R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–6054
1. Introduction and background
On 13 November 2002, a Bahamian registered oil tanker,the ‘Prestige’, containing 77,000 t of crude oil was severelydamaged in a storm 250 kms off the Galician coast in North-West Spain. The ship split in two, sank and spilt thousands oftons of heavy and toxic oil, much of which was blown land-wards arriving on the Galician beaches as well as the coasts ofnorthern Spain and France. Having sunk, the tanker continuedto discharge large quantities of oil for months afterwards. Thesinking of the ‘Prestige’ is generally acknowledged to havebeen the most serious ecological disaster to have affected Eur-ope. While the environmental damage was immediate, the so-cial and psychological impacts on the population are still beingfelt 1 year later and in some cases are now only becomingmanifest.
The oil spill had a significant effect, not only from an eco-logical and human point of view, but also on the economy ofthe region. Galicia is highly dependant on the sea because fish-ing and fishing-related activities are a vital part of the GDP ofthe region. Furthermore, the coast is also important for tourism.Government figures indicate that tourism in Galicia had recov-ered by Easter 2003 to 80% of the level of the previous year;however, some organizations maintain their distrust in respectof this figure. The fact that the number of visitors and touristsdecreased, the more general impact on the perception of thearea as a potential holiday destination (the main percentage ofvisitors come from other parts of Spain) as well as the impactof the oil spill on the catching and selling of seafood, all con-tributed to the concerns of the inhabitants of Galicia. The scaleof its ecological impact was demonstrated by Carlota Viada,Director of the Conservation Department at SEO/BirdLife,who made a conservative estimate that the 23,000 birds col-lected in Spain, France and Portugal only comprise 10–20%of the birds affected by the Prestige disaster (RSPB, 2003).
Galicia has been the victim of previous oil tanker accidents(e.g. Urquiola, Aegean Sea, Casón, among others), but the so-cial and political response was considerably more muted onthis occasion. One key aspect of the social response comesfrom both the scarcity of information, and the unreliability ofavailable information. From the time the tanker was in troubleand approached the Galician coast looking for refuge, until itfinally sunk in the ocean, neither the national nor regional gov-ernment nor the ship owners or insurers provided informationon the risk to the population and/or the ecosystem. The im-mediate response of both the national and the regional Govern-ments from the outset was focused entirely on playing downthe level of the risk involved, denying the possibility of any oilspill, and underestimating the importance of the magnitude ofthe tragedy. It was announced that everything was under con-trol, and the public should have no cause for concern. Thisstrategy was maintained even in the face of evidence to thecontrary, supplied for example by the mass media. This re-sulted in a loss of credibility in the government as well as otherofficial institutions. The only scientific information available tothe local population about the nature of the contaminationcoming from Prestige oil tanker or its associated health effects
on citizens came from non-governmental organizations such asecological groups, the mass-media, and several French andPortuguese research institutions. The Spanish National Scienti-fic Research Center (CSIC) as well as some Universities alsoproduced reports contradicting the Government’s analysis ofthe potential risk and impact of the capsizing of the oil tanker.All these reports were ignored or openly criticized.
As time went by, it became clear to the general public thatthe information provided by both the national and regionalgovernments was unreliable, and that there were real ecologicaland economic threats to the local community. This was laterconfirmed with the breaking up and sinking of the tanker whileit was being towed away from the coast, a decision which waswidely criticized by experts because it could only extend thedamage to a wider area; this indeed happened. At this stagesocial action was initiated by the community both in terms oftaking practical action to mitigate the effects of environmentaldamage, but also political action in the form of public protestagainst government complacency. Several citizens’ organiza-tions as well as members of local private companies and afew co-coordinators from different non-governmental organiza-tions assumed the management of the crisis along the entirestretch of the Galician coast. The most striking example of col-lective action was the presence of thousands of people whovolunteered to clean up the coast by hand (i.e. with the mostbasic and inadequate tools and without appropriate safety pro-tection) and who received neither institutional nor financialsupport.
