<i>The Epistemic Role of Consciousness</i>
Elijah Chudnoff
Abstract
Epistemology and philosophy of mind are areas of philosophy where one expects to find significant cross-fertilization. We know with our minds. So, knowledge is the sort of thing that our minds can have. And our minds are the sort of thing that can have knowledge. What one will make of all that, however, turns on the positions taken in some of the main debates that structure the epistemological playing field.Both internalists and externalists will think it matters that we perceive, remember, reason, and so on. But why does it matter? And how much need an epistemologist worry about the details? Externalists will think the various mental processes matter because of their differential contributions to the reliability with which we form beliefs. And they are in a good position to see opportunities for rich engagement with cognitive science and related fields insofar as such reliability is a function of how our mental mechanisms actually work.The answers for internalists are less clear. In his ambitious new book, The Epistemic Role of Consciousness, Declan Smithies pursues the following line. What happens in our minds matters for epistemology because we are phenomenally conscious. Philosophical zombies who are physical and functional duplicates of us but lack phenomenal consciousness cannot know anything. They cannot know anything because they cannot have any epistemic justification for their beliefs. And that is because all epistemic justification derives from mental states that are, or are disposed to cause, phenomenally conscious mental states.Smithies has written a long book that is chock-full of careful definitions, thorough arguments, and illuminating discussions of the literature. It is an impressive achievement that is both uncommonly wide-ranging and admirably systematic. I found it to be a fascinating and instructive read throughout. However, while I share a number of Smithies’s key starting assumptions, the view he ultimately develops is one I found myself disagreeing with in almost every particular.Here is how I think it is all supposed to fit together.The first part of the book, titled “Philosophy of Mind,” builds an intuitive case for the view Smithies calls Phenomenal Mentalism: necessarily, which propositions you have epistemic justification to believe at any given time is determined solely by your phenomenally individuated mental states at that time. Phenomenally individuated states include both current experiences which are defined by their phenomenal character and standing states which are defined by their dispositions to cause experiences—for example, standing beliefs, which Smithies takes to be defined by their dispositions to cause conscious judgments. The arguments for Phenomenal Mentalism culminate in chapter 6, and they are variations on themes from two well-known challenges to externalism. The first is Stewart Cohen’s new evil demon challenge. The second is Laurence Bonjour’s Norman the clairvoyant challenge, which Smithies crosses with blindsight cases in ways I think make for some of the most convincing arguments in the book. The previous chapters of part 1 deal with preliminaries about consciousness (chapter 1) and mental representation (chapter 2), and also defend specific views about the epistemology of three key mental phenomena: perception in chapter 3, cognition in chapter 4, and introspection in chapter 5. I return to some of the details below.The second part of the book, titled “Epistemology,” develops theoretical arguments for the view Smithies calls Accessibilism: epistemic justification is luminously accessible in the sense that, necessarily, you’re always in a position to know which doxastic attitudes you have epistemic justification to hold at any given time. Supposing this view is true, then how does it relate to Phenomenal Mentalism? In chapter 7, Smithies argues that Phenomenal Mentalism is the best explanation of Accessibilism and that Accessibilism generates theoretical support for the intuitions motivating Phenomenal Mentalism. So, provided the arguments for Accessibilism developed in later chapters are compelling and independent of Phenomenal Mentalism, Smithies’s overall view exhibits a pleasing form of reflective equilibrium. Those later arguments focus on the following JJ Principle: necessarily, you have justification to believe that p if and only if you have higher-order justification to believe that you have justification to believe that p. One is an argument from the stability of justification under reflection (chapter 8); the other is an argument from the irrationality of epistemic akrasia (chapters 9 and 10). More on these in a moment. The book ends with two chapters that compare Smithies’s view with prominent forms of externalism (chapter 11) and a rival form of internalism, according to which all epistemic justification is grounded in seemings (chapter 12).The first crack in the edifice is that Phenomenal Mentalism—which is intuitive—does not explain Accessibilism on its own. To explain Accessibilism, Smithies relies on a series of claims about the evidence provided by perception, cognition, introspection, and reason—all of which claims are themselves extremely counter-intuitive. Let’s take them in reverse order of appearance.Reason tells us about logic and evidential support. Intuitively, justification from reason can be partial and misleading, and different thinkers will exhibit differences along these dimensions depending on how fully and competently they’ve done their reasoning. According to Smithies, everyone has justification for all and only correct beliefs about logic and evidential support.Introspection tells us about our