¹ Fairbank, , “The Meaning of Vietnam,” New York Review of Books, XXII, June 12, 1975, p. 31.Google Scholar

² Vandenberg, Arthur H. Jr, ed., The Private Papers of Senator Vandenberg (Boston: Houghton Mifflin 1952), 1.Google Scholar

³ We recognize that there are a number of drawbacks to the use of mailed question naires. There is validity in the observation that elite interviews may provide more revealing information about attitudes and beliefs than questionnaires. This point is particularly relevant when the population of interest can be clearly defined, when it is quite limited in number, and when it is geographically concentrated rather than dispersed. However, our goal was to gain access to the occupants of a broad range of top leadership roles; for that purpose, interviews seemed out of the question. Given limited research resources, the trade-off is between a small number of interviews and much broader coverage by means of a mailed questionnaire. Choice of the latter strategy does not, of course, preclude interviews of a small sample in the future.

⁴ A bibliography of these items appears in Appendix A of Holsti and Rosenau, “The ‘Lessons’ of Vietnam: A Study of American Leadership,” a paper prepared for the 17th Annual Meeting of the International Studies Association, Toronto, 1976.

⁵ Rielly, John E., ed., American Public Opinion and US. Foreign Policy 1975 (Chicago: Chicago Council on Foreign Relations 1975)Google Scholar; Sussman, Barry, Elites in America (Washington, D. C.: The Washington Post Publishing Co., 1976)Google Scholar; Russett, Bruce M. and Hanson, Elizabeth, Interest and Ideology (San Francisco: W. H. Freeman 1975)Google Scholar; Barton, Allen H., “Consensus and Conflict Among American Leaders,” Public Opinion Quarterly, XXXVIII (Winter 1974–1975), 508–30.Google Scholar

⁶ News conference, May 6, 1975.

⁷ Converse, Philip E., “The Nature of Belief Systems in Mass Publics,” in Apter, David E., ed., Ideology and Discontent (New York: Free Press 1964), 207.Google Scholar

It might well be argued that, since the hypotheses all pertain to the single issue of Vietnam, neither they nor the data used to test them are sufficient to justify the derivation of coherent belief systems—that, while the issue is an important one, the internal consistency of beliefs will break down once analysis moves beyond the issue of Vietnam. This argument is valid, but we have attempted to answer it elsewhere: our questionnaire included a number of items on foreign policy issues other than Vietnam that are analyzed in Holsti, and Rosenau, , “America's Foreign Policy Agenda: The Post-Vietnam Beliefs of American Leaders,” in Kegley, Charles W. and McGowan, Patrick J., eds., Challenges to America: U. S. Foreign Policy in the 1980s (Beverly Hills: Sage International Yearbook of Foreign Policy Studies, IV, 1979), 231–68.Google Scholar

⁸ Virtually all of these hypotheses were formulated before the questionnaire was mailed out. Some examples may be found in Holsti and Rosenau (fn. 4).

⁹ Although the data have been subjected to various statistical tests, we shall avoid burdening the reader with more than the minimum discussion of them. To determine whether differences among the seven groups identified in Tables 3, 5, and 7 are statistically significant, we have undertaken analyses of variance (ANOVA). The choice of a criterion of significance is always an arbitrary one. Because our sample is large, the familiar .05 or .01 levels seemed too lax. We therefore selected the more stringent .001 level—that is, we accept differences among the groups as significant if such results could have occurred by chance less than once in a thousand times. The .001 level corresponds to an F ratio of 3.47. Because the use of parametric statistics on these data will not evoke universal approval, we have also done the same analyses using a non-parametric test (chi square). Both the analysis of variance and the chi square tests only indicate whether the differences between groups are sufficiently great to meet a specific level of significance (.001 in this case). Since our hypotheses also specify a direction of difference among the groups, they are rejected unless they meet both the statistical and directional criteria. The results of these tests are summarized in Table 9.

¹⁰ This theme may be found in Kennan's, American Diplomacy, 1900–1950 (Chicago: University of Chicago Press 1951)Google Scholar, and continues through his writings up to his most recent book, The Cloud of Danger (Boston: Atlantic-Little Brown 1977).

