Writing history - Immanuel Wallerstein

The problem about writing history can be seen in the very title of the colloquium, which exists in three language versions. In English it is "[Re]constructing the Past." This version indicates an ambivalence between construction and reconstruction, the latter term fitting in more with an evolutionary, cumulative concept of knowledge than the former. In French the title is "Le Passé Composé."
No reconstruction here, but the title permits an allusion to grammatical syntax, and refers to the verb tense that denotes a past that continues into the present and is not yet totally completed.
In French, this form is distinguished from the Preterite, which is sometimes called "Le Passé Historique." In everyday conversation, one normally uses "le passé composé." Finally in Dutch/Flemish, the title is "Het Verleden als Instrument," a far more structuralist title than the others.

I do not know if the organizers intended this ambiguity deliberately. But it is hard to speak about history, especially these days, unambiguously. Let me raise still another ambiguity. In English, "story" and "history" are separate words, and the distinction is thought to be not only clear but crucial. But in French and Dutch, "histoire" and "geschiedenis" can have both connotations. Is the distinction less clear in these linguistic traditions? I hesitate to answer. I do notice that the organizers have charged us collectively, at least in the Englishlanguage version of their announcement, with the task of conducting "a largescale meditation on the usefulness and disadvantages of history for life." This seems to me a wise startingpoint, since it recognizes that what we are about might not necessarily be useful; it might possibly be unuseful, actually disadvantageous for life. And a final comment on the title.
This is said to be a "Colloquium on History and Legitimization." Is the legitimating of something the instrumental goal that was mentioned in the Dutch title? Are we to be very Foucauldian, and assume that all knowledge is primarily an exercise in legitimating power? I am tempted to say, of course,
what else could it possibly be? But then it occurs to me that, if this is all that it were, it could not possibly serve its purpose very effectively, since knowledge is most likely to succeed in legitimating power if the people, that is those who consume this knowledge produced by historians, thought that it had independent truthvalue. It would follow that knowledge might be most useful to those in power if it were perceived as being at most only partially respondent to power's beck and call. But of course, on the other hand, it might not be useful at all if it were entirely antagonistic to power. So, from the point of view of those with power, the relation they might want to have with intellectuals purporting to write history is an intricate, mediated, and delicate one.

I propose to discuss what are, what can be, the lines between four kinds of knowledge production: fictional tales, propaganda, journalism, and history as written by persons called historians. And then I wish to relate that to remembering and forgetting, to secrecy and publicity, to advocacy and refutation. Fictional tales are the earliest knowledge product to which most people are exposed. Children are told stories, or stories are read to them. Such stories convey messages. Parents and other adults consider these messages very important. There is considerable censorship by adults of what children may hear or read. Most people rate possible stories along a continuum running from taboo subjects to highly undesirable subjects to subjects that are considered innocent to tales with a virtuous moral. The form of such stories may vary, from those that are sweet and/or charming to those that are frightening and/or exciting. We frequently assess and reassess the effect of such stories on children, and adjust what we do in the light of such assessments. Such stories are of course fictional in the sense that a person named Cinderella is not thought by the adults telling it to have actually existed, and the place where the tale occurs cannot be located on a standard map. But the story is also considered to be about some reality perhaps the existence of mean adults in charge of a child's welfare, perhaps the existence of good adults (fairy godmothers) who counteract the mean adults, perhaps the reality of (or at least the legitimacy of) hope in difficult situations. Is children's fiction different from fiction that is said to be intended for adults? If we take a work by Balzac or by Dickens, by Dante or by Cervantes, by Shakespeare or by Goethe, we are aware that each is describing a social reality via invented characters. And we evaluate the quality of their works not merely on the beauty of the language employed, or the emotions collected, but on the ways in which the work leads us to reflect upon this social reality. There are those who claim that such fictional works are more effective in getting the reader to reflect more carefully about the social reality that is described than a work of social science analyzing the same topic. The intent of such a fictional work may well be to legitimate. Surely this was the object of the classic sagas of the Iliad, of the BhagavadGita. But of course the intent might on the contrary be to delegitimate. Or perhaps the intent of the author is irrelevant, since the text may run away from the author and the consequences of reading the work quite other than that which the author had hoped to achieve. Now many authors explicitly deny such social intent.

