By selecting Theodore J. Lowi to serve as its President, the American Political Science Association has recognized his extraordinary scholarly and professional contributions. Indeed, Lowi's work has played a singularly important role in the intellectual development of the profession during the past two decades. He was recognized in an American Political Science Association survey as the most significant political scientist of the 1970s, and continues in every study to be the single most frequently cited member of the profession. At the same time, as a truly gifted teacher, Lowi has inspired large numbers of graduate and undergraduate students, many of whom have followed in his professional footsteps.

Lowi was born in Gadsden, Alabama in 1931. Much of his family lived in Marks, Mississippi (a city named for Lowi's great-uncle), and Lowi may be said to have considered—but rejected—Marksism in his youth. He chose, instead, to enter Michigan State University on a music scholarship. At Michigan State, Lowi played the oboe in both the marching band and orchestra before receiving his B.A.

The End of Liberalism is an ambitious, engaging, and intellectually provocative work even if it is also occasionally ambiguous, ahistorical, assuming, and utopian. This work attempts to build a theory of the modern American political regime which captures its complexity, inherent contradictions, institutional dilemmas, and self-destructive public philosophy. Since America has yet to abandon the politics and public philosophy of interest group liberalism, Theodore Lowi's (1969 and 1979) polemic remains a useful and now comfortable perch from which to observe the continuing crisis of public authority and shout appropriate epitaphs. There is little doubt that The End of Liberalism continues to stimulate, define, and inform intellectual and political debate in America 20 years after its initial publication.

There are two ironies about The End of Liberalism which are now apparent two decades after its publication. First, if Lowi's work had been completely successful in laying to rest the demon of interest group liberalism and its spiritual compatriots, we might not be assessing the longevity and influence of this work two decades later.

The End of Liberalism by Theodore Lowi was and is an important book, large in conception, bold in execution, and substantial in influence. It is a provocative and influential statement of some of the problems of pluralism, or “interest-group liberalism.” One of those problems is the inadequately constrained delegation of legislative problems to administrative agencies, leading to “the atrophy of institutions of popular control” (p. 86). The atrophy occurs because administrators exercise their discretion in ways set by agency-group relationships rather than by popular choice (p. 90). To solve this problem, Lowi calls for a return to “juridical democracy,” which means, in part, limiting federal action “to those practices for which it is possible to develop a clear and authoritative rule of law, enacted democratically and implemented absolutely” (p. 271). One way of achieving that limit is by asking the Supreme Court to declare “invalid and unconstitutional any delegation of power to an administrative agency that is not accompanied by clear standards of implementation” (p. 298).

Earlier this had been the view of Albert V. Dicey, the great British legal scholar. Dicey argued against the emerging administrative law on grounds that, by removing the defense of individual rights from the common courts, it would permit the growth of the welfare state. Lowi does not mention Dicey, and there is no indication that he is opposed to the substance of the welfare state.

Why is The End of Liberalism important? This work is not without its flaws or critics, but it is indisputably important—not because it is right, although I believe it largely is—but because it is provocative, stimulating, and pertinent to both scholai and practitioner.

The Culprit

One of the more revealing and less pedantic kiss-and-tell books to come out of a presidential administration was David Stockman's (1987) The Triumph of Politics. In it, Stockman detailed his dismay over the failure of the Reagan Revolution—an effort to establish “minimalist government” (p. 9) incorporating “sweeping, wrenching change” that “would have hurt millions of people” (p. 11) by such actions as “elimination of subsidies to farmers and businesses … an immediate end to welfare for the able-bodied poor … [and] no right to draw more from the Social Security fund than retirees had actually contributed” (p. 12).

Lowi, like Gaul, can be divided into three parts: (1) the policy analysis (arenas of power) scheme; (2) democratic theory (juridical democracy, the critique of pluralism, rule of law); and (3) constitutional advocacy centering on designs for parliamentary government. It is in the second category, I believe, that he has made the most significant and lasting contribution.

