There are two principal approaches to understanding social structure. One approach, exemplified by social class analysis, consists of postulating an ideal model of class positions and then mapping social reality onto this model. The result is a set of structural relationships that explain, or seem to explain, the positions taken by various social groups at different times. This approach is therefore also an interpretation of politics, in the sense that actions and positions are judged relevant or irrelevant to the extent that they match or fail to match the expected behavior of social classes according to the ideal model.

This essay is devoted to a critique of the interpretation of the present Brazilian regime given in a recent work by Philippe Schmitter, an interpretation that is organized around the themes of “Bonapartism” and “restoration/Portugalization” (Schmitter 1973). The basic problem treated by means of these concepts is that of the relation between the state and social classes in Brazil. Our argument will be that Schmitter's usage rips these concepts out of their original historical context and attempts to utilize them in a context in which they do not really fit. They are therefore only of limited use for understanding the nature of the post-1964 regime, and in fact their application can easily lead us to lose sight of what is specifically new in the political economy of Brazil today. Having made our criticisms, we will advance a number of alternative hypotheses about how that regime's relation to the fundamental classes of Brazilian society should be understood.

Several years ago a single voice, representing the frustration of many of us trying to carry out research on women in Latin America, decried the “lack of core bibliography, methodological apparati, or thematic models” (item 182:125) as major problems besetting the study of women in Latin America. As of 1975, these research hurdles had not been completely overcome, but certainly steps have been and continue to be taken.

An annotated bibliography on women in Spanish America will soon be published and other bibliographic guides of a more limited scope have been or are in the process of being compiled. Both in North America and Latin America, conferences, seminars, and workshops have been held to discuss methodological problems as well as recent research efforts. Associations, committees, coalitions, and centers have organized to promote the cause, lend support, gather and distribute information, and generally represent a growing concern with women's issues.

Library cooperation on regional and statewide levels has helped to expand significantly the collections of individual research libraries through cooperative purchasing policies, cataloging networks, union lists, interlibrary loan services and other means. However, access to the contents of these materials is also essential if this vast body of information is to be utilized fully. Books on specific topics can be found with relative ease by searching library subject catalogs, printed bibliographies, and specialized data bases. On the other hand, the wealth of information that appears regularly in periodical literature is all but lost unless it is adequately indexed.

Increasingly, historians of and in the Latin American countries are turning to quantitative data and analysis. TePaske (1972, 1975), Smith (1973), and McGreevey (1972, 1974) comment on work that has been and is being done and on problems inherent in quantification. The problems that students face as quantifiers of the past may be summarized under the rubrics: (1) sources, (2) methodology, (3) training, and (4) financing. It is with the first of these that this article is concerned, especially with sources for quantifying the nineteenth century after independence, a period neglected almost as much as the seventeenth century used to be, at least insofar as the smaller countries are concerned, except for their politics and personages.