Christopher Kerwick

Metaphysical Realism and Epistemology Option 2 Chris Kerwick 2018 Maxim 3: Settle the realism issue before any epistemic or semantic issue (Michael Devitt, Realism and Truth, 1991, p. 4). In this essay, I challenge Devitt’s third maxim, specifically the part about settling the realism issue before any epistemic issue. I’ll call this Maxim 3E (E for epistemic) to distinguish it from Maxim 3S (semantic). The best one can do towards satisfying Maxim 3E is to assume realism before tackling epistemic issues. This is what Devitt does, and the suggestion that the realism issue has thus been satisfactorily settled leaves him open to Putnam’s accusation: “Devitt's dismissive attitude is as unphilosophical as Samuel Johnson's stone‐kicking” (cited in Devitt, 2010, p. 51). It is unfair, however, to suggest that Devitt does nothing more than kick stones and insist, for he does go on to make a solid argument for realism based on abduction. What he doesn’t seem to appreciate is that his abductive argument relies upon epistemological principles. The abductive argument itself is not enough. There is a need to articulate why this method of justifying the realist doctrine is compelling. Thus, the challenge of scepticism can be met rather than simply fobbed off, which is the way Devitt is inclined to treat it. The abductive argument is compelling, I suggest, because it responds to our psychological need for our world to make sense, and our feeling that, at least to some degree, it does make sense. While this may sound like a mere argument from wishful thinking, I will suggest that it actually strengthens the claim that realism is a rational position. In the following, I first analyse Devitt’s arguments in the fifth chapter of Realism and Truth: his negative and positive arguments for realism, and his implicit argument for Maxim 3E. I then go on to pay more attention to his positive argument, the argument from abduction. After then considering Devitt’s understanding of what epistemology should be, I look at two different conceptions of what is meant by knowledge (fallible and infallible) and why we care about it. Devitt’s conception of knowledge is surely of the fallibilist kind. However, human psychology demands we pay tribute to the infallibilist conception as well, with its demand for certainty. I conclude that in order to satisfactorily respond to this demand, which undergirds the sceptical challenge, Devitt’s abductive argument needs to be supplemented with a defence of the reasonableness of trusting in knowledge that is less than certain. Thus, contrary to Maxim 3E, it is necessary to do epistemology in order to settle the realism issue. Devitt’s implicit argument for Maxim 3E I’m not aware of anywhere in Realism and Truth that Devitt makes a straightforward argument for Maxim 3E. This is understandable, since his maxims appear as methodological guides, rather than conclusions. Nonetheless, I think Devitt does make an implicit argument for the maxim in his fifth chapter, “Why be a Common-Sense Realist?” (p. 60-82). In this chapter, Devitt argues for the realist doctrine that “tokens of most current observable common-sense and scientific physical types objectively exist independently of the mental” (p. 24. Henceforth, references to realism are to common-sense realism). Here, Devitt engages with a variety of historical figures – Descartes, Locke, Kant and others - whom he charges with trying to save realism in altogether the wrong way: by doing a priori epistemology. He then offers his own argument for realism, and finishes by presenting a program for naturalized, empirical epistemology. Below I set out Devitt’s arguments in standard form; first his arguments against scepticism and a priori epistemology, and then his positive arguments for realism. I then set out an argument for Maxim 3E which I take to be implicit in the whole chapter. The common starting point for both of Devitt’s explicit arguments (against a priori epistemology and for realism), is that our experience is as-if there were an external world, filled with the everyday objects of common sense. From there: Argument 1: against scepticism and a priori epistemology in general (p. 61-73) 1. Scepticism claims that realism about the external world is justified only if good reasons can be given for eliminating alternative hypotheses, such as that the apparent world is an illusion. 2. Undefeated scepticism is very costly, leaving us with nothing more than instantaneous solipsism. 3. Beating scepticism on its own terms requires a priori epistemology. 4. A priori attempts to save realism from scepticism by seeking an area of knowledge which is not open to doubt and which can serve as a foundation for other claims to knowledge (the foundationalism of Descartes, Locke, and others) all fail. 5. Kant’s a priori constructivism is to is unable to say anything about the real world, and thus also fails to save it from scepticism. 