INTRODUCTION

This collection brings together eleven essays written over the past fifteen years or so that either reflect on the current practice of comparative history or represent my own efforts to do this kind of work. The latter essays emerged from a topical interest in racism and antiracism that antedates my explicitly comparative scholarship. Most concern black-white relations in the United States and South Africa; but one, published here for the first time as chapter 6, deals with mid-nineteenth-century ideas about race and empire in Alexis de Tocqueville's writings on American democracy and the French colonization of Algeria. Earlier versions of the piece, called "Understanding Racism" (chapter 5), which attempted to define and historicize racism in a general way, appeared in places where few readers are likely to have seen them; this essay has been substantially expanded and reformulated.^[1]

This essay has a complicated history and has been in a constant state of revision for the past seven years. It originated as an address to a plenary session of the American Anthropological Association in November 1989. I was asked to speak generally about the historical meaning of racism, an invitation that forced me to think in broader terms than I had previously found necessary. Shortly thereafter I presented it as a paper at a Newberry Library seminar on race in American history. Publication was supposed to follow, but the projected book of contributions to the seminar never made it into print. In June 1991 I presented a revised version at an international conference on racism in Paris. The proceedings of this conference were published in French, and my paper appeared as "Une histoire comparée du racisme: réflexions générales" in Racisme et modernité, ed. Michel Wieviorka (Paris, 1993), 42-53. I gave it in a slightly different form as the keynote address to the first meeting of the American Studies Association of Southern Africa in June 1992 at Pietermaritzburg, South Africa. The USIA, which had helped to sponsor the conference, circulated the proceedings in mimeographed form. I made a lot of copies and gave them to anyone who was interested. Finally, the same text was published as "Reflections on the Comparative History and Sociology of Racism" in a remarkable little magazine devoted to antiracist opinion and scholarship—Race Traitor, no. 3 (Spring 1994), 83-98. When I came to revise it for this volume, I expanded it considerably in order to clarify and refine the argument and make it reflect my current thinking about race and racism as fully as possible.

The other essays either were published before 1997 or were originally prepared for symposia or edited books of essays and may appear in slightly different form in the resulting volumes. Except in the case of "From Black Power to Black Consciousness" (chapter 11), which has been condensed from a longer piece, these essays have not been significantly revised. Either they are recent enough to reflect my current thinking, or, as in the case of the survey of comparative historiography from 1980 (chapter 1), an essay is juxtaposed with a later essay of similar scope (chapter 3) to show how my assessment

of the field has evolved in light of subsequent scholarship, including my own.

Introducing this volume prompts some autobiographical reflections. Comparative historians are still rare enough to evoke curiosity about how they came to embrace that role. When I am asked, as I often am, how I first got interested in comparative history, I am never able to provide a very good answer because I cannot remember a time when I did not take a comparative approach to almost everything I was thinking seriously about. I instinctively assimilate new experience by first asking what familiar thing it resembles and then how it differs. Everyone does this to some extent, but in my case the urge to compare things may be especially intense and possibly a bit obsessive. Perhaps my comparative bent stems ultimately from the fact that my family made several major moves when I was child, and I was always contrasting where I was with where I had previously been.

As an undergraduate majoring in History and Literature at Harvard, I concentrated on what amounted to American studies, partly because at that time I did not have enough French to pursue my original plan to study the United States, England, and France since 1800. (I had the misfortune to attend a public high school that offered neither French nor German.) When I came to write a senior honors thesis, I chose a topic with a comparative dimension—an investigation of Norwegian-American writing that compelled me to assess the Old World background as well as the New World experiences of an articulate group of first- and second-generation immigrants. A chapter on the ethnic dimension of Thorstein Veblen's thought was later published in The American Quarterly .^[2]

George M. Fredrickson, "Thorstein Veblen: The Last Viking," American Quarterly 11 (1959), 403-415.

At this point I was uncertain whether to pursue historical or literary studies in graduate school. A Fulbright to Norway that was supposed to involve a further study of immigration history actually became an excursion into comparative literature. The fruit of this year was a published article on Ibsen's Hedda Gabler that compared the ideas at work in the play with Nietzsche's "neopagan" ethics. I also attempted a comparative study of Ibsen and Melville based on their shared experience when, as children, both writers suffered a loss of status and security when their fathers went bankrupt. My enterprise also failed, but its insolvency taught me the valuable lesson that comparison must be based on substantive rather than incidental or superficial similarities.^[3]

George M. Fredrickson, "Ibsen's Antichrist: A Study of Hedda Gabler," Symposium 12 (1958), 117-132.

