Rationality and transcendence
The narrativist philosophy of historiography in its early period, and most vocally in Ankersmit (1986; 1989a), argued for the rejection of the epistemological evaluation of historiography. This does not mean that the narrativist philosophy of historiography wanted to abandon the evaluation of narratives altogether or to accept some kind of 'anything goes' principle with regard to interpretation. It should also be noted that this does not automatically imply relativism, because the suggestion is not that epistemic evaluation is relative to some S, but that we ought not to use epistemological evaluation at all. Instead it might be said that the early narrativist philosophy of historiography recommended that we substitute one form of rational evaluation, the epistemic one, with another form of rational evaluation, namely that of assessment by aesthetic and moral criteria. Aesthetic and moral judgment naturally do not provide an algorithm on the basis of which to decide what historical interpretations or narratives are preferable, but the discussion of aesthetic values and moral goodness can still (arguably) be practiced on rational grounds. If these standards cannot provide unequivocal determination of the best interpretation, they do not make the process entirely arbitrary either.
In general, one can examine the narrativist analysis of historiography and reactions to it in terms of three different issues: (1) epistemic evaluation; (2) rationality; and (3) transcendence. Assessed with these parameters, the narrativist rejects the first, endorses the second and most likely rejects the third, that is, the option that the standards of evaluation could transcend our culture or community.
Postmodernism recommends, just like the narrativism, that we reject epistemic evaluation. Munslow writes that epistemology itself is a problem for a historian (Munslow 2010, 15; Jenkins 2003; cf. Jenkins 2008a, 69). Yet postmodernism can be seen to go further than this. As we saw earlier, the most radical form of postmodernism would commit to individualistic subjectivism, which evaluates historiographical interpretations only by assessing how they serve the interests of an individual and specifically his or her political emancipation (e.g. Jenkins 2003, 14, 18; Munslow 2007, 116). This would come very close to abandoning any kind of rational evaluation. In other words, 'anything goes' as long as it is beneficial for an individual in his or her attempts to make sense of and cope with the world. However, if interpretations are nevertheless seen as aesthetic and ethical artifacts, evaluation does need to go beyond the concerns of individuals. That is, aesthetic and moral values cannot arguably be entirely individualistic but must be socially and communally assessable. Their aesthetic value or moral worth needs thus to be evaluated on a non-arbitrary, principled, basis. There must be some standards to form a judgment that the Mona Lisa is aesthetically valuable or that a racist historiography is morally questionable. The existence of such standards does not imply any point-by-point evaluation, but entails the possibility of rational discourse nevertheless. This would mean a commitment to inter-subjective communal (non-epistemic) rules of evaluation. Naturally, no postmodernist would go further than this and suggest that aesthetic and moral standards could transcend the level of communal construction and communal validity.
Here it is worth making a distinction between those 'communitarians' who wish to employ epistemological standards and those who do not. Postmodernists belong to the latter group, as does Rorty. Rorty attempted to develop 'epistemological behaviorism' as an alternative to 'philosophy-as-epistemology', as Michael Williams called it (2009, xiv). For Rorty, communal conversation is the ultimate context that defines knowledge and within which it is to be evaluated. He specifically opposed what he called the 'objectivist tradition', which 'centers around the assumption that we must step outside our community. This tradition dreams of an ultimate community which will have transcended the distinction between the natural and the social, which will exhibit a solidarity which is not parochial' (Rorty 2011, 22). The former group includes those who would stick to epistemic standards, but only on a communal basis or within a community. For example, Tucker makes a principled and formal commitment to a community-based epistemic evaluation of historiography. According to Tucker, if a sufficiently large and heterogeneous group reaches an uncoerced consensus, the group consensus is a fallible but nevertheless reliable indicator of historical knowledge (Tucker 2009, 29-36). This communitarian suggestion is commented on positively by John Zammito (2013, 414).
The problem with the communitarian solution is that any agreement or knowledge becomes community-relative. Is there any guarantee that consensus has any cash value in another community? And is there any way of knowing whether this community-based consensus or 'knowledge' is of a good quality itself; that it is not merely communally endorsed. Everyone is familiar with consensus in the past regarding the fact that the Earth is flat or that certain women are witches (and therefore, deserve to be burnt as a punishment). Although communities may have reached a consensus on these matters, we would certainly not consider them to constitute knowledge nowadays. Tucker tries to deal with relativism by means of probabilistic thinking and by recourse to the knowledge-conducive values that have led to a particular consensus. It is obvious that this will not automatically give us anything like community transcendence. One would need to say why values are valuable as such or why consensually produced knowledge is valid also externally. As Tucker himself recognizes, if knowledge is defined in terms of consensus regarding beliefs, then reference to knowledge as an explanatory factor for consensus is both circular and vacuous. Tucker makes an attempt to avoid the issue by claiming that several notions of knowledge are compatible with his account. But this won't do. Provided that 'knowledge' is the pivotal (explanatory) notion in the scheme, the question of whether it implies truth (and in what sense) must also be tackled and specifically spelled out. Tucker indicates that knowledge entails truth according to the classical definition of truth (2004, 25). But this is problematic. It is not the case that uncoerced consensus by a large heterogeneous community indicates the truth of what is believed. Ancient astronomers believed that the earth was located at the centre of the universe and a high proportion of Americans in 2003 believed that Iraq had weapons of mass destruction without any perceptible coercion. If these are knowledge, then 'knowledge' seems to imply false beliefs. If 'knowledge' is understood in some other sense, the issue will be different. The definitional matter cannot be taken so light-heartedly. Furthermore, the three features (uncoerced, uniquely heterogeneous, sufficiently large) do not seem sufficient since communities that fill these criteria may still be possessed by a collective hysteria, for example, or otherwise just end up believing false beliefs. In brief, I don't know how one could assess likelihood or probability without adding something about the process itself that justifies the views that the communities have adopted.
