Attitudes towards anti-terror measures: the role of trust, political orientation and civil liberties support

Abstract

This article analyses Norwegian citizens' attitudes towards the use of strong prevention measures in the fight against terrorism. Norwegians stand out as being relatively supportive of such measures, despite having been relatively shielded from serious terrorism. Nevertheless, they are generally more critical towards keeping people in custody without trial. This analysis shows that trust, political orientation and civil liberties support have significant explanatory value, also when the effects of gender, age, education and occupational status are controlled for. However, trust turns out to be a complicated indicator; whereas high scores on social trust predict more sceptical attitudes, a strong belief in personal abilities and in the political system fosters attitudes that are less sceptical. The article concludes that the combination of low threat, limited previous experience and high levels of trust creates a favourable environment for preventive policies, and recommends a further exploration of these dimensions in different contexts.

Keywords:

• anti-terror measures
• preventive crisis management
• citizens’ attitudes
• trust
• political orientation
• civil liberties

Notes

1. ISSP is a continuous programme of cross-national collaboration running annual surveys on different social science topics in 44 member countries worldwide: http://www.gesis.org/en/services/data/survey-data/issp/

2. See T. Almenningen, 2007. Undersøkelse om samfunnsdeltakelse [Study on society participation]. Report 121. Bergen: NSD.

3. For a linguistic discussion of the two terms, see, for instance, Nunberg (2004). On the definition of terrorism, see Weinberg et al. (2004).

4. Nordenhaug and Engene (2008) further introduced different perceptions of terrorism present in this research. There is a distinction between state terrorism and terrorism performed by non-state actors, between international (transboundary) terrorism and internal (nationally based) terrorism. There are also different typologies based on the variations in cause and motivation behind the terrorist acts.

5. This alleged act of terror has never been solved, but is thought to be linked to the Islamic fatwa issued against Salman Rushdie and his publishers.

6. In 1993, proposals for a penal clause against terrorism were discussed, but renounced in a Government White Paper (NOU 1993, p. 3).

7. A process that aimed to reinforce the so-called unorthodox investigation methods and methods aimed at crime prevention had actually begun before 9/11. In January 2000, the police were allocated new powers in the area of phone tapping, communication control, and delayed notification when searching. In July 2001, a public commission (the ‘Police-method Commission’) was appointed to investigate and propose new crime-prevention methods for the police. The Commission delivered a report (NOU 2004) which resulted in a legislative proposal in 2005 that extended opportunities for police surveillance (Ot. prp. nr. 60 (2004–2005)).

8. Religion has been ascribed a central role in the ongoing war on terror, and some have even called it a religious war (Rojecki 2005). After 9/11, the Norwegian government identified Islamic extremists as the most likely potential terrorists, and religious fanatics were seen as the most damaging threat (St. meld. 17 (2001–2002), Nordenhaug and Engene 2008). It has further been argued that religion can be used as predictors of the degree of integration in society and that people who are more integrated in society are likely to be more supportive of the authorities (Christensen et al. 2007). Consequently, religion and religious activity could be essential explanatory variables when explaining variations in attitudes towards terror prevention. However, Norway is a very homogeneous society when it comes to religion, and religious conflicts are not prominent. Our survey includes only 21 respondents reporting that they are Muslim (1206 are defined as Christian, 48 as belonging to an ‘other’ religion and 111 as having ‘no religion’). A variable on religion would therefore appear more as a constant and is not included in our analysis.

9. Previous studies have shown that political orientation is related to how one views the authorities, and furthermore to general trust in government (Forster and Snyder 1989, Lægreid 1993, Christensen and Lægreid 2005).

10. About 85% preferred to let someone guilty go free over judging someone innocent.

11. Sollund (2007) discussed this particular measure in relation to the interaction between ethnic minorities and the Norwegian police.

12. Preliminary bivariate analysis was conducted to single out relevant variables. Further analysis of the dependent variables was conducted to ensure that there were no violations of necessary assumptions of normality, linearity, multicollinearity and homoskedasticity. The dependent variable ‘attitude towards phone tapping’ was not satisfactorily normally distributed. However, after an assessment of the advantages and disadvantages of using alternative analysis, linear regression was pursued. Three qualitative categorical explanatory variables (occupational status, political party affiliation and attitudes towards civil rights) were recoded into dummy variables (coded 0 and 1) in order to be included in the linear regression.

13. Different statistical testing techniques have been used to assess the bivariate relations between the three dependent and the various independent variables, owing to different levels of measurement.

14. People voting for the extreme left party (the Red Party) and people voting for the extreme right party (the Progress Party) are the groups of voters who are the farthest apart. People affiliated to the Red Party are more likely to be sceptical towards strong measures in the fight against terror (keeping people in custody without trial), while people voting for the Progress Party are more likely to be supportive of these measures.

15. A positive beta coefficient indicates that a high value on the independent variable corresponds to a high value/sceptical attitude on the dependent variable. A negative beta coefficient indicates that a high value on the independent variable corresponds to a low value/supportive attitude on the dependent variable.