They may say that they tell a tale in order to amuse the reader or to express themselves or indeed merely to earn money. But once again the intent of the author may be irrelevant, and we the analysts may come along and say that the fictional work did indeed have the consequence of either legitimating or delegitimating, of either forcing reflection upon the reader or of making such reflection more difficult. Indeed, such analyses of literature are constantly made. And then there are those works of fiction, which actually use historical characters, such as Tolstoy's War and Peace. Today, television techniques permit what are called docudramas, in which newsreel shots are interspersed with fictional sequences. Today, more people may in fact acquire their historical knowledge, such as it is, from such historical novels or films than from reading the works of certified historians. Is there any way we can hold the authors of such quasihistorical fictional works to the demands of something called historical objectivity? Should we want to? And what if these authors are recounting history in a way that historians consider quite false? This is not merely hypothetical. For example, there is much controversy, at least in the United States, about the role of Oliver Stone, whose films are said by some (but not by others) to falsify history in the expectation of delegitimating power. Or is it in the expectation of legitimating power more subtly, as some allege? When we come to propaganda, we presumably move beyond fiction. But how far? Propaganda is usually defined as making statements purported by the maker to be factual but which others consider to be false. Indeed, in some cases the maker of the statement knows them to be false, or at least exaggerated. Propaganda is an exercise in politics, an attempt to sway public opinion in favor of or against some policy. We should remember that the word comes from the practices of the Roman Catholic Church, seeking to propagate the faith. But of course the Church believes the faith to be the truth. Those who are not believers may choose either to refute the truth propagated by the Church with other truths, or to ignore (and thereby tolerate) their propagation. Propaganda in the public sphere of politics is a word we use to condemn the statements of someone on the other side. No political figure these days would say he is engaged in propaganda. He would doubtless refer to his own statements more positively, saying he is engaged in telling his side of the story. Telling one's side of the story has come to be seen as a legitimate activity in the context of the wide acceptance of the belief that in politics there is no absolute truth, that there are "two sides to every story." The Japanese film, Raton, illustrated the phenomenon that there exist multiple visions of the same event, a different one for each participant. It showed this so well that Roscommon has now become a code word for this reality. Anyone who has worked with public statements of political leaders knows that the line of division is not simply between someone like Gobbles telling the "big lie" and others telling the truth. There exist a whole series of intermediate possibilities. Indeed, in recent years, a word for these in-between practices has been invented in U.S. politics. It is called "putting a spin" on the news, which means explaining what has happened in such a way that it places maximum favorable light upon the teller or the group he represents. Thus, while we may agree that Tolstoy's War and Peace does not fully represent objective history, we also may feel that the statements of official spokesmen of political leaders are not significantly different in this regard. Journalism is supposed to represent a much higher degree of truth value than propaganda. Journalists tend to define themselves as persons who take the statements of various political actors (and others), check these statements against those of opponents, and then recount what they think actually happened, presumably from a somewhat more neutral point of view. They are supposed, at least in theory, to be searching out contradictory points of view, weighing them against whatever evidence seems to exist, and drafting an independent version of reality. But of course we know the many problems with this scenario. Some journalists are not free to tell the truth; others are not honest journalists. Even if we exclude these two possibilities, journalists who are honest and unconstrained by the authorities may nonetheless not have the access to necessary information, a problem made quite acute by the rapidity and timeboundedness of their activity. They are supposed to recount what happened yesterday, not 50 or 500 years ago.

This has the advantage that they may actually be able to interview participants, but it has the disadvantage of lack of time to acquire knowledge, not to speak of acquiring perspective. Thus, as we try to move up the ladder of objectivity, going from fictional tales to propaganda to journalism, we finally reach the level of the historians, that is, those persons who have prided themselves, at least since the so-called historiography revolution of the nineteenth century, on the fact that they follow Ranke and tell history wie es eigentlich gewesen ist. To fulfill this objective, most historians have accepted a set of rules, which, it is claimed, will maximize objectivity. They have sought to base their statements on data, which has tended to mean written documents, although in recent years historians have become willing to take other kinds of data into account as well. Not just any written documents, however! Historians, at least since the nineteenth century, have made a distinction between so-called primary and secondary documents, and have given pride of place to the former. A primary document is one written more or less at the time of the event under discussion. The presumption is that these documents were written for some immediate purpose, and therefore were not written with an eye to their possible uncovering by a historian some centuries later. Primary documents of course may be difficult to understand, because the language and the contextual allusions are those of some past moment. It is assumed therefore that a competent historian will be well immersed in the cultural ambiance of the time, the result of training and considerable general research. To be sure, reliance on these primary documents at most ensures that the documents themselves were not intended to deceive, or at least that they were only intended to deceive others living at the time. No doubt we have all sorts of problems about this. Perhaps they were truly intended to deceive, and the historian is unable to decipher this. Or even worse, perhaps the documents are fakes, that is, written later and deposited somewhere to make historians believe that they were written at the time. But even after we consider amply all these issues, there remains the question of the historian's own attitude towards the substantive issues he/she is analyzing. Will the historian bring biases to interpretations of the data? Here, aside from reliance on the ethical exhortations of the community of historians, it is assumed that there is a structural check, that historians are less likely to be emotionally involved in issues of the past than issues of the present.