¹¹ It should be noted that the Paris Peace Agreement was signed in January 1973, some time before the Nixon Administration's cover-up of the Watergate scandal started to unravel.

¹² That is, they did not meet both the substantive and statistical criteria (based on correlations and factor analysis) used to determine whether we were justified in treat ing items together. For a fuller discussion, see the conference paper cited in the acknowledgment footnote.

¹³ Recall the phrase “a decent respect for the opinion of mankind” in the Declaration of Independence.

¹⁴ Buchan, Alastair, “The Indochina War and World Politics,” Foreign Affairs, Vol. 53 (July 1975). 649.CrossRefGoogle Scholar

¹⁵ It should be noted, however, that a much smaller proportion of the respondents (56%) believed that the United States should operate with a scaled-down conception of its leadership role (see Table 7).

¹⁶ On January 26, 1976, two hundred members of the House of Representatives signed a resolution opposing American aid to Angola, and on February 18 the Senate passed a foreign aid bill that included a prohibition of covert military aid to any forces in Angola. The Senate vote was 60–30. Our questionnaires were distributed during the first week of February.

¹⁷ The Supporters appear to have been more consistent than the Critics in appraising Soviet goals and Soviet uses of détente. As noted earlier, however, this may be the result of ambiguities about the meaning of the term.

¹⁸ The overwhelmingly favorable response to this proposition perhaps confirms, in part, another of the lessons (discussed below): “The American people lack the patience for foreign policy undertakings that offer little prospect for success in the short run.”

¹⁹ A preliminary analysis of responses to the sources of failure and the lessons of Vietnam according to occupational groups also revealed a somewhat mixed pattern on the items dealing with the proper role of the United States in the world. See Holsti, and Rosenau, , “The Meaning of Vietnam: Belief Systems of American Leaders,” International Journal, XXXII (Summer 1977), 471.Google Scholar

²⁰ A number of prominent academic observers had proposed that such an intervention would be feasible, as well as necessary to avoid the costs of any future oil embargo. See Tucker, Robert W., “Oil: The Issue of American Intervention,” Commentary, Vol. 59 (January 1975), 21–31Google Scholar; Tucker, , “Further Reflections on Oil and Force,” Commentary, Vol. 59 (March 1975), 45–56Google Scholar; Friedland, Edward, Seabury, Paul, and Wildavsky, Aaron, The Great Detente Disaster: Oil and the Decline of American Foreign Policy (New York: Basic Books 1975).Google Scholar

²¹ The terms “grand design,” “grand strategy,” and “tactics” are borrowed from Alexander L. George of Stanford University.

²² We do not wish to imply, however, that analyses of these and other background attributes would not contribute to an understanding of policy preferences toward Vietnam, as well as of other isssues. Presumably such analyses would enhance comprehension, but perforce they must await treatment in future papers.

²³ Holsti and Rosenau, “Does Where You Stand Depend on When You Were Born? The Impact of Generation on Post-Vietnam Foreign Policy Beliefs,” Public Opinion Quarterly, forthcoming.

²⁴ For a similar conclusion based on different kinds of data, see Chace, James, “Is a Foreign Policy Consensus Possible?” Foreign Affairs, Vol. 57 (Fall 1978), 1–16.CrossRefGoogle Scholar

²⁵ That is not to say that belief systems only change suddenly and as a result of great calamities. Alexander L. George has shown that the beliefs of Harry Truman and many of his advisers during the early postwar years changed gradually, and initially at the level of beliefs about tactics—for example, on how best to secure Soviet postwar cooperation. It was only many months after the end of the war that Truman and his advisers came to question and then discard some of the more optimistic elements of Franklin D. Roosevelt's “grand design” for postwar cooperation among the United States, Russia, England, and China. See Alexander L. George, “The Role of Cognitive Beliefs in the Legitimation of a Long-Range Foreign Policy: The Case of F. D. Roosevelt's Plan for Postwar Cooperation with the Soviet Union” (paper presented at the Conference on Approaches to Decision-Making, Oslo, August 9–12, 1977).