As there was with historical description, there seems to be a prima facie weakness in Nietzsche’s rejection of historical motivational explanations, one internal to his own historiographical accounts. For it is everywhere evident that Nietzsche himself attributes all sorts of motivations to the agents whose actions and influences on contemporary life he endeavors to explain. “[St.] Paul wants to confound the ‘wisdom of the world,’” Hartmann’s goal is to lead the world into placidity, and Wagner wanted nothing other than to express Schopenhauer in music.102 Nietzsche often enough speaks as if groups and institutions have willed-motivations too - what do those ‘English psychologists’ want actually?103 In fact, Nietzsche even speaks of ideas as having direct causal efficacy on historical events. “The beginning of the slaves’ revolt in morality occurs when ressentiment itself turns creative and gives birth to values.. .”104 It thus seems that on the one hand Nietzsche denies that the world, including the historical world, is populated by freely willed agents whose motivations are discernible and whose actions are thereby explainable, but on the other is perfectly content to base major aspects of his philosophy on the force of his explanations as to why historical agents acted as they did.
Were Nietzsche a realist about the representation of historical objects in his mature work, as he was in his youthful historiography, then his explanations about how historical agents operated would be incoherent within the framework of his theory of agency. He could not consistently say - as he did with his philological explanations of the texts of Ermanarich, Theognis, and Diogenes - “this is the motivation by which ‘x’ brought about action, influence, or event ‘y,’” while maintaining that those motivations are either (his stronger position) non-existent or else (his weaker one) inscrutable. The same goes for origins. He cannot with consistency say “this is the origin ‘x’ that caused action, institution, value, or event ‘y’ to be as it was” - as he did with his account of the origin of tragic culture - while maintaining that “the origin of the emergence [Ursache der Entstehung] of a thing and its ultimate usefulness, its practical application and incorporation into a system of ends [System von Zwecken], are toto coelo separate.” Had he maintained an ontological realism and a representational realism, his mature historiography would be singularly ridiculous. And his numerous critiques of values and culture, a substantial portion of which are framed as historical arguments, would be similarly so.
In order to avoid the error, he would need to surrender one of the horns of the following dilemma. Either historians, including Nietzsche himself, really represent a real past when they explicate accounts that rely upon a free deliberative will as a causal principle, and thus those free deliberative wills exist and are explicable; or, what motivates historical agents to act as they do is genuinely inscrutable, and thus historical accounts, while they may be meaningful in a certain way, fail to represent the real past as it really was. I maintain that Nietzsche rejects the first horn and accepts the second, under the proviso that he thinks the meaningfulness of an historical claim consists in more than its referentiality. That is, I argue Nietzsche is an ontological realist but an anti-realist about historical representation.
Of course Nietzsche never calls himself an anti-realist anywhere in his writing. The term is admittedly anachronistic. But let us see if this contemporary position does adequately characterize what Nietzsche intended. By definition, an anti-realist holds that historiographical accounts do not represent the real past as it was, but are at least partly a construction within the mind of the present-day historian. A realist, on the other hand, maintains that the historian’s account really is a genuine re-presentation of that past as it actually existed. To an anti-realist like Michael Dummett, realism is inherently faulty because the only adequate way to adjudicate whether an account does re-present the past would be to verify its claims independently of the evidence relied upon for its construction. Since the past itself, independent of those traces in evidence, does not exist today in order that one could even in principle verify that the evidence at hand really evidences what it is claimed to, it seems that the realist cannot adequately verify his own truth conditions.
