Anil Gomes's The Practical Self

Abstract

Gomes's rich and compelling book revolves around an extended line of argument for the thesis that a self-conscious subject must be one object amongst many in an objective world: self-consciousness entails objectivity. Others have offered arguments for the same conclusion; but, in contrast with his own, Gomes finds all of these wanting. After setting out my own understanding of Gomes's central argument, I will raise a series of concerns about each of its key moves.

This is an original and significant line of argument. In what follows, I raise five critical questions about it: two each about premises 1 and 2, and one about premise 3. I develop two of these in detail, into fully-formed objections to Gomes's argument; the others I leave almost to fend for themselves against his case for the thesis that self-consciousness entails objectivity.

By way of clarification of the content of premise 1, Gomes contrasts the cognitive agency in question with a corresponding passivity in perception. Here is what I think he means in the perceptual case. We select and initiate projects, like counting the stripes on a zebra, discerning which chair seat is closest in colour to the carpet, and so on. We focus and modulate our attention appropriately over time in order to execute them as best we can, checking and going back if necessary as we proceed. But which specific experiences and beliefs we find ourselves with at the end of the day is entirely outside of our control; and so it should be, if the result is to be determined by how things really are rather than by our own preferences and prejudices. We set the question and direct our attention and capacities to pursue it; the facts settle the answer.

Isn't it just the same with paradigmatically intellectual projects too, though, such as counting the primes between 0 and 100, or working out how best to accommodate all a child's friends at a sleepover without provoking too much over-excitement or antagonism, and so on? We pursue the project and keep our attention on the relevant considerations, taking each stage in turn, checking and going back if necessary as we proceed. But, again, which beliefs we find ourselves with at the end of the day is entirely outside of our control; and so it should be if the result is to be determined by how things really are rather than by our own preferences and prejudices. Just as in the case of perception, we set the question and direct our attention and capacities to pursue it; the facts settle the answer.

Perceptually: ‘is there a dark blue chair in the room? … ‘yes'; and, analogously, intellectually: ‘is there a prime between 37 and 43?’ … ‘yes'. Realism about the domain of enquiry in both cases surely legislates in favour of our ultimate passivity with respect to the outcome. So it is unclear to me precisely what the cognitive activity that Gomes is interested in comes to. Certainly, the comparison with perception suggests similarity rather than difference in this respect.

At points, Gomes suggests that the demand for cognitive agency, and our acceptance of it, stems from our ongoing commitment to revise and adjust our perspective on the world in the light of new evidence, or simply in the light of reconsideration by reflecting on the propriety of our current views. But, again, it seems to me misleading to suggest that this involves anything like an active intervention directly on those views themselves, moving our beliefs around as we might rearrange our furniture, as it were. Rather, it's a matter of recommitting ourselves to attending carefully and systematically to all the considerations, perceptual or intellectual, that are pertinent to having our beliefs either reconfirmed or revised in the light of the facts, in a way that is once again ultimately outside our control and settled by reality rather than by us.

I do not say there aren't many and various interesting differences between perceiving and thinking; but I don't find the contrast specifically illuminating in elaborating what Gomes takes to be our crucial cognitive agency as self-conscious beings. The direct objection arising out of Q1, then, is that Anil's relatively passing appeal to a contrast with the passivity of perception is insufficient as it stands to elaborate the target notion of cognitive agency. A more general challenge is for Anil to specify more precisely what it is that the first premise of his central argument asserts any self-conscious subject must accept about the acquisition and maintenance of their beliefs, either by developing the contrast between reasoning and perceiving further, or in some other way.

Here I have a relatively cheap point, but still one worth making, I think.

Given the difficulties we have just seen in formulating precisely the notion of cognitive agency that is supposed to be necessary for self-consciousness, it is surely correspondingly implausible to insist that every self-conscious subject must, simply as such, accept, understand, or know that they are a cognitive agent in just this sense. Of course, this raises the question precisely how intellectually demanding acceptance of cognitive agency is supposed to be. On the assumption suggested by my reading of Gomes's discussion at least, that this requires grasp of the proposition involved in asserting the cognitive agency in question, then the cheap point surely stands as a further challenge to the first premise of Anil's main argument for the thesis that self-consciousness entails objectivity.