Social pressure and protest, with the collaboration of themass media, not only forced the Government to take an interestin organizing volunteers (albeit 3 months later), but also toapprove an emergency plan to create new infrastructures anddevelop projects for the benefit of the social and economic de-velopment of Galicia such as high speed rail improvements,motorways, etc. Furthermore, most fishermen in the areas weresubsidized by the government. Both measures were partiallyeffective in muting the protests, as was confirmed by the localelections held shortly afterwards. Whereas there was a strongpolitical reaction in most Galician cities (where the govern-mental party lost political control of all councils, with onlyone exception), such a reaction did not happened in the ef-fected rural areas where all the councils were retained.
When we speak about a toxic contamination event there arethree stages through which a risk situation passes: a) non-issue,b) public issue, and c) political issue (Reich, 1991; McGee,1999). Reich defined non-issue as the phase prior to the publicidentification of the contamination agent. In the public issuephase the situation moves from being a disaster at an individuallevel to a disaster at a group level; victims try to organize andexpand the scope of collective action. The disaster moves frombeing a public issue to a political issue when it gets into thepolitical domain, involving other governmental or non-govern-mental organizations, political parties, social movements, andthe mass media. The Prestige case moved very quickly intothis third phase. More importantly, a year later, it still remainsthere.
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–60 55
2. Objectives
Following studies by Pol and Moreno (2002) on social im-pact evaluation, San Juan (2001) on the psychology of emer-gencies, and Uzzell et al. (2002) on environmental evaluation,it was felt appropriate to evaluate the social impact of the Pres-tige disaster on the population immediately after it happenedand a year later. We were also interested in examining the dif-ferential perceptions and evaluations of the inhabitants and thevolunteers to the developing environmental and political situa-tion. A distinction was made between responses from inhabi-tants and those from volunteers, on the basis that volunteersdiffer from inhabitants in many respects (i.e. higher commit-ment, better information about the situation, more objectiveview, not economically affected). Volunteers’ opinions couldonly be gathered in the first wave of interviews, because thecleaning work had finished long before the second wave ofinterviews was carried out.
The study seeks to assess the changes in attitudes experi-enced by the population in respect of:
● the degree to which the inhabitants and volunteers were af-fected by the catastrophe;
● the inhabitants’ and volunteers’ understanding of what hap-pened;
● the inhabitants’ and volunteers’ attribution of responsibility;● the inhabitants’ and volunteers’ assessment of consequencesof the disaster;
● the inhabitants’ and volunteers’ assessment of the credibilityand degree of satisfaction with political institutions, organi-zations, and the media.
3. Method
3.1. Sample and instruments
A random sample of 1491 respondents (51.2% males and48.8% females) was surveyed in December 2002, 1 monthafter the disaster. Of these, 1246 were inhabitants in the ef-fected areas, and 245 were volunteers from other parts ofSpain. The second wave of interviews, carried out in December2003 and January 2004, comprised another random sample of1504 subjects (49.5% males and 50.5% females), all of whomwere inhabitants in the effected areas. For both studies, thereliability was equal to 99.7% and the sample error was lessthan 4%. The criteria for sampling were that respondentsshould normally be resident in the Autonomous Communityof Galicia and aged over 18 years old. We used a social impactevaluation protocol specially designed for this study which in-cluded information on:
● socio-demographical variables;● perception and understanding of the impact and duration ofthe threat;
● the evaluation of received information;● trust and credibility of the regional and national govern-ment, media and other organizations;
● attribution of causes and responsibilities;● perception of consequences;● degree of satisfaction with response given to the disaster bydifferent public and private organizations;
● evaluation of the efficiency of response and remediation ac-tions by the regional and national government, and others;
● impact on health;● implications for changes in behavior.
The questions were of both a closed and open response for-mat; some of the closed format questions required categoricalresponses (e.g. attribution of responsibility) while others reliedon the completion of Likert-type rating scales (e.g. estimationsof the differential impact of the disaster). All interviewers werefully trained prior to conducting the face-to-face interviews.
