Paul Oskar Kristeller and the Status of Historical Knowledge

Part of a Series on the Philosophy of History

Nick Nielsen
12 min readMay 23, 2024

It is the 119th anniversary of the birth of Paul Oskar Kristeller (22 May 1905–07 June 1999), who was born in Berlin on this date in 1905.

Kristeller is most widely known for a popular anthology of renaissance philosophy — Renaissance Thought: the Classic, Scholastic, and Humanist Strains — and I quoted Kristeller in my episode about renaissance philosopher Franciscus Patricius. Here is Kristeller on the historiography of renaissance humanists:

“The humanist works on history have a number of peculiarities, both good and bad. They are often written in a highly rhetorical Latin, and they show the influence of classical historiography in the use of fictitious speeches. Since they were usually commissioned by the very state or city whose history was to be written, there is an element of eulogy and of regional or dynastic bias, something which I understand is not entirely absent from the national histories of modern times. On the other hand, the humanists usually did not place much credence in miracles and avoided theological speculations, and they tend to account for historical events on a strictly rational basis. Moreover, they often had access to the archives and original documents illustrating the subject matter of their history, and employed more exacting standards of documentation and historical criticism than had been the custom during the preceding centuries. Valla’s treatise on the Donation of Constantine is a famous example of historical criticism in the fifteenth century; in the sixteenth, we might single out humanists such as Sigonius, who must be considered in their erudition and critical acumen as the direct forerunners of the great historians of the seventeenth and eighteenth centuries.”

Kristeller wasn’t exclusively a renaissance scholar. He was also fully engaged in the philosophy of his time. Kristeller edited and wrote the introduction to Siegfried Kracauer’s book on philosophy of history, History, the Last Things Before the Last. In my episode on Kracauer I quoted from Kristeller’s introduction to Kracauer. Of Kracauer’s book Kristeller wrote, “I profoundly agree with the spirit out of which all of his work was written,” so at least some of what he says about Kracauer applies to his own conception of history. In his Foreword to Kracauer’s book Kristeller wrote:

“On several important issues, Kracauer hesitates to give a definitive solution, but rather formulates a problem and thus lays the ground for further thought. The discrepancy between general and special history, or as he calls it, macro and micro history, represents a serious dilemma. Kracauer seems to think that the results of special research are so complicated and so resistant to generalization that most of them must be ignored by the general historian. I tend to be slightly more hopeful, and to believe that the results of special research, after some lapse of time, will penetrate the general histories, and that the diversity of contradictory details can be handled in terms of comparative and qualified statements.”


“Another basic dilemma that Kracauer forcefully presents but does not resolve is that between chronological time and ‘shaped’ time, or between the general sequence of all events occurring at a given period, and the specific sequences peculiar to one particular area or tradition. Again I should be inclined to be more optimistic, and try to lay the stress on the pluralism of cultural history, while maintaining the concept of universal cultural history at least as a regulative idea in the Kantian sense.”

Thus Kristeller presents himself as being more sympathetic to the possibility of a philosophically informed history (or what we could call a macro-historical account) eventually coming into being, than was Kracauer, but Kristeller is also willing to grant significant concessions to the idiographic conception of history implicitly held by Kracauer. In another place in this Foreword, Kristeller notes that Kracauer was less interested in philosophers discussing history than in the actual practices of historians, so we shouldn’t be surprised at Kracauer’s philosophical skepticism, which Kristeller does not fully share, though he implies that his point of departure in philosophy of history is similar to that of Kracauer’s.

As I discussed in my episode on Morton White, there is a tension in those philosophers of history that claim to follow the lead of the practicing historian, because they want to borrow the authority of the historian within his own discipline, but they rarely want to follow the historian’s lead in all matters theoretical. And many historians seem to leave no theoretical lead for the philosopher to follow, or it is a very slim lead at best. This might be better expressed as there being an overlap between the practicing historian and the philosopher who takes his inspiration from the practicing historian. Where exactly is the overlap between the practicing historian and the philosopher of history, and how extensive ought it to be?

More often than not, the overlap is an idiographic orientation toward historical knowledge, along with the conviction that philosophers seeking nomothetic explanations have gotten lost in specifically philosophical problems that aren’t firmly rooted in history and the study of history. On the other hand, there is a perceived gap between historians and philosophers of history who make no pretense of taking their talking points form historians, which is what I just called philosophers discussing history, in contradistinction to philosophers discussing the work of practicing historians.