Because there are logically distinguishable versions of both realism and anti-realism, one can be both a realist in a certain sense and an anti-realist in a different sense simultaneously. The two varieties of realists are ontological realists and representational realists. A commonsense ontological realist believes that there was a past, one filled with agents and events. The documents, statues, and archeological sites we see before us really have persisted from the past into the present. To reject commonsense ontological realism would be to accept Russell’s famous thought-experiment that the world “sprang into being five minutes ago, exactly as it then was, with a population that ‘remembered’ a wholly unreal past.”1 see nothing at all in Nietzsche’s writing to suggest he thinks reality is, so to say, brand-new. In fact, a rejection ofcommonsense ontological realism would render his many claims about the historical development of values bizarrely incoherent. Nietzsche speaks often enough of the influences of ideas, values, and institutions upon people, ofpower relationships between religions, cultures, institutions, of types like the ‘priest,’ the ‘scholar,’ the ‘slave,’ etc., none of which would make much sense if he denied each of their existences.
But while Nietzsche may be a tacit commonsense ontological realist, he is not a representational realist, that is, one who holds that a historiographical representation is true if and only if it corresponds to the past as it really was.
In Maurice Mandelbaum’s well-known illustration of historiographical realism, “‘Caesar crossed the Rubicon,’ is true if the relation which it expresses did in fact hold of the objects with which it is concerned, if the action which it states was done was actually done.” Just as Mandelbaum has it, Nietzsche’s philological articles were to be considered true insofar as they adequately expressed who Ermanarich was, what Theognis and Diogenes really wrote. His account of tragedy was to be considered true insofar as it expressed the real Idea of the tragic inner character of the world. But since the mature Nietzsche denies the correspondential verifiability between an historiographical explanation and the way the past really was independent ofjudgments that in some sense misrepresent reality, we have more evidence that there is a marked and meaningful transition from his earlier to his later philosophy of history.
Although a commonsense ontological realist is certainly no ontological anti-realist, this does not entail he is necessarily a representational realist. One can maintain a belief in the reality of the past and also deny that the historian’s account of it re-presents the past as it really was. An ontological realist can still be a representational anti-realist, i.e., one who denies a correspondence between the account and that to which the account allegedly adequates. This is the position I ascribe to Nietzsche. And if I am right, then he was the first to hold what has become a popular view in contemporary philosophy of history. For today’s most well-known proponent, Frank Ankersmit, our propositional models “do not refer to things in or aspects of the past.” “For the ‘historical landscape’ is not given to the historian; he has to construct it ... The structure of the narration is a structure lent to or pressed on the past and not the reflection of a kindred structure objectively present in the past itself.”1 will say more about the influence of this position below.
Admittedly, my characterization of Nietzsche’s historiography is anachronistic. It does, however, both help to make clear Nietzsche’s meta-history and to distinguish it from rival positions and also presents at least a logically viable account of how Nietzsche can on the one hand criticize positivistic notions of description and explanation as depending upon an epistemologically naive concept of human judgment and agency, and yet at the same time offer historical descriptions and explanations that he believes are nevertheless meaningful. What I must show now is that Nietzsche actually holds such a position.
Besides those passages already examined and apart from the common- sense appeal of not writing about history in a way that overtly contradicts what he says are the limits of writing about history, there are two other clear pieces of evidence that Nietzsche holds a representational anti-realism with respect to historical judgment. 
First from Beyond Good and Evil: “We are the ones who invented causation, succession, for-each-other, relativity, compulsion, numbers, law, freedom, grounds, purpose; and ifwe project and inscribe this symbol- world [Zeichen-Welt] onto things as an ‘in-itself,’ then we deal with things as we always have, namely mythologically.”117 Most of these terms are historiographical stock in trade: causation, succession, purpose, etc. And once again these descriptive and explanatory devices are said to be symbolic rather than referential, rendering any account which employs them meaningful in a way other than its adequation with any ontologically real past. The passage following this one elucidates how these projections can be the sort of “bad tricks of interpretation” that “an old philologist like” Nietzsche “cannot help maliciously putting his finger on.” These symbolic projections are not to be found in the world - “not a matter of fact, not a ‘text’ but instead only a naive humanitarian correction and distortion.” Said otherwise, Nietzsche thinks the most essential representations in historiography do not represent anything real at all, though they are essential for making what genuinely is real in the past meaningful for agents like us.