Although it is clear to me now that my interest in literature was historical rather than aesthetic, I seriously considered doing graduate

work in comparative literature. But I was again foiled by language requirements. Although I now had some French and Norwegian, the latter was not in demand, and I therefore lacked the second "major" European language and the Greek or Latin that was then also required. I decided eventually on the American Civilization program at Harvard because it allowed me to defer the choice between literary and historical studies and also seemed to have a comparative aspect in that secondary fields were required in the history or literature of two other societies. It soon became apparent, however, that this requirement was pro forma and not meant as an invitation to sustained comparative work. I therefore became a thoroughgoing Americanist and put my cross-national interests on the back burner: my dissertation was on that seemingly most American of subjects—the Civil War.^[4]

George M. Fredrickson, The Inner Civil War: Northern Intellectuals and the Crisis of the Union (New York, 1965; second ed., Urbana, 1993).

My one significant foray into comparative history as a graduate student occurred when I took Louis Hartz's seminar on comparative political thought. In it Hartz expounded at length the ideas that would inform his Founding of New Societies. It was my first exposure to a frame of reference that would include both South Africa and the United States. My seminar paper, however, compared the Utopian socialism of Edward Bellamy and William Morris on the very Hartzian basis that the lack of a tradition of communalism handed down from the Middle Ages made American radical thought more individualistic than the English equivalent. There may have been some validity to the argument, but I am glad that I did not attempt to publish this first attempt at comparative history. I had made the mistake of taking Hartz's concept of American exceptionalism as an established truth to be applied and exemplified rather than as a debatable hypothesis to be tested.^[5]

I note in chapter 3 that a reading of Hartz's The Founding of New Societies: Studies in the History of the United States, Latin America, South Africa, Canada, and Australia (New York, 1964) first piqued my interest in comparisons of the United States and South Africa. The book included essays by specialists on each of the cases treated, and Leonard Thompson's piece on South Africa introduced me to aspects of that country's history that lent themselves to comparison with the United States. But, as indicated later in this introduction, it was Pierre van den Berghe's Race and Racism that brought South African white supremacy into my field of vision in a way that compelled me to think seriously about it. My current views on American exceptionalism can also be found in chapter 3.

When I left graduate school, I had seemingly found a professional identity as an intellectual historian of nineteenth-century America with a special interest in the issues raised by the North–South conflict and the Civil War. My second book concentrated on the great unresolved issue arising from the war—the place of emancipated blacks in the American republic. What turned out to be a lifelong commitment to the study of race and racism was driven by the strong emotions aroused in me by the civil rights struggle of the 1960s and its disappointing aftermath. White supremacy struck me then—and still strikes me today—as America's most persistent problem and its greatest shame, the most glaring and lamentable contradiction of what were supposed to be its fundamental values. I resolved to probe the intellectual and ideological foundations of what Gunnar Myrdal called "the American dilemma."

This field of study also invited comparison with slavery and race relations in other societies.

The Black Image in the White Mind , published in 1971, was not comparative in any sustained way. But a reading of sociologist Pierre van den Berghe's incisive cross-cultural study Race and Racism provided me with the concept of "Herrenvolk democracy," an illuminating and provocative key to the question of how white Americans, particularly defenders of slavery and segregation, had managed to resolve in their own minds the conflict between the egalitarian values of the Declaration of Independence and the practice of racial domination. I was struck by the fact that van den Berghe had found a similar pattern in South Africa, and at some point the thought occurred to me that a full historical comparison of the development of white supremacist attitudes, ideologies, and policies in the two countries might prove valuable.^[6]

George M. Fredrickson, The Black Image in the White Mind: The Debate on Afro-American Character and Destiny, 1817-1914 (New York, 1971), 61; Pierre L. van den Berghe, Race and Racism: A Comparative Perspective (New York, 1967).

The decision to devote nearly a decade to this project came only after I had been persuaded to give a paper comparing white supremacy in the United States and South Africa at a session on comparative race relations at the 1972 convention of the American Historical Association. As I plunged into South African historiography, I became fascinated with the extent to which the two national experiences were similar in some respects and different in others. I soon realized that a conference paper would barely scratch the surface of the subject and that I would have to undertake research in South Africa to do it justice. After two years of exploring South African history in the incomparable Herskovitz collection at the Northwestern University Library, I went to South Africa for five months in 1975. I had temporarily narrowed the subject to race relations in the U.S. South (hereafter referred to as "the South") and the Cape Colony of South Africa before 1910, and I therefore spent most of that time doing research in the libraries and archives of Cape Town. The parallels between the growth of white supremacy in the South and the Cape of Good Hope in the seventeenth, eighteenth, and nineteenth centuries were quite strong, and the comparisons could be controlled and precise. As a general rule, comparison works best when the two cases being considered show a demonstrably high degree of similarity.

