” February 15, 2003b. http://news.bbc.co.uk/1/hi/uk_politics/2765763.stm (accessed February 15, 2016).

Blair, Tony. “Prime Minister's Monthly Press Conference.” February 18, 2003c. http://webarchive.nationalarchives.gov.uk/20061004051823/http://number10.gov.uk/page3006 (accessed August 28, 2015).

Blair, Tony. A Journey. London: Hutchinson, 2010.

Blair, Tony, and Dick Cheney. “Press conference.” March 11, 2002. http://webarchive.nationalarchives.gov.uk/20020926165953/http://www.pm.gov.uk:80/textonly/Page4598.asp (accessed February 15, 2016).

Blumer, Herbert. “Public opinion and public opinion polling.” American Sociological Review 13, no. 5 (1948): 542-549.

Bourdieu, Pierre. “Public Opinion does not Exist.” In Communication and Class Struggle: 1. Capitalism, Imperialism, by A. Mattelart and S. Siegelaub, 124-130. New York: I.G. Editions, Inc, 1979.

Bratberg, Oivind. “Ideas, traditions and norm entrepreneurs: Retracing guiding principles of foreign policy in Blair and Chirac's speechs on Iraq.” Review of International Studies 37, no. 1 (2011): 327-348.

Brookes, Rod, Justin Lewis, and Karin Wahl-Jorgensen. “The Media Representation of Public Opinion: British Television News Coverage of the 2001 General Election.” Media, Culture and Society, 2004: 63-80.

24

Bueno de Mesquita, Bruce, and Randolph Siverson. “War and the Survival of Political Leaders: A Comparative Strudy of Regime Types and Political Accountability.” American Political Science Review 89, no. 4 (1995): 841-855.

Bush, George W. “The State of the Union.” January 29, 2002a. http://georgewbush-whitehouse.archives.gov/news/releases/2002/01/20020129-11.html (accessed February 15, 2016).

Bush, George W. “Press conference with Tony Blair.” April 6, 2002b. http://webarchive.nationalarchives.gov.uk/+/http://www.number10.gov.uk/Page1711 (accessed October 23, 2015).

Campbell, Alastair. “Evidence to the Chilcot Inquiry (morning session).” January 12, 2010a. http://www.iraqinquiry.org.uk/media/42384/20100112am-campbell-final.pdf (accessed February 15, 2016).

Campbell, Alastair. “Evidence to the Chilcot Inquiry (afternoon session).” January 12, 2010b. http://www.iraqinquiry.org.uk/media/42387/20100112pm-campbell-final.pdf (accessed February 15, 2016).

Campbell, Alastair, and Richard Stott. The Blair Years: Extracts from the Alastair Campbell Diaries. London: Arrow Books, 2007.

Campbell, David. Writing Security: United States Foreign Policy and the Politics of Identity. Manchester: Manchester University Press, 1992.

Caspary, William. “The “mood theory”: A study of public opinion and foreign policy.” American Political Science Review 64, no. 2 (1970): 536-547.

Chan, Steve, and William Safran. “Public Opinion as a Constraint Against War: Democracies' Responses to Operation Iraqi Freedom.” Foreign Policy Analysis 2, no. 2 (2006): 137-156.

Chirac, Jacques. “Interview with TF1 and France 2.” 9 March 2003. https://www.youtube.com/watch?v=7R9Q9yopHAU (accessed August 28, 2015).

Cohen, Bernard. The Public's Impact on Foreign Policy. Boston: Little, Brown and Company, 1973.

Cohen, Bernard. Democracies and Foreign Policy: Public Participation in the United States and the Netherlands. Madison: University of Wisconsin Press, 1995.

Converse, Philip. “The nature of belief systems in mass publics.” In Ideology and Discontent, by David Apter, 206-261. New York: Free Press, 1964.

Converse, Philip. “Changing conceptions of public opinion in the political process.” Public Opinion Quarterly 51, no. 2 (1987): 12-24.

