How rational is "the rational public"? Evidence from U.S. public opinion on military spending.

In the last decade, there has been a revolution in the scholarly evaluation of U.S. public opinion on foreign policy. The traditional view held that most Americans react to foreign issues in an emotional and ill-informed way, creating the potential for public opinion to hinder pursuit of the national interest. This perspective has been vigorously challenged in recent research, and a new, revisionist consensus has emerged. At the most positive end of the revisionist work, Page and Shapiro even use the phrase "the rational public" to describe the public's capabilities (Shapiro and Page 1988; Page and Shapiro 1992). The revisionists have put forward several different definitions of what they mean by rationality, but the most common claims are that public judgments are "sensible" or "reasonable" (Nincic 1992; Hinckley 1992, 131).

The new consensus is premature. The revisionists have not yet addressed all of the key claims in the classical literature. Indeed, some of the evidence presented in defense of the rational public perspective is not clearly inconsistent with the observations used by an earlier generation of commentators to denounce public opinion. Because the classical view has not been fully tested, it cannot be rejected unless more analysis is done. This article begins that analysis by examining public opinion about military spending during the cold war. On balance, the results are more favorable to the rational public interpretation. Some caveats, however, suggest a need for further tests before researchers can confidently dismiss the dangers that worried the traditionalists.

THE CLASSICAL VIEW

The essence of the traditional critique of U.S. public opinion on foreign policy is contained in three widely quoted passages. I reproduce them here to make clear that they are the basis for my depiction of the classical perspective. The first comes from George Kennan ([1951] 1984):

I sometimes wonder whether... a democracy is not uncomfortably similar to

one of those prehistoric monsters with a body as long as this room and a

brain the size of a pin: he lies there in his comfortable primeval mud and

pays little attention to his environment; he is slow to wrath--in fact,

you practically have to whack his tail off to make him aware that his

interests are being disturbed; but, once he grasps this, he lays about him

with such blind determination that he not only destroys his adversary but

largely wrecks his native habitat. (p. 66)

Although this passage does not explicitly mention public opinion, further discussion (pp. 72-73) makes it clear that this is what Kennan saw as the root of the problem.

The second passage is from Walter Lippmann's Essays in the Public Philosophy (1955):

The unhappy truth is that the prevailing public opinion has been

destructively wrong at the critical junctures. The people have imposed a

veto upon the judgments of informed and responsible officials. They have

compelled the governments, which usually knew what would have been wiser,

or was necessary, or was more expedient, to be too late with too little,

or too long with too much, too pacifist in peace and too bellicose in war,

too neutralist or appeasing in negotiation or too intransigent. (p. 20)

The final quotation comes from two passages in Gabriel Almond's The American People and Foreign Policy ([1950] 1960):

The characteristic response to questions of foreign policy is one of

indifference. A foreign policy crisis, short of war, may transform

indifference to vague apprehension, to fatalism, to anger; but the

reaction is still a mood, a superficial and fluctuating response .... This

is to say that foreign policy attitudes among most Americans lack

intellectual structure and factual content. Such superficial psychic

states are bound to be unstable. (pp. 53, 69)

These statements suggest that what has been labeled the Almond-Lippmann consensus (Holsti 1992) in fact contains two separate elements: an empirical observation and a theory of public opinion that provides a possible explanation for that observation. The empirical observation, made by Kennan and Lippmann, concerns how the public responds to foreign threats: the public typically reacts too slowly, and then overreacts. The potential explanation for the public's ills, developed by Almond, has been dubbed "the mood theory." Because the public does not understand and tends to ignore foreign policy, it prefers the United States to remain uninvolved unless danger is imminent. Then, the sudden realization of external threat induces panic, thus causing the swings between extremes observed by the authors mentioned above. The revisionist school has effectively critiqued this second element of the classical view. So far, however, it has failed to address the description that makes up the first component.

The traditional critique of the public has two different elements because it arose from two distinct sources. Kennan and Lippmann were acting as advocates for foreign policy realism in the works quoted above. The writings of such postwar realists are the main source of the view that the public responds too slowly and then too strongly to foreign threats (see also Morgenthau [1948] 1960, 146, 567).

The second body of research that contributed to the classical view came from social science studies of public opinion. Some of Lippmann's earlier work (1922, 1925) was an important forerunner of this research, so revisionists do often cite these earlier studies. However, Almond's book and a follow-up by James Rosenau (1961) were the most explicit to link indifference and lack of structure in public attitudes to dramatic swings in foreign policy. Their arguments received an important boost from Philip Converse's 1964 essay on "The Nature of Belief Systems in Mass Publics." About not just foreign but domestic policy as well, Converse found most Americans exhibited little ideological consistency across issues. He also found most Americans did not hold consistent preferences over time, leading him to dismiss much of public opinion as "nonattitudes" (Converse 1970).

The revisionist studies have focused largely on refuting the arguments of Almond and Converse. Unfortunately, this has led to a research agenda that is not well suited for the evaluation of the empirical claims of the realists. This is so for two reasons. On one hand, Almond's ([1950] 1960) mood theory is not the only possible explanation for the supposed empirical problem of opinion fluctuation. Another well-known line of thought suggests America's liberal political culture (Hartz 1955) leads to a regularly alternating cycle between global activism and withdrawal (Klingberg 1952; Holmes 1985). The existence of this ideology-based explanation means that refuting Almond's emotion-based explanation would not allow one to reject the basic observation of the classical perspective, since other theories might account for the realists' description of public responses.