The first chapter of White Supremacy , which deals with the expansion of white settlement in the South and the Cape Colony before 1840, would have fit well into the study as I conceived it in 1975 (as would the subsequent two chapters on slavery and race mixture). Since chapter 1 deals with the relations of European settlers to indigenous peoples,

its American component addresses the experience of Native Americans rather than that of African Americans. I compared the process leading to the forced removal of the "civilized tribes" from the southern states in the 1830s with the events that led up to the Great Trek in South Africa during the same decade. Although I did not pursue the history of Native American–white relations in the remainder of the book, this research and writing aroused my interest in the comparative study of frontiers, which I addressed briefly in the 1980 survey of comparative history. It also formed the subject of a review essay on recent work on American and South African frontiers, published in 1982, which appears as chapter 2 of this collection. Not a great deal has been done on this subject since, except for James Gump's illuminating comparison of the context and consequences of frontier wars in the 1870s—the conflict between the Sioux and the United States Army on the Great Plains and the clash of the Zulu and British regulars in eastern South Africa.^[7]

George M. Fredrickson, White Supremacy: A Comparative Study in American and South African History (New York, 1981), 3-53; James O. Gump, The Dust Rose Like Smoke: The Subjugation of the Zulu and the Sioux (Lincoln, Neb., 1994).

As the work on White Supremacy progressed, however, I concluded that a broader geographical and chronological canvas, one that necessarily involved some sharp contrasts, would more effectively illuminate the historical roots of contemporary U.S. and South African situations than would a more narrow focus on the South and the Cape before 1910. As a result, the last half of the book covers the United States and South Africa as a whole and extends the chronology to the present. The massacre of African teenagers at Soweto in 1976 and the murder of Steve Biko in 1978 horrified much of the world and contributed to my decision to deal more directly with the roots of the current crisis than the more limited comparison would have allowed. I wanted a chance to address more fully the origins and evolution of apartheid and to examine how it compared or contrasted with "the strange career of Jim Crow" in the United States. Although I found that the differences almost overwhelmed the commonalities, I felt that such a study in contrast was a way of countering illusions about the possibility that apartheid might be reformed out of existence, as Jim Crow had been.

In studying a subject like white supremacy it is difficult to avoid thinking about contemporary implications; thus it was a desire to be relevant and topical that, more than anything else, caused the book to overflow the dikes that I had originally constructed to keep it under control. The first three chapters, which were based in part on my intensive research in Cape Town and my prior work on racism in the United States before 1914, were written with more authority and a deeper knowledge of the sources than the last three. The book as a

whole may have been less solid and definitive than the more focused study that I had originally projected, but it was clearly more provocative and of greater general interest.

I do not regret this decision to address contemporary concerns at the expense of some historical precision and authoritativeness. The search for a "usable past" and a willingness to generalize about it in ways that subsequent research may call into question does not have to be an invitation to distortion and propaganda, provided that one is asking open-ended questions of the historical record rather than seeking evidence to support a preconceived political position and ignoring whatever contradicts it. The questions are ours and are properly influenced by what we consider important to know right now, even when the answers must be tentative or provisional. The most valuable findings are the ones that convey a realistic sense of possibilities and not false assurances that may bolster our morale while they feed our illusions. No essay in this volume demonstrates more fully my desire for a usable past than "Understanding Racism," although contemporary concerns are evident in most of the pieces—especially in my overview of American and South African freedom struggles (chapter 8) and the review essay on some black American and South African leaders of the 1960s (chapter 9).

When I started working on White Supremacy in the early 1970s, I had few models to follow, having given little thought to what comparative history was and how it should be done. As part of a generation of historians who disdained theory and simply went about their business of examining the sources and drawing conclusions from them, I went to work without worrying much about theory and method. But I did have to know what I was writing about, and I turned to the sociological literature on race and racism in search of definitions and concepts.

It was not until I had finished a first draft and began to think about an introduction that I gave serious thought to comparative methodology. What I had been doing felt right to me, but I realized that others would want to know about my procedures and assumptions. If I could not justify what I was doing, I would have to go back and rethink the whole enterprise. One conclusion that I drew was that my methodology differed significantly from that of the comparative historical studies then being produced by sociologists and political scientists. I did not begin with a well-defined model or hypothesis with a fixed number of variables. Rather I put my study in an interpretive context that historians and others had established to expose and explain the differing patterns

of race relations in the United States and Latin America. But this was merely a point of departure. I also found new themes dictated by the American and South African materials and sought variations on them. Social theory came into the picture mainly as a heuristic device whenever it seemed applicable. I described and attempted to justify this "historicist" approach to comparative studies both in the introduction to White Supremacy and in my 1980 survey of work in comparative history written for a volume summing up the recent endeavors of American historians. The latter essay is reprinted as chapter 1 of this book.

In that essay I was trying to say that comparative history can be done in a manner quite compatible with normal or traditional historical method. It can make use of narrative as well as analysis and manifest a concern for the particular as well as (if not more than) a search for general truths about human experience.

