Epistemic authority and truth
The problem with truth-makers and truth-bearers is of course not necessarily the end of the road for someone who insists that we ought to view integrative historiographical theses as true. Perhaps 'truth'-claims can be explicated without them. One might think that truth consists of something other than correspondence. For example, it might be possible to take 'truth' as an epistemic notion and think that 'ideally justified' is the same as 'true'. Or maybe a deflationary definition can save truth. I discuss a pragmatist attempt to redefine 'truth' in more detail in the next section. Let me, however, suggest now that what is fundamentally at stake in these discussions is the epistemic authority without which scholarly historiography would not make much sense.
There is no doubt that the notion of 'truth' serves a pivotal function in many scholarly and non-scholarly discourses. This becomes painfully evident whenever it is suggested that something cannot be true in an absolute sense. Often these kinds of claims trigger accusations of 'postmodernism', 'nihilism', 'relativism', or an 'anything-goes' attitude. It is noteworthy that these accusations are mostly meant not as an opening for further philosophical analysis of what is at stake, but as signals of a disapproval of the view. It seems that this is also, in reverse, the case with claims that include the term 'true'. That is, truth-claims convey the message 'believe this', 'accept this', 'this is belief-worthy', etc. And the claim that is attributed with 'truth' is deemed epistemically authoritative and expected to be promptly assented to. Consequently, the person or institution that is seen to deliver truths is given an epistemi- cally authoritative status. For example, Raymond Martin argues that it is the discovery of truth that historians are after. But then he says that 'discovering truth' means deciding among competing interpretations on the basis of reasons and evidence. What Martin does is to try to find a way to attribute an epistemically authoritative status to a historiographical interpretation, but he does this without providing any substantial content to 'truth'. The source of the epistemic warrant in Martin's case is in the rational warrant that an interpretation possesses (he writes about 'relevant reasons' and 'evidence') (Martin 1993, 29).
Epistemic authority is something that can be attributed to a cognitive entity such as a theory, a belief or a historical interpretation. To say that P is true is one way in which one can attribute such authority. To say that P is true because P corresponds to a fact would be one specific truth-functional way of providing epistemic authority to P. Epistemic authority thus yields a principled reason to accept an entity as epistemically trustworthy and compelling. Tucker (2004) has suggested that belief formation should be 'uncoerced'.2 Indeed, appropriate epistemic authority attributed to a belief should be compelling due to cognitive qualities without external coercion to accept the belief.3 If all candidates were on a par, or none had a higher epistemic authority than another, it would not make any difference what to endorse. This would indeed be a case of the dreaded 'anything goes' attitude. In actual historiographical practice not all interpretations are taken as equally worthy. And, while it might be possible to discriminate between different interpretations also by other than epis- temic means, the focus now is on cognitively principled discrimination, whether the principle can be provided truth-functionally or not.
It is important to make a terminological distinction at this point. The notion of epistemic authority has not been broadly discussed in philosophy but, insofar as it has, it typically refers to someone's personal authority, that is, to a person that is seen as an authoritative source to rely on in forming one's beliefs (cf. Zagzebski 2012). This kind of person is often called an 'expert' and discussions have led to a consideration of the role of testimony in epistemology. I do not refer to people when speaking of 'epistemic authority' in this book; or rather, reference to people is at most a sub-category of the wider meaning. 'Epistemic authority' refers to any property that is attributed to an epistemic entity in order to provide it with epistemically authoritative status. In some cases this can of course be personal authority, but that is not the typical case.4 I am more interested in the kinds of qualities that an assertion itself should possess in order to be believed in historiography. If the epistemically authoritative property is that of truth, then we need to ask what 'truth' means in such cases.
The correspondence theory has been criticized innumerable times and for various reasons in the past. I hope that my reason to reject it in the case of synthesizing historical theses has become clear. It should be mentioned in passing that this does not necessarily mean discounting it in all contexts and in all functional roles (see below). What about attempts to redefine 'truth'? Could 'truth' be taken to mean something other than correspondence?