Attribution Theory and Arab Images of the Gulf War Author(s): Daniel Heradstveit and G.

Matthew Bonham Source: Political Psychology, Vol. 17, No. 2 (Jun., 1996), pp. 271-292 Published by: International Society of Political Psychology Stable URL: http://www.jstor.org/stable/3791811 Accessed: 02/04/2009 20:00
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ispp. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

International Society of Political Psychology is collaborating with JSTOR to digitize, preserve and extend access to Political Psychology.

http://www.jstor.org

Political Psychology, Vol. 17, No. 2, 1996

AttributionTheoryand Arab Imagesof the Gulf War

Daniel Heradstveit
Norwegian Institute of International Affairs, Oslo University of Bergen

G. Matthew Bonham
Maxwell School of Citizenship and Public Affairs Syracuse University

This article describes the resultsof an in-depthinterviewstudyof Arab elites in the wake of the Gulf war, research that only partially supports a crucial finding of research on actor-observerdifferencesin social attribution.Although the aggregated explanationsof these Arab respondentswere consistentwith the predictionsof the theory,the attributions of Iraq's behavior by Egyptian elites, and attributionsof the Coalition's behavior by Moroccan and Tunisianelites were notfully consistentwith the hypothesis.These results were interpretedto suggest the importanceof distinguishing,in applicationsof attribution theory,betweencomplexpolitical situationsand simplesocial situations,theperspectiveof actors versus observers,and culturaldifferencesin discursivepractices.
KEY WORDS: Arab elites; attributiontheory; dispositional attributions;fundamentalattribution error;Gulf war;perception;SaddamHussein;and situationalattributions

I. INTRODUCTION On 2 August 1990 IraqinvadedKuwaitand declaredthe countryto be Iraq's 19thprovince.Twelve U.N. SecurityCouncilresolutionsdemandedthe immediate and unconditionalwithdrawalof Iraq,first imposing sanctionsand finally authorizing the use of force. Iraq,underthe leadershipof SaddamHussein, refused to back down, setting in motion on 17 January1991 the militaryforces of a coalition led by the UnitedStatesandjoined by morethan30 othercountries,includingArab countries,to force Iraqto comply with the U.N. resolutions.The warended with a

271
0162-895X ? 1996 International Society of Political Psychology Publishedby Blackwell Publishers,238 Main Street,Cambridge,MA 02142, USA, and 108 Cowley Road, Oxford, OX4 1JF, UK.

272

Heradstveit and Bonham

ceasefire on 28 February. Iraqiforces were defeated,Iraq'seconomic infrastructure was severely damaged,and many Iraqicivilians were left dead, injured,or homeless.1 of Despite the participation Arabcountriesin the U.S.-led Coalition,people in the Arab world did not view the situation with one mind. In North Africa, for in example, the onset of fighting resultedin protestsand demonstrations support of Iraq. "It is doubtfulthat many North Africans actually approvedof the Iraqi invasion of Kuwait .... [but] many were drawnto SaddamHussein because he to or represented, at least pretended represent,a challengeto prevailingpatternsof political economy,"as well as the UnitedStatesandits allies who "wereperceived to be operatingin defense of the statusquo"(Tessler, 1991, pp. 8-9). Both Tunisia and Morocco, countrieswhose governmentssupported Coalition,experienced the substantial public support for Iraq and Saddam Hussein (Seddon, 1991, pp. 110-111). Even in Egypt, which contributednearly 30,000 troops to the Coalitionforce, the adventof the groundwartriggeredstudentdemonstrations and calls for a ceasefire (Haseeb& Rouchdy, 1991, pp. 77-78).2 White (1994) argues that attribution theory and one of its central ideas, the "fundamental attribution error"hypothesis,can help us to understand why some with SaddamHussein during the Gulf war. This hypothesis people empathized states that actors explain their own behaviorin terms of situationalconstraint("I act this way becauseI have no choice"),while observersexplainthe same behavior in terms of innatedisposition("Sheacts like thatbecause it is her nature"). In this articlewe will use attribution theoryto study how people in the Arab world viewed the Gulf war. Previously, attributiontheory has been applied to situationsthat were seen in the Arab world as being black-and-white, such as the conflict between the Arabsand Israelis(Heradstveit,1979). The Gulf war is quite different, however, because many Arabs were ambivalentabout the issues involved. While therewas sympathyfor Iraq'spoliticalclaims, which were seen as rectifying old injusticesdone to Iraqby Westernpowers, there was little support for its methodsand even less afterhe was defeatedand became a "loser." Before describing our empirical application,which involved field work in Egypt, Morocco, andTunisia,we will first discuss the claims of attribution theory with a view to exploringthe efficacy of this approach outside a laboratory setting. Because there are so few applicationsof attributiontheory to complex social situationsinvolving people with differentidentitycommitments,we approach this task with serious misgivings about the theoreticaland methodologicalproblems thatwe may encounter,as well as aboutthe relevanceof a theorythatis grounded in an object-oriented view of language(Bonham& Shapiro,1997).

1For a thoroughdescriptionand analysis of the Gulf war, see Freedmanand Karsh(1993).

2 For more discussion of Arabsupportfor SaddamHussein, see Khalidi(1991) and Khouri(1991).

Attribution Theory and Arab Images of the Gulf War

273

The Study of Social Attribution3 The study of attribution a separatearea within social psychology can be as in tracedback to Heider (1958), who examinedthe process of "causalattribution the perceptionof others."Since the publicationof Heider'searly work, attribution theorists have studied the efforts of people to explain and draw inferences from behavior-their own behaviorandthebehaviorof others.The individual,according thinker" "naivescientist"who searchesfor or to this viewpoint, is a "constructive the causes of events and drawsconclusions aboutpeople and their circumstances as a basis for action. Often,the searchfor causalexplanationis aidedby schemata, that or cognitive struckires represent "organized knowledgeabouta given concept or type of stimulus"(Fiske & Taylor, 1984, p. 140). Schematainclude general structures,based either on standardknowledge or an individual's direct prior aboutthe self thatarederivedfrompast experience,self-schemata,generalizations of experience, and persona, which are representations the "personaltraits and characteristic behavior or particular human types" (Vertzberger, 1990, pp. 156-159). In the field of internationalrelations, Jervis (1976) has applied attribution theory to foreign policy decision-making,Heradstveit(1979) has studied how Arabs and Israelisperceivedthe causes of the MiddleEast conflict, Larson(1985) has used attribution theory to explain the containmentpolicies of the Cold War, and Heradstveitand Bonham (1986) have analyzedthe attributions American of andNorwegianpolicy officials withrespectto Soviet activitiesin northern Norway. The "naivescientist"framework not the only way of thinkingaboutattribuis tion. Attributions serve otherfunctionsin additionto cognitivemastery."Therange of motives hypothesized to affect attributionsincludes the need to protect or enhance one's self-esteem (e.g., Miller, 1976; Zuckerman, 1979), to create a favorable impression to others (e.g., Bradley, 1978; Orvis et al., 1976), and to believe in a just world(e.g., Lerner& Miller, 1978)"(Tetlock& Levi, 1982, p. 69). This researchsuggeststhe operation self-esteemmotivesor "self-servingbiases" of in attribution (Bradley,1978). Self-esteemtheory,for example,predictsthatpeople will make internal attributionsfor success and external attributionsfor failure (Tetlock & Levi, 1982; Bradley, 1978). This tendency may be especially strong when people feel emotionalaboutan issue. The presenceof an audience,moreover, may affect the attributionspeople communicate to others. "According to the self-presentationposition, people often communicateattributions designed, consciously or unconsciously, to gain public approvaland to avoid embarrassment" (Tetlock & Levi, 1982, p. 78).