In explaining my own practice in 1980 and 1981, I may have offered too restrictive a view of comparative history. At a 1985 conference at Northwestern University I gave a paper—parts of which were eventually incorporated into the 1995 historiographic essay that became chapter 3 of this book—that distinguished between comparative history in the grand manner and other historical writing employing a comparative method. In this essay I defined the former as book-length studies of developments over time in two or more societies in which each case is treated with a similar degree of depth and specificity. Although systematic comparisons within a single nation have been made between regions, communities, and even the biographies of major historical actors or thinkers, I chose not to consider these because, as the published essay makes clear, I was really interested in "cross-national comparative history." From my perspective, historical comparison is not merely a method or procedure but also an antidote to the parochialism that may accompany a fixation on the history of one nation. For me, comparative analysis is a means of viewing history from a cosmopolitan or international perspective, which in turn makes the experience of individual nations more meaningful.

In-depth bilateral comparison is not, however, the only way to accomplish this objective. Increasingly historians have been utilizing what I would call "comparative perspective" as a tool of analysis. This approach employs general knowledge of an external example or examples of a phenomenon to determine what is characteristic or distinctive about the manifestation of that phenomenon within a single society.

This body of work is extremely valuable and undoubtedly represents the type of comparison that is most compatible with the expectation that academic historians will specialize in the past of one country. As I pointed out in the 1980 essay, such career patterns present a major obstacle to the type of cross-national comparative history that seeks to provide new insights into the history of two or more nations. This is still generally the case, although—as I show in the 1995 historiographic survey (chapter 3) and the review essay of two of the most important comparative works of the past few years (chapter 4)—the number and quality of systematic cross-national histories are on the rise.^[8]

Two good examples of the effective use of comparative perspective to illuminate one case are Eric Foner, Nothing but Freedom: Emancipation and Its Legacy (Baton Rouge, 1983), which relates American slave emancipation to a range of transformations from forced to free labor in other parts of the world; and Alan Dawley, Struggles for Justice: Social Responsibility and the Liberal State (Cambridge, Mass., 1991), which highlights special features of the rise of the American liberal state in early to mid-twentieth-century American by comparing it to the growth of state responsibilities in Germany during the same period.

I was of course an Americanist before I was a comparativist and for most professional purposes am still one. Not only do I involve the United States in whatever comparison I am attempting, but so do most of the works that I discuss in the "historiographic explorations" in part one of this book. The essays in part two, especially chapters 5 and 7, explicitly privilege the American experience over those with which it is being compared and are closer in method and spirit to "comparative perspective" than to "cross-national comparative history." "Understanding Racism" and "Black-White Relations since Emancipation" use the experience of Brazil and South Africa mainly to illuminate the American pattern rather than for their own sake, although they try to establish some general principles that would apply to all three cases and perhaps to others as well. Part three reverts to detailed bilateral comparison between the United States and South Africa in which neither case is privileged.

In the 1985 paper I also distinguished between "historicist" and "structuralist" approaches to comparative history. The former, which is closer to the normal disciplinary practice of historians, stresses differences, peculiarities, and multiple, continuous causation (i.e., that which is not easily reducible to a fixed set of variables.) It tends to focus more on culture and ideology than on institutions or structures. The aim of this approach is the better understanding of particular societies rather than the establishment or testing of universally, or at least broadly, applicable social scientific theories. The second approach tends to be more congenial to historical sociologists and political scientists. It identifies a limited number of structural or institutional variables operating in a small number of historical situations and uses comparison to isolate the ones that account for similarities and differences.^[9]

A classic example of comparative history of the second kind is Theda Skocpol, State and Social Revolutions: A Comparative Analysis of France, Russia, and China (Cambridge, U.K., 1979). For discussions of the methodology employed, see Theda Skocpol, ed., Vision and Method in Historical Sociology (Cambridge, U.K., 1984).

In the 1985 paper I stressed the gulf between the two methods and presented myself, for the most part, as a spokesman for the historicist approach. But a decade later when I revised and updated it for publication

as "From Exceptionalism to Variability," the essay included here, I determined on the basis of new comparative work, including my own, that there is a middle ground, a type of comparative history that combines elements of cultural contrast and structural analysis. I concluded therefore that the historical social scientists and the historians had a lot to learn from each other and that the work of both could and should involve the interaction between the peculiarities of culture and ideology on the one hand and recurrent and generalizable structural factors on the other.

The research and writing for Black Liberation, my second extended treatment of the United States and South Africa, made it clear that a natural and persuasive way to do comparative history was to establish a common international context and keep it in mind throughout the study. Awareness of the larger international struggle against imperialism and racism in the twentieth century permitted me to treat black resistance to white supremacy in the United States and South Africa as case studies of a more general phenomenon. The essays in part three of this collection, especially chapters 8, 10, and 11, exemplify this contextualist approach. Beyond the insights into the development of each of the societies being examined, comparative work contributes to a general understanding of the dynamics of world history.^[10]

George M. Fredrickson, Black Liberation: A Comparative History of Black Ideologies in the United States and South Africa (New York, 1995).