This was one of the classical arguments for restricting the research of historians to past eras. We have always known how shaky all this assumptions are. But we tended to handle the shakiness by denial. In recent years, a large number of scholars have openly challenged the veracity of the knowledge put forward by historians. Some have gone to the end of this road by arguing that veracity is intrinsically impossible, but others simply argue that we should be very cautious in assertions of truth value, since every analysis involves an interpretation, one that is colored by the social and personal biography of the interpreter and by the pressures of the moment at which the interpretation is being made. I have engaged thus far in an easy task. I have been illustrating the fact that there is no hard and fast, simple line which separates fiction from fact, fable from truth. The line from children's tale to professional history is a continuous blur that mixes reality, political argument, and utopian fantasy. Intrepid is the scholar who would engage in some triage between legitimate and illegitimate historiography. But this is easy to show, as I said. It is far from satisfying. For we each of us rely every day on "reality testing" in our individual attempts to cope with a very real world. And we rely on help from others to make this adequate. Historians are persons engaged in the social task of making plausible interpretations of social reality which, it is hoped, we will all find useful, not only individually but collectively. Why bother, unless we are ready to devote ourselves to designing such plausible interpretations, whatever the difficulties? We have to assume the risk. So, we come to the knotty question, what is a plausible interpretation? Clearly, there is a question of internal coherence, which is the easiest to judge, if not to achieve. I do not have to agree at all with the interpretation of someone else in order to assess whether the internal logic of the argument put forth seems tight or very dubious. And I personally feel free to pay no further attention to arguments that are not coherent. But this is far from sufficient.

I have also to feel that the questions being answered by the analysis are important questions. And I have the need to feel that the unit of analysis is appropriate to the question being answered. And finally I need to feel that no significant factors have been omitted from the analysis. There are however no simple criteria, widely agreed upon by most historians or most people, as to what are important questions, what is the appropriate unit of analysis for given questions, and what are the factors that are significant? These are in a sense all a priori decisions. What we can do with a priori decisions is one of two things. We can say that choosing between them is impossible, that they reflect basic philosophical and/or political options about which we can only agree to disagree. Or we can try to communicate and debate across these philosophical/political divides by analyzing the a priori decisions in terms of what Max Weber called "substantive rationality" (Rationalität materiel) by which we can only mean analyses that both seem to account for a larger degree of the variance on empirical questions and seek to speak to the principal philosophical/political questions of our times. Perhaps this only pushes the impossible divide back to another prior level from a debate about the plausibility of the interpretation of some limited question to a debate about what are the principal philosophical/political questions of our time. But if so, this is at least a shift which clarifies the underlying discussion, and makes it possible for more than professional historians to enter into it. Take for example the question of memory. In recent years, there has been much discussion of memory, of what we remember and ought to remember and what we forget and ought to forget. It is obvious that these are social decisions, and they are constantly being made collectively. Furthermore, the decisions are never permanent.
Even if at a given moment we decide we must collectively remember some past reality, thirty years later it is quite possible that we prefer to forget this same reality. Why then are we discussing memory so much as an issue these days? It is propelled, quite obviously by recent historical events. The issue was first put forward because of the Nazi systematic extermination of European Jewry, what has come to be called the Holocaust. It has been argued that it is vital not to forget what happened in order that it not happen again, and that therefore historians should write about it and teach this history. This view of the role of historians in creating and preserving collective memory spread rapidly. Armenians have argued that it applied to the 1915 slaughter of Armenians in Turkey. I have in my office a poster created in Argentina shortly after the ouster of the military, which reads is big letters "Nunca mas,"
and which denounces the disappearances, the tortures, the fear, the humiliations, the moral and material misery, the lies, the censure, and the silence. Above all, the silence. And we know how the celebration of the bicentennial of the French Revolution reopened the question of appropriate memory in France. Finally, we know how much debate there has been in east/central Europe and the former Soviet Union about what it is useful and what it is disadvantageous to remember.