3 This section is a revisionof the literature review and

and critiquefoundin Heradstveit Bonham(1986).

274

Heradstveit Bonham and

The Fundamental Attribution Error JonesandNisbett(1972) haveobservedthatattributors emphasizedispositions (abilities,traits,or motives) when explainingthe behaviorof others,while they use to situationalfactors(externalpressuresand constraints) explaintheirown behavior. Ross and otherwritershave referredto this phenomenonas "thefundamental "to the attribution error,"a tendencyin social attribution underestimate impactof situationalfactorsandto overestimate role of dispositionalfactorsin controlling the behavior"(1977, p. 83). The persistenceof the tendencyto ascribe"causalimportance personsat the to attribution expense of circumstances"suggests to some that the "fundamental error" may be "deeplyrootedin the individualpsyche,"ratherthana consequence of "the complexity and ambiguityof 'reality"'(Renshon, 1993, p. 72). Behavior to attributed the inherentnatureof othershas the advantageof imposingregularity thus andpredictability, enhancingthefeeling of control(Miller,Norman,& Wright, 1978). This creates a dilemma,however, because the feeling of control provided by dispositionalexplanations"hasto be balancedagainstotherneeds, such as the need for veridicality.These otherneeds may result... in self-serving situational 1990, p. 162). explanationsof others' behavior"(Vertzberger, attribution error" beenconfirmedin laboratory has Althoughthe "fundamental research,it has not been found in all situations(Cheng & Novick, 1990, p. 547). is Moreover,one can arguethatthe term"error" a misnomer.Is an involved actor in a betterposition to identifythe cause of his or her own behaviorthanthatof an uninvolvedobserver? MonsonandSnyder(1977) have statedthe case for the actor. First,they arguethatactorshave knowledgeof theirown innerstates,attitudes,and dispositions. Such informationis normallynot availableto observers.Moreover, actorsareusuallymoreknowledgeableabouttheirown behaviorin othersituations and at othertimes thanobservers.Theiranalysis"suggeststhatthe actor's attributions of cause would be more often 'correct'thanthose of the observer"(Monson & Snyder, 1977, p. 94). Not all psychologists agree thatactorsare awareof theirhigher-order mental processes. Nisbett and Wilson (1977) argue convincingly, however, that people, when asked to report how a particular"stimulus"influenced a particular"response,"applyor generate"causaltheoriesaboutthe effects of thattype of stimulus on that type of response."Sources for these causal theoriesinclude the cultureor subculture,empiricalobservation,and shared"networksof connotativerelations surroundingthe stimulus descriptionand the response description"(Nisbett & Wilson, 1977, p. 248). Otherresearchemphasizesthe importance the evaluativeaspect-the social of desirabilityof behavior-in the attribution process.Taylorand Koivumaki(1976) found thatthe actordoes not generallyview his or her behaviormore situationally thanthe behaviorof others.Instead,the actorexplainshis or herpositive behavior in termsof dispositions(e.g., "I won becauseI'm strong"),but dismisses negative

Attribution TheoryandArabImagesof the GulfWar

275

behavioras being caused by the situation(e.g., "I was outnumbered"). Likewise, Heradstveit(1979) found little evidence for the "fundamental attribution error" until the evaluative aspect of behaviorwas taken into account. Arab and Israeli respondents,however, were overwhelminglydispositionalwhen observing their own good behavior (and their opponent's bad behavior) and situational when attributingtheir own bad behavior(and their opponent's good behavior).These findings seem to be consistent with motivationalor functionalapproachesto the study of attribution. Social Attribution: The "British School" Approach Because our researchdesign takes into account societal and culturaldifferences, the work of Europeanscholars like Antaki (1981) and Hewstone (1989), which accountsfor groupandsocietalcircumstances, highly relevant.Hewstone is and other representativesof the "Britishschool" approach(Weary et al., 1989, or p. 192) have an interestin "socialrepresentations" sharedknowledgestructures (Hewstone 1989, p. 207), and how people make and communicateattributions about social events such as poverty and wealth, unemployment,riots, and racial equality (Hewstone 1989, pp. 217-231). This approachhas been influenced by Ichheiser (1943), who focused on "collectively (ideologically) conditioned patternsof misinterpretation" 145) (p. that were typical of people living within a group or a society. Ichheiser first identifiedthe "fundamental attribution error," noting that "success"and "failby "merit"and "blame"were collective determinations "basedon the belief in ure," determination behaviorandwere builtinto the ideology of our society" of personal (Hewstone, 1989, p. 212). Researchon cultureandattribution a closely relatedapproach. example, is For Miller (1984) found thatHindusaremorelikely thanAmericansto attribute events to situationalor contextualfactors.Althoughher researchalone does not discredit the "fundamentalattributionerror"hypotheses, it does suggest that "different culturesconstruethe worldin ways thataretrulydifferentat base"(Ross & Nisbett, 1991, p. 186). People in the United States and most Western cultures may be socialized to expect that the actor is the cause of his or her behavior(Quattrone, 1982, p. 607), while in collectivist cultures,behavioris viewed more as reflecting to norms, success is attributed help from others,and failureis explainedby a lack of effort (Triandis,1993, p. 164). Cultural Stereotypes and Attribution The researchof BodenhausenandWyer (1985) has examinedculturalstereobehaviorthatis consistypes and attribution. They foundthatsocially undesirable