It is worth mentioning in this context that there is a temperamental difference among those philosophers who are impatient with other philosophers who they see as getting too deep in the weeds on specifically philosophical problems, and those philosophers who are impatient with other philosophers who get too deep in the weeds on specific disciplinary questions that are not part of philosophy proper. At least they have their impatience with each other in common. For my part, it makes perfectly good sense that philosophers should discuss specifically philosophical problems, whether related to history or not, but there is a certain mauvais foi about this at present.

We can see more hints of Kristeller’s idiographic epistemic orientation in a 1943 paper co-authored with Lincoln Reis, Some Remarks on the Method of History. In this paper the authors wade into the venerable question of historical methodology, seeming to come down on the side of Dilthey, Windelband, Rickert, and Isaiah Berlin, inter alia, that history has a method all its own, and this is a method of particularity, not one of general laws, i.e., idiographic rather than nomothetic:

“…the special contribution of the historian is the reconstruction and interpretation of specific events. It results from the combination of elements united by the historian and made accessible to the reader. Every element of that combination must be supported by the proper evidence, and thus the whole combination must be verifiable. It is a picture rather than a causal generalization. Though it includes causal relations, it is itself not a causal relation. For this reason, narrative rather than generalization is the only possible mode of presentation of the historical event.”

However, Kristeller goes a little bit further than most philosophers of history who contend that the historical method is not that of the natural sciences. He places this unique historical method and its narrative products in a larger context that includes rather than rejects natural science and philosophy of history:

“We have said that universal laws or rules have no place in the actual work of the historian, which consists in fact-finding and interpretation, but that they do have their place before and after this work, and outside history. They belong to a different field which we may prefer to call sociology or rather the philosophy of history. Hence the philosophy of history will make use of the work of the historian, and to a certain extent it will provide him with the guiding ideas for his work. It is this latter fact which explains why the historian himself very often is also a philosopher of history, and rightly so. This accounts for the point we have discussed above, that is, that some people have taken the philosophical statements of great historians as authoritative statements about the actual method of historical investigation. The main requirement for such a philosophy of history is that it should be critical, not dogmatic, to use the Kantian terms. What it says about general laws of history should be tentative and flexible, since the precise manner in which these rules should be applied is a matter of debate, and it should be subjected to the test of further historical research. Otherwise it is as harmful to historical research as general metaphysical preconceptions are and have been to the progress of the natural sciences.”

Kristeller is saying here that philosophy of history is distinct from history itself, and part of its task is to place history in a larger epistemological framework, which latter would include the methodologies excluded from narrowly historical inquiry. However, I don’t think that many historians are likely to take kindly to the idea of philosophers of history providing them with “guiding ideas” for their work, which he implies is another task of philosophy of history, though Kristeller has formulated this a little ambiguously, so that we could give his sentence the opposite meaning, namely that the philosopher of history should take his guiding ideas from history.

At the same time Kristeller holds that the historian is also a philosopher of history, at least in some cases, or at some times. This is a theme I’ve discussed in several episodes, quoting Hayden White on every history involving an implicit philosophy of history. One of the distinctions we can draw within the philosophy of history is between those who hold that philosophical ideas distort and contaminate history, which ideally should be a bare account of facts, so that true history is innocent of all philosophy, and those who hold that all history involves philosophical presuppositions, whether or not an historian wants to acknowledge his use of them.

An overlapping distinction is that between the historian as historian and the historian as philosopher of history. One and the same person might put on the hat of an historian and write history, and then swap his historian’s hat for the hat of a philosopher of history, and in this way provide his own guidance, and integrate his strictly historical work into a large epistemic framework. In this way, a philosopher could provide the guiding ideas for a history, because they are his own ideas. And in this context, when then historian and the philosopher are the same person by turns taking different roles, this is the same thing as historians providing the guiding ideas for philosophers.

The same year as Kristeller’s paper with Lincoln Reis, 1943, Morton White responded directly to Reis and Kristeller in “A Note on the Method of History,” and he, too, goes a little beyond the familiar dialectic and keeps moving the ball forward:

“In general, I think, it is rare to find disciplines distinguished from each other solely on the basis of the number of logical concepts they employ. In this case, so far as I am able to see, Reis and Kristeller would have us rigidly separate two disciplines simply because one of them employs the logical operation of generalization or quantification, whereas the other does not. On the whole, I think, the history of the empirical sciences teaches us that this divorce is a rare one. Moreover, the history of methodology teaches us that it is a dangerous one. For these reasons, I think that Reis and Kristeller are wrong in their analysis of the nature of historical statements, and not only that — historians who practice as if Reis and Kristeller were right would do well to change their procedures.”