25

Cook, Robin. The Point of Departure. London: Pocket Books, 2003.

Cornwell, Rupert. “Saddam is next in the firing line. It's just a matter of when.” Independent on Sunday, February 17, 2002.

Daddow, Oliver. “'Tony's war'? Blair, Kosovo and the interventionist impulse in British foreign policy.” International Affairs 85, no. 3 (2009): 547-560.

Daddow, Oliver, and Jamie Gaskarth. British Foreign Policy: The New Labour Years. Basingstoke: Palgrave MacMillan, 2011.

Daily Mail. “Leading Article.” September 6, 2002.

Daily Mail. “Leading Article.” February 15, 2003.

Daily Telegraph. “Leading Article.” March 18, 2003.

Doty, Roxanne. “Foreign policy as social construction: A post-positivist analysis of US counterinsurgency policy in the Philippines.” International Studies Quarterly 37, no. 3 (1993): 297-320.

Dunne, Tim. “Britain and the Gathering Storm Over Iraq.” In Foreign Policy: Theories, Actors, Cases, by S. Smith, A. Hadfield and T. Dunne. Oxford: Oxford University Press, 2008.

Eichenberg, Richard. “Victory Has Many Friends: US Public Opinion and the Use of Military Force 1981-2005.” International Security 30, no. 1 (2005): 140-177.

Ellam, Dennis. “Protest virgins.” The Sunday Mirror, February 16, 2003.

Entman, Robert. Projections of Power. Chicago, IL: University of Chicago Press, 2004.

Eshbaugh-Soha, Matthew, and Christopher Linebarger. “Presidential and media leadership of public opinion on Iraq.” Foreign Policy Analysis 10, no. 4 (2014): 351-369.

Evans, Harold. “Saddam's pimp.” The Daily Mail, February 15, 2003.

Everts, Philip. “Public opinion after the Cold War: A paradigm shift.” In Decision Making in a Glass House: Mass Media, Public Opinion, and American and European Foreign Policy in the 21st Century, by Brigitte Nacos, Robert Shapiro and Pierangelo Isernia. Lanham, MD: Rowman & Littlefield, 2000.

Everts, Philip, and Pierangelo Isernia. Public Opinion and the International Use of Force. New York: Routledge, 2001.

Foyle, Douglas. “Public Opinion and Foreign Policy: Elite Beliefs as a Mediating Variable.” International Studies Quarterly 41, no. 1 (1997): 141-169.

26

Freedland, Jonathan. “Saddam and al-Qaeda? There's more chance of Will marrying Vanessa.” The Mirror, March 14, 2002.

Gamson, William, and Andre Modigliani. “Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach.” The American Journal of Sociology 95, no. 1 (1989): 1-37.

Gaubatz, Kurt. “Intervention and intransitivity: Public opinion, social choice, and the use of military force abroad.” World Politics 47, no. 4 (1995): 534-554.

Gelpi, Christopher, Jason Reifler, and Peter Feaver. “Iraq the Vote: Retrospective and Prospective Judgements on Candidate Choice and Casualty Tolerance.” Political Behavior 29 (2007): 151-174.

Gelpi, Christopher, Peter Feaver, and Jason Reifler. “Success Matters: Casualty Sensitivity and the War in Iraq.” International Security 30, no. 3 (2005): 7-46.

Ginsberg, Benjamin. The Captive Public: How Mass Opinion Promotes State Power. New York, NY: Basic Books, 1986.

Graham, Thomas. “Public Opinion and US Foreign Policy Decision Making.” Chap. 9 in The New Politics of American Foreign Policy, by David Deese. New York: St Martin's Press, 1994.

Greenstock, Jeremy. “Evidence to the Chilcot Inquiry.” November 27, 2009. http://www.iraqinquiry.org.uk/media/40456/20091127am-final.pdf (accessed August 28, 2015).