4. Results
4.1. Degree affected, understanding of the problemand attribution of responsibility
The analysis of the responses given by the volunteers andthe inhabitants at the different times showed a marked shift inthe responses given by the latter. The inhabitants’ responses in2002 were quite similar to those of the volunteers; 84% de-clared themselves to be quite affected or very affected by thedisaster (volunteers: 90%), 75% had a good or perfect under-standing of the problem (volunteers: 81%), and 83% attributedresponsibility for the disaster to the regional or national gov-ernments (volunteers: 93%). One year later, only 55% of theinhabitants declared themselves to be quite or very affected,only 69% said that they had a good or perfect understandingof the problem (when the circumstances of the accident hadalready been thoroughly studied), and only 45% attributed re-sponsibility for the disaster to the regional or national govern-ments.
In order to test whether the differences between the twogroups were significant, a hierarchical loglinear analysis wasperformed (Knoke and Burke, 1990; Powers and Xie, 2000).Four categorical variables were included in the analysis: degreeof impact (EFFECT), understanding of the problem (UNDER-STAND), attribution of responsibility (RESPONSIBILITY),and sample interviewed (POPULATION). The results showedno fourth or third order significant effects for the variables in-cluded. The best fitting model (chi square = 453.83; df = 480;P = 0.799; see Table 1) included six second order highly sig-nificant effects, and all four first order effects. Three of thesecond order effects (EFFECT * POPULATION, UNDER-STAND * POPULATION and RESPONSIBILITY * POPU-LATION) correspond to interactions between assessmentsmade about the disaster and the type of sample interviewed,thus showing significant differences in the responses made byvolunteers, inhabitants at 2002 and inhabitants at 2003. Withrespect to the remaining three effects, the first of them (EF-FECT * UNDERSTAND) revealed that a lower degree of im-pact was associated with also a lower degree of understanding,and vice versa. The second effect (EFFECT * RESPONSIBIL-
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–6056
ITY) revealed that a lower degree of impact was associatedwith non-political responsibilities (nobody in particular, fate,the captain of the oil tanker, the harbor technicians, the shipowner), whereas a higher degree of impact was associated withresponsibilities at the national and regional levels. The thirdsignificant effect (UNDERSTAND * RESPONSIBILITY)showed that a lower degree of understanding was also asso-ciated with non-political responsibilities, whereas a higher de-gree of understanding was associated with political responsibil-ities, both at national and regional levels.
In summary, it seems clear that, a year later inhabitants ofthe effected areas seemed much less concerned about the cir-cumstances and consequences of the disaster. Even if the re-sults of the cleaning performed during 2003 had lowered thedegree of impact on the local population, they cannot explainthe reduced degree of understanding and reduced attribution ofpolitical responsibility. A year later, and after a thorough inves-tigation of the disaster, both the degree of understanding andthe attribution of political responsibility should have beenhigher, not lower.
4.2. Assessment of the consequences, credibilityand satisfaction issues
Given the differences between the samples found above, itwas expected that such differences would also be found in theassessments that subjects made about the consequences of thedisaster, as well as their assessment of the credibility and de-gree of satisfaction with political institutions, organizations andthe media. To test the significance of these differences, three
Table 1Best fitting hierarchical loglinear model for the four-way table (EFFECT * UNDER
Effect dfUNDERSTAND * POPULATION 8
EFFECT * POPULATION 8
RESPONSIBILITY * POPULATION 14
EFFECT * UNDERSTANDING 16
EFFECT * RESPONSIBILITY 28
UNDERSTAND * RESPONSIBILITY 28
Likelihood-ratio chi square = 453.82747; df = 480; P = 0.799.
Table 2Analysis of variance for the different groups (volunteers 2002, inhabitants 2002 andconsequences of the disaster
Dependent (consequences) Sum of squares df Mean sqEconomical 592,107 2 296,053
Ecological 86,496 2 43,248
Social 446,465 2 223,232
Psychological 402,191 2 201,096
Employment 687,330 2 343,665
Emigration 864,344 2 432,172
Delinquency 340,920 2 170,460
Community life 208,973 2 104,486
Tourism 419,606 2 209,803
Image of Galicia in Spain 109,532 2 54,766
Image of Galicia in the world 112,762 2 56,381
multivariate analyses of variance were performed for thesethree sets of assessments.
