1 For Freud, ambivalence defines the mixing of love and hatred for the same person. More generally the term describes contradictory feelings or attitudes toward the same person, object, event or goal. In his Dementia Praecox; or The group of Schizophrenias, trans, by Zinkin, J. (New York: International Universities series, 1950), 34Google Scholar, Eugen Bleuler distinguishes three forms of ambivalence: (a) affective ambivalence (contradictory feelings for the same person), (b) ambivalence of the will (the desire to do mixed with the desire not to do something), and (c) ambivalence of the intellect (possessing two contradictory ideas).

2 See Kaplan, Kalman, “On the Ambivalence—Indifference Problem in Attitude Theory and Measurement: A Suggested Modification of the Semantic Differential Techniques,” Psychological Bulletin 77 (1972), N.S., 361–72CrossRefGoogle Scholar. For an earlier criticism of the assumption of unidimensionality between the polar opposites of Osgood's SD, see Green, Russell F. and Goldfried, Marvin R., “On the Bipolarity of Semantic Space,” Psychological Monographs, 79 (No. 6, Whole No. 599, 1965)CrossRefGoogle Scholar. For the importance of treating frustration derived from ambivalence as a specific drive, see Brown, Judson S. and Farber, I. E., “Emotions Conceptualized as Intervening Variables; With Suggestions toward a Theory of Frustration,” Psychological Bulletin 48 (1951), 465–95CrossRefGoogle Scholar. For the tendency to define one's variables as a function of either the positive or the negative side of a dichotomy, see Laponce, J. A., “Of Gods, Devils, Monsters and One-Eyed Variables,” this Journal 7 (1974), 199–209Google Scholar. See also Anderson, Norman H., “On the Quantification of Miller's Conflict Theory,” Psychological Review 69 (1962), 400–14CrossRefGoogle ScholarPubMed, and Cantril, Hadley, “Sentio, Ergo Sum: Motivation Reconsidered,” Journal of Psychology 65 (1967), 91–107CrossRefGoogle ScholarPubMed.

3 An excellent critique and refinement of the classical measures of cross-pressures is in Sperlich, Peter W., Conflict and Harmony in Human Affairs: A Study of Cross-Pressures and Political Behaviour (Chicago: Rand McNally, 1971)Google Scholar.

4 Semantic theory indicates that a positive stimulus is more likely to “enter” the neutral “chamber” than a negative one. For example, the question “how good was the play?” is more likely to be interpreted as neutral than the question “how bad was the play?” See Sapir, E., “Grading: A Study in Semantics,” Philosophy of Science II (1944), 93–116CrossRefGoogle Scholar; see also an experiment by Clark, H. H., “Influence of Language in Solving the Term Series Problems,” Journal of Experimental Psychology 82 (1969), 205–15CrossRefGoogle ScholarPubMed. In terms of our three chamber model, the positive question has a greater likelihood than a negative question of “arousing” information from the other two chambers. In this paper I consider only the problem of ambivalence in the organism; the study of ambivalence of the stimulus is a different and equally fascinating subject. For the latter see, for example, studies of the Ad'dad: i.e., those concepts thought by some linguists to have been used in classical Arabic to mean something and its opposite at the very same time, although thought by others to have acquired these opposite meanings at different times or in different contexts. See in particular Berque, Jacques and Charney, Jean-Paul, Ambivalance dans la culture arabe (Paris: Editions Anthropos, 1967)Google Scholar.

5 That the high scorers on both the liking and disliking measures are a distinct and fascinating group to study is shown in an experiment conducted by Katz, Glass and Cohen who verified in laboratory conditions that the denigration of subjects to whom one has inflicted a punishment is more intense for those punishers who had for the victim high positive and negative feelings than for those who had more exclusively high negative or positive feelings. They explain their findings by theorizing that ambivalence leads to guilt which is released by scapegoating. See Katz, Irwin, Glass, David C. and Cohen, Sheldon, “Ambivalence, Guilt and the Scapegoating of Minority Group Victims,” Journal of Experimental Social Psychology 9 (1973), 423–36CrossRefGoogle Scholar. The authors studied attitudes of White subjects toward Blacks; on this problem and on the role of ambivalence in the American political culture, see Myrdal, G., An American Dilemma (New York: Harper, 1944)Google Scholar.

6 I apologize to the reader for this abstract reporting of the findings. Although the questionnaire identified the four major federal parties by name, I prefer, in this report, to leave them anonymous; the number of subjects interviewed, while more than adequate to study psychological mechanisms, is not sufficient (and was not intended) to enable us to draw conclusions regarding the political parties.

Using a two-point difference to measure ambivalence is, of course, arbitrary. In our case it is justified by the shape of the scattergrams but might be inappropriate in other cases. Using a three-point difference criterion results in the following measures: A 43 per cent, B 45 per cent, C 66 per cent, D 57 per cent. The As and Bs maintain lower scores than the Cs and Ds, but the latter obtains a lower score than C. Small as it is, the difference provides an argument against relying on single indicators.

7 Another measure of ambivalence derived by Scott from that proposed by Kaplan gives scores ranging from 0 to 1. The formula is where S is the smaller score and L the larger one. See Scott, William A., “Brief Report: Measures of Cognitive Structure,” Multivariate Behavioral Research 1 (1966), 391–95CrossRefGoogle ScholarPubMed. The Scott index gives the following statistics: party A .44; party B .48.

8 Other techniques than those used in this note are available to test ambivalence. An especially attractive possibility would be to combine verbal and non-verbal stimuli. Recent work on the different functions and “memories” of the two hemispheres of the brain indicates that the reactions to a verbal and to a visual stimulus are sometimes markedly different. Ambivalence could be measured not only by testing positive and negative verbal reactions, but also by triggering both “rational” and “intuitive” answers, for example by testing the reactions to a name (such as Trudeau, working class, or religion) and to a picture. Ambivalence not apparent at either the verbal or the nonverbal level might result from a clash between the two. See Gazzaniga, M., “The Split Brain in Man,” Scientific American 217 (1967), 24–29CrossRefGoogle Scholar. A popular presentation of recent findings on hemispheric specialization is in Pines, M., The Brain Changers: Scientists and the New Mind Control (New York: Harcourt, Brace, Jovanovich, 1974)Google Scholar.