Grice, Andrew. “The week in politics.” The Independent, March 15, 2003.

Guardian. “Leading Article.” September 6, 2002.

Habermas, Jurgen. The Structural Transformation of the Public Sphere. Translated by Thomas Burger. Cambridge: Polity, 1962.

Hagan, Joe. Political Opposition and Foreign Policy in Comparative Perspective. London: Rienner, 1990.

Hansard. House of Commons Debates. 6th Ser. Vol. 401. March 18, 2003.

Hansen, Lene. Security as Practice: Discourse Analysis and the Bosnian War. Abingdon: Routledge, 2006.

Herbst, Susan. “Surveys in the Public Sphere: Applying Bourdieu's Critique of Opinion Polls.” International Journal of Public Opinion Research 4, no. 3 (1992): 220-229.

27

Herbst, Susan. Numbered Voices: How Opinion Polling Has Shaped American Politics. Chicago, IL: University of Chicago Press, 1993.

Hill, Christopher. “Public Opinion and British Foreign Policy Since 1945: Research in Progress?” Millennium: Journal of International Studies 10, no. 1 (1981): 53-62.

Hill, Christopher. “Bringing War Home: Foreign Policy Making in Multicultural Societies.” International Relations 21, no. 3 (2007): 259-283.

Hilsman, Roger. The Politics of Policymaking in Defense and Foreign Affairs: Conceptual Models and Bureaucratic Politics. Englewood Cliffs: Prentice-Hall, 1987.

Holsti, Ole. “Public Opinion and Foreign Policy: Challenges to the Almond-Lippmann Consensus.” International Studies Quarterly 36, no. 4 (1992): 439-466.

Holsti, Ole. Public Opinion and American Foreign Policy. Ann Arbor: University of Michigan Press, 1996.

Houghton, David Patrick. “Reinvigorating the study of foreign policy decision making: Toward a constructivist approach.” Foreign Policy Analysis 3, no. 1 (2007): 24-45.

House of Commons. “Early Day Motion no. 927.” March 4, 2002. http://www.parliament.uk/edm/2001-02/927 (accessed February 15, 2016).

Hurwitz, Jon, and Mark Peffley. “How are foreign policy attitudes structured? A hierarchical model.” American Political Science Review 81, no. 4 (1987): 1099-1120.

Independent. “Leading Article.” September 6, 2002.

Isernia, Pierangelo, Zoltan Juhasz, and Hans Rattinger. “Foreign Policy and the Rational Public in Comparative Perspective.” The Journal of Conflict Resolution 46, no. 2 (2002): 201-224.

Jacobsen, John. “Why do States Bother to Deceive? Managing Trust at Home and Abroad.” Review of International Studies 34 (2008): 337-361.

Jervis, Robert. Perception and Misperception in International Politics. Princeton, NJ: Princeton University Press, 1976.

Key, Vladimer. “The politically relevant in opinion surveys.” Public Opinion Quarterly 24, no. 1 (1960): 54-61.

Key, Vladimer. Public Opinion and American Democracy. New York: Knopf, 1961.

Klarevas, Louis. “The 'Essential Domino' of Military Operations: American Public Opinion and the Use of Force.” International Studies Perspectives 3 (2002): 417-437.

28

Klarevas, Louis. “Correspondence: Casualties, Polls, and the Iraq War.” International Security 31, no. 2 (2006): 186-198.

Koinova, Maria. “Four types of diaspora mobilization: Albanian diaspora activism for Kosovo independence in the US and the UK.” Foreign Policy Analysis 9, no. 4 (2013): 433-453.

Krippendorff, Klaus. “Reliability in content analysis: Some common misconceptions and recommendations.” Human Communication Research 30, no. 3 (2004): 411-433.

Lacina, Bethany, and Charlotte Lee. “Culture clash or democratic peace?: Results of a survey experiment on the effect of religious culture and regime type on foreign policy opinion formation.” Foreign Policy Analysis 9, no. 2 (2013): 143-170.

