II—Persistent Philosophical Disagreement

According to some philosophers, there are two awkward facts about philosophy. First, no philosophical problems have been solved; second, philosophers cannot agree about anything beyond that. The contrast with the natural sciences is then evident: many scientific problems have been solved, and there is appreciable agreement between scientists in scientific matters. But whatever we happen to think about the extent of progress and agreement in philosophy compared with science, the main issue is: to the extent that there is persistent disagreement in philosophy, what would explain it?

I

Introduction. According to some philosophers, there are two awkward facts about philosophy. First, no philosophical problems have been solved; second, philosophers cannot agree about anything beyond that. The contrast with the natural sciences is then evident: many scientific problems have been solved, and there is appreciable agreement between scientists in scientific matters. What is wrong with philosophy and what is wrong with philosophers?¹

There are various closely related issues here that do not quite come to the same thing. There’s the issue of why no philosophical problems have been solved. There’s also the issue of why there’s been no appreciable progress in philosophy. Those two issues differ. If there’s been no appreciable progress in philosophy, no problems have been solved, but the converse does not hold. There might be progress in philosophy, but it might also be asymptotic: always approaching solutions, but never reaching them. Then there is the issue of why philosophers persistently disagree. That is a further issue. It might be that there is progress in philosophy, and even that some of its problems have been solved. Disagreement would remain if there were philosophers with entrenched opinions who, for whatever reason, refused to fall in line. Consider mathematics: Cantor achieved a tremendous advance in our understanding of infinity despite the likes of benighted holdouts such as Kronecker.

Still, despite the differences between these issues, it will serve our purposes here to consider all of them. There is persistent philosophical disagreement. Why is this? Without a suitable explanation, it seems very implausible to suppose that, despite this, some philosophical problems have been solved or are approaching solution. Again, in the absence of a suitable alternative explanation, the fact that there is persistent philosophical disagreement seems to be evidence against the claim that there is substantive progress in philosophy (see Brennan 2010 and Chalmers 2015, §2; but see also Ballantyne 2014 and Hanna 2015).

Two staunch proponents of these pessimistic conclusions are Peter van Inwagen and William Lycan:

In metaphysics there is no information and there are no established facts to be learned. More exactly, there is no information and there are no facts to be learned besides information and facts about what certain people think, or once thought, concerning various metaphysical questions. (van Inwagen 2008, p. 10)²

But are there any successful philosophical arguments [where a successful argument that p is one that would convince any perfectly rational, intelligent and neutral party that p]? I know of none. (That is, I know of none for any substantive philosophical thesis.) … If any reasonably well-known philosophical argument for a substantive conclusion had the power to convert an unbiased ideal audience to its conclusion (given that it was presented to the audience under ideal conditions), then, to a high probability, assent to the conclusion of that argument would be more widespread among philosophers than assent to any substantive philosophical thesis actually is. (van Inwagen 2006, pp. 52–3)

Philosophical consensus is far more the result of Zeitgeist, fad, fashion, and careerism than of accumulation of probative argument. (Lycan 2013, pp. 116–17)

So, is there distinctively philosophical knowledge? Yes, I am forced to agree, but only dribs and drabs, here and there. Far less than Gutting has maintained, and nothing to write a song about. (Lycan 2013, p. 120)

Lycan is somewhat more concessive than van Inwagen. According to van Inwagen, the only knowledge to be had from philosophy is historical knowledge: knowledge of which philosopher said what and why. Lycan claims that, besides knowledge of various positions that might be taken on a given issue, we also know, of some of these positions, which ones are viable (i.e. not obviously false) and which ones are not (Lycan 2013, pp. 117–20). Beyond this, however, he denies that we have knowledge of philosophical facts, specifically, knowledge of which positions are the correct ones and which ones are not.

