Narrative skepticism and narrative essentialism
It is worthwhile to outline the overall conceptual landscape before delving into a detailed analysis of sample historiographies. The narrativ- ists have identified 'narrative' as the synthesizing element of historical works. It is now time to discuss in more detail what 'narrative' is, what a commitment to it entails and whether it is an inherent feature of any historiographical presentation. Even before that, however, it is useful to take a step back and briefly consider what it would mean to question the narrativist insight; that is, to think that a proper historical account could be devoid of any synthesizing elements.
Much of Ankersmit's early philosophy is based on criticism of the earlier philosophy of historiography for its failure to recognize and analyze the textual and other integrative features of historiography. In other words, as already discussed in the preceding chapters, he chided the philosophers of previous generations for focusing on statements as atomistic units and for not adequately differentiating between the historical accounts as a set of atomistic statements and as narratives composed of an ordered set of those statements. Questioning the narra- tivist insight would thus amount to thinking that a set of statements without connections or integration would pass as an acceptable form of historiography. What would such histories look like? Are there any examples?
The most obvious example is undoubtedly to be found in medieval annals that seem to merely record a year and an event or events without offering commentary or links between events. Another example would be the chronicle, which similarly lists events in a chronological order, although some see it as already being more structured in that it contains a 'central subject' (e.g. White 1980). In the historiography of science, we find a modern programmatic effort to 'merely describe' without imposing any pre-empirical structures on phenomena. Latour and Woolgar's (1986) anthropological studies of science require observers to empty their minds of any preconceptions regarding what science is and pretend ignorance with respect to it. Rather than seeking any larger frameworks or other integrative principles, anthropologists should 'just describe the state of affairs at hand' (Latour 2005, 144). According to Latour, 'the name of the game is to go back to empiricism' (Latour 2005, 144). Historiography based on anthropological studies of science would thus aim at only localized descriptions of social practice purified of all preconceptions of the described practice - something that has indeed found some applications in Latour and others.3 However, it is doubtful, whether even radical anthropology can avoid imposing any generalizing structure on its textual material. Latour and Woolgar's Laboratory Life itself is certainly not merely a sequence of statements without any commentary that would connect the statements. It seems that advocates of radical anthropology rather commit to an ideal of descriptivism that promises the possibility of avoiding narrative or other synthesizing features without actually implementing it themselves.4
Are we then left only with annals as a candidate for an account of history devoid of any integrative elements? There are two things to say. White has suggested that even annals imply some minimal narrativist plotting. I will discuss this suggestion below. A more important issue is that annals cannot be said to provide us with a proper historical account. No book of history is just a collection of descriptive statements and no student could pass a course in history if the final course work amounted to a random list of claims. Modern historiography requires coherence, consistency, integration and putting forward some views or theses on the past.
Let us next imagine a narrative skeptic and a narrative essentialist. They both accept that a work of history has to contain some kind of integrative element. The former is skeptical as to whether narrative form is necessary for historiography, implying that a non-narrative historiography is possible and reasonable, while the latter thinks that narrative is essential for any proper historical account. The narrativists are typically narrative essentialists, but here we can distinguish at least two different views, already mentioned in Chapter 2. Phenomenological narrativists, such as Paul Ricoeur and David Carr, assume that human experiencing itself must take a narrative form. This implies that narrativity is a transcendental condition of any kind of human experiencing and historical reality (as a human reality) always takes narrative form. Linguistic narr- ativists like Hayden White and Frank Ankersmit commit to a weaker claim that historical reality itself has no form, but that its intelligibility requires imposing a narrative order onto it.5 Further, Mink famously stated that to claim that 'the qualities of narrative are transferred to art from life seems a hysteron proteron. Stories are not lived but told ... it seems truer to say that narrative qualities are transferred from art to life' (1970, 557-558). Now, those who claim something about the conditions of human experiencing engage in a metaphysical debate, which arguably also links to the concerns of evolutionary biology and cognitive psychology (how was it that humans developed such an alleged transcendental condition of experiencing? How is this condition rooted in human cognitive capacities?). That debate is beyond the scope of this book and my interests, as the focus here is on the disciplinary nature of historiography. Narrativists of both kinds are nevertheless united in their belief that historiography necessarily requires narratives and is inherently a narrativist kind of practice. It is necessary to examine these claims.
Narrative essentialists form a large group in the theory and philosophy of history. We saw in the previous chapter that Croce thought that historiography necessarily takes a narrativist form that requires 'telling the facts.' Peter Gay states that 'historical analysis without narration is incomplete' (Gay 1974, 189). White called narrative a 'metacode' for transmitting transcultural messages (see note 5). Historical theorist Nancy Partner says that 'history is narrative in form, virtually by definition, because narrative is what brings the seriatim stream of time under control for intelligible, meaningful comprehension' (Partner 2013, 2, cf. Mink 1970, 547). Even Ankersmit thinks that books of history typically adopt a narrative mode (2005, xiiv).
White mentions Tocqueville, Burckhardt, Huizinga and Braudel as examples of scholars who have attempted to draft a non-narrative or even an anti-narrative form of historical knowledge. They are thus potential narrative skeptics. However, White's claim boils down to a view that, although they did not 'tell a story with well-marked beginning, middle, and end phases' (1980, 6), they nevertheless 'narrated' their accounts in some sense. The hardest case for the narrative essentialist seems to be annals and their practice of mere recording mentioned above, because there does not seem to be any 'necessary connection' (White 1980, 11) between the events in their bookkeeping. And although it is possible to deny that their 'history writing' amounts to proper historical accounts, White argues that in fact even the annals contain 'surely a plot - if by "plot" we mean a structure of relationships by which the events contained in the account are endowed with a meaning by being identified as parts of an integrated whole' (1980, 13). For White, the registration of the events in the list of dates confers coherence and fullness, and in this way also 'meaning' to them (White 1980, 13). Further, he says that even a singular entry in the annals can be seen as a narrative. For example, the event of 1056, 'The Emperor Henry Died; and his son Henry succeeded to the rule', is a narrative according to White (1980, 18). Thus, although the annals and the chronicle fail to 'narra- tivize reality adequately', even they are not totally devoid of narrative elements.