Lai, B., and D. Reiter. “Rally 'round the Union Jack? Public opinion and the use of force in the United Kingdom, 1948-2001.” International Studies Quarterly 49, no. 2 (2005): 255-272.

Lippmann, Walter. Public Opinion. New York: Harcourt, Brace and Co., 1922.

Mirror. “Leading Article.” March 11, 2002.

Mueller, John. “Presidential popularity from Truman to Johnson.” American Political Science Review 64, no. 1 (1970): 18-33.

Mueller, John. War, Presidents and Public Opinion. New York: Wiley, 1973.

News of the World. “Poll results: Iraq.” March 16, 2003.

Nincic, Miroslav. Democracy and Foreign Policy: The Fallacy of Political Realism. New York: Columbia University Press, 1992.

Nincic, Miroslav, and Jennifer Ramos. “The sources of patriotism: Survey and experimental evidence.” Foreign Policy Analysis 8, no. 4 (2012): 373-388.

Noelle-Neumann, Elizabeth. “Public opinion and the classical tradition: A re-evaluation.” Public Opinion Quarterly 43, no. 2 (1979): 143-156.

Orr, Deborah. The Independent, March 14, 2003.

Osborne, Thomas, and Nikolas Rose. “Do the Social Sciences Create Phenomena? The Example of Public Opinion Research.” British Journal of Sociology 50, no. 3 (1999): 367-396.

Page, Benjamin, and Robert Shapiro. The Rational Public: Fifty Years of Trends in Americans' Policy Preferences. Chicago: University of Chicago Press, 1992.

29

Partell, Peter. “Executive Constraints and Success in International Crises.” Political Research Quarterly 50, no. 3 (1997): 503-528.

Peer, Limor. “The Practice of Opinion Polling as a Disciplinary Mechanism: A Foucauldian Perspective.” International Journal of Public Opinion Research 4, no. 3 (1992): 230-242.

Powell, Jonathan. The New Machiavelli: How to Wield Power in the Modern World. London: The Bodley Head, 2010.

Powlick, Philip. “The Sources of Public Opinion for American Foreign Policy Officials.” International Studies Quarterly 39, no. 4 (1995): 427-451.

Powlick, Philip, and Andrew Katz. “Defining the American Public Opinion/Foreign Policy Nexus.” Mershon International Studies Review 42, no. 1 (1998): 29-61.

Rawnsley, Andrew. The Observer, March 2, 2003.

Rees-Mogg, William. “In all honesty, they were still Saddam's useful idiots.” The Times, February 17, 2003.

Risse-Kappen, Thomas. “Public Opinion, Domestic Structure, and Foreign Policy in Liberal Democracies.” World Politics 43, no. 4 (1991): 479-512.

Risse-Kappen, Thomas. “Masses and Leaders: Public Opinion, Domestic Structures, and Foreign Policy.” Chap. 11 in The New Politics of American Foreign Policy, by David Deese. New York: St Martin's Press, 1994.

Rosenau, James. Public Opinion and Foreign Policy: An Operational Formulation. New York: Random House, 1961.

Ross, Liat Radcliffe. “Muslim interest groups and foreign policy in the United States, Canada and the United Kingdom: Identity, interests and action.” Foreign Policy Analysis 9, no. 3 (2013): 287-306.

Rycroft, Matthew. “Iraq: Prime Minister's Meeting, 23 July.” The Sunday Times. July 23, 2002a. http://web.archive.org/web/20110723222004/http://www.timesonline.co.uk/tol/news/uk/article387374.ece (accessed August 26, 2015).

Rycroft, Matthew. “Press Conference: Iraq: Difficult Questions - Memo to Tony Blair.” September 2, 2002b. http://www.iraqinquiry.org.uk/media/51613/2September2002-RycrofttoPrimeMinister-minutePressConference-Iraq-difficultquestions.pdf (accessed February 15, 2016).