Van Inwagen remarks that ‘it would not be the whole truth to say that by definition there is no body of philosophical fact because it is a defining characteristic of philosophy that it has no information to offer’ (van Inwagen 2008, p. 11). Suppose this is at least part of the truth. All the same, we might still sensibly ask why this is a defining characteristic of philosophy. For consider: it is a defining characteristic of war that it is terrible, but there are informative things we can say about why war is terrible. Similarly, there might be other features of philosophy that explain why it has no information to offer (if in fact it does not).³ So it need not be misguided to try to find such an explanation.

Now, perhaps Lycan, van Inwagen and others are exaggerating the amount of agreement there is in science and the amount of disagreement that there is in philosophy. Perhaps they tend to focus on the big questions of philosophy and notice a marked lack of agreement on them, and then to turn to science and focus on the little questions and notice a marked abundance of agreement. I do not take sides on the issue of the extent of our knowledge in philosophy or the issue of whether there are any established facts in philosophy to be learnt.⁴ For the purposes of discussion, I will take Lycan’s mildly concessive view as read. There has been progress and corresponding agreement in recognizing many and various views as viable. (Two views might each be viable but also mutually inconsistent. So agreeing that a view is viable isn’t to agree that it is correct.) But nothing has been achieved beyond that. Those are the terms of the current debate. Van Inwagen and Lycan’s views are stark and vivid, and they bring the main issue into sharp focus. But whatever we happen to think about the extent of progress and agreement in philosophy, the main issue is: to the extent that there is persistent disagreement in philosophy, what would explain it? I will now consider three accounts of why there isn’t appreciable progress in philosophy.

II

Russell on Philosophical Progress. Russell closed his 1918 lectures on logical atomism with the following declaration:

I believe that the only difference between science and philosophy is that science is what you more or less know and philosophy is what you do not know. Philosophy is that part of science which at present people choose to have opinions about, but which they have no knowledge about. (Russell 1956, p. 281)⁵

Here Russell is pithy but cursory. First, there are scientific issues about which scientists have opinions but lack knowledge. Informed speculation is not the prerogative of philosophers. In Russell’s day scientists speculated about the nature of matter; in our own day they speculate about dark matter and dark energy. Second, Russell presents philosophy as having a stock of problems that, over time, are taken off its hands and solved by science: ‘every advance in knowledge robs philosophy of some of the problems which formerly it had [and] a number of problems which had belonged to philosophy will cease to belong to philosophy and will belong to science’ (Russell 1956, p. 281). Far from explaining the lack of progress in philosophy, this entails that there is progress, where philosophical progress is measured by the number of philosophical problems that have been solved. What remains on Russell’s view, however, is a lack of progress by philosophers in solving problems. Nothing he says explains why there is such a lack of progress. Equivalently, despite however many problems that have swapped over from philosophy to science, there are many that have not and which remain resolutely philosophical. Those problems remain unsolved, and Russell has no explanation of why that is so (see van Inwagen 2008, pp. 11–12).

III

MacBride on Philosophical Progress. The next account I will consider is taken from Fraser MacBride. He offers the following account of why there isn’t appreciable progress in philosophy:

What makes these problems so resilient is the fact that they are general and pluriform. We cannot expect them to receive a definitive resolution until the Epistemic End of Days … [To solve a philosophical problem requires understanding many things.] But we cannot acquire such pervasive understanding without relying upon our grasp of other concepts that may also turn out to be problematic and drawing upon the results of other disciplines. And we don’t know in advance what other questions will be thrown up by future developments within these disciplines or by the efforts of philosophers to integrate them into a unified scheme. But this isn’t a scandal to philosophy. It’s a consequence of the encompassing and compounding character of the problems with which philosophy deals that their resolution requires of us a synoptic understanding. (MacBride 2014, p. 231)