276

Heradstveit and Bonham

tent with a culturalstereotypeis attributed observersto dispositionalfactors.In by such cases the observers are also more punitive in their interpretations. Such behaviorthatis contrary a cultural to on the otherhand,will be attributed stereotype, to transitory situational or factors,andobserverswill be morelenient.Forexample, in a study carriedout in southernIndia,Taylorand Jaggi (1974) found thatHindu undesirable behaviorsperformed Hindusto externalfactors, subjectsattributed by while they explainedthe undesirable behaviorof Muslims in the same situationby referringto internalcharacteristics. Afterreviewingthe literature cultural on Hewstone stereotypesandattribution, finds clear evidence for the "tendency to attributenegative outgroup (1989) behavior to personal causes within the actor" (p. 190). There are exceptions, however, especially in the case of "low statusgroups, againstthe backgroundof political and culturaltension"(Hewstone, 1989, p. 176). For example, a study in Malaysia found evidence of ethnocentricattributionfor Malays but not among membersof the minorityChinese community(Hewstone& Ward, 1985). Theoretical and Methodological Problems The study of attribution hampered the absenceof a "falsifiablecognitive is by that specifies when people will use particularrules of causal theory-a theory inference or the extent to which priorbeliefs will influence the interpretation of evidence" (Tetlock & Levi, 1982, p. 74). There is no comprehensive incoming theoreticalstructureto explain the "fundamental attribution error"or any other heuristics.Although plausible explanationsare offered by researchers, cognitive is "plausibility" largely a function of some formal or informalrule of judgment. For example, the "fundamental attribution error"has been attributed the repreto sentativeness heuristic: "Actors (and their dispositions) may be a more representativecause of behaviorthanaresituations,becauseit is, afterall, the actorwho does the acting"(Nisbett & Ross, 1980, p. 122). The same researchershave also explained the phenomenonby the availabilityheuristic:"The actor is an easily available explanationof this action because of his perceptualproximity to his action"(p. 122). Othershave suggestedthat self-servingbiases explain the social attribution process: Individualstend "to accept responsibilityfor positive behavioral outcomes and to deny responsibility for negative behavioral outcomes" (Bradley, 1978, p. 68). Methodologicalproblemshave also hamperedresearchon social attribution. Findings are based, for the most part,on laboratory experimentsin which captive are populations,such as college undergraduates, asked to explain their actions or the behavior of others in hypotheticalor trivial situations. Because the generalizabilityto "realworld"settings of many of these experimentsis "questionable" (Olson & Ross, 1985, p. 287), "muchmoreempiricalworkusing more naturalistic stimulus materials(ratherthan or in additionto linguistic materials)to.represent

Attribution Theory and Arab Images of the Gulf War

277

social stimuli will be necessary . . ." (Weary et al., 1989, p. 61). Such research, conductedin naturalsettings and focusing on complex situations,may reveal that "people do not simply attributecauses and responsibility;they also offer more elaborateaccountsand stories"(Wearyet al., 1989, p. 194). Forced-choice,closed-endedscales are almost always used to recordattribushow a lack of convertions, but studies using multitrait-multimethod approaches such scales. This implies "eitherthatthe scales areunreliableor that gence among they are measuringdifferentthings"(Milleret al., 1981, p. 83). Researchbased on the use of free-choiceand open-endedscales, on the other to hand,which allows respondents expresstheirown views aboutcause andblame, is time-consumingand open to coding problems(Howard,1987, p. 50). Coding is often done on the basis of the form of the responserather thanthe content(Van der Plight, 1981). For example, the statement"Johndoesn't want to go to the soccer match because of hooliganism"might be coded as a situationalattribution. However, the response "Johnwill not go because he is afraidof getting mixed up in a (Van der Plight, 1981, p. 99). fight" might be coded as a dispositionalattribution This coding problemis frequentlyencountered becausethe situational-dispositional distinctionis "notreallya dichotomy.Most situationalexplanationsimply assumptionsaboutrelevantdispositions"(Monson& Snyder, 1977, p. 20). "Each covers more than one type of thing, and either can have the same role in some instances"(White, 1991, p. 266). For example, "He did it for the money" can be restatedas the dispositionalattribution "He did it because he is money hungry." distinction"mayreflectdifferencesin language Hence, the situational-dispositional ratherthanthought"(Monson& Snyder, 1977, p. 20). Ross (1977, p. 5) has attemptedto solve the problem by formulatingthe following definitionsof situationaland dispositionalattributions: SituationalAttributions. Those explanations stateor imply no dispositions that on the partof the actorbeyond those typical of all or most actors. Those explanationsthat state or imply something Dispositional Attributions. uniqueor distinguishingaboutthe actor. Although the proposal by Ross may not have solved the problem (Lau & distincRussell, 1980), it does offer some guidance.The situational-dispositional tion may not representa dichotomousclassification,but we can makejudgments and code the perceiver'sweighing of the relativeimportanceof each (Monson & Snyder, 1977). An explanation or attributionis dispositional only when it focuses on the It idiosyncraticor the particular. may be useful to call dispositional attributions attributions as thenthe contentstandsout moreclearly.We may instead, "personal" thus ask whetherthe respondentis explainingan event by referringto the personto ality traitsandpeculiaritiesof the actor,or by referring a responsethatis thought to be caused by stimuli in the actor'senvironment.

278

Heradstveit Bonham and

Hypotheses researchon actor-observer Attribution differencesin attritheory,particularly us to understand some of the conclusionsthatArabsdrew about butions,may help the actorswho were involved in the Gulf war. Observers,as contrastedto actors, will attribute factorsandsocially socially desirablebehaviorof othersto situational undesirablebehaviorof others to dispositionalcauses (Monson & Snyder, 1977, p. 103). Hence, Arab elites in our interviewsample, observersof behaviorin the Gulf war, should be inclined to explain the actions of the Coalition in situational terms that imply no dispositions beyond those typical of most actors, (e.g., "the Coalition had to move against aggression"),while attributingIraqi actions to dispositionalor personalfactorsthat are idiosyncraticor unique (e.g., "Iraqhas a war-likeculture"). Sampling and Interviewing Procedures For this researchthe characteristics the statisticaluniverse are unknown, of which makesit difficult,if not impossible,to obtaina "representative" sample.We were looking for membersof the "elite"who were helping to influence policy in the conflict. Nobody has yet developedgeneralcriteriathatcan be used to define who is and who is not an elite politicalactor.Even if the relevantelite groupswere clearly defined, it would still be difficult to specify which persons ought to be includedin the theoretical universe.Sucha specificationwouldrequiremuchcostly field work and complex analysismerely to delineatethe theoreticaluniverse. Instead, we tried to find what we might, ratherimprecisely,call "leadersof opinion"in the context of the conflict-people we suspectedof having a director indirectinfluence on the policy of the main actors. In other words, we were not interested in those with a high position but no well-articulatedpoint of view. Belonging to the sociopoliticalelite was not enough;we emphasizedthe person's reputation. Our choice was made from these categories:active politicians,civil servants in the foreignministry,membersof universityfaculties,andjournalistsandeditors. We employed two samplingstrategies.On the basis of our generalknowledge of the Arab world, we drew up a list of people we consideredinfluential.Here we consulted experts, includingthe NorwegianEmbassyin Egypt and Morocco, and the Swedish Embassyin Tunisia.Once in the field, we used a secondmethod.After we having intervieweda respondent, askedthatpersonfor namesof othersit might be useful to interviewandwho wouldbe willing to talkto us. This technique,called "chainselection,"was very helpfulfor gettingus appointments. the otherhand, On with this method,the personsrecommended probablybe those the respondent will likes or agrees with, which means that the sample becomes self-selecting. It was thereforeimportant combine the two methods. to