With respect to the assessment of the consequences of thedisaster, a significant multivariate effect for the type of sample(volunteers 2002, inhabitants 2002 and inhabitants 2003) wasobtained (Wilk’s lambda = 0.743; F = 83.413; df hyp. = 22; dferror = 5438; P < .001). At the univariate level, significant dif-ferences were also found for each of the consequences assessed(see Table 2). An inspection of the mean scores for the threesamples revealed that the assessments made by the 2003 waveof the inhabitants’ sample had the lowest mean scores, whereasassessments made by volunteers in 2002 had the highestmeans. Additionally, paired comparison tests (not reportedhere) showed significant differences between the assessmentsmade by the 2003 sample (inhabitants) and the two 2002 (in-habitants and volunteers) samples. Thus, it can be seen that, ayear later the inhabitants of the effected areas tended to mini-mize the impact of the disaster.
Another significant multivariate effect (Wilk’s lamb-da = 0.898; F = 10.257; df hyp. = 26; df error = 4830;P < 0.001) was found for the type of sample when assessingthe credibility of political institutions, community organiza-tions and the media. At the univariate level though, no signifi-cant differences were found for the assessments of the credibil-ity of local politicians (all below the mid-point of the scale),fishermen associations (all high), a private TV channel (allhigh) and family and friends (all high, see Table 3). A furtherinspection of the mean credibility scores for the three groupsrevealed that the highest means were invariably a feature of the2003 population, and the lowest means were a consistent re-
STAND * RESPONSIBILITY * POPULATION)
L-R chi square change P Iteration31,287 0.0001 5
314,405 0.0000 4
143,627 0.0000 5
184,599 0.0000 6
126,020 0.0000 5
101,894 0.0000 5
inhabitants 2003). Univariate tests for the attribution of the seriousness of the
uare F P Eta square Power151,945 0.000 0.100 1000
64,547 0.000 0.045 1000
174,819 0.000 0.114 1000
84,696 0.000 0.058 1000
287,260 0.000 0.174 1000
275,950 0.000 0.168 1000
134,326 0.000 0.090 1000
79,253 0.000 0.055 1000
140,958 0.000 0.094 1000
31,560 0.000 0.023 1000
30,748 0.000 0.022 1000
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–60 57
sponse of the 2002 volunteers; with respect to the formergroup, the means were always close to the mid-point of thescale. Paired comparison tests (not reported here) showed that,with the four aforementioned exceptions, there were alwayssignificant differences between, at least, both groups.
These results reveal the agreement between groups when thecredibility was already high (fishermen associations, T5 TV,and family/friends) or neutral (local politicians) in 2002. Dif-ferences only appeared when the source of credibility wasclearly low. In these cases, the 2003 respondents tended tokeep their assessments as neutral as possible.
Finally, another significant multivariate effect (Wilk’s lamb-da = 0.776; F = 29.747; df hyp. = 22; df error = 4848;P < 0.001) was found for the three samples interviewed whenassessing their degree of satisfaction with political institutionsand other organizations. The differences were also significantat the univariate level for all the items assessed (see Table 4).However, paired comparisons tests (not reported here) revealedthat the pattern of the differences was not the same for all as-sessments. When satisfaction was related to political institu-tions, the pattern was as expected (with the exception of localpoliticians): the highest mean scores for 2003 inhabitants, andlowest mean scores for 2002 volunteers. But when assessingsatisfaction levels with other organizations (fishermen associa-
Table 3Analysis of variance for the different groups (volunteers 2002, inhabitants 2002institutions, community organizations, and the media
Dependent (credibility) Sum of squares df Mean squareLocal politicians 7,693 2 3847
Regional politicians 154,107 2 77,054
National politicians 173,632 2 86,816
Fishermen associations 7,645 2 3823
Local press 22,487 2 11,243
National press 62,063 2 31,032
Public national TV 118,306 2 59,153
Public regional TV 217,239 2 108,619
A3 TV 118,479 2 59,240
T5 TV 1,833 2 916
Public radio stations 67,690 2 33,845
Private radio stations 27,108 2 13,554
Friends and family 4203 2 2102
Table 4Analysis of variance for the different groups (volunteers 2002, inhabitants 2002 apolitical institutions and other organizations
Dependent (satisfaction with role played by…) Sum of squares dfEU 107,762 2
National government 135,127 2
Regional government 150,816 2
Local government 8455 2
Fishermen associations 51,855 2
Volunteers 10,274 2
Ecologist organizations 9091 2
Ship owners 87,071 2
Local companies 121,434 2
Neighbors 366,528 2
tions, ecologist associations, volunteers, local companies, andneighbors), the highest mean scores corresponded to 2002 vo-lunteers and, in most cases, the lowest mean scores corre-sponded to 2003 inhabitants. In almost all cases, 2002 volun-teers and 2003 inhabitants were at opposite poles in theseassessments.