The first and earliest segment, the policy analysis scheme, was developed in articles, in and after the seminal 1964 book review in World Politics (Lowi 1964, 1972, 1976). Lowi's students have directly employed the policy typology in work on national institutions and policy development (Spitzer 1983; Sanders 1981). However, Lowi himself has not fully developed the implications of the typology in such a way that it could generate an integrated set of testable hypotheses and has failed to clear up essential definitions. The words used to label the four cells of the typology (distributive, regulatory, redistributive, and constituent) have fallen into common currency, but are used in a muddle of ways not consistent with the meanings I believe (as an alumna of Lowi's graduate public policy workshop) that he wants to attach to them. Much of this confusion, and the failure of this set of tantalizing ideas to bear much theoretical fruit, must be laid at the creator's doorstep for neglecting to draw out at length and in detail the typology's implications.

The End of Liberalism is one of the most widely read books in political science. As book review editor of the American Political Science Review, it interests me that the reviewers of neither the first nor second edition of the book predicted its enormous success. While the reviews were clearly positive in tone, they were far from effusive, and none suggested that the book would become required reading for a generation of political science students. My own copy of the book is annotated by three sets of underlines and comments—my own and those of my two daughters who were assigned the book as undergraduates. Why has the book become so important? What can we learn from it about the ingredients of a great book? Given the broad readership that has responded to Lowi's book, how relevant are the criteria that political scientists use to evaluate their own literature? What qualities should disciplinary reviewers watch for if they hope to recognize what will become a classic?

In the December 1989 issue of PS, Thomas Dye and Harmon Zeigler argue that socialist nations are inherently more militaristic than capitalist ones because they are “driven by the bureaucratic imperatives of a society organized on collective coercive principles.” In contrast, “capitalist states, whether authoritarian or democratic, rely principally upon markets to organize societal activity. Markets organize people for private enterprise and rely upon voluntary exchange to carry on these enterprises” therefore, capitalist nations are inherently less militaristic (Dye and Zeigler 1989: 812). Below are some major differences I have with their position.

Socialism is not merely a way of organizing an economy; it is also a secular religion. And as a religion it cannot be falsified; it cannot be dis-confirmed by empirical evidence in the minds of its adherents. Non-believers might think that the spectacular collapse of socialism around the world—its demonstrated unpopularity and illegitimacy—would impress Marxists in American universities. But academic Marxists remain insulated from world events, spinning out theoretical webs undisturbed by critical evidence.

Recent events provide more dramatic evidence of the centrality of military force to socialist regimes. The collapse of socialism in Eastern Europe came about as a direct result of Mikhail Gorbachev's decision not to use military force to sustain these regimes; socialism survives in the People's Republic of China only because of the violent repression of Tiananmen Square.

Professors Dye and Zeigler (D&Z) take the easy way out. They invite the reader to think of their present critic as one of those “academic Marxists” in the grip of a “secular religion” “insulated from world events, spinning out theoretical webs.” This approach enables them to rest secure in their ideological assumption that, while their critic is a hopeless ideologue, they themselves are value-free social scientists arguing closely from evidence.

Yet they continue to ignore evidence they do not find agreeable agreeable—as with the numerous examples of capitalist militarism and aggression I proffered. I noted other variables that could measure militarism, including treaties, overseas bases, striking power, mobility, levels of technology, etc. Given this opportunity to respond to these specifics, D&Z choose not to. One is tempted to conclude that they fail to answer because they have no answers.

The management of large amounts of data is an important part of many research efforts. Quantitative data are inputted, stored, manipulated, and analyzed using such familiar tools as spreadsheets, database management programs like dBASE or Paradox, and statistics packages. However, some text based data such as bibliographies, text for content analysis, committee hearing transcripts, and historical information are not easily processed using such software. Statistics packages can handle only single words as string variables, and spreadsheets can usefully manage only short segments of text. Traditional database management programs can store and manipulate longer text segments, but many are limited to roughly three lines of text per entry or field unless they have a feature called a memo field to be discussed below.

When we covered this topic for PS three years ago, we concluded that high end word processing packages should be considered first by most scholars who work with text data. We noted that advanced word processors contain search and sort capabilities that may be sufficient for many needs.