Indeed, Tucker eventually comes around and confesses to believing that the 'hypothetical descriptions' of past events, that is, explanations of evidence in historiography, are probably true, but 'not true in an absolute sense' (Tucker 2004, 258). However, this is not really helpful. Does Tucker mean merely to pronounce his opinion of their truth-value ('probably true') or rather to suggest that the hypotheses are 'truth-like' or 'approximately true'? The latter could be the opposite of 'absolute truth', although a better phrasing would be 'strictly true', since 'relative' pairs even better with absolute. And if this is not a correct reading, then it is not entirely clear how truth itself (not an assessment of truth-value) could be probable (cf. Rescher 1977, 191). When Tucker comments soon after this that 'most of history is and always will be unknown and unknowable', he introduces yet another, epistemic, interpretation. On the other hand, Tucker correctly observes that historiography may contain various parts, which he calls 'traditional' and 'scientific,' and whose epistemic standing differs (2004, e.g. 183-184; 254). I agree with Tucker that some parts, such as the claim that George Washington was the first president of the United States, are on very solid ground; they may be described as objective (in the 'ontological' sense) in my vocabulary, although it is a different matter how important a role they play in actual historiographical knowledge production, as discussed above. Tucker's reference to common causes would seem to work only in some cases. For example, the use of colligatory notions, such as the 'Thaw', cannot be traced back to any common cause in the external world other than someone's initial colligation and others' subsequent participation in the discourse.
Rorty comments on the threat of relativism emanating from commu- nitarianism as follows:
To say that unforced agreement is enough raises the spectre of relativism. For those who say that a pragmatic view of rationality is unwholesomely relativistic ask: 'Unforced agreement among whom? Us? The Nazis? Any arbitrary culture or group?' The answer, of course, is 'us'. This necessarily ethnocentric answer simply says that we must work by our own lights. (Rorty 2011, 38)
But if our only principle is to evaluate others' suggestions 'by our own lights', are we not locked into some kind of dogmatic imperialism of our own culture? Again, what guarantee is there that our values are any good? Why should members of any other group, the Nazis for example (arguably not 'our group'), listen to what we have concluded (on, say, the worth and epistemic standing of racist historiography)? The bottom line is that the fact that it is our conversation or judgment does not give it sufficient epistemic authority. We need something more. We need to answer the question of what it is 'about good instances of historical thinking [that] make[s] them good historical thinking', as Goldstein put it (1976, 215-216).
I find Rorty flimsy in his attitude towards objectivity. On the one hand, he is against objectivity as a 'desire to escape the limitations of one's community'. Instead he as a pragmatist desires as much 'intersubjective agreement as possible' and hopes to 'extend the reference of "us" as far as we can' (Rorty 2011, 22-23; similarly 38). But if we widen our community so that evaluative standards are applicable both in our and in what was previously a foreign community, do we not transcend our community by going beyond the borders of the community and by creating a new enlarged community? And if one keeps working in this manner, ideally, in the end, one would transcend all community borders and evaluative standards would be applicable in all communities. Or perhaps not; but the idea of community transcendence raises the prospect that some values and standards gain intersubjective and inter-communal validity. If our habits are exposed to the habits of another community, and vice versa, the most striking idiosyncrasies are arguably eliminated. If we repeat this kind of round several times, then more and more communally/cultur- ally specific idiosyncrasies are removed. It is worth considering directly, philosophically, which practices might be community-transcending and gain inter-communal validity? And why?
Realists would have an answer to the problem of community transcendence: all true propositions, narratives and historiographical theses have universal epistemic authority, in all communities, because of the (absolute) truth (as correspondence). Rorty seems to think that truth as correspondence is the only way to provide rationality. Truth as correspondence does not in any case work, provided the status of synthesizing expression in historiography, for the reasons discussed at length in this book. To repeat, the higher-order constructs in historiography typically lack truth-makers in the object world that would make them true and false. The central suggestion of the book is that historiography is about making rational, argumentative, speech acts. This is based on the idea that well-performed argumentative speech acts are rationally persuasive and that historiography is ultimately a rational practice. This takes us to the final theme of the book, which is of fundamental importance: it is rationality itself that provides the prospect for community- transcendence and the inter-communal validity of historiographical arguments. Alternatively expressed, a good historiographical argument has epistemic authority due to the rational properties it contains. In the previous chapter, historiography-specific rationality was already considered. It is worth saying here, at the very end of the final substantial chapter of this book, a few words on the notion of rationality in general and its capacity for community transcendence in historiography.