If I am, as Gomes insists, an agent in working out how best to accommodate all a child's friends at a sleepover, say, then I take it that that is something that I am doing intentionally, and so something that I know that I am doing in the way in which I generally know what I am doing intentionally. I certainly don't have a worked-out account of such knowledge in intention. But it is a widely accepted phenomenon. And why is it not the source of my acceptance of my cognitive agency whenever I am actively thinking in the relevant sense?

Perhaps Gomes distinguishes specific such knowledge, in any particular case of our exercise of cognitive agency, from the acceptance that he is primarily concerned with, namely, that I am, quite generally, the agent of my thinking. But even acknowledging that distinction, it might be replied that any such general acceptance we arrive at is derived from the more specific individual instances that we have through our knowledge in intention in specific cases. So, without an independent objection to that route, the critical point remains.

As I understand it, Gomes's argument here is this.

I can see reasons to doubt both substantive premises: (F2) & (F3).

Against F2, we can surely take on projects that we do not explicitly accept as attainable (by me/us, here, now, in this way,…?) at all. Even the weaker claim that taking on a project depends upon not regarding it as absolutely unattainable may be too strong in certain cases. Crash-landing into the sea in an aeroplane, for example, I may commit myself to the project of swimming to safety on a distant, just-visible island, however unattainable I take my successfully reaching safety to be. In any case, the gap between taking on a project and positively accepting it as attainable surely increases the more extended, demanding, and complex the project in question happens to be. Examples that help to make the point might be solving the P vs NP problem; or arriving at a stable solution to the Middle East Crisis. In such cases, the subject may well embark on an extended, complex, and extremely demanding enterprise without accepting its attainability, by them, or even by anyone, and certainly not necessarily its attainability by them in the way they are going about it. Still, this is the project they take on. So extended complex cases of this kind constitute counterexamples to Anil's F2.

A first reply might be to attempt to close the gap between taking on such projects and the subject's acceptance of their attainability by insisting that the project in question in such cases is only ever to try – to solve the P vs NP problem; or to arrive at a stable solution to the Middle East Crisis – and trying may well be something that the subject accepts is attainable. But this strikes me as forced. Trying just is taking on the project itself – of actually solving the P vs NP problem, or of actually arriving at a stable solution to the Middle East crisis and doing everything possible to attain it, whether or not this is eventually successful. So, the gap remains, especially, as I say, in such extended and complex cases.

In the light of cases of this kind, a second reply would be to insist that premise F2 is supposed to apply only to brief, simple projects, and so only to each of a series of successive manageable sub-projects of any more complex and extended overall project of the kind we have been considering. It is not obvious to me why we should accept F2 even in this qualified form; but in order to set out all the pieces of my challenge to Gomes's case for the claim that self-conscious thinking depends upon faith in cognitive agency, I turn instead at this point to consider F3.

Given that cognitive agency is understood as what is involved in self-conscious thinking but absent in perceiving, and that we often accept that the ends of perceptual projects are attainable, accepting that the ends of such projects are attainable cannot in general require our acceptance of our agency in the relevant sense. Suppose I set myself the project of discerning the colour of the eyes of the person in front of me. In suitable circumstances, of good lighting, of the person being close enough to me with eyes wide open, and so on, I may surely accept that this is attainable. Yet, since the project is purely perceptual, it involves no cognitive agency of the kind that Gomes is concerned with. Hence my taking it on cannot involve any acceptance of such cognitive agency – at least on the assumption that I have been working with throughout, following Gomes, that such acceptance is factive. This case therefore constitutes a counterexample to the general claim a subject's accepting that the end of a project they take on is attainable depends upon their accepting their cognitive agency in doing so.