Again, we can see a tendency of 2003 inhabitants to try and“smooth” their assessments, thus locating far from both thehighest and lowest scored items for the 2002 wave of inter-views. It should be noted that univariate effects for single de-pendent variables were tested after checking that there was amultivariate effect (i.e. for all dependent variables simulta-neously). Although we found some high correlations (e.g. be-tween assessments for regional and national politicians), thesewere not sufficiently high as to imply the risk of multicollinear-ity.
5. Discussion and conclusion
It is not always possible to organize a collective response toa disaster. While some actions are evidence of an organiza-tional response to environmental contamination (cf. McGee,1999; Abbot Wade, 1991; Brown and Mikkelsen, 1990; Bul-lard, 1990; Edelstein, 1988; Erikson, 1990; Levine, 1982),others exemplify the difficulties of organizing social opposition
and inhabitants 2003). Univariate tests for credibility of different political
F P Eta square Power2397 0.091 0.002 486
56,536 0.000 0.045 1000
67,177 0.000 0.052 1000
2812 0.060 0.002 554
8662 0.000 0.007 969
25,430 0.000 0.021 1000
37,189 0.000 0.030 1000
61,740 0.000 0.048 1000
40,825 0.000 0.033 1000
723 0.486 0.001 173
22,995 0.000 0.019 1000
10,580 0.000 0.009 989
1796 0.166 0.001 377
nd inhabitants 2003). Univariate tests for satisfaction with the role played by
Mean square F P Eta square Power53,881 55,092 0.000 0.043 1000
67,563 58,711 0.000 0.046 1000
75,408 61,508 0.000 0.048 1000
4,227 3111 0.045 0.003 600
25,927 20,636 0.000 0.017 1000
5137 11,987 0.000 0.010 995
4546 3658 0.026 0.003 676
43,535 30,055 0.000 0.024 1000
60,717 48,316 0.000 0.038 1000
183,264 122,988 0.000 0.092 1000
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–6058
(Brown and Mikkelsen, 1990; Couch and Kroll, 1994; Wi-saeth, 1994; McGee, 1999). In the case of the Prestige disaster,the social response was organized quickly and spontaneously.The role played by volunteers was decisive, not only in termsof coping with the consequences of the disaster, but also inobtaining a response from both institutions and the govern-ment.
Another important aspect in the environmental managementof a crisis is the citizens’ credibility and trust. This is an essen-tial requirement in order to carry out the effective managementof hazardous waste sites (Williams et al., 1999). According toWilliams et al., if there is no trust, it will be very difficult forGovernments to persuade citizens in a convincing way thatplaces are safe and can be used again, in this case for providingthe authority to encourage and permit fishing again. In thePrestige case, the lack of trust and credibility, together withthe low levels of perceived effectiveness and satisfaction, poseda high load on governmental institutions. Trust and publiccredibility is a dynamic construct (Greenberg et al., 1991)and most studies suggest that its determinants are very com-plex. It has been demonstrated that trust or credibility in gov-ernment is a function of public perceptions of knowledge andexpertise, the degree of information disclosure, information re-ceipt, openness and honesty, and media sensitization (Peters etal., 1997; Williams et al., 1999). Assessments made by bothvolunteers and inhabitants in respect of public institutions in2002 were consistently low, largely because the degree of in-formation, knowledge and expertise they provided was as-sessed as being deficient.