MacBride’s account is pioneering and thought-provoking. I have two comments on it. First, he emphasizes the need for philosophical inquiries to achieve overall coherence. But, as Duhem (1906, part ii, ch. vi, §2) helped show, this is a general feature of inquiry, and not a special characteristic of philosophy. In advancing a hypothesis, an inquirer relies upon a background of accepted hypotheses, and so upon her grasp of the concepts figuring in these hypotheses, ‘concepts that may also turn out to be problematic and [that draw] upon the results of other disciplines’. For example, in framing a hypothesis about the observational consequences of certain astronomical facts, a scientist relies upon accepted hypotheses about instrument design, the margins for experimental error, optics, the workings of our perceptual system, and so on. The concepts figuring in these hypotheses may be problematic for philosophical or scientific reasons, and they draw upon the results of other disciplines, such as engineering, statistics, optics and physiology. Since we cannot predict future developments in an inquiry, nor how its results are best integrated in a simple and comprehensive fashion, we should not expect a definitive resolution of any line of inquiry until ‘the Epistemic End of Days’. Any well-conducted inquiry will have the features MacBride cites, namely, using problematic concepts, utilizing the results of other disciplines, ignorance of what new questions will arise from developments in these disciplines, seeking coherence. So he has not identified a special feature of philosophy, and a fortiori has not identified a feature which explains the lack of appreciable progress in philosophy in contrast to science.

Second, there can be a lack of appreciable progress in philosophy even in cases where these features are not markedly present. Consider some debate in philosophical logic, such as about the nature of truth or alethic modality or entailment. And then consider debates in (say) metaethics about the nature of moral obligation or in (say) aesthetics about whether moral failings can enhance aesthetic value. I find it very doubtful that any of the debates mentioned in philosophical logic require for their resolution any of the debates mentioned in metaethics or aesthetics, or that a greater understanding of the latter debates and notions peculiar to them will advance our understanding of the notions of concern in philosophical logic. In fact, in so far as there has been progress in philosophical logic during the past century, it was due to increasingly specialized work that was markedly independent of developments in the particular areas of metaethics and aesthetics mentioned. Nor has progress in philosophical logic been affected by developments in other disciplines such as the sciences. (See Williamson 2013, p. 423, on how little evidence there is that the results of any branch of science would be of much help to research in modal logic.) The persistence of philosophical disagreement need not trace back to MacBride’s claim that to get a solution for any one philosophical problem we need something like a simultaneous solution for a raft of philosophical problems. Certainly many philosophical issues are interconnected, but the claim in question is overstated.

IV

Cognitive Closure and Philosophical Progress. The third account I will look at appeals to the notion of cognitive closure.⁶ A type of mind is defined as cognitively closed with respect to a property F iff the concept-forming capacity of a mind of this type cannot extend to an understanding of F. This third account thinks that there will be no end to philosophical disagreement, and the reason it gives is that our minds are cognitively closed to the solution of philosophical problems. That is, we lack the cognitive capacity to solve philosophical problems and to establish true and illuminating philosophical theories (McGinn 1993, chs. 1 and 9; van Inwagen 1996, pp. 255–6):

[O]ur minds are not cognitively tuned to these problems [i.e. the problems of philosophy]. This is, as it were, just a piece of bad luck on our part, analogous to the lack of a language module in the brain of a dog. (McGinn 1993, p. 13)

Our cognitive capacities, although they are very well fitted to the task of figuring out how cell division and rainbows work, are not at all fitted to the task of figuring out how consciousness and freewill work. ‘Scientific questions’ are just those general, theoretical questions that we are cognitively properly fitted out to answer, and ‘philosophical questions’ are just those that we are not. (van Inwagen 1996, p. 254)

As with Russell’s account, this account needs to explain why we are good at answering scientific questions but not philosophical ones. McGinn’s explanation is that we are good at using concepts which involve spatial thinking—and these are the concepts used in science—whereas we are not good at using concepts which don’t involve spatial thinking—and these are the concepts used in philosophy. Yet what about mathematics? We are good at doing mathematics—or at least as good as we are at doing science—but mathematical concepts do not involve spatial thinking. McGinn acknowledges the point, and says that mathematical thinking involves thinking about ‘quasi-spatial’ dimensions (McGinn 1993, p. 20). That is to say, such thinking can be formally represented as being about a space. But, by the same reckoning, we could say that philosophy involves quasi-spatial thinking. For we can say that the entailment and probability relations between propositions impose an ordering on them, and this can be represented as a quasi-spatial lattice. Here is what McGinn says about modes of thought that, he thinks, we do excel at. Such a mode concerns

an array of primitive elements [which] is subject to specified principles of combination which generate determinate relations between complexes of those elements. (McGinn 1993, p. 18)