Although White Supremacy located American and South African developments within the larger context of the expansion of Europe, it moved quickly to American and South African peculiarities. In a piece published in my 1988 collection of essays on slavery and race, I attempted to contextualize these developments more fully by relating them to the larger history of Western colonialism and racism. In two of the essays in this book, "Understanding Racism" and "Race and Empire in Liberal Thought: The Legacy of Tocqueville," I attempt to use cross-national comparisons (my own or Tocqueville's) to exemplify some general tendencies of recent world history. I now feel much more strongly than I did that comparative history, besides showing the special features of particular national experiences, can contribute much to a broad understanding of the forces that have shaped the modern world as a whole—although the last section of the racism essay betrays again my special concern as an American to provide useful insights into our particular version of racial injustice.^[11]

See George M. Fredrickson, "Colonialism and Racism: The United States and South Africa in Comparative Perspective," in George M. Fredrickson, The Arrogance of Race: Historical Perspectives on Slavery, Racism, and Social Inequality (Middletown, Conn., 1988).

Students and others frequently ask my advice about the practice of comparative history. I have learned a few things from my twenty-five

years as a comparative historian that might be of practical help to students or established historians who are considering doing work of this type (and at the same time perhaps make it easier for serious readers to appreciate and evaluate comparative histories). First of all, it is probably advisable to master one nation's history before undertaking extensive cross-national comparisons. Unless a graduate student has prior training in a second case, say an M.A. in a national history other than the chosen area of specialization for the Ph.D., it would be prudent to choose a dissertation topic that involves at most "historical perspective" rather than full-blown bilateral comparison. But excellent comparative dissertations in the fullest sense have been written by those who have made the extra effort to complete what amounts to a double major. It can be done, but its difficulties should not be underestimated.

Besides learning a second history, would-be comparativists originally trained as conventional historians need to become interdisciplinary. They should immerse themselves in the literature on their thematic interest (race relations in my case) produced by sociologists, political scientists, anthropologists, and even those in the new and controversial field of cultural studies. One does not have to become a theorist to use theory productively as a way of developing hypotheses or provisional generalizations.

One should also be willing to learn the languages essential to indepth historical research. For White Supremacy, I taught myself a reading knowledge of Dutch and Afrikaans. For Black Liberation, however, I did not find it essential to learn any of the African languages, because English was the lingua franca of black protest in South Africa in the twentieth century. Some sociologists and political scientists have written comparative histories involving countries whose languages they do not read, but the depth of cultural penetration that sophisticated historical work normally requires obliges the comparative historian to read the languages of scholarship and public discourse employed in both cases.

I have been preoccupied with two-sided comparisons because historians can do them with the best effect. Another piece of advice, therefore, is to avoid spreading oneself too thin by trying to deal extensively and in detail with more than two cases. I once attempted a three-way comparison of emancipation in the U.S. South, the Cape Colony of South Africa, and Jamaica in an essay that appeared in an earlier collection, and chapters 5 and 7 of this volume triangulate the histories of

race relations in the United States, South Africa, and Brazil. But I would not pretend that my knowledge of Brazilian or Caribbean history would suffice for more than a very general and schematic comparison, suitable for an essay but not for a book-length study. If the aim is to have one's work respected and found useful by scholars in all the fields that one treats comparatively, it is generally advisable to stick to two cases. A third might be possible, but to do it with the same depth as the other two would be a formidable task, and organizing the finished study would be difficult. There are good reasons, I think, that those who profess history as a discipline have rarely attempted to treat three or more cases in a detailed and sustained way.^[12]

George M. Fredrickson, "White Responses to Emancipation: The American South, Jamaica, and the Cape of Good Hope," in Arrogance of Race, 236-253.

The need for such depth might lead some readers to conclude that comparative history is so difficult and time-consuming that no sane person would attempt it. I do not wish to underestimate the effort required, but it is nevertheless possible to exaggerate it. Although it is essential to master the relevant historiography on one's topic, a good comparative history need not involve extensive research in archival or primary sources on each case. If the topic is broad, and the secondary literature voluminous and of high quality, there is no reason not to rely heavily upon it. White Supremacy made substantial use of secondary sources (at some points, especially in the last half of the book, they were virtually the only sources used to support an argument). My previous knowledge of primary materials on the United States and what I could learn from published memoirs, travel accounts, and official records or reports reflecting the evolving racial attitudes and policies in South Africa added a great deal to my critical synthesis of the two historiographies, especially in the first three chapters. I was nevertheless impressed with how much could be done with only the secondary sources. In a seminal article on comparative history written forty years ago, the eminent economic historian Fritz Redlich described very cogently the special relationship between primary and secondary sources in comparative historical work:

Comparative historiography's main sources are by necessity historical monographs, i.e. the results of preceding generations of historians. While up to now a historical work to be recognized had to be based on primary sources … comparative historiography demands as its source material a welter of monographs. It would go to the archives to fill gaps in our knowledge and check to see if the monograph writers really saw those aspects of historical fact or development in which we are interested. One might put it this way: while traditional monographic historiography begins in the archives, comparative historiography ends it there.^[13]

Fritz Redlich, "Toward a Comparative Historiography: Background and Problems," Kyklos 11, no. 3 (1958), 386.