The duration of threat has also been found to be an impor-tant factor in the response of the community (Evans and Co-hen, 1987; Otway and von Winterfeldt, 1982; Slovic, 1987,1993). Studies by Levi et al. (2001) also reveal that the riskimpact on inhabitants’ lives depends on the duration of theeffects of the danger. In the same way, it has been suggestedthat the longer-lasting the negative consequences for the vic-tims the more serious the subsequent adverse effects, leading toa reduction in individual, social or community well-being(Baum et al., 1983; Edelstein, 1988). Levi et al. (2001) arguedthat, from a community point of view, long term disasters canhave a double impact—that which arises from the disaster it-self, and the social problems generated subsequently. In thecase of Prestige disaster, the inhabitants and the volunteerswere quite pessimistic about the duration of threat, and thissuggests that the negative impact could be reinforced and ag-gravated over time rather than reduced. The incidence of healthproblems and the changes in daily habits are also indicators ofthe social and individual impact of the disaster, mainly for theinhabitants.
We have to acknowledge that people develop their own be-liefs system about the nature of the threat after a disaster as partof the coping process (Vyner, 1988; Kroll-Smith and Couch,1993; Rochford and Blocker, 1991). Beliefs are socially con-structed (Berger and Luckmann, 1966) and are influenced bythe interaction of individual, social, cultural, political and eco-nomical factors in addition to the characteristics of the disasterevent or contamination itself. In this study, we focused on be-
liefs about the duration of contamination, the effects on health,or the scope of the disaster, but we also explored the attribu-tions that people made concerning the actions and effectivenessof the Government and other institutions to manage the crisis.One consequence of the subsequent management of the disas-ter was that in addition to a social fracture, there was a percep-tion of distrust and lack of institutional credibility. Such out-comes are not limited to an individual response, but emerge aspart of a subjective and socially constructed process, or from acombination of both (Peters and Slovic, 1996; Williams et al.,1999). In the case of the Prestige disaster, further highly salientsocio-economic and political factors may have been mediatingthis process.
Technological disasters are different from natural disastersin various ways that are salient to both risk perception, themanagement of the disasters and the response by the publicto the disaster management. Technological disasters can be ea-sier to predict. Technological disasters similar to the sinking ofthe Prestige have happened in the past and therefore they couldhave been anticipated. In such situations, the damage to boththe ecosystem and the inhabitants could have been avoided or,at least, minimized. The recovery from social and psychologi-cal trauma, once the damage is done, only serves to add toother stressors like dealing with insurance companies, lawyers,contractors, and politicians to obtain economic or legal com-pensation (Blaustein, 1991; Levi et al., 2001).
The effective management of the chronic consequences ofdisasters such as the Prestige requires the development andimplementation not only of support services but also economicand technological, as well as psychological and social strate-gies. The importance of this integrative approach is supportedby previous research (e.g. Bolin, 1988), where psychosocialrecovery is related to a positive perception of government as-sistance and support. On the contrary, a negative perception ofinstitutional support only sustains the detrimental and dama-ging psychosocial impact. The effect of a crisis as the resultof a disaster, according to San Juan (2001), depends on threefactors: a) the nature of the disaster and its psychologicalmeaning; b) the state of vulnerability and resources in relationto the individual, group and community; and c) the kind ofhelp that the individual, group or community can receive. Allthese factors were very negative in the management of thePrestige crisis.
How can we explain the apparent complacency of the popu-lation in response to the disaster? Initially it appears difficult toaccount for such a shift in public opinion, especially as most ofthe subsequent damage was not easily healed. One might drawon a number of psychological theories to account for this re-sponse (Uzzell, 2000). For example, minimizing the perceivedimpact may be a coping strategy to deal with serious detrimen-tal effects to the environment. Likewise, denial and failure toact occur when a person perceives that a threat is uncontrolla-ble. This, in turn, may lead to the reduction of fear and anxietylevels and lessen the negative feelings consequent upon thelack of perceived control over the situation (Perloff and Fetzer,1986). Some environmental stressors such as noise or pollutionmay be personally manageable because the stressor is poten-
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–60 59
tially within the bounds of a person’s immediate control andpersonal powers. However, some environmental problems, inparticular environmental catastrophes, may be perceived to beoutside the individual’s or even the community’s immediatecontrol and socio-political powers.