But this characterization is satisfied by propositions. They are subject to specified principles of combination—for example, there are the principles governing the logical constants, and then there are the principles specified by Kripke in his model theory for modal discourse—that determine relations between complexes formed from propositions, including relations of entailment and of logical independence. For these reasons, then, the supposed contrast between mathematics and philosophy collapses. (This response does not require that those combinatoric facts exhaust the subject matter of philosophy. The point is that combinatoric facts are shared between philosophy and mathematics, and so McGinn cannot distinguish between these disciplines on these terms.)

Nor do appeals to evolution provide the explanation that the proponent of cognitive closure needs. Suppose it is claimed that humans are (comparatively) poor at philosophy because evolution did not equip us with minds that were up to the task: ‘Perhaps human metaphysicians … work by taking human intellectual capacities designed for purposes quite unrelated to questions about ultimate reality and pushing these capacities to their limits’ (van Inwagen 2008, p. 14); see also McGinn (1993, p. 5) and Ladyman and Ross (2007, §§1.1 and 1.2). But then it would be puzzling why our species struggles with metaphysics but excels in mathematics. And if it is claimed that our mathematical ability is a fortuitous offshoot of how the human brain evolved under environmental pressures, an explanation would be needed of why this does not carry over to the case of philosophy.

James Ladyman addresses essentially this point when he compares natural science (rather than mathematics) with metaphysics:

In response, though, note that even if it is granted that natural selection cannot explain how natural scientific knowledge is possible, we have plenty of good reasons for thinking that we do have such knowledge. On the other hand, we have no good reasons for thinking that metaphysical knowledge is possible. (Ladyman 2007, p. 183)

To address Ladyman’s point, I think we need to review the dialectic up to now. The original challenge was that since there is no evolutionary explanation of (purported) philosophical knowledge, there is reason to think there is no philosophical knowledge. The point then made was that there is equally no evolutionary explanation of (purported) scientific knowledge. In the above passage, Ladyman apparently concedes this point. He then responds that ‘we have no good reasons for thinking that metaphysical knowledge is possible’. Perhaps that claim is correct, but Ladyman makes it on the basis of considerations that are independent of evolutionary theory. The original challenge has then gone by the board.

McGinn himself finds the suggestion that our philosophical ability is a by-product of evolution ‘much too sanguine’:

First, we should be a good deal more surprised by the by-product story than we tend to be, regarding it as far more puzzling than is customary. It really is quite astonishing, and not at all predictable, that a faculty with the biological function of reason should be capable of the feats of which it is capable. (McGinn 1993, p. 134)

But what is surprising and puzzling is not specifically the idea that our ability to do philosophy is an evolutionary by-product. What is surprising and puzzling is the by-product story itself, the view that our higher cognitive abilities and our culture are a spin-off from the cognitive powers that we evolved solely as a matter of natural selection. Either the by-product story is credible or it is not. If it is credible, and it can account for (say) our mathematical ability, then, on the face of it, it can equally account for our philosophical ability. If, however, the by-product story is not credible, then not only is there a question of why it is that human beings should have philosophical ability, despite the fact that it is not selected for, there is the broader question of why it is that human beings should have higher cognitive powers—powers to theorize about a wealth of topics not directly concerned with survival. This is not to say that the by-product story is unpromising—far from it. It is that there seems to be no special puzzle about how it is that human beings can do philosophy (as opposed to any other higher cognitive task) despite their evolutionary heritage.

