Has philosophy made progress?

A few of my philosopher friends on social media have been sharing this interview with Tim Maudlin (NYU), where Maudlin contends that philosophy has made plenty of progress. This issue comes up time and again, and Daniel Stoljar wrote a recent book defending the existence of philosophical progress. On the other hand, there are a number of works arguing that philosophy has made no real progress and has no good claim to truth-aptness. See for example:

• Jason Brennan (2010). Scepticism about philosophy. Ratio.
• Eric Dietrich (2011). There is no progress in philosophy. Essays in Philosophy.
• Peter Slezak (2018). Is there progress in philosophy? The case for taking history seriously. Philosophy.

The last of these works, for what it's worth, directly addresses Stoljar's work. I'm curious what readers think, especially given broad and persistent divisions over what the most defensible answers are to most (if not all) basic philosophical questions.

My own view is that while I think some real progress has been made, it's probably a lot less than defenders of philosophical progress seem to think. Moreover, how one views the matter depends, presumably, on one's own substantive views. For example, Maudlin sides with the the 59% of philosophers or so today who report leaning toward or accept compatibilism about free will--even going so far as to say, "I think Locke and Hume nailed free will, and since then there has been no interesting debate about it." Notice, first, that this only indicates that majority of philosophers lean toward compatibilism (hardly a ringing endorsement), leaving a full 41% of philosophers who think it's not the right answer. Second, surveys of ordinary laypeople show a similar split. Consequently, there are at least two possibilities here: (1) philosophy has made real progress on free will, or (2) about 60% of people are psychologically inclined to compatibilism, and that's just a psychosocial fact about human beings, not a form of real philosophical progress. How are we to discriminate between these two hypotheses, especially when some of us suspect that compatibilism may be the worst of all answers to the free will problem?

Similarly, consider moral philosophy. Does it make progress? Well, here are some relevant philpapers results:

Meta-ethics: moral realism or moral anti-realism?

Accept or lean toward: moral realism	525 / 931 (56.4%)
Accept or lean toward: moral anti-realism	258 / 931 (27.7%)
Other	148 / 931 (15.9%)

Normative ethics: deontology, consequentialism, or virtue ethics?

Other	301 / 931 (32.3%)
Accept or lean toward: deontology	241 / 931 (25.9%)
Accept or lean toward: consequentialism	220 / 931 (23.6%)
Accept or lean toward: virtue ethics	169 / 931 (18.2%)

Metaphilosophy: naturalism or non-naturalism?

Accept or lean toward: naturalism	464 / 931 (49.8%)
Accept or lean toward: non-naturalism	241 / 931 (25.9%)
Other	226 / 931 (24.3%)

By my lights, its hard to even know what to make of these results. First, in the best cases (questions 1 and 3) the leading positions in these polls only indicate that approximately 50% of philosophers lean toward the view. Second, the views all seem in tension with each other. Moral realism is generally understood to to be the position that there are objective moral facts that don't depend upon us for their truth--thus constituting a form of non-naturalism. Yet, the answer to (3) indicates that nearly 50% of philosophers lean toward naturalism, and the answer to (2) shows that philosophers are widely split on what the moral facts are (deontology, virtue ethics, and consequentialism being vastly different normative positions). 

Let me be clear, I don't mean to say that there's no moral progress. I wouldn't do philosophy myself, writing books and articles defending new theories, if I didn't think there to be some truth the arguments. What I am inclined to think, however, is that confident pronouncements about philosophical progress go too far. We should admit, as Russell does in the passage below, that philosophy has serious epistemic limitations:

Philosophy, as I shall understand the word, is something intermediate between theology and science. Like theology, it consists of speculations on matters as to which definite knowledge has, so far, been unascertainable; but like science, it appeals to human reason rather than to authority…All definite knowledge—so I should contend—belongs to science; all dogma as to what surpasses definite knowledge belongs to theology. But between theology and science there is a No Man’s Land, exposed to attack on all sides; this No Man’s Land is philosophy. (Russell 1945: xiii)

We should then aim to improve upon those limitations by doing better metaphilosophy and putting that metaphilosophy into practice, including testing empirically whether the kinds of premises philosophers appeal to really are ones that people in general take to be true.