30

Scarlett, John. “Evidence to the Chilcot Inquiry.” December 8, 2009. http://www.iraqinquiry.org.uk/media/40665/20091208pmscarlett-final.pdf (accessed February 15, 2016).

Schuster, Jurgen, and Herbert Maier. “The Rift: Explaining Europe's Divergent Iraq Policies in the Run-Up of the American-Led War on Iraq.” Foreign Policy Analysis 2, no. 3 (2006): 223-244.

Seldon, Anthony, Peter Snowdon, and Daniel Collings. Blair Unbound. London: Simon and Schuster, 2007.

Seymour, David, and Oonagh Blackman. The Mirror. September 2, 2002.

Shapiro, Robert, and Lawrence Jacobs. “Who Leads and Who Follows? US Presidents, Public Opinion, and Foreign Policy.” In Decision Making in a Glass House: Mass Media, Public Opinion, and American and European Foreign Policy in the 21st Century, by Brigitte Nacos, Robert Shapiro and Pierangelo Isernia. Lanham, Maryland: Rowman & Littlefield, 2000.

Short, Clare. “Evidence to the Chilcot Inquiry.” February 2, 2010. http://www.iraqinquiry.org.uk/media/44771/20100202am-short-final.pdf (accessed August 20, 2015).

Smith, David, and Nick Speed. “French stance tilts voters towards war.” The Sunday Times, March 16, 2003.

Sobel, Richard. The Impact of Public Opinion on US Foreign Policy Since Vietnam: Constraining the Colossus. New York: Oxford University Press, 2001.

Solomon, Ty. “Social Logics and Normalisation in the War on Terror.” Millennium: Journal of International Studies 38, no. 2 (2009): 269-294.

Splichal, Slavko. Public Opinion: Developments and Controversies in the Twentieth Century. Oxford: Rowman and Littlefield, 1999.

Straw, Jack. “Memo to the Prime Minister.” March 25, 2002. http://www.iraqinquiry.org.uk/media/50151/straw-to-blair-25march2002-letter.pdf (accessed February 15, 2016).

Straw, Jack. “Evidence to the Chilcot Inquiry: Written memorandum.” January 21, 2010. http://www.iraqinquiry.org.uk/media/43119/jackstraw-memorandum.pdf (accessed February 15, 2016).

Strong, James. “Two level games beyond the United States: International indexing in Britain during the wars in Afghanistan, Iraq and Libya.” Forthcoming, 2016.

The Guardian. “Leading Article.” October 10, 2001.

31

The Sun. “The Sun says: Leading Article.” March 6, 2003.

The Times. “Leading Article.” March 11, 2002a.

The Times. “Leading Article.” March 12, 2002b.

The Times. “Leading Article.” March 6, 2003.

Tilly, Charles. “Speaking your mind without elections, surveys or social movements.” Public Opinion Quarterly 47, no. 4 (1983): 461-478.

Verba, Sidney. “The citizen as respondent: Sample surveys and American democracy.” American Political Science Review 90, no. 1 (1996): 1-7.

Verba, Sidney, et al. “Public Opinion and the War in Vietnam.” American Political Science Review 61, no. 2 (1967): 317-333.

Vertzberger, Yaacov. “Foreign policy decisionmakers as practical-intuitive historians: Applied history and its shortcomings.” International Studies Quarterly 30, no. 2 (1986): 223-247.

Weldes, Jutta. “Constructing national interest.” European Journal of International Relations 2, no. 2 (1996): 275-318.

Zaller, John. The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press, 1992.

Zaller, John. “Coming to grips with V.O. Key's concept of latent opinion.” In Electoral Democracy, by Michael MacKuen and George Rabinowitz, 311-336. Ann Arbor, MI: University of Michigan Press, 2003.

Zaller, John, and Stanley Feldman. “A simple theory of the survey response: Answering questions versus revealing preferences.” American Journal of Political Science 36, no. 3 (1992): 579-616.