6. It is not clear that there is any such thing as a priori knowledge in any case. 7. Therefore, scepticism is unanswerable (by 3, 4, 5, 6). 8. Therefore, scepticism might be right (by 7). 9. Scepticism is uninteresting. 10. We don’t have to accept the sceptical challenge set out in 1 (by 9). 11. We should set scepticism aside (by 9, 10). Argument 2: for realism (p. 73-79) 12. We are spontaneous common-sense realists. 13. There are no plausible alternatives to realism (by Argument 1). 14. The most plausible explanation for the fact that experience is as-if there were an external world, filled with the everyday objects of common sense, is that there is an external world filled with the everyday objects of common sense. 15. We should be realists (by 12, 13, 14). Argument 3: for Maxim 3E 16. It’s not necessary to do epistemology to settle the realism issue (by Argument 1, Argument 2). 17. We should forget about a priori epistemology, which may not be possible in any case. 18. We should turn our efforts to empirical (naturalized) epistemology. 19. Naturalized epistemology proceeds from realism, and thus benefits from the realism issue being settled. 20. We should settle realism issue (by 16, 18, 19). Each of the three arguments has some weaknesses. In Argument 1, there is absolutely no justification for the claim in 9, and thus 10 and 11 are also undermined. If we are to disarm the sceptical challenge, we need to do better than simply claim that it is uninteresting. Perhaps Devitt thinks that the claim is justified by the conclusion in 7, that scepticism is unanswerable, but it doesn’t follow that it’s uninteresting. My own hunch is that the dismissive claim in 9 is motivated by horror at the prospect of instantaneous solipsism, and lack of a better tool with which to dispatch of it. Better tools, however, are available. While Devitt may have shown that the sceptical challenge is unanswerable according to its own standards, this doesn’t show that it can’t be answered in other ways. I think it can, and must be, if we are to be responsible realists. There are also weak points in Argument 2. 12 and 13 don’t offer much to support the claim in 15. The strongest justification for the realist conclusion in 15 is the abductive argument in 14. 14 is correct, but itself needs to be justified by appeal to epistemological principles, elaborating on why realism is plausible. Making this elaboration not only strengthens the argument for realism, but also gives us the tools for a more adequate answer to the sceptical challenge. Much of the remainder of the essay will be taken up with this task. In Argument 3, 16 fails because of the weaknesses in Arguments 1 and 2. Devitt hasn’t established that realism can be settled without doing epistemology. I also doubt the strength of Devitt’s point in 17. Nonetheless, the positive value of the argument in 18, 19 and 20 is not much harmed, and thus at one point below I also briefly consider Devitt’s agenda for naturalized epistemology. The argument from abduction According to Devitt, “the argument for realism starts from folk theory and scientific theory” (p. 73), which posit the entities of common-sense. We are confident of most of these posits, allowing for the fact that we might be wrong about some of them. There is, he acknowledges, some folk epistemology here. Folk epistemology holds that these things exist independently of our knowing them. “That is how we arrive at our realism (p. 73).” This, of course, is not really an argument for realism at all. It is just a restatement of what the folk believe. Devitt, one suspects, would prefer to leave things there. He acknowledges, however, that many people aren’t satisfied. How can the doctrine be defended? Devitt’s first move is to point out that he has shown the implausibility of anti-realism (we have granted him this) and that, thus, realism is the only plausible doctrine left. But again, he acknowledges some will want a demonstration of its plausibility. In looking for a way to defend the plausibility of realism, Devitt appeals to abduction. Abduction is often described as ‘inference to the best explanation.’ In the Stanford Encyclopedia of Philosophy, Igor Douven (2017) defines abduction as a non-necessary inference (which distinguishes it from deduction) that makes an implicit or explicit appeal to explanatory considerations (which distinguishes it from induction, which appeals only to statistical considerations). What is it exactly that abduction is supposed to explain, in this case? Devitt says that, intuitively, what we want realism to explain is “the facts of experience” (p. 74). The difficulty we have is that our ordinary language already bears the hue of an assumed realism, replete with real things, before we even try to defend it. Nonetheless, Devitt acknowledges that it is possible to speak carefully in such a way as to distinguish things from our experience of them. This careful language he refers to as “as-if” language (p. 74). He shows the difference between ordinary language and as-if language with the following examples. Ordinary language: There is a black raven on the lawn. As-if language: It is as if there is a black raven on the lawn. The realist, Devitt says, is willing to make both statements. If we begin with the as-if statement, and then proceed to ask why it holds (eg. Why is it as if there is a black raven on the lawn?), the best explanation is that it holds because realism is true (ie. There is, indeed, a black raven on the lawn). Realism therefore explains why our experience is as-if there is an external world full of the objects of common sense. I find Devitt’s position here to be persuasive. Realism is the best explanation for the way the world seems. And yet an important question has been left untouched. Is it reasonable to allow ourselves to be persuaded? Abduction is not conclusive in the way that deduction is. Just because the existence of a real external world is a plausible explanation for the way things seem, doesn’t mean it couldn’t be wrong. I will try to respond to this worry below. But first, let me look briefly at Devitt’s attempt to set the worry aside and move on. Assuming that realism has been settled, he goes on to consider what epistemology should be doing when it’s not occupied with justifying our realist beliefs. Naturalized Epistemology At the beginning of a subsection titled “Naturalized Epistemology” (p. 75), Deviitt claims that our best science shows that the standards of knowledge set by the sceptic cannot be met, short of instantaneous solipsism. Such standards are too high, can never be met, are literally incredible and should be ignored as uninteresting. Exactly what Devitt means by science showing this is not clear. Such conclusions surely haven’t been arrived at by simple observation. It is, I suspect, philosophy of science, rather than science itself, which has reached these conclusions. There have been inferences drawn, and the reasonableness of such inferences needs some elaboration which I doubt science itself can provide. Nevertheless, we can grant that sceptical standards of knowledge have not been found methodologically useful when it come to doing science. So, having dismissed the quest for certainty, Devitt goes on to ask what role remains for epistemology. His answer is that “it is left with the task of explaining our coming to know science (and common sense)” (p. 76). Devitt draws an attractive picture of some of naturalized epistemology’s agenda (p. 76-77). (1) A naturalized epistemology recognises that human beings are born with dispositions to respond selectively to stimuli at our surfaces, to order our experience in certain ways, and thus arrive at certain beliefs. For Devitt, our arriving at such beliefs is pre-reflective and does not involve inference. Others have defined inference differently, suggesting that abductive inference to the best explanation is often a subconscious process (see Douven), and this does not seem to contradict Devitt’s account of how we arrive at our initial realist beliefs. In any case, the first job of naturalized epistemology is to uncover how we in fact arrive at our beliefs. (2) The second job of naturalized epistemology is to help us to weigh and discriminate among our beliefs, becoming critical and scientific. (3) In order to achieve 2, naturalized epistemology takes on a normative task, identifying what makes a belief-producing procedure good or bad. A good belief forming procedure leads to justified beliefs, and thus is characterized by its tendency to lead to truth. But truth, here, is not the truth of realism. It is truth within the world to which realism is committed. It is not so much the truth that cats and dogs exist, as the truth about the cats and dogs we already believe in. As Devitt says, naturalized epistemology “takes science, and hence its posits pretty much for granted,” (p. 76). The same, of course, goes for the posits of common sense. It is within the constraints of these assumptions that naturalized epistemology conceives of truth. What is true is what is true about world which we already take to be real independently of us, and what makes for good belief forming procedures are analysed within naturalized epistemology within this setting. But if we wish to justify our belief in the external world in the first place, naturalized epistemology does not