Second, most fishermen receive a subsidy from the govern-ment because they receive a low and irregular income fromfishing which is normally, even leaving aside the present dis-aster, highly risky. This part of the Galician coast is known asCosta da Morte—the ‘Death Coast’, because of its dangers toshipping. Subsidies from the government meet the needs ofmost fishermen and in some cases may even provide them witha wage in excess of what they could earn if they were relianton a reasonable income solely from the sea. The kind of fishingthat is practiced in this area (i.e. very profitable and difficult toobtain seafood) is of a depredatory nature in which the bene-ficial fruits of the environment take priority over its preserva-tion. When visual clues of pollution (e.g. oil slicks) are nolonger visible even though they may still be present, for exam-ple, on the seabed, and signs of life are seen again on the rocksand beaches, it is easier to forget what happened.
Finally, one might explain this shift in the inhabitants’ opi-nion by reference to the socio-political and cultural setting.These are rural, small and closed societies. They are also veryconservative. Being critical of or challenging authority doesnot come naturally and is not easy to sustain over a long periodof time. Once the protest had been made and placatory mes-sages of support had been sent from the government there wasa feeling that everything should be left in the hands of theauthorities; once this stage had been reached other social andeconomic pressures start to have an effect on people’s attitudes.
The damage from the sinking of the Prestige and its subse-quent oil spill was substantial—one of the worst ecologicaldisasters to affect Europe. It is clear from our research, how-ever, that clearly it was highly salient for the public too, affect-ing livelihoods and the quality of community life. But what isparticularly interesting from this research is that the public’sevaluation changed quite quickly over time. We have sug-gested that the cause of this might be economic, that is, thepopulation are so financially dependent upon the environmentand the government that one coping strategy—both practicaland psychological—is to minimize the perceived impact ofthe disaster as quickly as possible. The second explanation iscultural, a function of the history and traditions of the peopleand their relationship with authority. Clearly this is an area thatwarrants further investigation in the context of other kinds ofdisasters and in other socio-political and cultural settings.
6. Curriculum vitae
Ricardo García-Mira is Associate Professor of Social andEnvironmental Psychology and directs the People–Environ-ment Research Unit in the Department of Psychology, Univer-sity of Corunna, Spain.
J. Eulogio Real is Associate Professor of Methodology ofBehavioral Sciences in the Department of Social Psychology,University of Santiago de Compostela, Spain.
David L. Uzzell is Professor of Environmental Psychologyand directs the Environmental Psychology Research Group inthe Department of Psychology, University of Surrey, UK.
César San Juan is Associate Professor of Social Psychol-ogy in the Department of Social Psychology, University of theBasque Country, Spain.
Enric Pol, is Professor of Social and Environmental Psy-chology and directs the Master’s Program on EnvironmentalIntervention in the Department of Social Psychology, Univer-sity of Barcelona, Spain.
References
Abbot Wade, D., 1991. Kurnell residents take on the giant Bayer. In: Baldry,E., Vinston, T. (Eds.), Actions Speak: Strategies and Lessons from Austra-lian Social and Community Action. Longman Cheshire, Melbourne, Aus-tralia, pp. 37–55.
Baum, A., Fleming, R., Davidson, L., 1983. Natural disaster and technologicalcatastrophe. Environment and Behavior 15 (3), 333–354.
Berger, P.L., Luckmann, T., 1966. The social construction of reality: a treatisein the sociology of knowledge. Penguin, New York.
Blaustein, M., 1991. Overview: natural disasters and psychiatric response.Psychiatric Annals 21, 518–520.
Bolin, R., 1988. Response to natural disasters. In: Lystad, M. (Ed.), MentalHealth Responses to Mass Emergencies. Bruner Mazel, New York.
Brown, P., Mikkelsen, E.J., 1990. No safe place: toxic waste, leukaemia, andcommunity action. University of California Press, Berkeley.
Bullard, R.D., 1990. Dumping in Dixie: Race, Class and Environmental Qual-ity. Westview, Boulder, CO.
Edelstein, M.R., 1988. Contaminated communities: the social and psychologi-cal impacts of residential toxic exposure. Westview, Boulder, CO.
Erikson, K., 1990. Toxic Reckoning: Business Faces a New Kind of Fear. Har-vard Business Review 90, 118–126.