In October 1998, in South Africa, a five-volume report was published, issued by a body called The Truth and Reconciliation Commission (TRC). This body, chaired by Archbishop Desmond Tutu, was constituted by the post apartheid government and charged with elucidating the truth on violations of human rights in the period 19601994. The decision was made to link three questions: truth, reconciliation, and amnesty. In order to arrive at "truth," they offered the chance of amnesty for crimes to anyone who would recount in detail and publicly what crimes he/she had committed. They said they found the concept of truth very complex, and that they had come up with four notions of truth: factual or forensic truth, personal or narrative truth, social or "dialogue" truth, and healing or restorative truth. Factual truth, as they defined it, was more or less what positivist historians would call truth "factual, corroborated evidence,...obtaining accurate information through reliable (impartial, objective) procedures...." They said that their findings at this level served to "reduce the number of lies that can be circulated unchallenged in public discourse," and that this was socially useful. By personal truth they meant the truth of victims telling their stories. These stories were "insights into pain" and created a sort of "narrative truth." It was an act of "restoring memory."Social truth was however closest to the goal of the Commission, they said. By interaction and debate, the Commission sought "to transcend the divisions of the past by listening carefully to the complex motives and persp

ectives of all those involved." It was seen as "a basis for affirming human dignity and integrity." Finally, healing truth is "the kind of truth that places facts and what they mean within the context of human relationships both amongst citizens and between the state and its citizens." It was for this reasons that the Commission insisted not merely on knowledge but on acknowledgment. "Acknowledgment is an affirmation that a person's pain is real and worthy of attention. It is thus central to the restoration of the dignity of victims." Is the report of this Commission history or is it a document to be used by historians with all their customary cautions? This is of course a question that not only historians have to pose themselves. The four categories of truth the Commission used is in fact a modification of the four kinds of truth put forward by Justice Albie Sachs of the Constitutional Court of South Africa. Sachs is not a historian but a jurist. He is also a militant of the African National Congress, who lost his arm in a bomb attack by agents of the apartheid regime. He is thus also a victim. Sachs asserts his puzzlement "as a lawyer and a judge" about truth. He calls factual truth "microscopic truth," and notes this is normally the prime concern of a court of law: "whether a certain person is guilty of wrongfully and intentionally killing another at a particular time and in a particular manner." It is detailed, focused truth. His second truth he calls "logical truth" "the generalized truth of propositions, the logic inherent in certain statements...arrived at by deductive and inferential processes...." Sachs has thus spelled out here the distinction between idiographic truth and nomothetic truth, long the subject of a Methodenstreit among social scientists. Sachs's third truth is "experiential truth," which seems close to the Commission's "personal or narrative truth," but not quite. Sachs says he took the name from Mohandas Gandhi's book, My Experiments with Truth. Sachs says he came to realize that Gandhi was not experimenting in the sense of a laboratory scientist, but rather that "he was testing himself, not an idea of the world out there." It was an effort to look at one's subjective experience objectively, "in a truly unprejudiced way." Sachs says law courts will hear nothing of this kind of truth. It "embarrasses" them. Should it embarrass historians? And finally, he talks of dialogical truth, the concept taken over from him by the Commission. It embodies elements of microscopic, experiential, and logical truth, "but it assumes and thrives on the notion of a community of many voices and multiple perspectives. In the case of South Africa, there is no uniquely correct way of describing how the gross violations of human rights took place, there is no single narrator who can claim to have a definitive perspective that must be the right one." Now there is a challenge to Rankean historiography. But note this is not a postmodernist suggestion that objective truth does not exist. It is rather a suggestion that the road to such truth is through very intensive, often very emotional dialogue tempered by careful sifting of the evidence, in order to arrive at a multivoice, multiple perspective version of the truth. To remember and to forget, to keep secrets or expose them to public glare, is to advocate and refute.

