Several years ago Professor Robert Dahl argued that the traditional concern over the Supreme Court's power of judicial review was largely unfounded. Dahl demonstrated that seldom, if ever, had the Court been successful in blocking the will of a law-making majority. This paper argues that, had Dahl considered his data from a different perspective, he would have discovered that, by virtue of the recruitment process, the Court will rarely even attempt to thwart a law-making majority. Examining Dahl's data in the context of the Survey Research Center's election classification scheme, the paper focuses on the Court's relation to patterns of partisan change to show that the traditional philosophic concern with the counter-majoritarian nature of judicial review is largely divorced from empirical reality and has relevance only during periods of partisan realignment within the political system as a whole. The paper buttresses the argument that the Court's “yea-saying” power is more important than its “nay-saying” power, a realization which can serve as the premise from which a logically consistent justification of the Court's power of judicial review may be dialectically constructed.

An explanatory model for the outcomes of midterm congressional elections is developed. Midterms are a referendum on the performance of the President and his administration's management of the economy. The explanatory model of midterm congressional elections is sufficiently powerful so as to yield honest and accurate pre-election predictions of the national two-party vote in midterm elections. These predictions have usually outperformed pre-election forecasts based on survey data. The model is extended by considering the translation of votes into seats, models of the electorate as a whole and of the individual voter, and the causes of the off-year loss by the President's party.

A common hypothesis about the sources of international conflict holds that war and turmoil will be an inevitable consequence of the widening “gap” between the developed and underdeveloped states. This view is based on a common Western image of underdeveloped countries which assumes that striving for economic betterment is universal in all underdeveloped countries, and is primarily a grass-roots phenomenon. This essay challenges the hypothesis and the assumptions upon which it is based. It argues that the images of underdevelopment generated by economists using aggregate data are in many cases incorrect or distorted. Studies by anthropologists which are based on micro- rather than macrodata produce quite different impressions of the underdeveloped society. The human costs involved are for the most part overlooked in development schemes, and the wholesale importation of Western economic development strategies has led in many cases not only to a poor allocation of resources, but also to many of the problems the developed societies are now facing, including urban congestion, rising crime rates, higher incidence of mental breakdown, and the like.

The paper concludes with a critical review of common liberal solutions to development problems, and suggests that one strategy possible for some developing countries is increased isolation from the international system. International conflict may result not only because the underdeveloped states wish to close the “gap,” but because some may choose deliberately to reduce their dependency on the West. Conflict may be generated through isolation as well as through increased interaction.

Positive or rational choice theorists have tended to suppress under the rubric of “winning” elections a critical distinction in ths goals of political parties (or candidates)—the distinction between the primary goal of office and the goal of the benefits derived from the control of office. The distinction, however, has strategic consequences. Logically, the office-seeker should follow the vote-maximization strategy put forth by Downs, whereas the benefit-seeker should find Riker's minimal winning coalition most congenial. The distinction in goals and strategies also implies divergent ways of organizing political parties. A concern for benefits logically leads to the development of structures designed to insure that the party's officeholders will deliver the desired benefits. The office-seeking goal implies structures which free the party and the office-seeker to maneuver in response to electoral needs. Thus there are two positive theories resting upon two primary political goals. In their differences we find an explanation of the tensions in democratic parties.

Following Walker, investigators have assumed that state executives frequently seek policy guidance from other states. This study expands the “diffusion of innovation” literature by demonstrating that levels of elite consensus about which American states have the better agencies vary by policy area. It is hypothesized that state administrators in policy areas characterized by general agreement about the better state programs are more influential in their own state's policy-making process than are state administrators in policy areas where there are no acknowledged leaders. Data gathered by mail questionnaire from upper-level state executives in ten American states provide support for the hypothesis. Finally, results from state expanditure studies, which also indicate that different mixes of actors are influential in state policy making depending upon the policy area involved, are found to be consistent with this interpretation.

This paper is a partial systematic test of Morton A. Kaplan's “theory” of alliance behavior in balance of power international systems first proposed in his well-known System and Process in International Politics (1957). Three hypotheses are inferred from Kaplan's writings predicting that in a stable balance of power system, (a) alliances will occur randomly with respect to time; (b) the time intervals between alliances will also be randomly distributed; and (c) a decline in systemic alliance formation rates precedes system changing events, such as general war. We check these hypotheses by applying probability theory, specifically a Poisson model, to the analysis of new data on fifty-five alliances among the five major European powers during the period 1814–1914. Because our research questions are so general, our findings should not be regarded as definitive; however, the data very strongly support our hypotheses. We conclude that Kaplan's verbal model of a balance of power international system has had its credibility enhanced as a result of this paper.