How should we then understand 'narrative' itself? There is no universal definition, but a number of features are typically associated with it. One point of confusion is that 'narrative' is sometimes equated with 'story' (cf. Chapter 2), which was also Ankersmit's reason to reject the notion of 'narrative' in favor of 'representation.' It is clear in any case that 'narrative' for the narrativist philosophers of historiography is not 'just a story'; it is something more abstract.6. White writes that 'by common consent' narrativity is an essential attribute of 'history proper', which requires that 'the events must not only be registered with the chronological framework of their original occurrence' but also be 'revealed as possessing a structure, an order of meaning, which they do not possess as mere sequence' (White 1980, 9; original emphasis).7 While narrativity itself does not require chronological ordering for White, a chronological sequence is a condition for the narrativization of a set of events in historiography. We can also consider Ankersmit's 'narrative realist's' idea of narrativity as depicting a consecutive set of events like in a film (see Chapter 3). And this same seriatim principle with regard to narrative may be said to characterize the 'narrrative idealist's' mode of presentation. Further, the early narrativists, such as Danto (1962, 1968) and Mandelbaum (1967), also linked narrativity with chronological ordering.
However, if the chronological presentation of real events were enough by itself, then the annals and chronicles would count as proper narratives, which they do not. Yet opinion seems to be split as to what other features narratives must possess. It is nevertheless clear that the narra- tivist philosophers of historiography, particularly White and Ankersmit, require that a narrative must reveal events 'as possessing a structure, and order of meaning, which they do not possess as mere sequence' (White 1980, 9). This structure and meaning thus constitute features that the past itself does not have and which is imposed by the historian. Further, it seems clear that White, Ankersmit and the early narra- tivists more generally (see Chapter 2) require that there is some kind of central subject (such as some 'narrative substance' or 'presented'), which is the principle that provides importance or significance to events and around which the narrative is organized. White thinks that even chronicles require a central subject: 'the capacity to envision a set of events as belonging to the same order of meaning requires a metaphysical principle by which to translate difference to similarity. In other words, it requires a "subject" common to all of the referents of the various sentences that register events as having occurred' (White 1980, 19).8 Finally, some have also suggested that a narrative (or 'plot') must include causal links between events (e.g. M. White 1965, 223-224; Foster 1927, 60; Carroll 2001, 126).
The narrative essentialist's commitment to narratives and their central subjects has a philosophically significant consequence: holism. We just saw how 'plot', according to White, amounts to an integrated whole that yields a specific meaning to the events subsumed. In accordance with this, Partner writes that the 'plot' must be 'intelligibly connected, every component standing in some logical relation to the others' (Partner 2013, 503; my emphasis; similarly 502). The tenet of holism in general was discussed in detail in Chapter 3, where it became clear that this commitment entails some philosophical problems (undecomposability, analyticity and unfalsifiability). Now it is time to move closer to the actual books of history. It will soon become obvious that there is a price to pay also on the practical level.
A consequence of the union between narrativism and holism is the belief that everything in a narrative defines the narrative and the narrative amounts to historical knowledge only as an irreducible whole. The claim is thus that, without seeing all the events as part of one whole (narrative), understanding of a historiographical thesis is not achievable and, further, that historical knowledge is the whole written narrative from the beginning to the end.9 Let me first say that this view sets the bar very high for the historian. The historian is expected to have the skill of a novelist and to design a book so that it amounts to one comprehensive narrative in which everything mentioned in the narrative finds it place. But this conception is very demanding with regard to the reader as well. In order to comprehend the historian's narrative and acquire historical knowledge, the reader should be able to know the whole narrative and all the events that are part of it. All the events and statements of those events are definitionally part of the narrative: without all the statements of events there is no narrative, and without the entire narrative there is no adequate historiographical comprehension of those statements.
We are ready to state, then, what the double commitment to narrative and holism by the narrative essentialist entails. With regard to narrative, the commitment implies that:
- 1. The only acceptable form of presentation in historiography is the narrative form, which implies at least chronological ordering and the holistic endowment of meaning to all parts of the narrative.
- 2. Historiography takes a fundamentally descriptive mode in the sense that it attempts to order, link and give meaning to a sequence of events in the object world of historical reality.
With regard to holism, the commitment implies that:
A. Every event or a statement of the event has a definitional role in the narrative.
B. No historian's cognitive message can be understood without understanding the whole narrative. That is, no part of a narrative, statement, etc., can be ignored.
Next, I test these presumptions of narrative essentialism by examining some actual historiographical examples. This takes the discussion onto a more tangible level and in this way provides ingredients for abstract evaluation at the same time. It might also be said that the following discussion is a search for two kinds of answers. On the one hand, the questions deal with the nature of historiography: What kind of practice is it fundamentally? How should it be characterized? Or simply, what is the point of historiography? On the other hand, this involves the relations of different elements in a book of history: What is the relation between lower-order statements of knowledge and the higher-order historical thesis? Is a work of history a whole? Can some parts of it be understood without understanding some others?