Attribution Theory and Arab Images of the Gulf War

279

Our interviews were conductedin Egypt and two Maghrebicountriesin the Desert Storm,but before world attentionshiftedto springof 1991, afterOperation the Kurdishinsurrectionand its aftermath. Thirty-two"elite"respondents,12 in Egypt, 12 in Morocco, and eight in Tunisia,were asked to accountfor the causes of the Gulf war and the motives of the antagonists.The interviewswere conducted in English or Frenchand translated English. See the Appendixfor a list of the into in of interviewquestionsand the distribution occupationsrepresented the sample. The results in the tables (shown in the Appendix) are meant as descriptive summaries-we arenot in any way generalizingfromsampleto universe.Samples that are neitherrandomnorrepresentative demand,of course, thatwe use the data with caution. Therefore, the results must, in the stringent scientific sense, be regardedas tentative. II. RESULTS The responses of our Arabelite samples were coded on the basis of the type of causes of Iraqiand Coalition behaviorthey cited, dispositionalor situational. The resultsof this analysisfor all respondents, well as a breakdown Egyptian as for and Moroccan/Tunisiansamples, are shown in Table I. A separate coding of whichdistinguishes betweenexplanations Western(U.S.A.) of Egyptian respondents, behavior and Egyptian behavior, is shown in Table II. In addition, Table III summarizesour respondents'views of SaddamHussein as a person in terms of positive and negative characteristics. Explanations of the Coalition's Behavior If we look at the resultsshown in TableI, we see that,generallyspeaking,they conform to the predictions of attributiontheory. We hypothesized that Arab explanationsof the Coalition's behaviorwould be largely situational,and this is the case; the interpretation the Coalition's actions is sought mainly in the of situation in which it found itself: 68% of statementswere situational,while only 32% were dispositional. In their explanationsof the Coalition's behavior in the Gulf war, our Arab respondents viewed the conflict largely from a realist perspective. Given that perspective,they were not surprisedto see the United States promotethe balance of power in the region throughits militarypresenceand the destructionof Iraq's nuclearpotential.For most of them,therewas nothingunusualaboutthis behavior because it is to be expected that a hegemonic power will assert itself to thwart challengers to system stability. From a realist viewpoint, these are clearly situational attributions.On the other hand, it would be out of characterfor Iraq to develop the ambitions of a hegemonic power by developing nuclear weapons.

280

Heradstveit Bonham and

to Iraq'sbehaviorwas contrary expectations.SaddamHussein was acting"toobig for his breeches!" The geopolitical ambitionsof the United States were highlightedby many of ourArabrespondents. Accordingto some, the mainmotive for the warwas not the liberationof KuwaitbutcrushingIraqas a nation,bothmilitarilyandeconomically. For the United States the primarygoal was to prevent Iraq from becoming an actorin the MiddleEast. Nor could the Gulf oil monarchiesacceptIraqi important hegemony.They were afraidof SaddamHussein and very pleasedto see an end to his ambitionsin the Gulf. Especiallythreatening Iraq'sabilityto makenuclear was weapons. The Coalition, especially the United States, wanted to fight a war that would destroyIraq'snuclearpotential. Some Arabrespondentsalso claimedthatthe Coalitionhad anotherimportant aim: to secure the oil resourcesof the Gulf. The war enabledthe United States to establish a direct military and political presence in an area of great economic significance. This was a vital interestfor the domestic Americaneconomy, and it was also necessaryif the United Stateswas to retainits hegemonyfor Europeand Japan. The war gave the U.S. importanteconomic agreements and contracts, including those for Kuwaiti reconstruction.Oil was the world's prime energy source, and the United Stateswas extremelyunwillingto allow a countrylike Iraq to controla largeproportion it, and,of course,oil has alwaysbeen a majormotive of for foreign interventionin the region. Although a large majority of Arab attributionsabout the Coalition were situational, some attributionspresentedthe United States as a unique actor in international politics-a powerthatactsdifferentlythanmost others.The argument is that Americansare naturallyarrogant thatthey do not toleratethe exercise and of authorityotherthantheirown. Americanswill not allow ThirdWorldcountries to be self-sufficient;they are hostile to any such countries,like Iraq,that reach a certainlevel. madefew attributions aboutGeorgeBush Surprisingly,ourArabrespondents as an individual,except that he needed to display strengthand leadership.Some also statedthatGeorgeBush would nottoleratea militarychallengefroma dictator. Explanations of Iraq's Behavior with respect to Generally,we also found the expected patternof attribution behavior;that is, explanationstended to be more dispositional than situIraq's ational. For all respondents64% of attributions were dispositional,comparedto 36% situational. Dispositional attributionsmade by the Arab elites in our sample were not confined to Saddam Hussein. Iraq was portrayedas a country that does not necessarilyact the same way as othercountries;for example,it was too ambitious. From a realist perspective,Iraq,by tryingto act like a superpower,was behaving

Attribution Theory and Arab Images of the Gulf War

281

countriesin the region.Behavior in a mannerthatwas not typicalof most "normal" like the United States, is for that might be "situational" a hegemonic superpower, for "dispositional" Iraq,a small countryin a regionalsetting. Some respondentsstatedthatIraqhas always had a warlikeculturewith very of SaddamHusseinwas reallyjust a product the Iraqiculture-he violent traditions. was merely living up to the ambitionsof Iraqas a nation.Iraqis,accordingto some are towardKuwait. of the respondents, also vindictive,whichexplainstheirattitude A warlike nationwith violent traditionsand a thirstfor revenge is likely to make such decisions as invadingKuwait.Respondentsalso emphasizedIraq'sideologithe cal orientation; Ba'ath Partyis utterlyopposedto the presentpolitical borders of the Arab world; it aims at Arab unity and intends to achieve this by military force. Most of the dispositional attributions,however, did refer to the person of SaddamHussein, who was perceivedin a very negativelight. For all respondents, of of 82% of the attributions the personalcharacteristics SaddamHussein were while only 8% were positive and 10%neutral(see Table III). negative, Westernargumentsand assertionsabout Saddam,the Many of the recurrent in theresponsesof theArabelites whomwe interviewed.There individual,reappear referto SaddamHussein is, however,one majordifference:noneof ourrespondents who was ungrateful all the help Americahad given," as an "ungracious for upstart which was a typical theme in the U.S. press (Goertzel, 1993, p. 722). SaddamHussein's behaviorwas attributed manyof ourrespondentsto his by as character an autocratic authoritarian and leaderwho hadno interestin democratic values and rarelyconsultedanyone. Otherssaw him as a coward,concernedonly with hanging onto power or "saving his own skin." They thought he was quite convinced that Iraqwas destinedto enjoy hegemony over the Arab Middle East. When they explainedhis behaviorduringthe war,the consensuswas thathe acted on impulse and improvised.Strategywas made on the run-he was an "Oriental rug merchant."Lacking any long-term strategy, Saddam Hussein persistently He miscalculated,both locally and internationally. attackedIsraelin the hope that countrieslike Syria and Egypt would switch sides andjoin him. His most serious misjudgment,however, was the belief that the United States had given him the green light to invade Kuwait.That SaddamHussein could misjudgethe situation to thatextent was explainedby his small sphereof contactwith the outside world. He isolated himself and did not consultthe rightpeople. Respondents who made situationalattributionsabout the behavior of Iraq pointed out that Iraq had heavy debts and was in conflict with Kuwait over oil werealso emphasized,andthe argument Iraqhad that prices.Historicalantecedents never acceptedthe borderwith Kuwaitwas repeatedby some. War was waged to correcta historicalinjusticefor which the colonial powers were responsible.Iraq had historicalrightsto Kuwait,which neverexisted as an independent emirate. thatSaddamHusseinwas drivenby narrownational Respondentsemphasized interests. Iraq was acting undercompulsion, because the other Gulf states were

