Hume's Radical Scepticism and the Fate of Naturalized Epistemology

K. Meeker

• Publisher: Palgrave Macmillan
• Year: 2013
• ISBN:
  9781137025555
• Topic: Philosophy
• Subtopic: Philosophy of Religion

Treating David Hume as a partner in a continuing philosophical dialogue, this book tries to come to terms with Hume's influential thoughts on scepticism and naturalism in a way that sheds light on contemporary philosophy and its relationship to science.

1

A Tale of Two Interpretations

I. The rise of the naturalized interpretation

Quine once quipped, “The Humean predicament is the human predicament” (1969, 72). His point is that people rely on inductive reasoning that, when carefully examined, faces perplexing philosophical problems of the sort that David Hume famously discussed centuries ago. While Quine’s pithy remark borders on the hyperbolic, no one doubts that Hume had a penchant for putting his finger on intriguing and important philosophical matters. Indeed, it is difficult to overstate the impact of Hume’s writings on the present philosophical landscape.¹ In many senses, Hume has set the terms of debate to the extent that understanding his writings is almost essential for understanding the nuances of the contemporary situation, at least in the Anglo-American philosophical community. For example, contemporary discussions of the nature of causality, the relationship between descriptive and normative statements and so on are heavily indebted to his reflections on these matters. It is unsurprising, then, that those who offer Humean approaches to certain philosophical issues are legion today.

Interestingly enough, one of Hume’s main methodological prescriptions may be one of the most prescient parts of his philosophy. For his philosophy is, as he puts it in the famous subtitle to A Treatise of Human Nature, “An Attempt to introduce the Experimental Method of Reasoning into Moral Subjects”. This wording suggests that Hume was attempting to understand humans from the perspective of the natural sciences (see, e.g., Loeb 2006, 323). The past forty years have seen an explosion of philosophical interest in this general naturalized approach, sparked in part by Quine’s famous essay “Epistemology Naturalized”.² Coinciding with this naturalized movement, interpreters of Hume have, over the last several decades, following the course charted by the work of Norman Kemp Smith (see his 2005/1941), attacked the traditional reading of Hume as a sceptic³ and championed him as a naturalist who simply wanted a scientifically acceptable account of humans.

Although it has now fallen on hard times, the sceptical reading of Hume enjoyed the status of orthodoxy for well over one hundred years.⁴ James Beattie, Thomas Reid, Immanuel Kant, John Stuart Mill and others understood Hume mainly as a negative thinker who showed, roughly speaking, how embarking on the ship of empiricism constructed by Locke and Berkeley inevitably navigates one into the storm of destructive scepticism.⁵ With the rise of Hegelianism/idealism, this view of Hurricane David as having ravaged the vessel of empiricism became standard. The idealist T. H. Green argued that Hume’s empiricist method “ ... which began with professing to explain knowledge, showed knowledge to be impossible” (1874, 2). For many, Green presents a clear case of how not to interpret Hume. But there can be little doubt that views such as Green’s dominated for decades. Even Bertrand Russell, who partially rehabilitated some of Hume’s positive doctrines in the early twentieth century,⁶ still viewed Hume’s system as despairingly sceptical.⁷ The logical positivists also tried to develop a positive empiricist system along Hume’s lines, a system that avoided the scepticism that troubled Russell.⁸ But reading Hume as a protopositivist did not have much staying power. Generally speaking, the positivist reading of Hume seems to have drowned with the sinking of the positivist movement in general.⁹ Nevertheless, interpreters resisted any rip currents that might have pulled them back into a negative, sceptical reading of Hume; as we have mentioned, they began instead to build on the work of Norman Kemp Smith to provide a naturalistic reading of Hume. Despite the waves kicked up by the positivist interpretation, then, those who interpret Hume as a naturalist almost always use the traditional sceptical interpretation as their main foil.

In this book I challenge the most recent conventional wisdom and argue that we have good grounds to think that Hume is much more sceptical than the vast majority of current Hume interpreters allow. In the course of defending a radically sceptical reading of Hume, I do not mean to endorse all of the negative claims made by earlier sceptical interpreters. Instead, the goal of this book is to furnish a defensible, unified view of David Hume’s scepticism that resonates with Hume’s intentions and illuminates our own philosophical situation.

