Metaepistemology and Relativism

J. Carter

• Publisher: Palgrave Macmillan
• Year: 2016
• Print ISBN: 9781137336637
• eBook ISBN: 9781137336644
• Topic: Medios de comunicación y artes escénicas
• Subtopic: Música

Is knowledge relative? Many academics across the humanities say that it is. However those who work in mainstream epistemology generally consider that it is not. Metaepistemology and Relativism questions whether the kind of anti-relativistic background that underlies typical projects in mainstream epistemology can on closer inspection be vindicated.

1

Metaepistemology and Realism

Abstract. Metaepistemological commitments are revealed in first-order practice, though (unlike in metaethics) are not often given explicit expression. This chapter does two central things. The first is to develop a reasoned way of locating revealed second-order commitments in metaepistemology by looking straight to paradigmatic first-order disagreements, and to what is common ground to these disagreements. The second is to show that a pervasive element of the common ground in first-order epistemological debates turns out to be a commitment to at least a minimal form of metaepistemological realism. I conclude by considering the kinds of claims that mainstream metaepistemology tacitly excludes, in virtue of presupposing this kind of realism, and how we might best locate epistemic relativism within the picture proposed.

1.1 Introduction

Metaethics, as Sayre-McCord (2014) aptly puts it, attempts to ‘step back from particular substantive debates within morality to ask about the views, assumptions, and commitments that are shared by those who engage in the debate’. Think of metaepistemology as doing the same thing, but for epistemology.¹ One striking difference between metaethics and metaepistemology is that there are a lot of people actually doing the former. Accordingly, metaethical commitments are often articulated, explicitly and carefully. With a few outlying exceptions, the same isn’t so for metaepistemology. Epistemologists by and large carry on in their first-order projects without taking the time to articulate the more general commitments operating in the background.

An examination of metaepistemology ought to distinguish between two sources of metaepistemological commitment: on the one hand, one can take on a metaepistemological commitment by simply articulating such a view (as the metaethicists often do). Call these articulated metaepistemological commitments. A separate source of metaepistemological commitment is unarticulated, but revealed in first-order practice; call metaepistemological commitments of this latter variety revealed metaepistemological commitments. Given the inchoate state of metaepistemology, the focus here will be primarily on the latter: that is, we’ll look to the action in first-order debates in mainstream epistemology in the service of characterising what the second-order commitments are. The wider objective will be to argue for and characterise mainstream metaepistemology as a function of what characteristic debates in mainstream epistemology take for granted.

On first blush, we can envision two prima facie plausible ways to do this – viz., to locate revealed metaepistemological commitments by looking straight to first-order debates, the important disputes actually going on. The first way proceeds as follows: we look to some of the most striking disputes in first-order epistemology with an eye to mapping these first-order disputes on to more general second-order dividing lines. This is more or less the strategy that has been pursued by William Alston (1978) and Jonathan Dancy (1982) in their early attempts to carve up metaepistemological space.

An alternative strategy starts in the same place – viz., first-order debates in epistemology – but, rather than trying to map first-order disagreements on to second-order disagreements, the second strategy we can envision tries to reason from these first-order disagreements to background second-order agreements – that is, by locating the common ground lying behind first-order disagreements and with reference to which we can best explain why first-order disagreements take the characteristic shapes that they do in practice. I’ll be opting for the second strategy in what follows, and in doing so, it will be shown further how what’s taken as common ground connects with certain features of realism; the more general picture I’ll develop here is one where participants in mainstream epistemology’s paradigmatic debates can be understood as by and large revealing a tacit background commitment to at least a minimal form of metaepistemological realism.

Here is the plan. In §1.2, I’ll show why Alston’s and Dancy’s respective attempts to map first-order disagreements on to second-order disagreements in epistemology aren’t compelling. In §1.3, I’ll sketch an argument for doing things the other way around. In particular, I’ll outline a plausible picture on which the notion of presupposition – and in particular pragmatic presupposition – can be connected to both metaepistemological commitment and as well as to disagreement; what emerges is a rationale for locating metaepistemological commitments in the common ground of first-order disputes. In §1.4 I sharpen this picture by using, as a case study, the perennial debate between Moore and the sceptic; with reference to the Stalnaker/Grice model of common ground, we’ll see that Moore and the sceptic are (despite little first-order agreement) presupposing a certain shared background commitment to epistemic facts with an objective profile. And from Moore and the sceptic we can generalise. The upshot of §§1.2–1.4 will be that paradigmatic disputes in mainstream epistemology (much like the one between Moore and the sceptic) betray a revealed metaepistemological commitment to – by specifically taking for granted – the objectivity of epistemic facts under discussion. The remainder of the chapter shows how the particular picture of epistemic facts we find in the common ground (of paradigmatic first-order debates) de facto satisfies plausible conditions of a realist picture of epistemic facts. Toward this end, the approach taken in §1.5–1.6 defines and clarifies a kind of generic realism – as a combination of (suitably articulated) existence and independence theses, and defines (with reference to this working general picture of realism) metaepistemological realism, as constituted by a conjunction of existence and independence theses about epistemic facts. In §1.7 I show how mainstream metaepistemology’s commitment to epistemic facts (as developed and defended in §§1.2–1.4) constitutes a minimal form of metaepistemological realism – mainstream metaepistemological realism. In §1.8 I conclude by charting a range of metaepistemological anti-realist views that are precluded by the kind of realist picture first-order epistemologists take for granted.