282

Heradstveit and Bonham

for againstit. Kuwait'sdemandfor repayment loans was waging economic warfare Faced with this situation,and remembertakenby Iraqas an extremeprovocation. Kuwaitwas, Iraqhad no choice. Objectivelyspeaking,the ing how "stiff-necked" was obliged to "retreat advancing."In a way, SaddamHussein was by country into this decision, and,in this context,externalforces, like the United manipulated States, pushedhim into war. Egyptian and Maghrebi (Moroccan and Tunisian) Attributions Telhami's study of Arabpublic opinionandthe Gulf war suggests substantial differences in attitudeswithin the Arab world, depending on distance from the removedfrom the Gulf an Arabpolity was, the conflict. For example, "thefurther less botheredit was by the violation of Kuwaitisovereignty. . ." (Telhami, 1993, p. 194). of Similarly,furtheranalysis of the attributions the Arabelites in our sample reveals differencesbetween nationalsubsamples.Table I shows thatexplanations of the Coalition's behaviorby Maghrebirespondentswere much more situational (87%) than dispositional(13%). This result is highly consistent with researchon of differences.The attributions Egyptianrespondents,on the other actor-observer were much more evenly balancedbetween situational(56%) and disposihand, tional (44%). A differentpatternwas found for explanationsof Iraq'sbehaviorin of the Gulf war. As Table I shows, the attributions the Egyptianrespondentswere were disposiconsistent with the hypothesis:90% of Egyptianattributions very while only 10% were situational.For the respondentsfrom Morocco and tional, aboutIraq'sbehaviorwereevenly balancedbetween Tunisia,however,attributions situational(47%) and dispositional(53%). It should be noted, however, that the sample size for this analysis is small: 20 Maghrebi respondentsand only 12 Egyptianrespondents. III. DISCUSSION: IMPLICATIONS FOR ATTRIBUTION THEORY Although our interviews of Arab elites provided general support for the furtheranalysis differencesin attribution, findings of researchon actor-observer that were relatedto the nationalityof the responrevealed patternsof attribution dents. The Egyptian elites conformed to the hypothesis with respect to Iraq's behaviorbutnotthe Coalition'sbehavior,while theresultsforMoroccoandTunisia were consistentwith the hypothesisfor the Coalition'sbehaviorbut not for Iraq's these results? behavior.How should we interpret

Attribution Theory and Arab Images of the Gulf War

283

The Egyptian Respondents fromthe othertwo countries,were The Egyptianelites, unlikethe respondents unwilling to make statementsaboutthe behaviorof the Coalition,as such. Instead, betweenthe behaviorof theWest (U.S.) andEgyptin theconflict they differentiated with Iraq. For the Egyptians the motives of the Coalition members were quite different.The United States was not a friendin need; rather,the Americanswere of againstthe Arabs.The United States supporters Israel,and they were prejudiced had a "one-track mind,"namely,to extend its influencein the Arabworld.Egypt, on the other hand, was in the uncomfortableposition of being a member of a coalition against an Arab state, but this was a temporaryalliance, which was dictatedmore by circumstancethandeliberatechoice. To capture this viewpoint, we recoded Egyptian statementsto distinguish between attributions the West's (U.S.) behaviorand attributions the actions of of of Egypt. The results are shown in Table II. Explanationsof Egyptian behavior were much more situational (78%) than dispositional (22%), a result that is consistentwith the "fundamental attribution error" hypothesis.However,explanations of the West's behavior were almost equally balancedbetween situational (42%) and dispositional(58%). This resultsuggests thatapplicationsof attribution theoryoutsidea laboratory setting, especially those that involve collective actorslike nation-states,ought to take into accountthe discursivepracticesof the observers.Some of them, such as the respondentsfrom Morocco and Tunisia, may constructthe world in terms of unitaryactors. The Coalition, regardlessof its membership,was behaving like a single unit. Others,because of their psychological involvementas membersof a collective, like the Egyptians,may differentiate single actorswithinthe collective. As a consequence of such differences in the discursive spaces of observers,the attributional patternsthatemerge in theirexplanationsmay be quite distinctive. The Maghrebi Respondents Now let us look moreclosely at ourMaghrebi First,althoughthey respondents. may not respect SaddamHussein as an individual,this does not mean that no one empathized with him and the position of his country. Respondents saw many positive aspects of the ideas Hussein standsfor, such as standingup to the West, punishing Kuwait for its lack of will to negotiate over real grievances, drawing attentionto the Palestinianproblem,and so forth. This is reflected in Table III, which shows that25%of the attributions the personalcharacteristics Saddam of of Hussein by MoroccanandTunisianrespondentswere eitherpositive or neutral,as contrastedwith only 8% of the attributions the Egyptians. of There is some evidence in the attributionliteraturethat when observers empathizewith actors, they are no more likely to make dispositionalattributions

284

Heradstveit Bonham and

than the actors themselves (Weary et al., 1989, p. 37). For example, Regan and Totten (1975) found in a laboratory experimentthat"takingthe perspectiveof the that actor]throughempathyresultedin attributions were relativelymore target[an observers" situationaland less dispositionalthanattributions providedby standard In anotherexperiment,"instructionsto empathize led to dispositional (p. 850). attributionsfor success and situational attributionsfor failure, while standard observationinstructionsresultedin dispositionalcasual attributions regardlessof outcome" (Gould & Sigall, 1977, p. 480). If empathicobservers and actors are functionally equivalent, then the Maghrebi respondentswho empathized with Saddam Hussein and his people might have acted like Iraqis themselves and attributedthe plight of Iraq to situationalfactors, thus resulting in the mix of shown in Table I. situationaland dispositionalattributions attribution treatment "actors" "observers" of and does not do Second, theory's to the humanabilityto distinguishbetweena political leaderand his or her justice country, between personal characteristicsand policies, or between the aims of policies and the means employed to achieve them. People might consider that a leader acted as he or she did because of dispositionalor "personal" factorswhich were distasteful, while at the same time claiming that the countrywas in such a position thatsomethinghad to be done, if not exactly whatthe leaderactuallydid. OurMaghrebirespondentsmakea cleardistinctionbetweenSaddamHussein as a person,on the one hand,andthe politicalideas he standsfor, on the other.This results in their disliking both SaddamHussein and the policy of the Coalition.If the so, it is easy to understand Maghrebirespondents' emphasison both situational anddispositionalattributions. Theyfeel the needto explainthe logic of the situation thathas led Iraqintoconflictandwar,at the sametime as theyfeel a needto distance themselves from Saddam Hussein the man. Since the Maghrebi countries are relatively open societies, moreover,there seems to be a need to distance oneself from the allegedly "thuggish" Iraqipoliticalcultureas well. This distancingmotive may be seen in the context of our previousdiscussion of the "evaluative"aspect of social attribution. the Arab-Israelistudy (HeradIn were overwhelminglydispositionalwhen observingthe stveit, 1979), respondents when observingthe opponent's"good" behavior,andsituational opponent's"bad" behavior. This suggests a large number of dispositional attributionsby the MaghrebirespondentsconcerningSaddamand Iraqmight reflect a deep ambivalence as to whose interestswere really furthered the Gulf crisis. by OurMaghrebirespondents of impliedthatthe personalcharacteristics Saddam Hussein ("miscalculations") the Americansdown on his head,with helpedto bring unfortunate consequencesfor both the Gulf and the Arab world as a whole. His standingup to theWest was, in principle,a good thingandwas, therefore, explained situationally, but this praiseworthyself-assertion failed, and the causes of this failure are explained dispositionally.In short, SaddamHussein was a bad man tryingto do a good thing and failing, in part,throughhis own badness.