This task will be quite difficult, at least partly because the race between the sceptical reading and the naturalist reading of Hume is not much of a contest now because the latter currently dominates the scene. In fact, any radically sceptical interpretation of Hume is considered beyond the pale by some, such as João Paulo Monteiro: “ ... the myth of a skeptical Hume – more radical than he really is in his own, admitted ‘mitigated scepticism’ – has been discredited in recent years ... ” (2000, 139–40; emphasis added). Along similar lines, Louis Loeb claims, “The evidence against the skeptical interpretation has steadily accumulated since the 1970s, to the point that the interpretation has effectively been demolished. [note omitted] This is not to say that there are no hold-outs, but that holding out is now untenable” (Loeb 2006, 323; emphasis added).¹⁰ Given these bleak appraisals, one should ask: what, precisely, accounts for the alleged intellectual collapse of this view?

Presumably the accumulating evidence against this interpretation does not include some major new discoveries of Humean texts. Moreover, what separates the two camps is not a disagreement about the relative influence of preceding philosophers, with sceptical interpreters emphasizing the importance of, say, Locke and Berkeley and naturalistic philosophers pointing to, say, Newton and Hutcheson as more important influences. To be sure, through many groundbreaking books and articles we have a much better handle today on Hume’s historical context. Yet while there are still some major disagreements in this area, these disagreements are usually not at issue in the sceptic versus naturalist debate. The key difference between the sceptical reading and the naturalist reading centres, then, not on some new-found texts or some differing emphasis on preceding philosophers but, rather, on the argumentative support of certain key texts.¹¹ That is, the ascendancy of the naturalist interpretation is due to the undermining of the arguments for the sceptical reading of Hume’s texts, as well as the textual evidence that has been marshalled for the naturalist reading.

Given this situation, I shall simply assume the historical context presented by naturalist interpreters of Hume and focus on the textual and philosophical/argumentative issues that lie at the heart of the matter. After all, delving into contexts that are accepted by all sides in the dispute runs the risk of overcontextualizing. As Julia Annas points out, in at least some cases of interpreting philosophical texts: “ ... added knowledge about cultural and historical background may not be helpful for a philosophical understanding of the texts. At most such background can position someone to focus on the philosophy; sometimes it can even be a distraction” (2004, 30). This is not to say, though, that I shall ignore such issues. For example, I shall argue that we need a better and historically richer notion of Hume’s intellectual context to understand how he uses the term “evidence”. But I shall broach such topics only when it is necessary to advance the debate.

II. Issues in interpretation

Before continuing, we need to address two crucial interpretive questions. First, how precisely should we understand the current epistemic status of the sceptical interpretation of Hume? Second, what interpretive principles should we employ in trying to come to grips with Hume’s texts? With regard to the first question, recall that while Loeb concedes that some people cling to the sceptical interpretation, he claims that such a position is “untenable” (Loeb 2006, 323). As I understand it, then, Loeb is claiming that the argumentative/textual support for the naturalistic reading of Hume and against the sceptical reading of Hume is so strong that those who oppose it are cognitively deficient in some sense. But what sense, exactly? Suppose Kristilin submits a paper to a journal that argues for a sceptical reading of Hume but ignores countervailing arguments from the likes of Don Garrett and others. It is plausible that the existence of such a large body of scholarly dissent undermines the justification she has for her thesis, even if she is somehow not aware of the existence of such dissent. After all, she should have been aware of this dissent.¹² Hume interpretation is not a solitary exercise; it is a communal, social endeavour that requires some awareness of views that are contrary to one’s own. In such a situation the social context could understandably render a sceptical interpretation untenable, as Loeb suggests.

But what about sceptical interpretations that do explicitly take into account major naturalist alternatives? Would they be untenable? It’s difficult to say in the abstract. Presumably Loeb would not contend that the accumulated arguments against the sceptical interpretation amount to a deductive proof, so that it is impossible for the conclusion to be false if the premises are true. I am not aware of any substantive interpretive reasoning in the history of philosophy that is so strong that it rules out any possible future argument supporting a contrary conclusion. But then how should we understand Loeb’s assessment of the epistemic status of the sceptical interpretation of Hume, given the current scholarly climate?

I propose that we explore this question further in the context of discussing the second main issue concerning which interpretive principles we should employ. Unfortunately, although philosophers have investigated the nature of interpretation, they have generally not explicitly stated the interpretive principles they use when reading philosophical texts. My suggestion is that building an interpretative framework is analogous to constructing a theory in science. Of course, just as scientific theories are built upon particular perceptions, so, too, interpretive frameworks are based upon particular textual readings. And presumably we operate with some kind of principles (most often implicit) that connect the particular texts with the framework as a whole. As a result, evaluating the reading of a particular text must eventually fit with how we assess the interpretive framework as a whole.

Let us descend from such abstractions and examine a more specific proposal. It seems that most readers at least implicitly employ the following Primary Interpretive Principle:

(PIP) If a text appears to assert X, then one is prima facie justified in interpreting the text as asserting X.

