¹ Armor, David J., Giaquinta, Joseph B., McIntosh, R. Gorden & Russell, Diana E. H., “Professors' Attitudes Toward the Vietnam War,” Public Opinion Quarterly, 31 (Summer, 1967), 159–175CrossRefGoogle Scholar.

² See for example, Lipset, Seymour M., “Democracy and Working-class Authoritarianism,” American Sociological Review, 24 (08, 1959), 482–501CrossRefGoogle Scholar; Lipset, Seymour M., Political Man (Garden City: Doubleday, 1960), Chapter 4Google Scholar.

³ Gallup Opinion Index, Report 13 (Princeton, New Jersey: Gallup International, Inc., 06, 1966), p. 6Google Scholar.

⁴ Gallup Index, Report 21, 03, 1967, p. 5Google Scholar.

⁵ Gallup Index, Report 32, 02, 1968, p. 18Google Scholar.

⁶ The New York Times, 09 27, 1970, p. 18Google Scholar.

⁷ Hamilton, Richard F., “A Research Note on the Mass Support for ‘Tough’ Military Initiatives,” American Sociological Review, 33 (06, 1968), 439–445CrossRefGoogle ScholarPubMed.

⁸ Hahn, Harlan, “Correlates of Public Sentiment about War: Local Referenda on the Vietnam Issue,” American Political Science Review, 64 (12, 1970), 1186–1198CrossRefGoogle Scholar.

⁹ The belief that those lower in SES favor tougher policies cannot quite be dismissed as a complete myth. Some findings indicate that they are more prone to support extremely belligerent policies; e.g., the less educated were slightly more prone to endorse both the bombing of cities in North Vietnam (Gallup Index, Report 9, February, 1966), and the introduction of atomic weapons into the fighting (Gallup Index, Report 34, April, 1968). Such policies receive little support in any quarter (about 25 per cent nationally), and the slightly greater support for them among those lower in SES probably reflects a lesser sensitivity to the massive civilian casualties associated with them.

¹⁰ Hamilton, , “Mass Support for ‘Tough’ Military Initiatives,” pp. 439–445Google Scholar.

¹¹ These were: #302 (April, 1951), #307 (May, 1951), #314 (November, 1951), #315 (December, 1951), #317 (February, 1952), #320 (March, 1952), #327 (June, 1952), #329 (August, 1952), and #332 (October, 1952). The sample sizes ranged from 1200 to 1300.

¹² Here, as on all items, “don't know” was not explicitly offered as an alternative to the respondent, but a “don't know” response was recorded as such.

¹³ On surveys #317 and 320 the question was worded slightly differently as follows: “If the Chinese Communists attack any other country in Asia, do you think the United States should stay out of it, or should we help defend that country? The effect was to increase very slightly (4 to 6 per cent) sentiment to stay out.

¹⁴ Note that a respondent received a score of “1” either by expecting agreement within a month or by favoring continuation of the talks. In effect, it was assumed that respondents expecting agreement soon would want to continue the talks.

¹⁵ Scoring was as follows: Education—(1) no formal schooling to 11th grade, (2) 12th grade to college; Occupational Status—(1) unskilled workers, (2) semiskilled workers, (3) skilled workers, (4) office workers, (5) business executives and managers, (6) professionals and semiprofessionals: Economic Level—(1) low or below average, (2) medium or average, (3) high or above average; Race—(1) nonwhite, (2) white. It might be noted that education was dichotomized only after analyses using more refined breakdowns revealed that no curvilinear relationships (or other interpretable departures from linearity) would be obscured as a consequence.

The above scoring was used when examining the separate effects of these variables. In the creation of the composite SES Index, all four variables were scored dichotomously: occupational status was dichotomized between blue and white collar, and economic level was dichotomized between low and medium. Scores on the summed index thus ranged from 4 to 8 and were recoded as follows for purpose of analysis: (1) low = scores of 4 or 5, (2) medium-low = score of 6, (3) medium-high = score of 7, (4) high = score of 8. The decision to treat race as an indicator of SES and, thus, to include it in the composite index was a departure from conventional practices about which some may have certain legitimate reservations. Although race clearly is associated with both economic and social status in this society, it is also associated with certain other important factors (such as cultural background and a history of discrimination) that have little to do with the conventional meaning of SES. While the inclusion of race might add a small amount of unwanted noise to our operational index of SES, nonetheless, the concept this index is intended to assess is the standard one, and throughout this paper the term “SES” is used in its conventional sense. To aid the reader who may wish to separate the effect of race from those of the more standard indicators, the data tables that follow will always report findings based on the SES index together with parallel findings based on its separate components.