The critical point here is especially clear with respect to the simple component sub-projects of any more extended and complex task, where the analogy between cases of thinking and perceiving is stronger. That's to say, it may be that a subject's accepting that some complex project is attainable involves commitment to the idea that they have some control over the order and initiation of the various sub-projects that will, all being well, eventually accomplish it. But, in relation to its individual simple sub-projects, their commitment to their attainability seems to presuppose little more than the conviction that, if they direct their attention to these in the right way, the correct answer will be forthcoming, and will arrive out of their control, determined by the relevant facts rather than by their own preferences and prejudices, as it should be.

It may be objected to this discussion that it depends upon something like an early modern conception of demonstration, in the sense of extended, directed reasoning, as an appropriate sequence of intuitions, in the sense of simple intellectual perceptions. It does indeed depend on a conception of thinking along these lines, with the distinction between extended complex projects and their manageable simple sub-projects. But, first, this early modern distinction between simple intuition, and extended complex demonstration, in these senses, seems to me independently defensible, although it may be applicable only relative to an individual subject on a given occasion of reasoning. Second, and more importantly for present purposes, the basic distinction has been introduced precisely to make the best possible case individually for each of Gomes's F2 and F3. The fundamental problem for his overall argument for the claim that self-conscious thinking depends upon faith in cognitive agency is that this evidently rests on an equivocation as a result.

For atomic sub-tasks, premise F2 may be correct; but premise F3 is surely false. Our taking them on may presuppose acceptance that they are at least likely to come off: we will find ourselves with a conviction one way or another settling the question at hand. But such acceptance involves no commitment to our agency in respect of that specific outcome. Indeed, it is more likely to involve recognition that the answer will just come to us.

For complex extended projects composed of multiple such components, on the other hand, premise F2 is surely false. We can, and do, take on many such ambitious projects without antecedently accepting that they are attainable (by us, here, now, in this way, at all, …?). Furthermore, the only acceptance of agency required by acceptance of attainability is that equally involved in extended perceptual tasks that are precisely the contrast with cases of active thinking that Gomes aims to establish: acceptance, that is to say, that we have some control over the order and initiation of their various component sub-activities.

So, although there may be some projects for which both F2 and F3 are true, they are not both true for all projects in which we aim to settle some specific question in the light of the relevant considerations for and against, as Gomes's argument requires. His argument for the claim that self-conscious thinking depends upon faith in cognitive agency is therefore invalid.

These reflections, and those under my first question above – whether self-consciousness really requires cognitive agency in Gomes's sense at all – suggest that the problem may lie in his attempt to articulate the cognitive agency he is interested in by contrast with perception. Perhaps the point is just that, in both cases, we actively select the questions we pose, and pursue them by attending in perception and in thought to the relevant considerations by which they are passively settled, in successful cases, by the facts. It is far from clear that any of this requires an acceptance of our agency in this regard that can only be based upon faith in the way Gomes suggests, though.

The alternative, that Gomes apparently leaves entirely open, is that faith in cognitive agency, in those self-conscious beings who are related to an objective world of other thinkers, may in some cases be sustained by their interpersonal practices of holding each other accountable for their thinking. That he does indeed endorse something close to this view himself is suggested by his characterization of objective engagement as ‘central’, although not strictly necessary, for the faith in cognitive agency that he argues is in turn essential to self-consciousness. In that case, any relation to an objective world to be derived from a thinker's being self-conscious is that assumed from the start in regarding them as de facto situated in an interpersonal world of other such thinkers whom they hold accountable, and who hold them accountable, for their thinking.

Gomes is scrupulous in subjecting others' arguments from self-consciousness to objectivity to rigorous objection. Yet, if this understanding of his account is accurate, then his own argument is simply invalid. A self-conscious thinker may in principle have faith in their cognitive agency unsustained by any engagement with an objective world of other thinkers, which constitutes a direct counterexample to the final stage of that argument. So, by the standards he applies to others, his own case is also wanting.