Attribution Theory and Arab Images of the Gulf War

285

There is, on the other hand, no comparablecognitive dissonanceor ambivalence when it comes to explainingthe Coalition'sbehavior.Nobody seems to feel a need to demonstratea mirror-image propositionthat George Bush was a good man doing a bad thing, and attributions thereforemore uniformlysituational. are The West does what is expected of it, whoever is presidentof the United States, and there is no need for complex special pleadingto overcome a sense of discomfort. What we may be seeing here is a need in our Maghrebirespondentsnot only to complainaboutWestern"imperialism," also to exculpatethe Arabworldfor but (so surprisinglyandunpredictably) mismanagedits affairs.Herethe villain having of the piece is not George Bush but SaddamHussein.In the context of explaining the catastropheof 1990-1991, Saddamis the real "opponent"; though,of course, in the widercontext of whatan Arabstatesmanshouldbe tryingto do, he is not the are Dispositionalattributions associatedwith responsibilopponentbutthe "hero." and Saddamis held responsiblefor fightingthe wrong war at the wrong time. ity, CONCLUSIONS Our interview study only partiallysupportsa crucial finding of researchon actor-observer differencesin social attribution. Althoughthe aggregatedexplanations of these Arabrespondentswere consistentwith the predictionsof the theory, the attributionsof Iraq's behavior by Egyptian elites and attributionsof the Coalition's behavior by Moroccanand Tunisianelites were not fully consistent with the hypothesis. In our view cognitive attribution theoryfalls short,in part,becauseit takes its from explanationsof an isolated "event."Events in international starting point politics, however, are generally a matterof sequences of events, or cumulative development over time, where the actor who startsa particular sequence (in our case Saddam)necessarily "steals a march"on other possible candidatesfor that honor. This does not mean that he has "caused"the whole development,whose roots may stretchback centuries. What, then, does it mean to "explainan event"?Respondents,who know the it historyof the regionbetterthanoutsiders,or at least have internalized more,may "takeaccountof events moreextendedin time"(Monson& Snyder,1977, p. 101). They attemptto "understand, categorize,andexplainactionsandevents by placing them in typical sequences of the kinds discussed in researchon scripts and story (Lamb& Lalljee, 1992, p. 31). understanding" Presentattribution becauseit does theoryfails to accountfor suchexplanations not "provide a sufficient analysis of the knowledge used nor of the cognitive processes involved in makingthe necessaryinferences"(Read, 1987, p. 288). To capturethe background knowledgethatpeopleutilizeto explainsocial andpolitical events, "we must investigate,in detail, people's scenariosand the knowledgethat

286

and Heradstveit Bonham

is used to constructthem"(Read, 1987, p. 295). Individualsand groups produce in "stories"aboutcomplex situations,which can be represented the form of causal event chains (Pennington& Hastie, 1986) and analyzed using techniques like cognitive mapping.4 The stories that individualsand groups produceare not merely exercises in cognitive mastery, however, because they also take into account an audience, whetheractualor potential.The literature self-servingbiases tells us thatpeople on of tailor their stories, based on theirunderstanding the audience,to be convincing and make themselves look good. People are not unbiased and value-neutral, especially when they feel emotionalabouta situation,such as in the conflict in the Gulf. also applies to the content of an attribution, The problemof value-neutrality which may look quitedifferentto the codersandthe personmakingthe attribution. We argued above that a statementabout John's not wanting to go to the soccer matchmay be coded as eithersituational(becauseof the violence) or dispositional (because he is afraid).In fact, of course, any explanationof John's behaviormust involve boththe situationandthe way his personaldispositioninteractswith it, but any particularinterview response may come out sounding like either one or the other, but not both. Two respondentsmight formulatewhat is (if "unpacked") actually the same explanationin two differentways, leading coders to think that This is particularly the one is attributing dispositionallyandthe othersituationally. when emotive languageis used: one man might say thatthe soccer stadium likely was full of hooligans (situational),and anotherthatJohn was "chicken"(disposithe tional).To code such attributions, coderneedsto focus on the discursivespaces of the respondents(which might be relatedto age, subculture,mood, and so on), as we pointedout above. Similarly,differencesin interpretation might arisefromthe culturalaffiliation of the coders themselves. A large proportion the attributions the Coalition's of of behaviorthat we coded as "situational" were formulatedin terms of the need for oil. Within our own theoreticalparadigmthis is an entirely"rational" motivation, but in this context-within the Arabcognitiveframework-one mightalso classify it as an attribution that carries a heavy load of blame. For a Westerner,"oil" is for shorthand a set of technologicalandeconomic propositions,whereasfor Arabs it may be shorthandfor Westernimperialexploitation.That is, a response along the lines of "because of the oil" sounds situational to Westerners,evoking a paradigmof rationaleconomic manin searchof utility,whereasit mightwell have been meantas a referenceto the propensity(disposition)of the West to go to other people's countriesand removetheirresourceswithouttheirconsent.The tendency of our Arab respondentsto explain the Coalition's behavior"situationally" need not, therefore,be takento indicatesympathyor unwillingnessto apportionblame.
4 Cognitivemappingis described Axelrod(1976), Shapiro in andBonham(1973), Bonhamet al. (1976),

Bonham et al. (1979), Bonham et al. (1978), and Bonham and Shapiro(1986). For an example of attribution researchthatmakes use of cognitive mapping,see Heradstveit Bonham(1986). and

Attribution TheoryandArabImagesof the GulfWar

287

Attributionswhich, on the surface, seem to refer merely to features of the of situationcan thuscarryundertones or associationsto an (evil) disposition.These undertonescan be communicatedeffectively within a given culture but are not always "heard"across cultural boundaries.This is another indication that the theoreticalparadigmon which cognitive attribution theoryrests is too simple and needs to be stretchedby other perspectives (Shapiro et al., 1988). Ratherthan thinkingabouthow individualsconstrueanevent or situation,psychologically,one that representcollective reprecan treat construals as the pre-understandings "obviates et al., 1992). This approach sentationsof a cultureor subculture (Bonham the need for recourse to the interiorityof a conscious, meaning-giving subject, either in terms of psychological and cognitive characteristicsof individuals or shared mental templatesof social collectives" (Doty, 1993, p. 302). Instead,the source of meaning in a particularsituationmay reside, in part, in the linguistic or practicesthat are common to a group,organization, culture. Future research within this linguistic tradition would "attend less to the referentialfunction of language performancesthan to the culture of self-understanding, the sharedmeaning commitments. . . that are institutionalizedin language practices"(Bonham& Shapiro,1997). Within a language-basedontology, researchwould focus on the historically(and culturally)developed practicesthat reside in the grammatical,rhetorical,and narrativestructurethat comprises discourse on events like the Gulf war. To study these practices,one might have to view of languagefavoredby objectivistsand emphasize evade the object-oriented the constitutivedimensionof languageas a discursivepractice.