¹⁶ The two items offering respondents three choices, ranging from “pull out” to escalate, had stronger associations with the items in the Disengagement cluster (medium gamma approximately .40), but were also associated with the items in the Escalation cluster (medium gamma approximately .20).

¹⁷ For a careful analysis of the extent to which a sample of academicians do, in fact, order military policies on a soft-to-tough dimension, see Brickman, Philip, Shaver, Phillip, & Archibald, Peter, “American Tactics and American Goals in Vietnam as Perceived by Social Scientists” in Vietnam: Issues and Alternatives, Isard, Walter, ed. (Cambridge, Mass.: Schenkman Publishing Co., 1969), pp. 80–104Google Scholar.

¹⁸ Item 4, which offered the respondent three alternatives, showed a slight curvilinear trend with education. Support for “pull out” decreased linearly, and support for “continue” increased linearly. Support for “attack,” however, peaked at 12th grade and fell away slightly at the college level (5 to 9 percentage points depending on the survey). This trend was not present in the similarly worked item 5, nor in any of the items belonging to the escalation cluster.

¹⁹ Two other demographic variables, occupational status and age, showed a pattern of association with military policy that resembled that of SES, but associations with the disengagement items were generally weaker and, in the case of age, disappeared entirely when education was controlled.

²⁰ See, for example, Hahn, “Correlates of Public Sentiment about War.”

²¹ Verba, Sydney, Brody, Richard, Parker, Edwin, Nie, Norman H., Polsby, Nelson W., Edman, Paul & Black, Gordon, “Public Opinion and the War in Vietnam,” American Political Science Review, 6 (06, 1967), 317–333CrossRefGoogle Scholar.

²² For two more complete but somewhat divergent statements of this model, see Galtung, Johan, “Foreign Policy Opinion as a Function of Social Position.” Journal of Peace Research, 1 (1964: 3–4), 206–231CrossRefGoogle Scholar; and Gamson, William A. and Modigliani, Andre, “Knowledge and Foreign Policy Opinions: Some Models for Consideration,” Public Opinion Quarterly, 30 (Summer, 1966), 187–199CrossRefGoogle Scholar.

²³ Couch, Arthur S. & Armor, David J., The Datatext System: Preliminary Manual. (Harvard University, Department of Social Relations, multilithed, 1967), 265–304Google Scholar.

²⁴ Caspary factor analyzes some of the same surveys starting from a correlation matrix of gamma coefficients and obtains nearly identical results. See Caspary, William R., “Dimensions of Attitudes on International Conflict: Internationalism and Military Offensive Action,” Peace Research Society (International) Papers, 13 (1970), 1–10Google Scholar.

²⁵ The loading of a particular item on a particular factor remained relatively stable across different surveys. In almost all cases these loadings were within .10 of their median value. In no case were they more than .18 from their median value.

²⁶ Though somewhat damaged by recent events in Southeast Asia, this premise probably continues to dominate our foreign policy even today as it has for the past quarter of a century. (See Caspary, William R. “The ‘Mood Theory’: A Study of Public Opinion on Foreign Policy,” American Political Science Review, 64 [06, 1970], 536–547CrossRefGoogle Scholar). The following excerpt from a recent speech about Vietnam by President Nixon illustrates well the special blend of altruism and selfrighteousness:

²⁷ See Lipset, Seymour M., “Three Decades of the Radical Right,” in The Radical Right, ed. Bell, Daniel (New York: Doubleday, 1963), pp. 313–377Google Scholar; and Rogin, Michael P., The Intellectuals and McCarthy (Cambridge, Mass.: MIT Press, 1967)Google Scholar.

²⁸ In April, 1953 almost six months after the last survey used in this study was conducted, fully 59 per cent of the public had no opinion on Senator McCarthy. Thus, it seems unlikely that he was responsible for creating the syndrome reflected by Factor II. (See Rogin, p. 232.)

