These metaphors are more than mere language: they structure thought. The public may think about a social problem in a way that disputes the expert evidence but you can be sure that people continue to reason according to this erroneous model, and to reason systematically, until a more compelling model displaces the faulty one. Ideas that do not “fit” into the metaphors we have accepted to structure our lives are rejected out of hand (Bales 2002:1).

Once received, the manner in which messages are processed also provide opportunities to affect attitude formation. Sears’ (1980) theory of symbolic politics suggests that reactions to some issues are based on strongly held values acquired early in life and occur almost without thought. His assertion that people’s reactions to issues are mostly irrational and “knee-jerk” in character, especially in relation to salient groups such as racial minorities, drug addicts, or welfare recipients, seem at odds with the findings of other researchers who emphasize the ambivalence and thoughtfulness that many study respondents show when considering issues (Doob et al. 1998, Zaller and Feldman 1992). Rather than an uncomplicated uni-dimensional system of thought, it appears more common for people to entertain a complex array of conflicting opinions and attitudes that are held loosely together with generalized, but deeply-held, overriding principles.

Feldman (1988) acknowledges that the public may easily absorb the major elements of the political culture through processes of socialization and continuous reinforcement but the mechanism of attitude formation is less irrational than it might seem. Instead of a knee-jerk reaction, he believes that a sort of constraint or consistency is achieved in people’s belief systems about specific policies by appealing to some superseding position that involves beliefs about big ideas like the nature of social justice or social change. It is the hierarchy of values associated with such a belief system that can be used as leverage to gain public acceptance for a specific social and political agenda.

For instance, in their study of Canadian attitudes to affirmative action, for example, Fletcher and Chalmers (1991) noted that questions about complex issues, a category that would surely include the best way to curb illicit drug use, touch on a recurring dilemma in contemporary life: that of resolving in a substantive way a conflict between two important, competing, abstract values. In the case of affirmative action, some respondents place greater value on the overriding principle of equality while others give primacy to merit. The authors found that a significant segment of respondents would change their original support or opposition to affirmative action policies when probed by making salient the values trade-offs involved in their decisions. Similarly, Doob (2000) found that a substantial number of respondents could be dissuaded from their original harsh views on sentencing to a more moderate position once the consequences of their choices were made salient to them. However, one can also assume that those holding more “progressive” views can also be similarly swayed. It appears that, generally, individuals most open to persuasion on issues are respondents whose opinions are not strongly rooted in their values--those who are said to have non-attitudes about the issue under consideration.

It is the public’s ambivalence on the acceptability or deviance of some behaviours that provides actors with the greatest opportunity to influence the public mood. In criminalizing some conduct and not others, the influence of dominant elites varies but their impact is greatest when public opinion is dormant or divided about the act in question (Hagan 1991). Their success at moving public opinion depends in large measure on their ability to couch their message in metaphors that resonate with widely-held and ingrained audience values in their efforts to make the considerations they think are important for judging a particular behaviour or group decisive to the thinking of a critical mass of their fellow citizens. Our window into what the public is thinking on any given issue is provided by public opinion polls. But rather than affording an unobstructed view, polls offer a view that is often refracted and distorted either inadvertently or with intent.