Or so I'm inclined to think. What do you all think?

Comments

[Nicolas Delon]

Are progress and convergence taken to be synonymous here? It could be that some fields have made progress toward the truth, or precision, or appropriate complexity of analysis, or whatever, without a corresponding trend toward convergence of views.

Btw, I strongly recommend this long interview with David Chalmers where they discuss, among other things, the Philpapers surveys and the notion of progress and convergence in philosophy:
https://80000hours.org/podcast/episodes/david-chalmers-nature-ethics-consciousness/

[Marcus Arvan]

Hey Nicolas: Thanks for weighing in. I agree that these are probably good discussions to have (viz. 'what is progress anyway?'). And I'll definitely read the Chalmers piece.

But here's my quick reaction. I've often heard some people in the profession say they don't think the point of philosophy is getting at the truth, but something like "understanding"--and insofar as that's the case, there's nothing really the matter with ongoing dissensus, since philosophy is always generating new theories and arguments that increase our understanding.

I've never found this answer very compelling, though, and for a couple of reasons. First, go back to ancient Greece--specifically, to the presocratics. Many of them (Anaximander, Anaximenes, Heraclitus, Thales) came up with fairly complex physical cosmologies--but as we know in retrospect, most of that was nonsense (in the colloquial sense of having little to with what science has since discovered). Or here are even better cases: astrology and alchemy. For generations, large numbers of investigators came up with really intricate theories for "understanding" astrological phenomena and how to blend physical elements. But, of course, as intricate and complex as those theories may have been, they didn't really amount to any real understanding, only the appearance thereof.

Anyway, here's one reason to think that convergence is important. Without it--or, to be more precise, without *methods* (such as those of science) that base philosophical inquiry on some firm evidential foundation that people can converge upon--we have little or no reason to think that whatever "understanding" we're developing is actual understanding bearing any relation whatsoever to the truth, as opposed to something akin to alchemy (viz. wishful thinking). This is, very broadly speaking, what my undergrad advisor, Dan Dennett, worried about many years ago in a paper many years ago entitled "Higher order truths about chmess": https://ase.tufts.edu/cogstud/dennett/papers/chmess.pdf

Now, I personally don't think philosophy is as bad off as all that. But (or so I've argued elsewhere), I don't think philosophy is nearly as *well* off as those who confidently pronounce progress seem to take it to be.

This brings me to my second concern, which really piggybacks on Dietrich's and Slezak's papers, which is that when I go back and look at historical sources, it often appears to me that "new arguments" and "progress" are simply old arguments dressed up in new clothing. I mean, take moral realism for example. TM Scanlon made 'reasons primitivism' seem like a new innovation in his 1999 book. But wait a minute: go look at Thomas Reid's moral philosophy from nearly 300 years ago. It's basically moral realism and normative reasons primitivism: https://plato.stanford.edu/entries/reid/#Mor (for my part, I think the view is as wrong now as it was then!, see https://www.princeton.edu/~msmith/mypapers/ParfitsMistakenMetaethics.pdf ).

Or take constitutivism in ethics. Kant first defended the view in 1785. Now, 235 years later, people are still defending constitutivism and it still doesn't seem to work:

https://philpapers.org/rec/BUKKAF
https://philpapers.org/rec/BUKACO
https://philpapers.org/rec/BUKSAI

Or take reflective equilibrium. It's not only commonly appealed to in moral philosophy but also as something like a general philosophical methodology (in metaphysics, etc.). Yet there are powerful arguments that it runs smack dab into the same fatal problems as epistemic coherentism:

https://philpapers.org/rec/DEMREA-4
https://philpapers.org/rec/WOOARE
https://philpapers.org/rec/MCPTMI-2
https://philpapers.org/rec/KELIRE
https://philpapers.org/rec/STRTAP-4

So, is progress being made, or are we mostly just continuing to run up our heads against the same philosophical walls over and over again?