²⁹ Stokes, Donald E., “Popular Evaluations of Government: An Empirical Assessment,” in Ethics and Bigness: Scientific, Academic, Religious, Political, and Military, ed. Cleveland, Harlan and Lasswell, Harold D. (New York: Harper, 1962), p. 64Google Scholar.

³⁰ Rosenberg, Morris, “Misanthropy and Political Ideology,” American Sociological Review, 2 (12, 1956), 690–695CrossRefGoogle Scholar.

³¹ See Litt, Edgar, “Political Cynicism and Political Futility,” The Journal of Politics, 23 (05, 1963), 320Google Scholar; and Aberback, Joel D. & Walker, Jack L., “Political Trust and Racial Ideology,” American Political Science Review, 64 (12, 1970), 1199–1219CrossRefGoogle Scholar. (See especially pp. 1204–1205.)

³² Survey #329 contained the following item, “Do you think most people can be trusted?” It had no association with Administration Distrust.

³³ Gamsom, William A., Power and Discontent (Homewood, Ill.: Dorsey, 1968), pp. 49–50Google Scholar. Gamson's discussion draws heavily on the work of Easton; Easton, David, A Systems Analysis of Political Life (New York: Wiley, 1965)Google Scholar.

³⁴ This item did have substantial loading on a third factor that emerged in three surveys. This factor reflected a general pessimism about the prospects of avoiding war and attaining a stable peace: e.g., another world war can be expected within 5 years, a new communist attack on another nation is likely soon, an agreement in Korea is unlikely to be reached soon, a long period of peace cannot be expected to follow an armistice in Korea, etc. This suggests that the desire to go to war with Russia reflected a sentiment to “get it over with,” since matters were likely to continue to deteriorate leading ultimately to world war anyway. See also Caspary, “Dimensions of Attitudes on International Conflict,” for other speculations on this item.

³⁵ Since this analysis is based on a selected portion of the sample, it should be clear that the results cannot be generalized to the entire population. It was conducted to see whether those segments of the public who best represented the four types did, in fact, have military preferences of the sort expected.

³⁶ Although this analysis was carried out on all surveys that had been factor analyzed, some contained too few Interventionism or Distrust items to create entirely reliable scales. The results to be reported here come from surveys #315, 317, and 332 all of which contained at least four items on each scale. Survey #302 is also used, even though its Distrust scale only had two items, because it contained some military policy items not found on the other three surveys. In general, the results from these surveys were similar to those that emerged on the other surveys containing less adequate scales.

³⁷ Items 1 and 2 offer the respondent a choice between disengagement and some version of current policy. Items 3 and 4 offer the full range of alternatives. Items 5 offers a choice between disengagement (phrased as an explicit rejection of current policy) or an explicit endorsement of current policy. Items 6–8 offer a choice between escalation and some version of current policy. In item 8, however, the escalation alternative is more ambiguous. Presumably, breaking off the truce would necessitate simply fighting it out in some manner. Still, this policy has certain overtones of disengagement which might explain its apparent appeal to the Isolationist groups.

³⁸ Hamilton, “Mass Support for Tough Military Initiatives.”

³⁹ For discussion of this feature, see Converse, Philip and Schuman, Howard, “‘Silent Majorities’ and the Vietnam War,” Scientific American, 222 (06, 1970), 21–22CrossRefGoogle Scholar; and Rosenberg, Milton J., Verba, Sydney and Converse, Philip E., Vietnam and the Silent Majority: The Dove's Guide (New York: Harper & Row, 1970), pp. 26–30Google Scholar.

⁴⁰ During the 1968 campaign, Distrusting-isolationists might well have accounted for the surprising number of voters who supported McCarthy or Kennedy in the primaries and then shifted to Wallace in November (Converse, Philip E., Miller, Warren E., Rusk, J. G. and Wolfe, A. “Continuity and Change in American Politics: Parties and Issues in the 1968 Election.” American Political Science Review, 63 (12, 1969) pp. 1083–1105CrossRefGoogle Scholar). All three candidates would have appealed to voters who distrusted President Johnson and wanted to end the war, though Wallace best embodied their military policy preferences with his “get out or do it right” platform. Consistent with this view, Rosenberg et al., report that, “Those who voted for McCarthy in the New Hampshire primary were more dissatisfied with Johnson's policies on Vietnam than those who voted for Johnson. But among McCarthy voters, those who were dissatisfied with Johnson for not pursuing a harder line outnumbered those who wanted a withdrawal by a margin of nearly three to two.” (Rosenberg et al., p. 49.)