Of course some types of comparative history can and should rely more heavily on primary sources than others. Black Liberation , essentially a work of intellectual history, was based mainly on the actual discourse of black protest in the two societies, rather than what historians had written about it. The texts I relied upon were primary sources, even if many of them could be found in published books at any good research library rather than in archival collections. The history of ideas and ideologies requires less digging and scurrying about than the history of institutions or patterns of behavior; it can be done comparatively without a heavy reliance on monographic literature. Comparative social history can also be based mainly on primary sources if the units being treated are small enough. There is a trend in comparative historiography, as some of the work discussed in chapter 3 would suggest, to focus on selected local experiences in two societies rather than a nationwide phenomenon. Cross-national comparisons of two cities or two organizations, for example, make it feasible to do extensive archival research on each case. But knowledge of the relevant secondary literature on each nation is still necessary to establish the broader contexts that give shape and meaning to the local experiences being treated in detail.

Besides reading the relevant works in their original language, the comparative historian would be well advised to establish direct contacts with historians working on the other, exotic case. Connections that I established with historians of South Africa in Britain, the United States, and South Africa itself during the 1970s alerted me to a major shift in historiography that surfaced initially in seminar papers and articles in relatively obscure journals rather than in widely distributed books and periodicals. I was able to learn from, and respond to, this emerging historiography in White Supremacy because correspondents and people I met on my travels informed me of what was going on.

My final word of advice to would-be comparativists concerns how the findings should be presented, once the basic research and analysis have been completed. To my way of thinking, it is much more effective—and genuinely comparative—to have topical chapters dealing interactively with both countries rather than separate chapters on each country. An argument against three-way comparisons is that it becomes very difficult to use this format without making excessive demands on the reader, who tends to feel like an inept juggler. The worst organization in my view is the one that has a long chapter on each case and then a last chapter that does the actual comparison. My two books on the United States and South Africa, as well as chapters 10 and 11 of this

volume (shortened and recast versions of the last two chapters of Black Liberation ), show how one can develop comparisons within each topical chapter without (I hope) giving the reader a headache. Separate treatment tends to result in parallel histories rather than genuinely comparative ones. Two narratives can be presented in such a way that the reader is always aware of both and senses that they are, as it were, constantly commenting on each other. Comparative history of this type is an extended conversation between two national histories in search of common themes and an understanding of the variations on them. It might be viewed as one aspect of the wider search of international or world historians and other scholars with cosmopolitan interests for a common language and a common set of concerns that cross international frontiers.

To the extent that it is a cosmopolitan or universalizing enterprise, however, comparative history may seem to many in the late 1990s to be intellectually problematic (or at least unfashionable) in light of the postmodernist stress on the particular and unique—in other words the incomparable or incommensurate. Can comparative history of the kind I practice, or any other kind for that matter, survive in the world as defined by postmodernists? Is it necessary to choose between comparative studies and what appears to be the dominant intellectual trend of the age? These questions are not confronted directly in the essays that follow (although "Understanding Racism" appropriates for my own purposes some of the language usually associated with postmodernism). But they have to be faced if the essays in this book are to be taken seriously.

The historical profession contains within it a strong trend that is consistent in some ways with the general thrust of postmodernism, without, in most cases, revealing much awareness of the affinity. This is the "revival of narrative history"—the growing emphasis on discrete storytelling at the expense of analysis and generalization. Most historians of course believe that their stories are factually based and not just fiction in another form, as some postmodern theorists do, but a heavy stress on the particularity or singularity of the human experiences that we treat and on the color and drama that our narratives about them can convey tends to aestheticize history, just as postmodernism often aestheticizes politics and philosophy.^[14]

A study of the intellectual roots of postmodernism that views aestheticism as its defining feature is Allan Megill, Prophets of Extremity: Neitzche, Heidegger, Foucault, Derrida (Berkeley, 1985).

I believe that narrative is an essential aspect of historical writing but that is not its be-all and end-all. Discrete narratives of particular events

that eschew the extraction of larger meanings are the antithesis of comparative history or any other kind of history that seeks to generalize beyond the immediate facts of the case at hand. If telling a good story in vivid detail becomes the sole or even the paramount aim of historians, if history becomes exclusively an art rather than a science, comparative history has no future. But we have not reached that point, if we ever will, and the desire to find a middle ground between narrative and analytical history remains the dominant aspiration of most working historians.

The treatment of change over time—the essence of historical scholarship—must in some basic sense take the form of storytelling. But the best storytellers have also been moralists. If the moral is not explicit, as in Aesop's fables, it is usually implicit and evident to the perceptive reader. Besides telling a good story, most great novels reveal something about human motivation and often make judgments on the social values and practices found in particular times and places. Historians' narratives are not, of course, freely invented but are based on sources that other historians can examine. The moral or meaning cannot be simply illustrated or personified, as in the case of fiction. It must be supported by a body of evidence.