I myself have hope for philosophy and for philosophical progress--but, for my part, I'm increasingly convinced (both for metaphilosophical and historical reasons) that philosophical progress primarily comes through something that many philosophers seem to despise and even call "not philosophy": namely, the *naturalization* of philosophical problems (viz. thoroughly blending philosophy with empirical science). I know I'm in the distinct minority here (thought there are a good number of naturalists out there), but to be forthright this isn't a position I came to easily. Rather, for most of my career, I *wasn't* a dyed-in-the-wool naturalist. I just became more and more concerned over time about philosophical methodology, and methodological naturalism increasingly seemed to me to be the only adequate answer.

[Justin Weinberg]

Yes, philosophy makes progress. However, that progress is not mainly in convergence on answers to philosophical questions, but in the creation of the questions themselves.

If you take a central aim of philosophy to be the identification of what we don't know, and you take good philosophical questions to be a way of specifying what we don't know, the development of new and better philosophical questions is philosophical progress. (I discuss this here: http://dailynous.com/2017/08/23/intellectual-achievement-creating-questions/ )

But what about all of the theories that are attempts to answer philosophical questions, one might ask? Are they useless? No. Perhaps such theories are intended as answers to philosophical questions, but they tend to function more as creators of philosophical questions--at least good theories do. Generally, attempts to answer philosophical questions are a key way of testing their quality as well as discovering new ones.

I understand this is not a popular view.

[Marcus Arvan]

Hey Justin: To the extent that I'm willing to move beyond convergence as a criterion of progress, yours is the account I think I'm most sympathetic to.

For example, I think that regardless of what one thinks about Kantian ethics, Kant raised really interesting new questions (about agency, free will, etc.). This is a large part of what I love about philosophy--namely, (1) considering the questions that former generations of philosophers raised, and (2) trying to come up with some new answers of my own! Indeed, something like this, I suspect, is why many of us gravitated to philosophy in the first place: we simply found the questions fascinating. I also agree with you that philosophy done well can help you realize what you don't know.

At the same time, I still have deep concerns about it as a conception of progress--or indeed, as reflective of what philosophers standardly take themselves to be doing or what they mean when they claim to make progress.

For example, most of the moral philosophers I know think there is some truth to their views--despite the fact that there are other philosophers in other traditions who think their view is precisely wrong. For example, as interesting as Kant's ethics is (I myself was a Kantian until about 5 years ago!), I myself think Kant led moral philosophy in exactly the wrong direction: toward thinking that moral reasons are special (i.e. categorical) and distinct from instrumental/prudential reasons in some fundamental way. Because I don't think this view is true (see https://philpapers.org/rec/FORATD-2 ), my own view is that a lot of philosophers think they know things about moral reasons that they have no good reason to believe. But, of course, I'm sure they'd say the same about me!

So this is my deeper concern: most of us doing philosophy, I take it, don't just think we are raising good questions or learning 'what we don't know'. We give what we take to be good answers--thereby committing ourselves (at least implicitly) to the proposition that philosophy *should* give good answers: namely, true ones. In which case I think we're right back at the convergence problem. Finally, I also don't think the "we ask good questions" and "learn what we don't know" claims can be substantiated *unless* we presuppose that philosophy is in some way getting at the truth. Why? Well look: alchemists thought they were raising good questions and constantly learning what they didn't know (how to get gold out of combining coal and bronze, for example). But they *weren't* raising good questions and were constantly taking themselves to know things that they didn't know!

[Mike]

You say:

"Moral realism is generally understood to to be the position that there are objective moral facts that don't depend upon us for their truth--thus constituting a form of non-naturalism. Yet, the answer to (3) indicates that nearly 50% of philosophers lean toward naturalism, and the answer to (2) shows that philosophers are widely split on what the moral facts are (deontology, virtue ethics, and consequentialism being vastly different normative positions)."