1.2 Metaepistemological dividing lines: two early attempts

1.2.1 Alston and the ‘fact/value’ divide

Let’s look first at how Alston tried to carve out what he took to be the most interesting metaepistemological dividing lines. Alston, it is crucial to note, understood the metaethical landscape (at least, in the late 1970s), as featuring three prominent positions.

(i)

metaethical noncognitivism, according to which ethical judgements – (e.g. Murder is wrong) – do not express truth-apt beliefs;^2 and, among cognitivist positions,

where such judgements are regarded as truth-apt:

(ii)		metaethical naturalism, according to which – at least, as Alston characterises it – ‘ethical terms (concepts, statements) can be defined, explicated, or analyzed in “factual” terms, that they, at bottom are factual terms, and that ethical questions are, at bottom, questions of fact, perhaps of an especially complicated sort’ (Alston 1978, 276).
(iii)		metaethical intuitionism, which ‘maintains that ethical concepts are sui generis, that they are of a distinctively and irreducibly normative or evaluative sort, not to be reduced to matters of fact, however complex’^3 (ibid., 276).

With reference to these three positions, Alston drew three corresponding metaepistemological conclusions:

(C1) that parallels with non-cognitivism in epistemology are rare;

(C2) that the epistemological parallel to ethical naturalism features in the background of the ‘causal theory of knowledge’⁴, according to which (roughly), S knows that p just in case S’s belief that p has a certain kind of causal history; and that

(C3) the epistemological parallel to ethical intuitionism features in the background of the JTB-style analysis of knowledge,⁵ where ‘justified’ is taken as an evaluative term.

Regarding the first of these three conclusions: Alston simply didn’t anticipate, at the time of writing, anything like contemporary versions of epistemic expressivism, which has been explored only in recent decades,⁶ but we can set this point aside. C2 and C3 reveal how Alston envisions the key metaepistemological dividing lines: there is metaepistemological naturalism (mapping on to the causal theory) and metaepistemological intuitionism (mapping on to the JTB theory).

Obviously, a prima facie objection already to this picture is that Alston – by focusing on just two kinds of rather specific positions in epistemology – was oversimplifying the contemporary epistemological landscape. This charge is fair, and we shall return to it. But for now, this oversimplification will be useful. It provides us a simple way to think about metaepistemological commitments as (at least, potentially) in connection with first-order commitments.

That said, it won’t be hard to see how the simple picture Alston gives us faces some problems. Consider (C2) and (C3). By these conclusions, Alston is committed to saying that (for example) Alvin Goldman (an arch-promoter of the most prominent contemporary version of the causal theory, reliabilism⁷) and Richard Feldman (an arch-promoter of a non-causal, internalist variety of the JTB account, evidentialism), are espousing not only different positions in (first-order) epistemology, but moreover, that they are importantly divided along second-order lines. Specifically, Alston must grant that (qua proponent of a causal theory) Goldman would be in a position to claim a remarkable advantage over Feldman’s evidentialist-variation of the JTB theory⁸ – namely, that his causal theory offers a theory about something, justification, which ‘can be defined, explicated, or analysed in ‘factual’ terms’ (1978, 276) whereas Feldman, in virtue of endorsing a variant of the JTB-view, is not.⁹ Unsurprisingly, this is not the way causal theorists of justification and knowledge actually in practice attempt to distinguish themselves from their opponents.

Furthermore, if Alston’s assessments in (C2) and (C3) were correct, then Feldman would find himself in a position to simply shift the focus to a different, (perhaps equally) remarkable, advantage that non-causal JTB accounts would have in comparison to causal theories. Specifically, Feldman could, by Alston’s lights, rightly insist that his favoured position – viz., evidentialism – is distinguished as a genuine theory of epistemic justification (or, justifiedness¹⁰), but – despite what Goldman professes – Goldman’s reliabilism, in virtue of its causal core, is not.¹¹ This is clearly wrong.

Goldman and Feldman are at odds about a lot in epistemology. But they at least take each other to be trying to do the same kind of thing when theorising about epistemic justification and knowledge. That’s one of the reasons they have debated back and forth, and why their debates make sense. Compare: their behaviour over the past 30 years would be utterly mysterious if they did not take each other to be doing (as regards any alleged fact/value divide which Alston (1978, 276) is adverting to in distinguishing naturalism and intuitionism) the same kind of thing. If indeed there is an interesting fact/value dividing line in metaepistemology, causal theorists and non-causal JTB theorists at least behave as though they regard each other to be on the same side of it.¹²

1.2.2 Dancy and the ‘monism/pluralism’ divide

Jonathan Dancy, writing in the early 1980s, thinks – like Alston did – that the first-order divide between causal theories and non-causal JTB theories (of knowledge and epistemic justification) directs us to an interesting metaepistemological division. How...

Table of contents

• Cover
• Title
• 1  Metaepistemology and Realism
• 2  Global Relativism
• 3  The Pyrrhonian Argument for Epistemic Relativism
• 4  Dialogic Arguments for Epistemic Relativism
• 5  Incommensurability, Circularity and Epistemic Relativism
• 6  Replacement Relativism: Boghossian, Kusch and Wright
• 7  A Different Kind of Epistemic Relativism
• 8  New Relativism: Epistemic Aftermath
• 9  Metaepistemology and Relativism
• Notes
• References
• Index