⁴¹ Erkine, Hazel, “The Polls: Is War a Mistake?” Public Opinion Quarterly, 34 (Spring, 1970), p. 138Google Scholar.

⁴² Caspary, “Mood Theory.”

⁴³ Gallup Index, Report 56, 02, 1970, pp. 8–16Google Scholar.

⁴⁴ The Distrust and Interventionism scales described in the preceding Methods Section were used in this analysis. Both were trichotomized at the terciles. The two scales proved to be as uncorrelated as the Factors on which they were based—their median intercorrelation across the eight surveys used was only −.07.

⁴⁵ Only those demographic variables with notable zero-order associations were used in the control analysis. Race was dropped because blacks were too few in number and too heavily concentrated in the Democratic Party. Controlling for political affiliation had no effect on the association between Interventionism and SES.

⁴⁶ For discussions of this premise, see: Converse, Philip E., “The Nature of Belief Systems in Mass Publics,” in Ideology and Discontent, ed. Apter, David, (New York: Free Press, 1964), 206–261Google Scholar; Gamson, and Modigliani, , “Knowledge and Foreign Policy Opinions,” pp. 190–197Google Scholar; and Sears, David O., Political Behavior, in Handbook of Social Psychology, ed. Lindzey, Gardner and Aronson, Elliot (Reading, Mass.: Addison-Wesley, 1969), 2nd edition, Vol. 4, 348–354Google Scholar.

⁴⁷ This explanation was suggested by Philip E. Converse (personal communication).

⁴⁸ Polsby, Nelson W., “Toward an Explanation of McCarthyism,” in Politics and Social Life, ed. Polsby, Nelson W., Dentler, Robert A., and Smith, Paul A. (Boston: Houghton-Miffiin, 1963), pp. 809–824Google Scholar.

⁴⁹ If we assume that the Distrust syndrome was capitalized on even if not created by McCarthy, our data bear directly on the continuing debate over the sources of McCarthy's support. Lipset and other theorists have argued that McCarthy's support was populist in nature—largely bi-partisan and concentrated among lower and middle SES sectors, especially among those subjected to status losses or status inconsistency. However, Polsby found only small associations with SES and a strong tendency for Republicans to be more pro-McCarthy. Rogin argues that McCarthy's appeal was not populist but elitist, resonating most with the traditional conservative wing of the Republican Party. He dismisses McCarthy's somewhat greater support among the less educated as a superficial endorsement of an anti-Communist crusader not based on any real support of his ideology. (See Lipset, “Three Decades of the Radical Right”; Bell, Radical Right; Polsby, “Explanation of McCarthyism,” and Rogin, Intellectuals and McCarthy.) If the Distrust syndrome reflects a receptiveness to McCarthy's ideology, then the present data give strong support to Rogin's argument: Distrust was greatest among high status Republicans.

⁵⁰ For a general discussion of attribution theory, see Kelley, H. H., “Attribution Theory in Social Psychology,” in Nebraska Symposium on Motivation, ed. Levine, David (Lincoln, Nebraska: University of Nebraska Press, 1967)Google Scholar.

⁵¹ Several are reviewed in Rotter, Julian B., “Generalized Expectancies for Internal versus External Control of Reinforcement,” Psychological Monographs, 80, No. 1, (1966)CrossRefGoogle Scholar.

⁵² For example: Communism is difficult to stop, hard to deal with; our aid program is not working; we are indirectly supplying the Communists with equipment; we can't seem to reach a solution in Korea; we need to take care of people and problems here at home first.

⁵³ For example: they are too soft, conciliatory, or defensive in dealing with Communist countries; they don't take strong enough action against Communist groups here at home.

⁵⁴ For example: too many Communists in Washington; some officials are trying to help the Communists; officials are corrupt, selfish, playing politics, stupid, disorganized, confused, etc.

