Values - Public Opinion - Politics Research with great ambitions and dangerous pitfalls

Although the research on values is thriving in the social sciences, the res earch is confined by some lasting problems. In this paper we argue that the se problems are partly embedded in the definition of values and in the comm only used value-model. The idea that all values are concepts about the desi rable shared by everybody is unsustainable. Some values are not unanimously shared, and what often seems to be a universal value takes on very differe nt meanings to different groups of people. It is also argued that the struc ture of value conflicts may shift as a result of political struggle. The ty pical causal-chain value model relies on strict theoretical distinction bet ween exogenous, intermediate and dependent variables. We argue that distinc tion between values and attitudes and between values and behavioural predis positions tend to get blurred by the introduction of concepts like apolitic al values,, and other concepts describing domain specific values in the mod el. Moreover, the unique capacity of value-theory to explain how politicall y unaware and uninterested citizens are capable of forming political prefer ences, is lost. More generally, the determinism of both psychological theor ies and functionalist models of value formation are criticised. Finally, it is argued that the value-theory is less well developed when it comes to ex plaining electoral behaviour.