On the other hand, it is also possible to derive from Almond's mood theory other empirical expectations besides the central realist observation. In fact, Almond and his successors did so, leading to a set of further propositions that sometimes differ markedly from the "too late, then too much" observation. Two of these propositions have been particular targets of the revisionists. The first is a prediction of volatility or "frequent alteration" (Almond [1950] 1960, 53); the second, an argument that public opinion is characterized by randomness or unpredictability (Converse 1964, 243; on both claims, see also Rosenau 1961, 36-37).

The key point is that these predictions are not the same as those of Kennan and Lippmann. Continuous instability is not what the realists expect. Rather, they expect some period in which the public holds (too long) to a stable preference for noninvolvement, before suddenly shifting to favor a moral crusade, which again remains its preference for some period of time beyond that which serves U.S. interests. Extreme mood swings, yes, but not the rapid fluctuation predicted by the behavioralists.

Likewise, the revisionists regularly attribute to earlier writers the view that public opinion is "fickle" (Wittkopf 1990, 219; Russett 1990, 92; Hinckley 1992, 130) or "capricious" (Page and Shapiro 1992, ch. 2). This is not an accurate characterization of the realist argument. Far from believing that opinion changes randomly, the classical writers forecast a very definite pattern. As Lippmann puts it, "The errors of public opinion in these matters have a common characteristic. The movement of opinion is slower than the movement of events" (1955, 20). This suggests opinion changes are neither random nor capricious. Instead, there is a pattern to public opinion so strong as to be easily predictable; unfortunately, it happens to be a pattern of repeated tragedy.(1)

REALIST PORTRAIT NOT ASSESSED

Within a broad traditionalist consensus there were thus some differences between foreign policy realists and social scientists studying public opinion. However, the revisionist literature has only assessed the claims put forward by the latter group. Thus, recent studies have shown that public opinion on foreign policy is actually structured by well-defined belief systems (Hurwitz and Peffley 1987; Wittkopf 1990; Hinckley 1992; Chittick, Billingsley, and Travis 1995), that voters do not remain indifferent to foreign policy when the issue is important (Aldrich, Sullivan, and Borgida 1989), and that stability of preferences rather than instability is the norm (Caspary 1970; Page and Shapiro 1992). Finally, in response to allegations that public opinion is capricious, revisionists have taken two approaches. Some studies have shown there are underlying heuristics that guide public judgments on certain foreign policy issues (Nincic 1988, 1997; Jentleson 1992). Others have sought to demonstrate that actual opinion changes can be explained as responses to relevant international developments (Page and Shapiro 1992; Peffley and Hurwitz 1992; Holsti 1996; but see Zaller 1992 for a partial defense of the Converse perspective).

These findings, although important, do not disprove the full set of charges traditionally leveled against public opinion on foreign policy. In particular, none of this research directly addresses the response pattern predicted by the classical realists. For example, Page and Shapiro (1992) show that most significant opinion shifts are preceded by real-world developments to which they are probably a reaction. The classical realists also expected opinion changes to be responses to identifiable, real-world events, however. But they predicted that the responses would come only after a dangerous delay and would be disproportionate to the triggering cause. Page and Shapiro (1992) do not consider these possibilities. In a similar vein, Yankelovich (1992; Yankelovich and Immerwahr 1994) argues that public opinion can be volatile and unreasonable when it first confronts an issue, but the public is capable of coming to a wise judgment over time. But he does not ask whether the length of time required might be problematic for U.S. diplomacy. Even the studies that identify underlying rules of thumb that govern public opinion (Nincic 1988, 1997; Jentleson 1992; Oneal, Lian, and Joyner 1996; Jentleson and Britton 1998) do not investigate whether there is any time delay before the public applies these heuristics or whether the resulting opinion change has a reasonable magnitude given the triggering cause. In sum, not a single revisionist study to date has explicitly assessed the response pattern predicted by Kennan and Lippmann. Until this is done, it is too soon to conclude that the unfavorable classical evaluation of the public's judgment should be rejected.

Evaluating the Kennan-Lippmann view is necessary, moreover, because of the research strategy the revisionists have adopted. Neither Page and Shapiro (1992) nor the other revisionist researchers actually seek to prove that public opinion fits the definition of rationality employed in decision theory or other rational actor approaches. In fact, many of the recent studies acknowledge important caveats on any positive depiction of public opinion, especially the public's extremely limited knowledge of world affairs (Jentleson 1992, 71; Holsti 1996, 215; Nincic 1997, 597). Instead of seeking to prove a positive, the revisionists are more concerned about disproving a negative. Their research strategy really aims to refute various propositions in earlier studies of public opinion that raise doubts about the public's competence in political matters. The revisionists' ultimate goal is thus to show that the public can safely be entrusted with its democratic responsibilities; by rational, the recent literature actually means trustworthy. Virtually every revisionist study, therefore, goes beyond the specific challenges to Almond, Converse, and the like. Nearly all include a section that argues that their refutations of these authors' claims mean public input into foreign policy making is not to be feared, so that democratic oversight of foreign policy can be defended (Hurwitz and Peffley 1987, 1114; Russett 1990, 115-18; Page and Shapiro 1992, ch. 10; Jentleson 1992, 50, 71; Holsti 1996, viii, 211-16).

Given the research strategy of identifying and rebutting specific propositions of earlier critics of public opinion, this larger conclusion is not yet justified (although one could of course still defend democratic participation on normative grounds). Although the revisionists have effectively challenged part of the traditional critique of public opinion, they have not challenged all the negative arguments. The research presented below seeks to be consistent with the revisionist research strategy. Thus, it does not aim to determine whether collective opinion fits one of the standard definitions of rationality. Rather, it tackles a piece of unfinished business in the revisionists' own research agenda. It takes the classical realist writings as the source of …