McGinn’s second response is that

if we take the by-product idea seriously we should be prepared to encounter limitations that derive from the primary purpose of the organ [i.e. the so-called organ of reason]—as we would be for any other biological organ. The inner nature of reason, as determined by its basic function, must to some degree constrain the kinds of side-effects it can have. (McGinn 1993, p. 134)

Again I concede McGinn’s premiss: our reasoning powers have biologically imposed limits. But this is only half of the by-product story. The story says that we cannot tell from the primary function of the organ of reason—namely, aiding our species’ survival—what other cognitive work we can put that organ to. The above passage leaves this an entirely open matter. So the passage does not raise a specific challenge for our capacity to do philosophy as against our presumed capacities to engage in other demanding cognitive tasks.⁷

No species can be good at everything: each species has a relatively fixed biological nature which confers on it certain skills although that same nature precludes it from having certain other skills. The structure of the cat’s eye enables the cat to see well in the dark, but makes it see poorly things that are close up. The case for humans being cognitively closed with respect to certain matters is sometimes made by analogy: ‘What is closed to the mind of a rat may be open to the mind of a monkey, and what is open to us may be closed to the monkey’ (McGinn 1989, p. 350). Since human beings form a species alongside the rat and the monkey, our minds too should be expected to have cognitive limits. Now, no doubt they do, but where do these limits lie? Not only do the rat and the monkey have cognitive limits, but one of their limitations is that they remain unaware of the fact that they have such limitations. The rat and the monkey cannot learn a language, but they also cannot recognize that they cannot. If humans cannot solve philosophical problems, how is it that they can recognize this? But perhaps this is pressing the wrong analogy. Perhaps a better analogy is between children and adults. A nine-year-old child cannot understand things which the adult can, but the former can also recognize that fact. By symmetry, there might be things that the adult can’t understand but can recognize that he or she can’t understand. Perhaps so, but are the solutions to philosophical problems among them? Thomas Nagel remarks that ‘people with a permanent age of nine cannot come to understand Maxwell’s equations or the general theory of relativity or Gödel’s theorem’ (Nagel 1986, p. 95). True enough, but then such people don’t understand any of the theories in these fields that have received serious consideration, including false ones such as Cartesian mechanics, Newtonian mechanics and Hilbert’s programme. By contrast, as a species, human beings are adept at understanding and criticizing a wealth of philosophical theories.

McGinn says that ‘cognitive deficits are apt to be the inevitable outcome of cognitive strengths along other dimensions: we are bad at philosophy because we are good at something else—rather as we are bad at breathing under water because we are good at breathing in the open air’ (McGinn 1993, p. 24 n.8). We have seen what McGinn takes this something else to be—that elements in our thinking can be combined by laws to form more complex elements—but found it inadequate at explaining our alleged inability to do philosophy well. Moreover, his analogy is unconvincing. No human can breathe at all under water, yet some people can do philosophy better than others. Maybe every human falls far short of being able to solve any philosophical problem, but the thesis of cognitive closure altogether fails to explain the comparative fact that some people are better at doing philosophy than others.

The puzzle facing those who appeal to cognitive closure is then to explain why we are unable to solve philosophical problems despite the fact that as a species we are inveterate philosophers. Of course, that might just be an unfortunate fact about where the gap between our aspirations and our cognitive abilities lies. Yet the appeal to cognitive closure was supposed to be more than a provocative speculation, but an explanatory empirical hypothesis (McGinn 1993, p. 151), and unless it can explain the above consequence—a purported fact about just where our minds become closed—it fails its own standard.⁸

Perhaps, though, the cognitive closure thesis can be supplemented to explain different levels of philosophical ability. Perhaps, for example, some humans have abnormal abilities relative to their species. And perhaps someone’s philosophical impairment is more severe if they lack compensatory abilities (such as high-powered memory, computational ability, and the like). This line of thought, however, departs markedly from the analogous cases used to support the cognitive closure thesis: human beings do not have different levels of ability at breathing underwater, nor rats in learning language. Differences of degree can matter, of course, but once the issue is opened out in this way it becomes less clear why the solution to any philosophical problem eludes every human’s ability. It seems as likely to have a ‘mixed case’ in which different humans can solve different philosophical problems. The truth of the matter is an empirical one, but, at the current stage of enquiry, the cognitive closure thesis seems no more compelling than its rivals.