It is a scientific, scholarly decision. It is a political decision. It is a moral decision. And we shall find no rapid consensus today or tomorrow among persons who call themselves historians about which decisions are the correct ones. All scholarship is an activity of the present, of an ever-evolving present. No scholar ever escapes the exigencies of the present. But the present is also the most evanescent of realities, since it is over in an instant. Therefore, all scholarship is about the past, and I firmly believe that all social science should be written in the past tense. History has no special claim to the past, since all science must be historical, in the sense of knowing that reality at any given point in time is the consequence of what happened at previous points in time, including of course all the radical disjunctures that have occurred. But since the past is infinite in detail, it is beyond the potential ability of anyone ever to take the entire past into account. We make selections, indeed we make a cascading set of selections. And we have as our best guide to what selections we make the knowledge we need to make sensible historical choices about the future. The first choice we have to make is the unit of analysis we shall utilize to make our selections. My own preference is very clear. I think we have to make our analyses within the framework of what I call historical systems, units of large-scale, long-term reality and social change that have some systemic quality, that is, have a life governed by some set of processes that we can analyze and are held together because they comprise a significant and continuing division of labor. All such systems are historical in that they constantly evolve, and all are systems in that they maintain some continuing features. This means two things above all: such historical systems have spatial boundaries, even if these are changing over time. And they have temporal boundaries, that is, they have beginnings, ongoing evolutions, and terminal crises. I believe, for example, that we are living today in a world system that I assert to be a "capitalist world economy." Today this world system covers the globe. When it originated some 500 years ago, it covered a relatively small segment of the globe. Why are we discussing Charles V? I cannot speak for others.

To me, Charles V is interesting because he symbolizes a major historical choice made in Western Europe in the sixteenth century. At the very beginning of the modern world system, there were forces that sought to consolidate the nascent capitalist world economy and forces that sought to transform it into a classical world empire. This tension has been a continuing one within the modern world system. Charles V failed in his attempt to create this kind of World Empire. Had he succeeded, we would not have known the modern world as we have seen it. I say this without any moral judgment. I am not at all sure the world is better off for the failure of Charles V. I simply note it as a turning point of some great importance. Analyzing Charles V reminds us of the unpredictability of historical choice. Systems in crisis come to chaotic periods and bifurcations. Choices are made. Once made, they result in the constitution of new systems, which then have a life of their own, with their cyclical rhythms and their secular trends. At a certain point in their life, when the secular trends lead the system far from equilibrium, the cyclical rhythms are no longer sufficient to maintain the system in reasonable working order, and the system enters into crisis. I believe we are there today in the case of our present system, although I will not argue the case for that now. Historians have an extra responsibility in times of systemic crisis. To be honest, what historians do in times of the normal functioning of historical social systems does not matter all that much. They may legitimate the system, or regimes. They may try to criticize them. They are likely to be largely ignored, or in any case neglected in favor of the preferences of more powerful forces. A certain amount of objectivity is asked of them, but not too much. Their ability to navigate well amidst the shoals of competing demands is very important to them and their self-esteem, no doubt. And it is important up to a point to political authorities. But a historian assessing the role of historians can only be skeptical about the role historians have historically played. But if indeed we are in systemic crisis today, then the situation is quite different. For, by definition, a system in crisis is quite different from a system that is functioning well: in the latter, fluctuations are relatively narrow and individual effort is limited in its effect, whereas in the former (a situation of crisis), fluctuations are great and therefore each individual effort has great impact, in the end determining which fork of the bifurcation we shall travel. Suddenly, what historians write becomes very consequential. Suddenly their "truths" affect people's decisions. Suddenly, the scientific tasks that are also political and moral tasks loom large. If we now compose or recompose the past, then indeed history is an instrument. Cui bon_?

I end with a statement that Pierre Chaunu put in the Avantpropos of his book on Charles Quint: "Cette Espagne de Charles Quint n'est, peut-être, totalement impartiale mais qu'est-ce que l'impartialité? du moins, cherchetelle à comprendre, à expliquer le passé par le présent, le présent par le passé, dans la solidarité des générations qui travaillent sur l'héritage. Nous nous sommes efforcés à la cohérence. Nous ne dissimulons pas notre sympathie." Historians should heed this call to coherence that does not hide its values and preferences. And historians should assume the task of contributing to dialogic truth.