Evans, G., Cohen, S., 1987. Environmental stress. In: Stokols, D., Altman, I.(Eds.), Handbook of Environmental Psychology. John Wiley, New York,pp. 571–610.
Greenberg, M., Spiro, H., McIntyre, R., 1991. Ethical oxymora for risk assess-ment practitioners. Accountability in Research 1, 245–257.
Knoke, D., Burke, P.J., 1990. Log-Linear Models. Sage, Beverly Hills, CA.Kroll-Smith, J.S., Couch, S.R., 1993. Technological hazards: social responses
as traumatic stressors. In: Wilson, J.P., Raphael, B. (Eds.), InternationalHandbook of Traumatic Stress Syndromes. Plenum, New York, pp. 79–91.
Levi, D., Kocher, S., Aboud, R., 2001. Technological disasters in natural andbuilt environments. Environment and Behavior 33, 78–92.
Levine, A.G., 1982. Love Canal: Science, Politics and People. D.C. Heath,Lexington, MA.
McGee, T., 1999. Private responses and individual action. Community re-sponses to chronic environmental lead contamination. Environment andBehavior 31 (1), 66–83.
Otway, H.J., von Winterfeldt, D.V., 1982. Beyond acceptable risk: on the so-cial acceptability of technologies. Policy Sciences 14, 247–256.
Peters, E., Slovic, P., 1996. The role of affect and worldviews as orientingdispositions in the perception and acceptance of nuclear power. Journal ofApplied Social Psychology 26, 1427–1453.
Perloff, L.S., Fetzer, B.K., 1986. Self-other judgements and perceived vulner-ability to victimisation. Journal of Personality and Social Psychology 50,502–510.
Pet
Po
Po
Re
Ro
RS
San
SloSlo
Uz
Uz
Vy
Wi
Wi
R. García-Mira et al. / Revue européenne de psychologie appliquée 56 (2006) 53–6060
ers, R.G., Covello, V.T., McCallum, D.B., 1997. The determinants of trustand credibility in environmental risk communication: an empirical study.Risk Analysis 17 (1), 43–54.l, E., Moreno, E., 2002. Social Impact Detection SID/BCN. PublicacionsUniversitat de Barcelona, Monografies Socio/Ambientals, Barcelona (21).wers, D.A., Xie, Y., 2000. Statistical Methods for Categorical Data Analy-sis. Academic Press, London.ich, M.R., 1991. Toxic politics: responding to chemical disasters. CornellUniversity Press, Ithaca, NY.chford, E.B., Blocker, T.J., 1991. Coping with “natural” hazards as stres-sors. Environment and Behavior 23, 171–194.PB, (2003) ‘Report confirms Prestige disaster’s devastating effect on sea-birds’, http://www.rspb.org.uk/international/prestigereport.asp.Juan, C., 2001. La intervención en desastres: bases conceptuales y opera-tivas. In: San Juan, C. (Ed.), Catástrofes y ayuda de emergencia. EditorialIcaria, Barcelona.
vic, P., 1987. Perception of risk. Science 236, 280–285.vic, P., 1993. Perceived risk, trust, and democracy: a systems perspective.Risk Analysis 13, 675–682.zell, D.L., Pol, E., Badenas, D., 2002. Place identification, social cohesionand environmental sustainability. Environment and Behavior 34, 1, 26–53.zell, D.L., 2000. The psycho-spatial dimension to global environmental pro-blems. Journal of Environmental Psychology 20, 4, 307–318.ner, H.M., 1988. Invisible Trauma: The Psychosocial Effects of InvisibleEnvironmental Contaminants. Lexington Books, Lexington, MA.lliams, B.L., Brown, S., Greenberg, M., 1999. Determinants of trust percep-tions among residents surrounding the Savannah River Nuclear WeaponsSite. Environment and Behavior 31, 354–371.saeth, L., 1994. Psychological and psychiatric aspects of technological dis-asters. In: Ursano, R.J., McCaughey, B.G., Fullerton, C.S. (Eds.), Indivi-dual and Community Responses to Trauma and Disaster: The Structure ofHuman Chaos. Cambridge University Press, New York.