own phenomenally individuated mental states, such as our current experiences and our stored beliefs. Intuitively, justification from introspection can also be partial and misleading, and if one momentarily consciously judges that p but has stored away an unconscious belief that not-p, then one can be rationally misled about the contents of one’s own unconscious mind. According to Smithies, everyone has justification for all and only correct beliefs about their own stored beliefs.Justification by cognition occurs when what one already believes makes a difference to what one has justification for believing. Intuitively, justification by cognition works via some kind of transmission of old justification for some beliefs to new justification for other beliefs. According to Smithies, justification by cognition works because in merely possessing a belief one thereby has defeasible justification for the possession of it.Perception tells us about our environment. Intuitively, if two people are looking at numerically distinct but qualitatively indiscernible objects, then each has perceptual justification for a belief about the seen object but not for a belief about its unseen duplicate. According to Smithies, each has perceptual justification for beliefs about both seen and unseen objects. Given his claims about perception, cognition, introspection, and reason, Smithies is able to produce a plausible story about how Accessibilism might be true. Given their lack of intuitive motivation, however, it is important that Smithies have good independent reasons for thinking that Accessibilism is true and therefore needs some such explanation.A second challenge to the overall project is that the reasons given for Accessibilism in the second half of the book lack the required independence. One of the main arguments is that justification is stable under reflection but wouldn’t be if it weren’t accessible. A key move in this argument is the following: you have justification to believe that p if and only if you have some basis on which you would believe that p after a fully justified process of reflection. But this claim is only plausible if we assume that a justified process of reflection cannot be misleading—that is, if we assume Smithies’s view about the evidence of reason. Another of his main arguments for Accessibilism is that it follows from the irrationality of epistemically akratic combinations of beliefs, such as the belief that p combined with the belief that one doesn’t have justification to believe that p. This argument ultimately hangs on commitment to the view that it is impossible to have misleading higher-order evidence about what one’s evidence supports. The story is involved and unfolds over two chapters. A key moment is Smithies’s assumption that the irrationality of a combination of beliefs must be explained by that combination’s conflict with one’s evidence. Given this assumption and the irrationality of epistemically akratic combinations of beliefs, it follows that one’s evidence never supports such akratic combinations. But for this to be true it must be impossible to have misleading higher-order evidence about what one’s evidence supports. The idea that this is impossible, however, is only tenable given Smithies’s counterintuitive claims about the various kinds of evidence, especially the evidence of introspection and reason.A big gap in my review so far is that I haven’t mentioned the distinctions between propositional and doxastic justification and between ideal and nonideal rationality. One might have propositional justification for believing p while lacking a doxastically justified belief that p, either because one doesn’t believe that p, or one does but in a way that is not appropriately sensitive to one’s evidence. Ideally rational agents form their beliefs in a way that is perfectly sensitive to their propositional justification; nonideally rational agents suffer impediments such as limited time and energy that render them imperfectly sensitive to their propositional justification.Let’s return to the claim that everyone has justification for all and only correct beliefs about logic and evidential support. As a claim about doxastic justification, the claim is obviously false, since no one believes all logical truths. As a claim about propositional justification, the claim is at least not open to the same objection. Suppose I consider a very complicated logical truth without examining a proof. Should I believe it? Plausibly no, but if not, why not, given that I have justification for doing so? Answer: I am not an ideally rational agent, and we have to tailor our epistemic shoulds and shouldn’ts to fit my nonideal rational capacities. Smithies makes similar moves throughout the book. One reason to remain dubious is that the distinctions between propositional and doxastic justification and between ideal and nonideal rationality can be deployed with more intuitive plausibility to undermine rather than support Smithies’s views. Take the case of epistemic akrasia. A natural way to describe what’s problematic about epistemically akratic combinations of beliefs is that while one could have propositional justification for such a combination of beliefs, one could never form a doxastically justified set of such beliefs on the basis of that justification, and that this is so even if one is an ideally rational epistemic agent. Final assessment of Smithies’s view hinges on determining how best to handle these two distinctions.While the jury is still out on that, I can confidently say that Smithies has enriched our understanding of the relevant issues, and in doing so has made an important contribution to the literature on epistemology and philosophy of mind.