seem to offer a lot of help. For that purpose, we need to go back to the abductive argument, and how it works to justify our realist commitments. To help uncover why abduction works as an argument, and why it is reasonable to allow ourselves to be persuaded by it, let me take a detour through some recent debates among epistemologists over the meaning and value of knowledge. What knowledge is and why it matters to us In recent decades, the doctrine of fallibilism has grown in popularity among epistemologists. Briefly, it is the view that knowing a claim to be true can be consistent with the evidence for that claim being less than conclusive and even with the claim’s being false. This is contrasted with the Cartesian, or infallibilist conception, according to which knowledge requires the highest possible level of justification, that is, justification which is conclusive and guarantees the truth of the known proposition (Hannon, p. 1119-20). Laurence Bonjour has presented a compelling case that common folk hold a strong, Cartesian conception of knowledge, and this is why, when presented with cases in which someone knows something by luck or in less than conclusive fashion, they are intuitively inclined to disallow that the person really knows (Bonjour, 63-77). Fallibilists, on the other hand, are able to point to the fact that people regularly and even usually make knowledge claims in cases where their evidence is less than conclusive. It seems harsh to suggest that people regularly speak falsely, and so fallibilists suggest that the correct conception of knowledge is a weaker, fallibilist conception. My own view is that the debate between fallibilists and infallibilists is misconceived if we think one side must be wrong and the other right. Both are correct in important ways. We do think of knowledge in its strong sense as applicable only when we know something which is (correspondence) true, and yet we often work with weaker conceptions of it in our daily lives, treating knowledge claims as responsible when the belief asserted is justified to what we take to be a reasonable degree, given the context and the pressures to which the belief is subject (eg. whether much is riding on whether we are wrong in a given circumstance). There is, thus, a certain duplicity in our conception of knowledge, a duplicity which I suggest has to do with our situated existence and our human psychology. Our situated existence means that we can only ever get a grip on the world as it appears to us. We can never get a God’s eye view, never get direct knowledge about how the world really is. We can only perceive what comes to us through our senses, and reason according to the principles that seem to us to hold. Whether they hold absolutely is something we can never objectively assess, since any arguments we appeal to will be arguments that make sense to us, according to the principles of logic as the appear to us. I believe there are principles that hold regardless of human nature, but again, I have no way of proving this except by appealing to them. So, I take it, there is no human knowledge which is guaranteed absolutely to be true, independently of us. This would seem to suggest that we should abandon the strong, Cartesian conception of knowledge as being impossible, and yet we don’t. It seems to be a given of human nature that we hunger to get an unshakeable grip on what is real, and at least at times, to strive for it. Sometimes this drive might derive from anxiety, a sense of insecurity that makes us hunger for something to anchor ourselves to. At other times, it might be to do with a certain fascination with the world, and an aspiration to be more and more intimately acquainted with it, to the point of being indubitably united with it. Either way, the value that we set upon it such certain knowledge, I suspect, has to do with how we feel, what we want, and what we feel we need. Much of the time though, our felt need for certainty fades into the background as we simply get on with our lives. We can enjoy living in our world, and make reasonable headway within it, without certainty. Thus, rather than focusing on when our justification for our beliefs is conclusive, we use our discretion to decide when it is appropriate to conclude the quest for further justification for our beliefs, to come to conclusions. For Devitt, the time to come to a conclusion about the reality of the external world and its everyday objects is at the beginning. He is not interested in playing the Cartesian game. He’s a fallibilist through and through. Let’s just assume that every day objects are real, and then get on with finding out more about them as the need or the fancy takes us. But Devitt is being slightly unrealistic here. As noted, the human drive for