Here is a further fact that needs explaining by the cognitive closure thesis. Take those many philosophical theories that it is granted we do understand. Then the question is: although we understand them, why is there persistent disagreement between us about their truth-values? According to the proponents of cognitive closure, none of these theories provides the solution to a philosophical problem since our minds are closed to such solutions, and so any theory that we devise will be inadequate to the task. Well, that tells us that none of those theories provides a solution, but it doesn’t tell us why we can’t agree about the theories. For that matter, why is there disagreement about the truth-value of the hypothesis that our minds are cognitively closed with respect to the solutions of philosophical problems? This is a hypothesis that we understand—otherwise we would not be debating it—yet opinions remain stubbornly divided about it. So even if the cognitive closure hypothesis explains why we have not solved any philosophical problems—something that I’ve anyhow questioned—it does not explain why there is persistent disagreement.

V

What’s Gone Wrong? What we have is a bona fide philosophical problem. It is something that can be stated in rough form easily enough. Yet it is not a straightforwardly empirical matter, and it repeatedly resists solution. With that on the table, let me offer a different diagnosis of why there is persistent philosophical disagreement. I think the fault doesn’t lie with the questions we ask—that they are somehow defective—or with our minds—that they are ill-suited or too limited—but with the methods and ambitions involved in our inquiry.

Let’s begin with our methods. The methods we use in philosophy are both too weak and too strong. They are too weak because, even where deductive arguments are used, issues arise about the clarity or the justification of the premisses used in the arguments. No deductive argument will settle any of these issues: it simply pushes the problem back by introducing new premisses subject to the same issues. Still, the same structure is found in other disciplines. For instance, in mathematics, chains of reasoning ultimately run back to axioms, and the question of their justification eventually arises (Maddy 2011). But this takes us to a special feature of philosophy, namely that the methods used in philosophy are also too strong: the same methods used to reach a conclusion from a premiss set can be turned back and applied to those premisses and to the inferential steps used in drawing the conclusion. Debate about the conclusion is then parlayed into debate about a premiss or an inferential step. To debate means to argue, and any argument provided will be open to the same scrutiny.

Does the answer lie in the devising of some ‘brave new method’ that will replace or supplement our existing methods? The possibility of methodological innovation should not be ruled out, but even so, we should not hold out for the prospect of it making a breakthrough in our philosophical fortunes. The considerations noted above that apply to our current methods would carry over. Issues about the new method’s justification, its reliability, and its standing relative to our other methods would need to be debated and resolved, and, on the face of it, there is no reason to think that these issues would be any more tractable in the case of any new method than they have been in the case of any of the methods that we have already got.

Some philosophers think that this emphasis on argument is misplaced, but I think it is paramount. David Chalmers reports that

Burton Dreben once memorably said to me … ‘Great philosophers don’t argue’. … A part of Dreben’s thought, as I understood it, was that since arguments are so easily rebutted, giving arguments is a sign of weakness. It’s better to simply assert and develop a thesis. Then one’s readers have to engage with the thesis itself, without the cheap distraction of rebutting arguments for the thesis. (Chalmers 2015, p. 21)

As a piece of advice that rather sounds like saying, ‘Since our attention is so easily distracted, paying attention to traffic is a sign of weakness. It’s better simply to step out into traffic and cross the road. Then one can engage with the business of getting across the road, without the cheap distraction of looking at which way the cars are coming from.’ Instead, I say: since theses are so easily rejected, it’s better to present arguments for your theses.