⁵⁵ The results in Table 6 are not a spurious consequence of the fact that Democrats are both lower in SES and less prone to be critical of administration officials. While Democrats as a whole are less likely than Republicans to see a lack of progress against Communism, those who do see it are no different than like-minded Republicans in their willingness to attribute it to disloyalty or incompetence.

⁵⁶ The “sophistication factor” may make it difficult to demonstrate the proposed association between status and attribution even when the question is not put in an open-ended manner as it was here. In many instances, to blame events on the personal shortcoming of actors is obviously more simple-minded than accounting for them in terms of environmental factors or more complex interactions. For example, Gallup asked the following question of a national sample in April 1963: “Who do you think is to blame for the present condition in which Negroes find themselves: white people or the Negroes themselves?” There was only a slight positive association between occupational status and propensity to hold whites responsible (Phi = .13, p = .05). Though counter to the prediction from the attribution argument, the weakness of the association is still noteworthy since the prediction is here pitted against both conceptual sophistication and racial prejudice. See Ross, Michael M., “Resistance to Social Change in the Urban North” (Unpublished Doctoral Dissertation, Harvard University, Department of Social Relations, 1972)Google Scholar.

A study by Sherwood and Nataupsky is especially significant because it controls for conceptual sophistication by dealing only with a sample of college-educated researchers (most with Ph.D.'s), all of whom had conducted research on Negro-White differences in performance on I.Q. tests. They found that researchers of higher SES background were more likely than those of lower SES background to conclude that Negro-White differences in I.Q. were due to innate rather than environmental factors. (Sherwood, John J. and Nataupsky, Mark, “Predicting the Conclusions of Negro-White Intelligence Research from Biographical Characteristics of the Investigator.” Journal of Personality and Social Psychology, 8 (01, 1968) pp. 53–58.CrossRefGoogle Scholar)

⁵⁷ Table 7, like Table 4, is based only on that portion of the sample that scored either high or low on both Interventionism and Distrust.

⁵⁸ Women tend to score much lower than men on measures of political knowledge even when education is controlled.

⁵⁹ Negroes were a markedly dovish sector of the public during both the Korean and Vietnam wars. Verba, et al., “Public Opinion and the War in Vietnam,” p. 331Google Scholar, attribute their Vietnam attitudes to a disenchantment with, and lack of commitment to, American society. While this explanation is certainly plausible, especially in view of the civil rights turmoil that coincided with the Vietnam War, the Korean data, at least, can be explained simply in terms of the social characteristics of Negroes. Not only were Negroes on the periphery of the society, and hence isolationist, but they were also heavily loyal to the Democratic party, and hence, Trusting.

⁶⁰ Verba, et al., “Public Opinion and the War in Vietnam,” p. 323Google Scholar.

⁶¹ Gallup Opinion Index, Report 56, 1970, pp. 8–16Google Scholar.

⁶² The problem is further complicated by the fact that the Johnson administration took a number of escalatory steps thus blurring the distinction between escalation and current policy. Presumably such steps would receive their greatest support from the politically distrusting prior to their implementation, but should be adopted by the politically trusting once implemented.

⁶³ The most interesting questions here center on the reversibility of trusting and distrusting attitudes. Do Republicans now become trusting while Democrats become distrusting, or are the dissidents likely to remain so until the war is settled? Another important issue concerns the role of Opposition leaders. If such leaders tend to favor disengagement rather than escalation, will this encourage distrusting persons to have similar preferences, or will the positive association between distrust and escalation continue to hold?

⁶⁴ For data bearing on the greater cost-orientation of grade-school educated, as opposed to college-educated, opponents of the War, Vietnam, see Cantril, Albert H., “The American People, Vietnam and the Presidency,” paper delivered at the Sixty-sixth Annual Meeting of the American Political Science Association, Los Angeles, California, 09, 1970Google Scholar, Appendix, Part D.

⁶⁵ See Robinson, John P. and Hefner, Robert, “Multidimensional Differences in Public and Academic Perceptions of Nations,” Journal of Personality and Social Psychology, 7 (11, 1967) 251–259CrossRefGoogle ScholarPubMed; and Wish, Myron, Deutsch, Morton and Biener, Lois, “Difference in Conceptual Structures of Nations: An Exploratory Study,” Journal of Personality and Social Psychology, 16 (11, 1970) 361–373CrossRefGoogle Scholar.