Both history and fiction aim to be truthful in some sense, but the historian's truth is subject to evaluation by a community of inquirers who have established rules to determine reliability, quality, and persuasiveness. Besides seeing if the historian has been true to her sources, a critic can evaluate a given narrative by seeing to what extent it is consistent with the stories that other historians have told about the same or similar events. The process can be compared to a jury's evaluation of the testimony of a number of witnesses to the same crime. Cross-national history takes two stories in two different contexts that are presumptively true by the normal standards of historical evaluation and juxtaposes them in the hope of finding a deeper meaning for each narrative and beyond that, if possible, a larger meaning that applies to both.

The revival of narrative history can have a salutary effect on comparative work by encouraging the idea that two stories are being compared, not simply two structures or conditions. Comparing the institution of slavery in what is taken to be its normal or typical form in two societies, for example, suspends time and is basically ahistorical. Comparing the process of enslavement, or the story of how the bonds of servitude were created and strengthened—or weakened—is a better way of conceiving the project. Radical historicity, the belief that time

changes everything and that nothing is truly permanent or "essential," is a characteristic of postmodernist thought that is congenial to the instincts of many historians.

But the comparative historian needs to know what slavery is in order to justify treating its evolution in two or more societies. One culture's word that translates into English as "slavery" may refer to something quite different from another's term for a state of servitude and extreme dependency. It is at this point that comparative history seems to clash directly with postmodernism. Is it justifiable to construct a general category or ideal type and then view two specific stories as examples of its presence? Pauline Marie Rosenau, a Canadian political scientist who has studied the implications of postmodernism for social science, argues that this new way of thinking invalidates all comparisons. After describing postmodernism's characteristic denial of the connection between the "signifier" and the "signified," its rejection of the possibility of representation, she goes on to show the devastating effect of this premise on comparative studies:

Dismissing the possibility of representations undermines social science methods in general, but its questioning of comparative analysis is particularly thoroughgoing. The very act of comparing, in an effort to uncover differences and similarities, is a meaningless activity because postmodern epistemology holds it impossible ever to define adequately the elements to be contrasted or likened.^[15]

Pauline Marie Rosenau, Post-Modernism and the Social Sciences: Insights, Inroads, and Intrusions (Princeton, 1992), 105.

If this is what postmodernists are really saying, and all that they are saying, then the comparative historian might be sorely tempted to join the ranks of those currently denouncing it as a form of nihilism that is threatening to undermine morality, rationality, and civilization as we know it. Some of the thinkers identified with postmodernism have undeniably espoused a radical skepticism of all claims to the establishment of truth, except of course their own claim that all truth is relative. The inherent contradiction of unqualified relativism, as its philosophical critics have often pointed out, is that its own assertion is either absolutely true or no more valid than various claims of a reality beyond language. In either case the result is incoherence.^[16]

My sense of the logical fallacies of extreme relativism and of radical postmodernism in general has been strongly influenced by Richard J. Bernstein, The New Constellation: The Ethical-Political Horizon of Modernity/Post-Modernity (Cambridge, Mass., 1993), as well as by various writings of Jürgen Habermas, especially The Philosophical Discourse of Modernity, trans. Frederick G. Lawrence (Cambridge, Mass., 1993).

But postmodern thought, restrained by a modicum of common sense and respect for what most human beings have taken to be good and true, may have something to offer the comparative historian. Some forms of social-scientific comparison, those inferring structures that can be diagramed from quantifiable data, may indeed be open to criticism for detaching the signifier from the signified. Do they in fact refer to any

reality other than one that exists on hard disks or in hyperspace? What is real for human beings must be present in their consciousness. To argue, for example, that southern slaveholders defended their peculiar institution to the death because modern economists have determined that it was profitable is not a convincing historical argument if the evidence shows that the slaveholders themselves believed it was unprofitable. The relationship between what historical actors firmly believed and an external set of facts that contradicts their beliefs is one of the most difficult problems of historical interpretation. I am not one of those who would endorse a purely subjectivist history by claiming that realities external to human consciousness are of no historical significance or (in the extreme postmodernist formulation) do not even exist. In the long run, brute reality can kill us whether we believe in it or not. The good news is that people do learn from experience. If patients die often enough from a particular illness, a traditional remedy once deemed efficacious is likely to be abandoned. The belief of American settlers of the Great Plains that rain followed the plow did not survive a decade-long drought. Truth, as William James and John Dewey taught us, is what works: when it no longer works, it is no longer true.