certainty, for an indubitable grounding in reality, always comes back to us at some time or other. And this drive, even if it cannot be satisfied, yet needs to be satisfactorily responded to. A two-step response to scepticism There are thus, I believe, two elements necessary in an adequate response to the sceptical challenge. The first is to show that, while our belief in the real world cannot be established with absolute certainty, an argument can be made for it. This is where the abductive argument comes in. Devitt is right; the most plausible reason for our experience being as-if there are cats and dogs and ravens and trees, is that there are cats and dogs and ravens and trees. Still, one suspects, the sceptic will not be satisfied. That realism is plausible doesn’t guarantee that it is true. That we can’t think of a better explanation also doesn’t guarantee that it’s true. At this point, Devitt seems close to despair. “We are confident that the sceptic who refuses to follow us in this will have a short life; but in the end we find ourselves with no argument to convince him” (p. 64). He resorts to emotivist dismissal: “scepticism is simply uninteresting” (p. 64). What is needed, however, is not dismissal, but a sober treatment of why it is reasonable to settle for less than certainty. Perhaps the folk most practiced at such treatments are the folk Devitt dismisses elsewhere, religious folk (p. 4). Devitt may have more in common with them than he realises. For it simply a fact of human life that all of us, at the end of the day, if we are not to be swallowed up by solipsistic despair, live by faith of some sort. Without being uncritical, we pay attention to our feelings, to what is aesthetically appealing, to what seems fitting, what makes sense to us, and then, without conclusive evidence, psychologically healthy people are able to use their discretion to come to a conclusion, and to rest on beliefs that are less than certain. Similar dialogues, of course, take place within science, with its talk of the simplicity, elegance and explanatory power of different scientific models. Various writers have suggested that our emotions play an important part in our rationality. Some such broad understanding of rationality is needed, I suggest, if a commitment to realism is to be defended as rational. With such wisdom to guide us towards responsible assent to beliefs, we are then able to get on with whatever we are about. Abduction to the best explanation relies, for its persuasive power, on such epistemological principles of responsible belief. I don’t have space to do justice to this thesis here, or defend it. I hope I have done enough, however, to show that some such treatment of human psychology and practical wisdom is needed to sure up the argument from abduction, and is a more satisfactory response to the sceptical challenge than the dismissive attitude that Devitt sometimes displays. Conclusion I began this essay with the contention that Maxim 3E is wrong, and that Devitt’s own argument for realism relies on epistemological principles. As we have seen, he makes an argument for realism from abductive inference to the best explanation, and expects his hearers to be persuaded by it. He seems not to appreciate, however, that the abductive argument relies for its persuasiveness on epistemological principles about what makes an argument appealing to us and when it is reasonable to come to a conclusion in the absence of certainty. Once such epistemological principles are brought into the open and incorporated into the argument, the sceptical challenge is more satisfactorily answered. Without being defeated on its own terms, it is shown to be based in psychological attitudes which are reasonably met by adopting different attitudes, attitudes which pay attention to our feeling for what make sense aesthetically and in other ways. Such attitudes dispose us rest responsibly with conclusions that are less than certain, conclusions which are no less rational for being uncertain, and which enable us to get on with our lives and with other important questions. Bibliography Primary Resource Devitt, Michael. Realism and Truth. Coursepack for PHIL40005, pages 89-279. (Page numbers given in the text are according to the internal numbering of the copied pages, not the page number of the coursepack). Other Resources Devitt, Michael. Putting Metaphysics First. Oxford: Oxford University Press, 2010. Bonjour, Laurence. “The Myth of Knowledge.” Philosophical Perspectives, Vol. 24, Epistemology (2010), pp. 57-83. Douven, Igor. “Abduction.” Internet Encyclopedia of Philosophy (2017). https://plato.stanford.edu/entries/abduction/ Retrieved 15 Oct 2018. Hannon, Michael. “Fallibilism and the value of knowledge.” Synthese, Vol. 191 (2014), pp-.