Cross-national research is plagued by several methodological problems that threaten to distort results and hence raise questions concerning the adequacy of substantive findings. The extent and impact of three of these methodological problems—measurement, sampling, and specification errors—are assessed for a recent model of civil conflict developed by Gurr and Duvall. Concepts in their model are measured with two distinct data sets to estimate measurement error; to assess sampling error, measures are applied to a sample of black African nations which were excluded from their analysis; and new concepts are introduced to assess error in theoretical specification. Although all forms of error are found to exist in their work and to have some distorting effects on their model, it is concluded that some of the general model is accurate. However, discovery of (1) the concepts and data sets that contain the greatest error, and (2) the sources of sampling and specification error, is used to improve substantive findings about the causes of manifest political conflict.

The concept “nondecision” was advanced by Bachrach and Baratz as a means of identifying certain areas of community power neglected by reputational and issue analysis approaches. While it is descriptively suggestive of certain possible areas of neglect, it has not been shaped into a useful analytic tool, and does not make any demands which cannot be met by decision-making analysis. Its terms are, in fact, somewhat confused since it fails to differentiate between nondecisions brought about by covert control, and those which may be more generally atlributable to a mobilization of bias. These together seem to comprise what Bachrach and Baratz describe as the “second face of power,” which has been neglected, they argue, because of inadequate consideration of the meaning of power and related concepts. But their own approach to these does not aid empirical analysis. It simply encourages concern for the minutiae of political action, whereas the simplified view of power adopted here suggests the advantages of a contextual approach.

In his Economic Theory of Democracy, Anthony Downs reached the conclusion that voting is usually an irrational act. Given that voting is costly in time and effort involved in getting to the polls and making a decision and given that the probability any one voter can affect the outcome diminishes very rapidly as the number of voters increases, it appears that a rational, utility-maximizing citizen will most probably abstain. Only if the citizen perceives a phenomenally large difference between (or among) the candidates or if the number of voters is small will he go to the polls.

Empirically, however, we know that substantially more than half of all eligible voters go to the polls in American presidential elections where the probability of any one voter's affecting the outcome is smaller than in any other election. If Downs is correct, at least half of the American electorate is irrational. But of what validity is a rational choice theory which characterizes most individuals as irrational? For this reason, theorists have attempted to modify Downs's theory so as to allow it to better explain why some rational citizens vote while others abstain.

It is continually demonstrated in the pages of the Review that formal theorists in political science have great difficulty in communicating with one another, in reaching consensus, e.g., on the rationality of minimal winning coalitions and the effects of vote-trading. It has always puzzled me how such confusion and controversy could surround simple arguments which are—or at least appear to be—mathematical. Reading the extraordinarily clear contribution of Ferejohn and Fiorina (“The Paradox of Not Voting: A Decision Theoretic Analysis,” APSR 68 [June, 1974], 525–536), I am led to the conjecture that a sort of natural selection is involved, that the more lucid papers are so patently silly that many of them fail to achieve publication, leaving the field to papers in which the nonsense is at least obscure.

In a recent article (“The Paradox of Not Voting: A Decision Theoretic Analysis,” APSR, 68 [June, 1974], 525–536) John Ferejohn and Morris Fiorina suggest that a (potential) voter may be rational in deciding whether to vote and yet not use the expected-utility calculus introduced for this purpose by Downs and Tullock and extended by Riker and Ordeshook. The authors suggest that a rational voter, in deciding whether to vote, may use the minimax regret criterion due to the late L. J. Savage. Define the regret of an Act A as the difference, if the true state of nature were known (in advance), between the best expected payoff that could be obtained and the expected payoff that would be obtained by act A. Then the minimax regret criterion requires the decision maker to act in such a way as to minimize the maximum regret that he can suffer.

They also discuss the fact that the minimax regret criterion does not require the voter to supply any (objective or subjective) estimates of the probabilities associated with the various outcomes of the election. Thus they feel that it is an attractive criterion for decision making under “uncertainty” as opposed to decision making under “risk.”

The introduction of decision making under uncertainty by Ferejohn and Fiorina is an interesting addition to the literature on rational theories of citizen participation. Decision making under uncertainty assumes, however, that the actor has no knowledge of the probabilities of the various outcomes; this is obviously no more true than the assumption of perfect information about these probabilities made in the decision making under risk model. Voters have some, but not perfect, information about the probabilities of at least some of the different possible outcomes.

Specifically, let us look at the two-party case. In Ferejohn's notation, (p3 + p4) is the probability of an individual's vote making a difference. We might expect a rational citizen to know that this probability is at most minuscule, even if he cannot calculate its exact value.

