Evidence, reasoning and argumentative structure
Now it is necessary to consider in more detail the role of those parts of a historiographical presentation that cannot reasonably be said to constitute the meaning of a thesis. In the background, there is the question of how presentations are structured. To put it differently, if the previous section rejected holism, now it is time to pay attention to the form of historical presentation. The most traditional suggestion is that historical presentations are narratives or take a narrative form, and it is often remarked that the narrative form can be found already in the works of Thucydides and other ancient writers of history. 'Narrative' was identified above as a structure that minimally entails chronological order and holistically endowed meaning. The latter feature has now been questioned with regard to the entire historiographical content. How about the chronological form of presentation? Is it necessary? More importantly, is it what historians must be committed to?
If the requirement of creating an all-inclusive narrative or plot elevated historians unrealistically to the league of master novelists, the expectation of necessary chronological ordering of one's presentation would seem to demean the skill of the historian. The narrative form in this traditional sense implies something like a 'descriptive mode' of presentation. The idea is that historians first describe events and then connect them to each other; they thus show what happened first, what happened next, and so on, ending with a final event at the end of the book. The descriptive mode relies on there being such 'events' in the object world, which are merely transferred to a chronologically ordered set of linguistic descriptions to produce the historian's narrative presentation. Naturally, a 'plot' could not be derived from the events themselves, but apart from that historical writing would seem take to adopt a realistic mode and the standpoint of an observer. As discussed in the previous chapter, it is instructive to remember that, despite their criticism of the copy theory of representation, both White and Ankersmit think that historiography necessarily creates re-presentations of the past, terminologically suggesting that the historian's narrative is a re-description of pre-given set of events in some way. White even stated very clearly that both historiography and literature are by necessity committed to 'representing reality realistically' (White 2011, 398).16 On another occasion, he even lamented that 'it is the historians themselves who have transformed narrativity from a manner of speaking into a paradigm of the form which reality itself displays' (1980, 27).
In this way we come back to the question of what kind of an activity historiography fundamentally is. Are historiographical presentations best characterized as being realistic narratives by virtue of their form? Is the structure of historical works like a temporally advancing set of events? It is important to emphasize here that the question deals with what historiography primarily is or what the governing function of historical presentation is.
My view is that it is degrading to suggest that historians merely report what happened first, what happened next, and thereafter, and still further on, etc., even if it is assumed that the chronologically advanced 'story' manages somehow to endow surplus meaning to the events. Some decades ago, Goldstein concluded that although the historian's presentation may take a narrative form, it 'need not and does not always' do so (Goldstein 1976, 141; see also 176). As an example, Goldstein argues that Ronald Syme's The Roman Revolution does not amount to a story that necessitates and makes all the events intelligible in followable order. Syme rather employs 'the method of accumulating examples' (Goldstein 1976, 172). Goldstein argues that Syme wrote his book in order to support the thesis that 'without a party a statesman is nothing' (Goldstein 1976, 178). Further, he writes that 'even a good deal of the more traditional products of historiography are not really narratives in any recognizable sense of that form' (1976, 177). Goldstein further thinks that 'it cannot be reasonable that the essential nature of the discipline is defined by the literary form' (Goldstein 1976, 142). Also Tucker has questioned narrativity as the defining feature of historiography (Tucker 2004, 139). However, although Goldstein and Tucker do thus not commit to narrative essentialism, they largely agree that typically historiography is presented in the narrative mode. By contrast, my point is that the nature of the 'superstructure', to use Goldstein's terms, is misunderstood. Further, the distinction between 'superstructure' and 'infrastructure' to distinguish the presentational mode from reasoning and the evidentiary mode is not firm.17 The presentation itself is a form of reasoning and part of the overall justification of a historical work.
I argue that historians use their critical and reasoning faculties more than is typically recognized. The American Heritage Dictionary of the English Language defines 'reasoning' as the 'use of reason, especially to form conclusions, inferences and judgments'. This is what historians do in their books: they present a view or views and reasons to accept it/them. Further, the concept of 'narrative' mischaracterizes the nature of knowledge production in historiography, which, in actuality, results in something more structured than just a set of descriptions of singular events. It also obscures the fact that the choices that the historian must make are far from self-evident. What is more, the narrative form of presentation would impose a kind of iron-cage model on historiography, implying that historians have to present their works temporally or chronologically. It is not too difficult to find examples of historians who do not use narrative plotting in any obvious sense, such as the US-style social science history (e.g. Fogel and Engerman 1974) and the French Annales School (e.g. Braudel 1996).18 It is not my intention to pass judgment on how successful these specific attempts are, but to merely mention them as some well-known cases of non-narrative historiography. And if they do not present their books narratively, then it is clearly not necessary to commit to a narrative-chronological presentation of history. However, it is more interesting to reverse White and others' approach to narrativize all historiography and ask whether even those who seem to subscribe to a narrative-chronological model can in fact be said to do so. In order to find out whether this is the case, the focus is shifted again to actual written historiographies.