Acknowledgments This is a substantiallyrevised version of a paperpresentedat the Sixteenth Annual Scientific Meeting of the International Society of Political Psychology, 6-11 July 1993. This researchwas supportedin part Cambridge,Massachusetts, by the Norwegian Oil IndustryAssociation and a grantfrom the NationalScience Foundation(DIR-9113599) to the MershonCenterResearchTrainingGroupon the Role of Cognition in Collective Political Decision Making at The Ohio State University.The authorswould like to thankReid Hastie and JerelRosati for their comments and suggestions, and FredericR. Fournierfor his researchassistance.

288

Heradstveit Bonham and

Appendix Interview Questions 1. According to your opinion, what were the principalaspects of the crisis and the events unfoldingin the PersianGulf since August 2nd of last year? 2. What were the causes of the crisis? 3. Accordingto your opinion, why do you thinkthe U.S. went to war? 4. Accordingto your opinion,whatdo you thinkwill happennext? Do you think thata similarcrisis could occur in the Gulf ? Can you specify in what way? 5. Do you think that the crisis in the Gulf has influenced the attitude of the Americanstowardthe Arabs,and, on the otherhand,the attitudeof the Arabs towardthe U.S.? If this is the case, why? 6. Do you thinkthatthe U.S. could have acted otherwiseagainstIraq? 7. What are the reasonsfor the reactionsof Iraq? 8. What do you think would have happened,if the U.S. had pursueda tougher line towardIraq? 9. If the U.S. had pursueda softer line towardsIraq,what do you think would have happened? 10. What should be the policies of the West in similarsituations? 11. Do you think the West could improve its relationswith the Middle East? In what way? Whatcould lead to a worseningof the relationsbetween the West and the Middle East? 12. Could you indicate more precisely what Western policy ought to be in the Middle East? The Occupational Backgrounds of the Respondents In the sampleof 38 respondents,13 were universityfacultymembers,10 were editors or journalists,nine were active politicians,and six were civil servants.

Attribution Theory and Arab Images of the Gulf War of Table I. Explanations the Behaviorof Iraqand the Coalition in the Gulf Conflict/War. = numberof statements). (n of Explanations the Coalition'sBehavior Situational All Respondents Morocco/Tunisia Egypt 68% (n= 133) 87 (n = 53) 56 (n = 80) 44 10 (n = 42) 13 47 (n = 85) 90 Dispositional 32% of Explanations Iraq'sBehavior Situational 36% (n= 142) 53

289

Dispositional 64%

Table II. EgyptianExplanations the Behaviorof the West (U.S.) of and Egypt. (n = numberof statements). of Explanations the West's (U.S.A.) Behavior Situational Egypt 42% (n = 48) Dispositional 58% of Explanations Egypt's Behavior Situational 78% (n = 32) Dispositional 22%

Table III. Attributions PersonalCharacteristics SaddamHussein. of of (n = numberof statements). Positive 8% 14 0 Neutral 10% 11 8 Negative 82% 74 92

All Respondents (n = 60) Morocco/Tunisia (n = 33) Egypt (n = 25)

290

Heradstveit and Bonham

REFERENCES
Antaki, C. (1985). Attributionand evaluation in ordinaryexplanationsof voting intention. British Journalof Social Psychology,24, 141-152. Axelrod, R. (1976). Thestructureof decision. Princeton,N.J.: PrincetonUniversityPress. and Bodenhausen,G. V. & R. S. Wyer,Jr. (1985). Effects of stereotypeson decision-making information-processingstrategies.Journalof Personalityand Social Psychology,48, 267-282. O. Bonham,G. M., D. Heradstveit, Narvesen,& M. J. Shapiro(1978). A cognitive model of decisionand Conflict,13, 77-92. making:applicationto Norwegianoil policy. Cooperation Bonham, G. M., V. M. Sergeev, & P. Parshin (1992). Building a shared reality in international for the negotiation: limitedtest-banagreement. Prepared deliveryatthe FifteenthAnnualScientific Society of PoliticalPsychology, San Francisco,4-8 July 1992. Meeting of the International of Bonham,G. M. & M. J. Shapiro(1986). Mappingstructures thought.In I. N. Gallhofer,W.E. Saris & M. Melman,(Eds.),Differenttextanalysisprocedures thestudyofpoliticaldecision-making. for Amsterdam: SociometricResearchFoundation. the Bonham,G. M. & M. J. Shapiro(1997). Representing discoursesof nationalism.In D. Sylvan and in J. Voss (Eds.) Problemrepresentation politicaldecision-making, forthcoming. Bonham, G. M., M. J. Shapiro& G. J. Nozicka (1976). A cognitive process model of foreign policy decision-making.Simulationand Games,7, 123-152. toward Bonham,G. M., M. J. Shapiro,& T. L. Trumble (1979). The Octoberwar.Changesin orientation the Middle East conflict. International StudiesQuarterly,23, 3-44. of Bradley, G. W. (1978). Self-serving biases in the attribution process: a re-examination the fact or fiction question.Journalof Personalityand Social Psychology,36, 56-71. Cheng, P. W. & L. R. Novick (1990). A probabilisticcontrastmodel of causal induction.Journal of Personalityand Social Psychology,58, 545-567. a Doty, R. L. (1993). Foreignpolicy as social construction: post-positivistanalysis of U.S. counterinStudiesQuarterly,37, 297-320. surgencypolicy in the Philippines.International Fisk, S. T. & S.E. Taylor(1984). Social cognition.Reading,MA: Addison-Wesley. Freedman,L. & E. Karsh(1993). The Gulfconflict 1990-1991. Princeton:PrincetonUniversityPress. Goertzel, T. (1993). The Gulf war as a mental disorder?A statisticaltest of DeMause's hypothesis. Political Psychology, 14, 711-724. an Gould, R. & H. Sigall (1977). The effects of empathyand outcome on attribution: examinationof the divergent-perspectives Social Psychology, 13, 480-491. hypothesis.Journalof Experimental Haseeb, D. & M. S. Rouchdy(1991). Egypt's speculationsin the Gulf crisis:the government'spolicies and the oppositionmovements.In H. Bresheeth& N. Yuval-Davis (Eds.) The Gulfwar and the new world order. London:Zed Books. Heider,F. (1958). Thepsychology of interpersonalrelations.New York:Wiley. Heradstveit,D. (1979). TheArab-Israeliconflict:psychological obstacles to peace. Oslo: Norwegian UniversityPress. attributions of Heradstveit,D. & G. M. Bonham (1986). Decision-makingin the face of uncertainty: Norwegianand Americanofficials. Journalof Peace Research,23, 339-356. Hewstone, M. (1989). Causalattribution; from cognitiveprocesses to collectivebeliefs. London:Basil Blackwell. and in Hewstone, M. & C. Ward(1985). Ethnocentricism causal attribution SoutheastAsia. Journalof Personalityand Social Psychology,48, 614-623. and of Journalof Experimental Howard,J. A. (1987). The conceptualization measurement attributions. Social Psychology, 23, 32-58. of Ichheiser,G. (1943). Misinterpretations personalityin everydaylife andthe psychologist's frameof reference.Characterand Personality,12, 145-160.

