A preferential response toward an attitude object cannot occur in a vacuum. It must necessarily be elicited within the context of some social situation, about which, as already noted, we also have attitudes. It is perhaps helpful to conceive of any particular attitude object as the figure and the situation in which it is encountered as the ground. How a person will behave with respect to an object within a situation will therefore depend, on the one hand, on the particular beliefs or predispositions activated by the attitude object and, on the other hand, by the beliefs or predispositions activated by the situation. We thus postulate that a person’s social behavior must always be mediated by at least two types of attitudes—one activated by the object, the other activated by the situation.
If one focuses only on attitude-toward-object one is bound to observe some inconsistency between attitude and behavior, or, at least, a lack of dependence of behavior on attitude. Most frequently mentioned as evidence in this connection are such studies as those by La Piere (1934) and by Kutner et al. (1952), in which there were found to be marked discrepancies among restaurant and hotel owners between their verbal expressions of discrimination toward Chinese and Negroes via letter or phone and their nondiscriminatory faceto-face behavior. One possible explanation of such apparent inconsistency is suggested by the present analysis: the investigators did not obtain all the relevant attitudinal information needed to make accurate predictions. The subjects not only had attitudes toward Chinese and Negroes but, being managers of an ongoing business, also had attitudes about how to conduct such a business properly. The investigator’s methods, however, are typically focused on obtaining data relevant to attitude-toward-object and are generally insensitive toward attitude-toward-situation.
One may thus readily agree with Krech, Crutchfield, and Ballachey when they say that behavior is determined by a number of attitudes, wants, and situational conditions rather than by a single attitude (1962). Their additional statement that “attitude test scores alone are usually not enough to predict behavior” (p. 163) does not necessarily follow from the preceding. As already suggested, a “situational condition” can psychologically be reformulated as “attitude-toward-situation” and assessed by methods similar to those employed in assessing attitude-toward-object. Unfortunately, however, only the latter kind of attitude has thus far been the focus of operational definition and measurement, even though attitudes have typically been more broadly defined as predispositions toward situations as well as toward objects.
However, it is not enough merely to assess in advance the two kinds of attitudes discussed. It is also necessary to recognize that attitude-toward-object and attitude-toward-situation will cognitively interact with one another and will have differing degrees of importance with respect to one another, thereby resulting in behavior that will be differentially influenced by the two sets of attitudes. In one case, an attitude object may activate relatively more powerful beliefs than those activated by the situation, thereby accounting for the generality of behavior with respect to an attitude object; or, the situation may activate the more powerful beliefs, thereby accounting for the specificity of behavior with respect to an attitude object. Campbell (1963) has shown that the threshold of discrimination toward Chinese seeking reservations for overnight lodging and restaurants is without exception lower —there is more discrimination—in non-face-to-face situations. He has similarly shown that the threshold of discrimination toward Negro miners by white miners is always lower in town than in the mines (Minard 1952). In pointing to “different situational thresholds,” Campbell is not only explaining away the apparent inconsistency between attitude and behavior, or between one behavior and another, but he is also suggesting that certain situations consistently activate discriminatory behavior with respect to a specific attitude object more than do other situations.
In the context of this discussion, one may fruitfully raise again Blumer’s criticism (1955). The state of present attitude theory is such that there are no rigorous criteria available for ascertaining when we are dealing with one attitude or with more than one attitude. We speak, for example, of an attitude toward the Negro, but also of an attitude toward desegregation of the Negro in education. In line with the present analysis, we would prefer to say that the way we feel toward desegregation of the Negro in the school involves the activation of at least two attitudes, one concerning the Negro, the other concerning a particular educational situation.