Scholars have argued that a passage like this presents a fictionalist or falsificationist epistemology. I do not label Nietzsche’s meta-history as falsificationist, on the one hand, because for Nietzsche there is no means by which to adjudicate whether judgments do or do not falsify the way the world is independent of the framework of meaning of he who judges it. To know one is falsifying the world by means of a judgment entails knowing what that world is actually like in-itself apart from our subjective intrusions upon it, something which Nietzsche denies. “It is true, there could be a metaphysical world; the absolute possibility of it can hardly be resisted. But we observe all things through the human head and cannot cut off this head.”1 do not consider the passage to be fictionalist, on the other hand, because the judgments follow as an unconscious function of the subject’s physiognomic facticities rather than a deliberate choice about how to narrate the content of experience generally, or of history specifically. The distinction will become clearer in the next two chapters.
But the key for us now, and why I hold this passage is evidence of a representational anti-realist position, is that Nietzsche emphasizes the communicative virtue of terms like these. We should use these terms every bit as much as we should use math, logic, or the welter of scientific concepts, albeit under the awareness that they function only as particularly meaningful symbols referring to something we cannot know outside of what we can possibly represent. Descriptive representations cannot be presumed to reflect the real, and as such cannot be relied upon to explain events in a real world apart from those representations. Nevertheless they are both very typical and useful ways of referring to it that reveal - not the world as it was - but the way that historians have been driven to represent it. If a description is false, then it should be corrected; if it is a fiction then it should not be relied upon. An anti-realist representation, however, is a necessary expression of a certain perspective and is informative about how that type views the world. “In history, one comes to know better the moving forces, not our ‘lovely’ ideas!”
A second passage that evidences my view of representational anti-realism with specific reference to historical judgments is from Daybreak:
What has happened! Yeah, what has happened has been made! [Facta! Ja Facta ficta!] A historiographer [Geschichtsschreiber] has to do, not with what actually happened, but only with events supposed to have happened: for only the latter have been efficacious. [...] His theme, so-called world history, consists in opinion about supposed actions and their supposed motives, which in turn give rise to further opinions and actions, the reality of which, however, is at once vaporized again and only as vapor [Dampf] is efficacious, - a continual generation and pregnancy of phantoms over the impenetrable mist of unfathomable reality. All historians [Historiker] speak of things which have never existed except in representation [Vorstellung].
Three aspects of this passage support my reading. First, Nietzsche plainly rejects representational realism. Historical accounts do not re-present ‘what actually happened,’ only what is ‘supposed to have happened’; historical accounts do not attend what the actions and motivations really were, but only what various historical minds ‘suppose’ they were; the representation has no reference to any reality beyond itself. Second, despite his rejection of representational realism, there is also a commonsense ontological realism presumed here. If reality is said to be ‘unfathomable,’ then it must exist external to us; temporally prior but real suppositions are said to produce temporally later but real effects; the real historians themselves are said to have built up their interpretations over time; their real activity is said to have been a ‘continual generation,’ a real dynamic change characterized as a ‘pregnancy.’ Each of these statements only makes sense if there is a real past in which these events took place. Third, and most interestingly, by framing his critique of historians in terms of an historical argument - this is, after all, an account of what historians ‘have done’ - Nietzsche himself is passing a historical judgment; in the same passage he claims that historical judgments fail to represent that past as it really was. And if the very passage where he most clearly rejects representational realism maintains the meaningfulness of its own historical judgment, then that meaningfulness cannot be a function of its correspondence to an extra-mental state of reality. His critique of how historians have historically failed to represent reality must itself be an anti-realist historical judgment: a judgment about the past that does not aim to present a past state of affairs as it really was but is nevertheless meaningful in a certain way to a certain audience. The question that now arises is precisely how a representationally anti-realist judgment can be meaningful even if its correspondential truth value is inscrutable.