Yesterday I sat in Le Metro, our local café, drinking chocolat chaud a l’ancien and re-reading the Treatise, Book I, Part IV, section ii, “Of scepticism with regard to the senses.” So much is going on in those thirty-two pages that one could teach almost an entire semester course on that one section alone. In an effort to subdue this complexity to manageable proportions, I am going to try to expound the central line of argument as it can be reconstructed from Hume’s text. Once again, I urge those who are seriously interested in Hume’s philosophy to read the original itself.
The difficulty of the text stems in part from the fact that Hume is what used to be called [and perhaps still is] a direct realist, not a causal realist. Direct realism is the view that we are immediately aware of independently existing real objects in space and time. Causal realism, by contrast, asserts that we are immediately aware only of subjective perceptions that are caused to occur in our minds by the interaction of independent objects with our sense organs. As Hume says in his charmingly quaint way, “children and peasants” and the rest of us in our unreflective moments are direct realists. It is only philosophers who are causal realists. Hume’s view is that we are all led by “permanent and irresistible” principles of the mind to believe direct realism, and it is only after we have arrived at these beliefs that further reflection brings us to the causal realist position.
The broad structure of Hume’s discussion of our belief in the external world mirrors his discussion of our belief in causal inference: First, a brief demonstration that the belief can be neither explained nor justified by appeal to reason [or to the immediate evidence of the senses], Then a lengthy, complex account of the propensities and dispositions that lead us to our quite ordinary belief that we are part of a world of objects. The analysis is fundamentally different, however, because we are led by our innate propensities to believe something that mere habit can never inculcate.
We have two quite separate beliefs about objects, Hume observes. The first is that they are independent of our perceiving them. The second is that they continue to exist when we are not perceiving them. Hume suggests that each belief entails the other, but he is only half right. It is certainly true that if objects continue to exist when I am not perceiving them, then [given Hume’s view of consciousness] they must exist independently of me. But it is not quite true that if they are independent of me, then they must continue to exist when I am not perceiving them. It is, after all, logically possible that objects pop in and out of existence precisely when I happen to be perceiving them, and yet exist independently of me in the sense that they could continue to exist, but just don’t. [Such a view would bear a striking similarity to the doctrine known as Occasionalism, which arose in early Arabic philosophy and was embraced most famously in
Consider, if you can, the way the world looks to a newborn infant. The infant is presented with a “buzzing, blooming confusion” of sense perceptions [to quote William James’ felicitous phrase], which come and go with great speed and variety. There are repetitions and patterns, but very little in the way of constancy. The baby looks up in its crib and has a visual impression of its father’s face [I now speak from experience]. For a few moments, there is a succession of visual impressions that quite closely resemble one another – that is to say, the father continues to look at the baby, his face turned toward the baby’s face. But then the father turns to pick up a fresh diaper, and the succession of resembling impressions is replaced by a new succession of impressions, these of the side of the father’s head. After a bit, yet another succession of impressions occurs, which strikingly resemble the first succession. The father has turned back to look again at the baby.
Now, it is all very well for us to describe the several series of impressions in this manner – first of the father’s face, then of the side of his head, then once more of his face. But there is no way that the baby can know that this is the correct description. How can it possibly arrive at the belief that the father’s face continues to exist in the interim when it is turned away And how can it form the belief that the impressions of the face and the impressions of the side of the head are the same object as seen from two different perspectives? Indeed, how can it form the idea that the multiplicity of impressions are all impressions of the identically same thing? What can it even mean to say that all of these impressions are one despite the fact that they are manifestly many?
It will not do to appeal easily to habit, as Hume did in the case of our belief in causal connexion. For, Hume points out, the propensity appealed to in the analysis of causal inference leads us to form the habit of expecting that the part of the world we have not yet observed will be relevantly like the part of the world we have observed. Fire is repeatedly conjoined with heat, and so we expect that it will be so conjoined in the future. But it is logically impossible ever to perceive a part of the world that I do not perceive. So I cannot, as it were, repeatedly observe that an object continues to exist when I am not observing it, until my innate natural propensity leads me to form the habit of believing that it will in the future go on existing when I am not observing it. Clearly, something new and rather extraordinary is going on here, and my entire fully fleshed out conception of the ordinary world of people and objects depends upon it.
I am going to do my best to summarize Hume’s account of the mental processes by which I form the conception of independently existing objects and repose belief in that conception, but before I begin, let me identify the core notion to which Hume is appealing. Briefly, it is this:
The mind has, as it were [my words, not Hume’s] a lust for regularity and order. The propensity to develop causal habits of association is one simple manifestation of this urge to impose order on its perceptual experience. But sense experience is in fact too gappy, too irregular, too lacking in reliable patterns of constant conjunction, to trigger in us a satisfying picture of a regular world order. Too often, for example, after I have observed several instances of conjunction of resembling instances, I encounter instances in which one, but not the other, is present. I can of course respond to this irregularity by weakening the habit of association. Perhaps, after all, fire does not cause heat. But instead, the mind yearns after order, and so posits the existence of the unobserved correlate. If I feel heat and do not see fire, rather than begin to doubt that fire causes heat, I assume that there is, unobserved by me, a fire causing the heat. Clearly this assumption arises out of a quite separate propensity. It cannot, in the nature of the case, be simply another instance of the propensity that led me to form the habit of associating fire with heat. So the baby lying in the crib brings greater order to its perceptions than they offer by positing the continued existence of the father’s face during its absence from the flow of perceptions. And so too, by an even more complicated mental process that we have not yet explored, it comes to suppose that the face and the side of the head of the father are two different parts of identically the same object.