Attribution Theory and Arab Images of the Gulf War

291

Jervis, R. (1976). Perception and misperceptionin international politics. Princeton,N. J.: Princeton UniversityPress. Jones, E. E. & R. E. Nisbett (1972). The actor and the observer:divergentperceptionsof the causes of behavior.Morristown,N. J.: GeneralLearningPress. Khalidi, W. (1991). Why some Arabs supportSaddam.In M.L. Sifry & C. Cerf (Eds.) The Gulf war reader. New York:Times Books. Khouri,R. G. (1991). The bitterfruitsof war.In M.L. Sifry & C. Cerf(Eds.) TheGulfwar reader.New York:Times Books. accounts. Lamb, R. & M. Lalljee (1992). The use of prototypical explanationsin first-andthird-person In M. L. McLaughlin,M. J. Cody, & S. Read(Eds.) Explainingone's self to others: reason-giving in a social context.Hillsdale,N.J.:LawrenceErlbaumAssociates. Larson, D. W. (1985). Origins of containment:a psychological explanation. Princeton:Princeton UniversityPress. in Lau, R. R. & D. Russell (1980). Attributions the sports pages. Journal of Personality and Social Psychology, 39, 29-38. Lerner,M. J. & D. T. Miller (1978). Justworld researchand the attribution process:looking back and ahead.Psychological Bulletin,85, 1030-1051. for Miller, D. T. (1976). Ego-involvementand attributions success and failure.Journalof Personality and Social Psychology, 34, 901-906. Miller, D. T., S. A. Norman,& E. Wright(1978). Distortionin personperceptionas a consequenceof the need for effective control.Journalof Personalityand Social Psychology, 36, 598-607. of Miller, F. D., E. R. Smith, & J. Uleman (1981). Measurementand interpretation situationaland Social Psychology, 17, 80-95. Journalof Experimental dispositionalattributions. Miller, J. (1984). Cultureand the developmentof everydaysocial explanation.Journal of Personality and Social Psychology,46, 961-978. Monson, T. C. & M. Snyder (1977). Actors, observers, and the attributionprocess. Journal of Social Psychology, 13, 89-111. Experimental Nisbett, R. & L. Ross (1980). Human inference: strategies and shortcomingsof social judgment. Englewood Cliffs, N. J.: Prentice-Hall. Nisbett, R. & T. D. Wilson (1977). Telling morethanwe can know:verbalreportson mentalprocesses. Psychological Review,84, 231-259. Olson, J. M. & M. Ross (1985). Attributionresearch:past contributions,currenttrends, and future basic issues and applications.Orlando: prospects.In J. H. Harvey& G. Weary(Eds.)Attribution; Academic Press. conflict in young couples. In J. H. Harvey, Orvis, B. R., H. H. Kelley & D. Butler(1976). Attributional W. J. Ickes & R. F. Kedd (Eds). New directions in attributionresearch, 1, Hillsdale, N. J.: L. ErlbaumAssociates. Pennington, N. & R. Hastie (1986). Evidence evaluation in complex decision making. Journal of Personalityand Social Psychology,51, 242-258. G. and when behaviorengulfs the person.Journal Quattrone, A. (1982). Overattribution unitformation: of Personalityand Social Psychology,42, 593-607. causal scenarios:a knowledge structure to Read, S. J. (1987). Constructing approach causal reasoning. Journal of Personalityand Social Psychology,52, 288-302. Regan, D. T. & J. Totten (1975). Empathyand attribution: turningobserversinto actors. Journal of Personalityand Social Psychology,32, 850-856. and Renshon,S. A. (1993). Goodjudgment, the lackthereof,in the Gulfwar:apreliminary psychological model with some Applications.In S. A. Renshon(Ed.) Thepolitical psychology qf the Gulfwar. Press. Pittsburgh: Universityof Pittsburgh

292

Heradstveit and Bonham

distortionsin the attribution Ross, L. (1977). The intuitivepsychologistand his shortcomings: process. In L. Berkowitz (Ed.) Advances in experimentalsocial psychology, 10, New York: Academic Press. Ross, L. & R. E. Nisbett (1991). The person and the situation;perspectives of social psychology. Philadelphia: Temple UP. Seddon, D. (1991). Politics and the Gulf crisis: governmentand popularresponsesin the Maghreb.In H. Bresheeth& N. Yuval-Davis(Eds.)TheGulfwarand thenewworldorder.London:Zed Books. Shapiro, M. J. & G. M. Bonham (1973). Cognitive process and foreign policy decision making. International StudiesQuarterly,17, 147-174. Shapiro,M. J., G. M. Bonham& D. Heradstveit(1988). A discursivepracticesapproachto collective StudiesQuarterly,32, 397-420. decision-making.International in and Taylor, D. M. & V. Jaggi (1974). Ethnocentricism causal attribution a South Indiancontext. Journal of Cross-Cultural Psychology,5, 162-171. of affect, and Taylor,S. E. & J. H. Koivumaki(1976). The perception self andothers;acquaintanceship, differences.Journalof Personalityand Social Psychology, 33, 403-408. actor-observer Telhami, S. (1993). Arab public opinion and the Gulf war. In S. A. Renshon, (Ed.) The political Press. psychology of the Gulfwar. Pittsburgh: Universityof Pittsburgh Tessler, M. (1991). Anger and Governance in the Arab World: Lessons from the Maghreb and Relations,13, 7-33. implicationsfor the West. TheJerusalemJournalof International P. E. & A. Levi (1982). Attribution bias: on the inconclusivenessof the cognition-motivation Tetlock, debate.Journalof Experimental Social Psychology, 18, 68-88. Triandis,H. (1993). Collectivismand individualismas culturalsyndromes.Cross-Cultural Research, 27, 155-180. Van der Plight, J. (1981). Actors' and observers' explanations:divergentperspectives or divergent evaluations?In C. Antaki,(Ed.) Psychologyof ordinaryexplanationof social behavior.London: Academic Press. Y. Vertzberger, (1990). Theworld in theirminds.Information processing, cognition,andperceptionin foreign policy decision making.Stanford:StanfordUniversityPress. New York:SpringerVerlag. Weary, G., M. A. Stanley, & J. H. Harvey(1989). Attribution. distinctionin causal attribution. Journal of White, P. A. (1991). Ambiguity in the internal/external Social Psychology,27, 259-270. Experimental with SaddamHussein-updated. In H. H. Blumberg& C. C. French White, R. K. (1994). Empathizing (Eds.) The Persian Gulf war: Viewsfrom the social and behavioral sciences. Lanham,MD: UniversityPress of America. M. of bias is alive and Zuckerman, (1979). Attribution success andfailurerevisited,or the motivational well in attribution theory.Journalof Personality,47, 245-281.