An illustration: Suppose that while shopping at a local health food store, I pick up an energy bar that has the following printed on it: “This is a vegan product.” By (PIP), I am justified in inferring that the bar contains no animal products. But my justification is only prima facie; it can be overridden. Imagine, then, that one day I see a sandwich in the office kitchen bearing the following label: “This is a vegan product.” It occurs to me that Jesmyn, one of my colleagues, oftentimes (for reasons unbeknownst to me) puts such labels on food that she intends to bring for lunch, even if the food contains meat. Normally, when a label bears the words “This is a vegan product”, one is justified in interpreting the wording as asserting that “This is a vegan product.” In this case, though, I am not justified in interpreting the label thus because of the real possibility that this is a misleading label. That is, my background knowledge defeats any justification that I might have for so interpreting the label. Interpretive justification, like epistemic and scientific justification, is thus defeasible.

Presumably the notion of being prima facie justified in interpreting a text is tolerably clear. But how should we understand the phrase “appears to assert”? That is a difficult question. On an intuitive level, the phrase “This is a vegan product” clearly appears to assert “This is a vegan product.” On the other hand, the phrase “This is a vegan product” does not appear to assert that “This food is filled with meat and cheese”. Can we go beyond an intuitive understanding? Perhaps. Above we suggested that interpretive principles connect particular texts with general frameworks. In other words, interpretations of particular texts are often offered in the context of a broader interpretive framework. These frameworks, and the way that we evaluate them, provide constraints on how a text “appears”. As with a good scientific theory, a good interpretive framework should display the virtues of simplicity and conservatism (see Quine and Ullian 1970, 66–73).¹³ The simplicity and conservatism of an interpretive framework help to determine how a text appears to one. In a health food store, it is simpler to assume that the vegan label on an energy bar is meant to convey information about the contents of the food rather than at which meal it should be consumed. By the same token, it would require revision of many of the beliefs in my overall interpretive framework to assume that a business seeking to make money and keep its customers would allow such blatantly misleading labelling, thus violating the conservatism principle.¹⁴

Although one could devote an entire book to elaborating on (PIP) and how we can evaluate interpretive frameworks, we have other issues that demand our attention. But before attending to them, we should return to our original discussion to see how (PIP) puts us in a position to understand the current epistemic status of the sceptical interpretation of Hume. The history of Hume interpretation, briefly recounted at the beginning of this chapter, strongly suggests that Hume’s texts appear to assert radical scepticism. If this were not a natural and plausible way to read Hume’s arguments, then it is difficult to explain how so many could have interpreted Hume as a radical sceptic for so long. Moreover, an appeal to (PIP) yields the conclusion that they were prima facie justified in interpreting Hume’s texts as asserting scepticism. The problem is that, since the 1970s, many intelligent and perceptive Hume scholars have offered extremely elegant and detailed naturalist readings of Hume. This plethora of naturalist readings constitutes a defeater (or a set of defeaters) for the justification one might possess for the traditional reading of Hume as a sceptic. In other words, the justification that the sceptical interpretation of Hume once enjoyed has been overridden by the sophisticated naturalist readings offered by commentators in the past forty years.¹⁵

Interestingly, the idea that an interpretation of a text can pose a defeater for other interpretations is implicit in several recent discussions of Hume. Let us briefly discuss two pertinent examples; first, this idea underlies Frederick Schmitt’s assessment of Louis Loeb’s stability reading of Hume: “Loeb’s case is so compelling ... that anyone who wants to defend an alternative interpretation will now have to assimilate or deflect the massive textual evidence in favor of stability” (2004, 297). Although I agree in general that the quantity and quality of current naturalist readings of Hume constitute a defeater for a sceptical interpretation, I do not think that Loeb’s reading in particular is intended to be as powerful as Schmitt claims that it is. For much of Loeb’s book is not an interpretation of Hume but a series of “amendments” (see Loeb 2004, 369). More specifically, Loeb admits that part of his project is “ ... to go beyond reconstruction and outline a series of amendments ... [that are not] in any way suggested by the texts or even lurking in the recesses of Hume’s mind” (2002, viii).¹⁶ Because many of t...

Table of contents

• Cover
• Title Page
• 1   A Tale of Two Interpretations
• 2   Fallibility Gains a Foothold: A Model for Understanding Humean Scepticism
• 3   Fallibility’s Ultimate Epistemic Consequence
• 4   Belief without Evidence
• 5   Endorsing Epistemic Egalitarianism
• 6   Scepticism and the “Nature” of Naturalized Epistemology
• 7   Hume’s Naturalistic Internalism
• 8   Philosophy after Scepticism
• Notes
• Bibliography
• Index