Both Kristeller and White are trying to get at the big picture, but each of them sees a slightly different big picture, hence there is a distinct role for history in each of their big pictures. As recently as 2018, Kristeller wrote in his paper “Philosophy and Its Historiography”:

“Philosophers who claim to explore the status of historical knowledge have written about general laws of history and about causal explanation. These topics may concern the philosopher of history and also the sociologist or anthropologist, but they are speculative and derivative, and at best marginal for the practicing historian or philologist.”

Again in it we see the idiographic orientation and the effective irrelevance of philosophy to history, though I’m not sure how Kristeller’s dismissive attitude to philosophy in relation to history is to be squared with his claim that historian could take the directives of philosophers for the guiding ideas for their work, but, as we saw, this claim can also be interpreted in the reverse sense, that philosophers of history should take their guiding ideas from historians. Kristeller’s views on philosophy of history did not substantively change from 1943 to 1985, as he is still making the same point, but here it has been shorn from Kristeller’s larger point from his earlier writings about the reunification of knowledge within a larger framework.

The previous quote from Kristeller was used by Fons Dewulf in his 2018 paper “Revisiting Hempel’s 1942 Contribution to the Philosophy of History.” Dewulf concludes this paper with the following observation:

“The disappearance of the Windelbandian problem left open what kind of philosophical questions one could ask about the historical sciences. Here, Hempel’s paper saw its most visible contribution. It suggested the centrality of a philosophical concept that had not yet received much analytic attention, namely ‘explanation.’ In Hempel’s and Oppenheim’s 1948 paper, ‘Studies in the Logic of Explanation,’ explanation became a central concept for understanding scientific activity in general, and thus a prime object of analysis for the philosophy of science as a discipline. This in turn was refracted into an analytic philosophy of history that in the 1950s and 1960s focused its debate around ‘historical explanation.’ Thus, when Dray looks back at four decades of analytic philosophy of history and lauds Hempel’s 1942 paper, he testifies to these shifts that radically changed the methodological and conceptual norms for a philosophy of history. And when Kristeller laments the path of the philosophy of history in the twentieth century, he also attests to these shifts, but evaluates them differently, since it is his methodological and conceptual voice that became lost in the disciplinary norms of the philosophy of science and the philosophy of history after the Second World War.”

The Windelbandian problem, as far as I can ascertain in this context, is a neo-Kantian understanding of logic as transcendental logic, i.e., determining the conditions of the possibility of knowledge, which in this case means the conditions of historical knowledge. We can find this in Windelband’s Theories in Logic, from which I quoted in the episode on Windelband.

Windelband’s conception of logic is much more comprehensive than that of Hempel. It was the disappearance of this comprehensive point of view in logic that led Anglo-American philosophers of history, following Hempel, to think of logic exclusively as an extentionalist modern formal logic, which is all about determining the conditions of valid inference. The extensionalist conception of logic, which is pervasive in contemporary mathematical logic, and in analytical philosophy built on mathematical logic, is prima facie less relevant to history than traditional logic. Whether this is a good thing or a bad thing is Dewulf’s point about the different valuations of this historical development made by Dray and Kristeller.

Logic has gotten trapped into its current extensionalist incarnation by its own precision, which no one wants to surrender, but that leaves a great deal of what we would like to think of as rational thought without the immediate assistance of logic. We can shift this problem a bit by recasting it from the problem of the role of logical inference in historical explanation, to the role of rationality in historical inquiry, or intelligibility of history. Whether or not historical explanation conforms to canons of extensionalist validity is a much narrower question than whether historical inquiry is a rational enterprise and whether history itself is intelligible.

Certainly, Morton White and Kristeller have somewhat divergent conceptions of scientific rationality, and therefore different conditions for the intelligibility of history. Dewulf formulates this in terms of the logic employed in historical explanation; above I characterized it in terms of a different conception of how history fits into the bigger picture of human knowledge. But the Windelbandian problem in the larger sense, in the sense of whether history has its own distinctive methodology, which is in turn the perennial problem of whether history is a science, has not disappeared.

This problem can be temporarily papered over, but it always returns, and it will continue to return until we formulate a conception of science and a conception of history sufficiently comprehensive that the two either fully coincide, and it can be shown that history is definitively a science, or that the two overlap to some degree, and the extent to which they overlap, and at what points they overlap, can be definitively shown. This addresses the point about the reunification of knowledge in a larger framework that I mentioned earlier. History and philosophy of history are incomplete until we can do this. However, logic and science are also incomplete until we can do this.