In most states it is permitted to write in the names on the ballot, and these write-in candidates can win if they assemble enough votes. Thus, in all elections in which I might conceivably participate, there are at least three candidates: the Republicans, the Democrats, and my own favorite candidate for president—Gordon Tullock. Indeed, using the notation of the Ferejohn and Fiorina article, assuming that we are talking about 1972, I would find k having a value of about .001. Under the circumstances, if I understand the article correctly, I should always vote, and the vote should always take the form of writing in my own name. Further, as far as I can see, this advice can be generalized. Everyone who would really like to be president should vote and write in his own name, because the minimum regret that they produce for three-candidate elections is also correct for the elections in which there are 30 million candidates. The only problem here is that it would, of course, amount to participating in a lottery, and my possible gain from writing in my own name might turn out to be less than the cost of writing it in. Granted that people are willing to buy lottery tickets, even when the various states offering them take very substantial rakeoffs, I do not imagine this would be much of a disadvantage.

This same conclusion could have been deduced from the Casstevens article, which Ferejohn and Fiorina quite properly criticize as being mathematically incorrect.

During the period in which our article (APSR vol. 68 [June 1974]) circulated in manuscript form it provoked an unusual amount of collegial reaction. Of course, we were quite prepared for a reaction from those who use decision-theoretic models in their research—they were our intended audience. More surprisingly, we also received comments from less directly involved bystanders—a medieval historian for example. All this correspondence indicates to us that nearly everyone has his own theory of how voters behave, and that most such theories do not agree with the one presented in our article. The comments of Professors Tullock, Beck, Mayer and Good, and Stephens further support this conclusion.

In an appendix to this note we have responded to the imaginative point raised by Tullock. As for the traditional questions raised by our other critics, however, we adopt a different line of rebuttal. Rather than conduct an unfruitful debate over the a priori plausibility of the minimax regret model we will do something that theorists too seldom do: examine some data. Before doing so we will make an important distinction between using a model prescriptively and using it descriptively. (Decision-theoretic types tend to move a bit too easily from one usage to the other.) Then, after reviewing the major point of our article we will turn to the data.

No rational egoist should bother voting because, as Skinner's Dr. Frazier notices, the probability of any one man casting the decisive ballot “is less than the chance he will be killed on the way to the polls.” No matter how deeply he cares about the electoral outcome, a man must realize that his vote is only one among very many. The larger the electorate, the smaller the probability of any one vote's changing the outcome, so in most modern polities the politically rational thing to do is to conserve on shoe leather.

Real-world voters, of course, do flock to the polls, which is usually explained in terms of a feeling of “civic duty.” The fact that men get some satisfaction from discharging their civic duty by voting might answer the question of why it is rational for them to go to the polls. Unfortunately, it leaves another—how do they vote once they get there? Presumably, they go ahead and vote according to the dictates of their egoistic interests.

In an article, “The Paradox of Vote Trading,” (APSR 67 [December, 1973]) William H. Riker and Steven J. Brams have argued that systematic logrolling among all members of a legislature produces a paradox: While each trade is individually rational, the effects of externalities offset the potential gains from exchanging votes and each voter finds himself worse off than he would have been by voting sincerely. We extend the results of Riker and Brams to a unanimity decision rule and find that a paradox of vote trading holds for that decision rule as well as for simple majority rule. Under a unanimity rule, however, trades which would be collectively rational (i.e., which would produce a Pareto optimal result) are not individually rational; the non-trader is the beneficiary under such a decision rule. Finally, we pose the question Riker and Brams suggested: Is the paradox of vote trading inescapable? Except under very restrictive conditions, we find that it is. However, given certain assumptions about the distributions of individual utilities, we present proofs of the necessary and sufficient conditions for the Pareto optimality of vote trading and argue that in actual legislative situations, when vote trading is Pareto optimal, learning behavior should serve to extricate the members from the paradox of vote trading.

This paper examines William Riker's thesis that only minimum winning coalitions form in n-person zero-sum symmetric games. It demonstrates that Riker's conclusion is false by identifying the conditions under which larger than minimum winning coalitions can form. Since these conditions are quite general it indicates that Riker's conclusion is valid only for a highly restricted class of games. This class of games is identified as those in which players not in a minimum winning coalition have no incentive to form any coalitions among themselves. These games are characterized as games inessential over coalitions of losers. Only in these games can minimum winning coalitions be expected to form exclusively. In all other games, larger than minimum winning coalitions are possible.

The Shapley value of the presidential election “game” is approximated by the method of multilinear extensions; the likely error in this approximation is computed by studying the error in the electoral college game. A comparison with the Banzhaf ratio is also given. The several biases inherent in the electoral system are analyzed from a mathematical point of view. It is shown that the system incorporates a strong bias in favor of the larger states, and a small bias in favor of the smallest states.