Sunday, October 14, 2007

Of Hume's Infamous Footnote

Chris Mathews has a nice weblog called Philosophical Misadventures (ht) which exists "to collect, document, and comment upon the various missteps, mistakes, and plain absurdities of prominent philosophers, from it’s earliest beginnings to the present day." This is a dangerous sort of endeavor, since history is filled with attributions of missteps, mistakes, and plain absurdities that are themselves missteps, mistakes, and plain absurdities (due, for instance, to a failure to take into account shifts in language, or precise phrasing of texts, etc.). But the attempt here is very sober and balanced, which is the best one can ask for in facing such issues.

One of the posts from several months ago discusses Hume's infamous footnote, which I have discussed several times before, and in particular looks at Eric Morton's paper, Race and Racism in the Works of David Hume, which I've recommended as a good beginning for inquiry on the subject, although criticizing it for its tendency to move from point to point without adequate development.

Morton claims that "Hume’s theory of knowledge is driven by Hume’s racism and the built-in racism in his philosophical and conceptual worldview." To this Mathews replies,

Morton may show that Hume’s racism taints his own conceptual worldview (hardly a difficult task, given the evidence) but fails to justify philosophically how “the conceptual framework of empiricism itself may be racist.” Is it not possible to simply apply the abstract principles of Humes’s philosophy without the empirical prejudices the philosopher himself held?


This is an interesting question to ask, but I think we need to make some distinctions if it is to be properly answered.

(1) There is a sense in which one can agree 'abstractly' with a philosopher but hold what is a very different philosophical position. A very good example of this, which I've pointed out a number of times, and which is always fascinating, is from Hume himself. Hume agrees 'abstractly' with a great many of Malebranche's principles in the theory of causation. You can even find Hume repeating Malebranche's arguments on the subject word-for-word (allowing for translation). But the 'abstract' level glosses over some serious differences in how the two understand those very principles. Malebranche makes his argument about balls hitting other balls, which Hume adapts into the billiard balls argument, to do in some sense what Hume later wants to do: namely, to show that we cannot perceive causation. But what a very vast chasm lies between them on both of these points. Malebranche is a rationalist: the sensory perception he has in mind has rationalist features. Hume is an empiricist, and the impression he has in mind is that of sensory impressions, which are related to certain kinds of 'ideas' examined by Locke and Berkeley. Malebranche is a devout Catholic who brings up these causal arguments as a twofold project of driving people to recognizing their dependence on God and purging philosophy and theology of elements of pagan philosophy. Hume's motivation, of course, is very different. Is Hume using the abstract principles of Malebranche's philosophy? In a sense it is undeniable that he is. But, on the other hand, there's a sense in which he means something different by the same words, and it's a legitimate question to ask -- and some of my academic research has been devoted to asking -- how different and how similar their views really are. So it's a legitimate question to ask what has to be changed in order to avoid a Humean conclusion.

And the question will get asked anyway. I cannot count the number of times I have been defeated in an attempt to explain to my colleagues that Malebranche's arguments are philosophical interesting despite and even because they are part of an ethical program for destroying idolatry and increasing piety, which they tend to regard as suspicious and unphilosophical -- and this, again, despite the fact that no one has any problem with Hume's arguments, many of which (particularly in Section VII of the Enquiry) are just taken over from Malebranche. The contexts in which people find philosophical arguments do make a big difference on how they perceive them, even where they themselves cannot give a clear account of what the difference of an apparently similar argument in two different contexts is. Malebranche makes the arguments as a Catholic reformer; Hume makes the arguments as a skeptic; they are often word-for-word the same but I am continually being forced to justify studying Malebranche's arguments in their context despite the fact that if you substitute 'Hume' for 'Malebranche' no one thinks the study needs justification -- it's just obviously relevant and contemporary and whatever other standard they apply for justification. The moral for this case is that at the very least we need a close examination of how the principles change and how they remain the same when they are taken out of the context of Hume's whole philosophy.

(2) We tend to divorce Hume's theory of knowledge from his ethical theory. It is clear that Hume himself saw no divorce, despite the fact that it's sometimes difficult for us to see precisely what he saw as the connection. But it's important to understand that Hume's investigations in Book I of the Treatise are not necessarily ethically innocent. This is because Book I can be seen as (in part) a portion of a larger attack on the rationalist worldview, which undermines not just the rationalist view of the world, but the rationalist view of reason -- the same view of reason that undergirds ethics of a rationalist type. If a rationalist comes to us and points to Hume's infamous footnote as an example of why you need to accept the rationalist view of reason -- as one that allows for a priori ethical principles that rule out racism out of hand in a way no empirical principles can -- this question is also worth taking seriously. One person's modus ponens is another person's modus tollens (as I half-jokingly noted with regard to an instance of Hume's sexism, since he unwittingly gives women an interesting argument for the immortality of the soul).

(3) It's fair enough, however, to ask whether Hume was consistent in making these racist claims. And this in turn needs to be divided into several possible answers. (3a) One might argue that Hume was consistent in making these claims in such a way that his principles in some way require him to have racist conclusions. (3b) One might argue that Hume was consistent in making these claims, only in that his principles taken all together gave him no way to reject these conclusions. (3c) One might argue that Hume was inconsistent in making these claims, in that his principles are naturally anti-racist. (3d) One might argue that Hume was inconsistent in making these claims, not because of any particular anti-racist bent to his principles, but because he simply had the empirical facts wrong, in which case his having them wrong was either (3d1) excusable because he couldn't have had the right facts; or (3d2) inexcusable because a reasonable person could easily have gathered the right facts. Beattie, I think, has shown conclusively that if we take the (3d) position we have to reject (3d1), because Beattie, a contemporary of Hume, attacks Hume, and rightly, on all his empirical claims and shows in a very vivid way how absurd they often are. The rest are all positions worth considering. Morton obviously accepts (3a). I've endorsed a counterpart of (3b) for some of Hume's sexism, and would therefore be inclined to accept it for his racism until some argument comes along to convince me otherwise. (It's interesting, to consider another issue of a similar sort that's worth comparing to this one, that I've argued that Heidegger's association with the Nazis is at least (3b)-like and very likely (3a)-like.)

All of these are interesting questions worth asking.