Galen Strawson shares something of Dreben’s attitude:

It’s often said that argument is the heart of philosophy, and especially of analytic philosophy, but I’m sure that’s not true, if argument is thought of as primarily a matter of formally arrayed premisses and conclusions … All arguments have premisses, after all, and not all premisses can be argued for on pain of never getting started. (Strawson 2008, p. 3)

Strawson’s argument fails. For consider: definition is the heart of dictionary compiling. Nonetheless, all definitions involve words, and not all words can be defined on pain of never getting started.

Still, even if argument is paramount in philosophy, why doesn’t the arguing ever come to an end? In philosophy, our claims outrun our evidence in two respects. First, even where we agree about the evidence, it is not apparent which claim the evidence provides the most support for. Where the evidence is rich in philosophy, it tends to be disparate and conflicting, and thereby hard to assess. And where the evidence is meagre, it provides little support for one philosophical claim over another. Second, in philosophy we often do not agree about the evidence. New evidence is always coming in just because new arguments are always being thought up. In debating about the new arguments we are debating whether they do provide evidence. And in the case of data besides argument, there is disagreement about whether such data as intuitions or phenomenology or parsimony principles are evidence or whether they are fundamental or whether they provide much support (see van Inwagen 2004, p. 335 n.4). Like a fractal, with every inferential step and with every appeal to evidence in philosophy, debate and argument can arise.⁹

Even so, why doesn’t this situation occur in other disciplines, since they too use inference and evidence? Here we turn to the ambitions of philosophical inquiry. For a large part of the answer to the question lies in philosophy’s ambitions: it seeks to identify the most fundamental level of epistemic justification for claims, and it aspires to an especially high degree of clarity and understanding.

Disciplines that lack philosophy’s ambitions make life easier for themselves, and consequently can be more successful, where success is measured by securing intradisciplinary consensus or satisfaction of the discipline’s internal standards. These disciplines can more readily achieve consensus about the evidential status of certain classes of claims, their relative weighting vis-à-vis each other, and the specification of their content. Moreover, they select from only a small menu of theoretical options and disregard the wider issue of the underdetermination of theory by data. Of course, these researchers may have criteria for selecting certain theories from an infinite set of options that are consistent with the data, but then these researchers simply acquiesce in the appropriateness of these criteria. Much less of this framework of consensus obtains in philosophy. Where proponents of one particular view agree on a body of data and theory, we will find proponents of a rival view that rejects at least some of that data or theory. Since so much is up for debate in philosophy, even within a given field, we find that philosophy has made a rod for its own back: its aspirations outrun what its methods can deliver.

Lastly, it might be wondered whether the conception of philosophy at work here harks back to a tradition that is at odds with contemporary naturalism. That depends, however, on what one takes ‘naturalism’ to mean, and its sense shifts between different philosophers’ usage. To fix ideas, the following passage from Quine picks out one important strand of naturalism:

[The naturalistic philosopher] begins his reasoning within the inherited world theory as a going concern. He tentatively believes all of it, but believes also that some unidentified portions are wrong. He tries to improve, clarify, and understand the system from within. He is the busy sailor adrift on Neurath’s boat. (Quine 1975, p. 72)

But if this is what naturalism amounts to—believing the total theory of the world that previous inquiry, both scientific and philosophical, has bequeathed to us, and updating it as new information comes in—then there seems to be no inconsistency between naturalism and the conception of philosophy outlined here. That conception places no restrictions on what philosophical claims can be revised or on what basis. In fact, given that it takes philosophy to be ‘argument without end’, that is precisely one of the consequences of the conception.¹⁰

1

Even if the alleged facts obtain, there may still be an upside for philosophy: see Decker (2015).

2

Van Inwagen goes on to claim that the situation in philosophy more widely is no different from the situation in metaphysics.

3

This might seem self-undermining: if the explanation is both successful and philosophical, then we would have philosophical knowledge of why we lack philosophical knowledge. At least two options are available. One is to provide a more refined statement of just which kinds of philosophical fact we lack knowledge of, while allowing that we have knowledge of some other kinds of philosophical fact (see van Inwagen 2006, p. 39). Another option is to say that an explanation of why there is no knowledge of philosophical fact that appeals to certain other features of philosophy need not be a specifically philosophical explanation. It might perhaps be an explanation taken from the sciences. The explanation offered in §iv of why there is no progress in philosophy purports to be a speculative empirical explanatory hypothesis.