One pragmatic test of new historical truth is the perception by the community of historians that the work has enriched our sense of the past and has raised new questions that are worthy of further investigation. No absolute truth can ever be attained, at least not until the end of the universe, but provisional truths that work for us here and now are within our grasp. Comparative historians may construct a working definition, or what Max Weber called an "ideal type," of whatever recurring category of human thought and action they want to compare. This abstract model or categorization is obviously an intellectual construction and not the direct representation of something that actually exists. But it is informed by knowledge of a range of conditions, institutions, ideologies, and other factors that share some defining features with the model or ideal type. The defining conception works if the cases to which it is applied manifest these characteristics. Once a basic and abstract resemblance has been established, it is possible to isolate some important differences and ask what accounts for them. This approach "works" if the differences can be correlated with other differences between the two societies, in which case a causal argument can be developed.

Nothing about this procedure necessarily involves a belief in the possibility of finding an absolute and final truth about anything significant.

The historian's representation of past human experience is provisional and subject to correction by fellow inquirers. It is also filtered through the historian's own consciousness and made to serve the historian's own purposes. The results are published and their significance is determined by how well they suit the purposes of an audience. Such an enterprise contributes to truth in the pragmatic sense to the degree that it is useful. Usefulness can be purely intellectual: it may stimulate and guide further inquiries that an intellectual community finds to be interesting and potentially rewarding. It may also be useful in a social and political sense by contributing to the construction of ideologies and the development of policies that provide more effective ways to deal with contemporary manifestations of whatever has been studied historically. Comparative history has a special ability to suggest new ways of thinking about old problems by calling our attention to alternative approaches to their management or solution. Understanding how another society has confronted challenges similar to our own can be a stimulus to social and political creativity.

My comparative work on race and racism in the United States and South Africa is therefore truthful to the degree that it stimulates inquiry and constructive debate within the community of historians and social scientists. It would become even more valid in a pragmatic sense if it added some wisdom or insight to efforts of concerned citizens and policy makers to respond in a just and equitable manner to the demands of formerly enslaved and disenfranchised people for equality of rights, status, and opportunity.

The pragmatic test of truth, as I conceive it, does not sanction amorality or a nonjudgmental relativism. What is useful should accord with some sense of how human beings deserve to be treated by virtue of being human. Phrases like "human rights" and "human dignity" may be hard to define precisely, but they obviously refer to a universal entitlement that we are bound to respect if we hope to make a world in which everyone will have a chance at happiness and self-fulfillment. An ethics based on utility or practicality in the broadest sense requires a conception of what is best for humanity as a whole. If we could ask every person in the world what would be of most practical value for him or her, we would receive many answers. But common to most of them, I imagine, would be the desire to be treated with respect, to be protected from torture and other forms of brutality whether administered by the state or by private parties, to be assured of the wherewithal to sustain physical existence with some degree of comfort and security,

and to be provided the opportunity to develop one's individual talent or genius freely to the full extent of one's abilities, as long as doing so does not impede the chances of others to do the same. To the degree that postmodernism is compatible with a pragmatic humanism of this type, I would welcome it. To the degree that it is not, I would resist it. Drawing a balance sheet is difficult and beyond the scope of this introduction. It is, however, both clear and incontestable that an intellectual encounter with postmodernism can stimulate new thought even if we end up modifying or rejecting some of its precepts.

These philosophical reflections may not seem necessary to conventional historians, but postmodernism is not going to go away, and it behooves historians, comparative and otherwise, to come to terms with it. David Harvey has recently made a strong case that postmodernity and the thinking that goes with it are the products of profound economic and social developments presently transforming world capitalism. From a pragmatic point of view, its assumptions are therefore becoming "true" for many people. But believers in human freedom and rationality have the right and the obligation to criticize many of these tendencies in the name of more humane and inclusive conceptions of practical truth. Scholars who value the legacy of the Enlightenment on whatever epistemological grounds should be prepared to explain and defend their work against the radical skeptics, but if they are true to the essence of the Enlightenment heritage, they must do so by reasoned argument rather than dogmatic assertions of absolute or revealed truth.^[17]

David Harvey, The Condition of Postmodernity (Cambridge, Mass., 1990). I am somewhat uncomfortable, however, with the deterministic implications of Harvey's essentially Marxian analysis.

I have made some effort in this introduction to conform to one of the healthier tendencies of postmodernism, the call for historians and other scholars to be more candid about where they are coming from, autobiographically and ideologically, rather than assuming an Olympian objectivity. One does not have to give up the hope for a common discourse, or what Jürgen Habermas called "communicative rationality," to be willing to reveal one's particular route to the public space available for the discussion of ideas. I could not, however, continue to do my work if I believed that my ascribed peculiarities—race, ethnicity, gender, and sexual orientation—predetermined what I had to say and provided sufficient grounds for accepting or rejecting it. I would not deny that being a heterosexual white male of Swedish-American ancestry has some effect on my view of the world. But I firmly believe that a willingness to participate in rational discussion and debate with others committed to the process—including some who might share none of the above attributes—can lead to transcendent perspectives and the staking out of

common ground. My conception of the character and function of comparative history will not be everybody's, but I hope that my version of it will stimulate or provoke those who see potential value in such an enterprise to understand my formulation, debate its merits, and try to provide better alternatives. This is the way that truth is made.