4

Some natural responses to van Inwagen and Lycan would fail to engage with their view. For instance, Prawitz (1997) takes the great leap forward from Aristotelian logic that Boole and Frege made to be an exemplary case of philosophical progress. Van Inwagen and Lycan, however, explicitly distinguish logic from philosophy, and emphasize that their sceptical thesis concerns philosophy alone.

In discussing van Inwagen (2006), Fischer and Tognazinni suggest that there are ways in which philosophical arguments can be successful other than by providing knowledge or established information:

For example, perhaps some philosophical arguments are ‘successful’ in virtue of making us (or many of us—or some of us) see a certain debate in a different way—as structured in a different way, or as requiring different presuppositions from what we had antecedently believed. Or maybe an argument can be successful by helping us to tease out certain distinctions, or clarify certain concepts, or encouraging us to think in new directions about an old problem. (Fischer and Tognazzini 2007, §ii)

It is not clear, however, why any of these features should count as a success unless it contributes to the provision of knowledge or the establishing of some information—and then we are back to facing van Inwagen and Lycan’s challenge. Otherwise, adopting a new perspective or acquiring a new concept or drawing a new distinction constitutes only a change in someone’s psychology, namely, by introducing added complexity to the stock of concepts or distinctions or views that they recognize.

Lastly, Wolterstorff claims that the key notion of epistemic appraisal in philosophical disagreements is that of entitlement, where ‘what makes a person not entitled to some belief is that there is some practice of inquiry that he failed to employ but ought to have employed with a seriousness and competence such that, had he done so, he would not have that belief’ (Wolterstorff 2014, p. 329). Now, a leading thesis of his paper is that ‘the sources of philosophical conviction and disagreement are in good measure hidden from us’ (Wolterstorff 2014, p. 323). It follows from this thesis that it is in good measure hidden from us whether, in forming her philosophical beliefs, each philosopher has failed to employ some practice of inquiry that she ought to have employed such that, had she done so, she would not have that belief. It further follows that it is in good measure hidden from us whether any philosopher is entitled to her philosophical beliefs, in Wolterstorff’s sense of entitlement. I do not know whether Wolterstorff would accept these consequences of his paper, consequences that seem germane to Lycan and van Inwagen’s scepticism.

5

For similar views, see James (1911, pp. 9–10), Lycan (2013, p. 117) and Chalmers (2015, p. 25).

6

Chomsky pioneered the idea that human beings are cognitively closed to the solution of certain problems (or ‘mysteries’, as he calls them); see Chomsky (1976, ch. 4). He is, however, cautious about saying which problems we are cognitively closed with respect to. For this reason I confine myself to discussing those authors who claim that we are cognitively closed with respect to the solution of philosophical problems.

7

The most developed case McGinn makes that we are cognitively closed with respect to some subject matter concerns consciousness. Brueckner and Beroukhim (2003) provide effective criticism of McGinn’s case.

8

McGinn further says that sometimes philosophical theories can be assented to given the reasons which support them, but that ‘often they can expect only to be taken seriously … [Often] enough the best we can do is to get ourselves into a position to regard the proposition with respect’ (1993, p. 1). In particular, he claims that the cognitive closure theory ‘may be true, and that much would make sense if it were’ (1993, p. 2). McGinn’s epistemic caution does not substantially affect the above assessment of his theory. If a theory fails to explain facts that we would reasonably expect it to explain, we should take the theory to be less likely, and—in that sense—we should take the theory less seriously.

9

This is something that MacBride’s account rightly captures.

10

Earlier versions of this paper were presented at the University of Manchester and at a meeting of the Aristotelian Society. I am very grateful to the audiences. I am especially grateful to Jason Decker, Dorothy Edgington and Guy Longworth for detailed and very helpful written comments.