I don't see any tensions here.

First, there are various versions of moral realism that take there to be objective moral facts that are (or are reducible to) natural facts: https://plato.stanford.edu/entries/naturalism-moral/

Second, I don't see why disagreement about what moral facts there are would be in tension with holding that there are such facts. There's significant disagreement about all sorts of issues in the sciences, history, religion, etc., but I don't feel any pull to being an anti-realist about those subject matters. Maybe there's some special reason to treat moral disagreement differently, but on its own, recognizing disagreement about what the right moral theory is doesn't seem difficult to square with moral realism (or naturalism, for that matter).

[Marcus Arvan]

Hey Mike: good point. I’m actually a naturalist moral realist myself! I think that’s an unfortunate ambiguity in the Philpapers survey, and I guess my thought was that when the typical philosopher thinks of "moral realism", they're probably thinking of Scanlon, Parfit, Enoch, Shafer-Landau type stuff.

In any case, I guess my concern then is that, from what I can tell, naturalist moral realists seem to be in the pretty clear minority. The vast majority of moral philosophers I’ve interacted with in recent years seem to think that naturalist moral realism violates the is-ought distinction and commits the naturalistic fallacy. If that’s right, then I think there’s still a tension: namely, that most moral philosophers seem to be non-naturalists, whereas most philosophers lean toward or accept naturalism. That seems to me to (at least anecdotally) to be a real tension in the discipline—though I agree that the survey results alone don’t entail it. I wonder if others have the same sense.

In terms of the other tension, I still think there is one. Sure, there is a lot of disagreement in other areas of inquiry, including science and history. But in science and history there are clear independent methods to serve as a broadly objective criterion for resolving disagreements--you know, for actually getting at the objective facts claimed to exist. For example, when relativity and Newtonian mechanics differed over Mercury’s perihelion, observations predicted by Einstein’s theory settled the matter. In contrast, in many areas of philosophy—including moral philosophy—people’s intuitions tend to serve as their own evidence. I think this is deeply problematic for reasons I outlined here (https://philosopherscocoon.typepad.com/blog/2019/06/against-commonsense.html) and discuss at length in the first chapter of my first book (https://www.marcusarvan.net/my-book ). When you have rampant disagreement about a set of phenomena and no independent criterion for resolving those disagreements, and those basic disagreements persist for thousands of years, you have a problem—one quite unlike science and history (for the most part): you have fundamental disagreement over what is true and no agreed upon way of determining what is actually true from what’s not.

Maybe there isn't a logical tension between believing there are objective facts and rampant disagreement--but still, I think there is an *epistemic* tension: namely, if you don't have any good method for distinguishing truth from fiction (to resolve the disagreement), what's your justification for believing there are facts to be found (as opposed to multiple incompatible fictions)?

[Travis Timmerman]

Thanks for this post Marcus. I am inclined to think there is real substantive progress in philosophy and that we can come to know the truth about various philosophical questions by engaging in philosophical discourse, reading and writing philosophy, and so on.

Here are a couple of quick condensed thoughts about the issues you raise here, which deserve much more attention.

First, I don’t think philosophical progress requires near universal agreement among experts. We can have sufficient justification (e.g. from understanding some sound or cogent arguments) to know contentious propositions, even propositions that are contentious among experts. Note that you cannot coherently believe this claim is false since doing so assumes a certain contentious position in the peer disagreement literature, one over which there appears to be peer disagreement. (See, some philosophical progress has already been made!)

Second, while I am a big fan of the PhilPapers surveys, I don’t think they provide good evidence that there is a lack of consensus about philosophical questions for a number of reasons.

(a) The survey questions are, I presume, selected in part because they’re issues over which there is thought to be substantial disagreement. The surveys actually often suggest there is more agreement on these broad issues that I would have otherwise assumed.

(b) The surveys go out to all philosophers and so are not assessing what the experts about each question think. Insofar as we should care about philosophers converging on some position, it should be among the experts intimately familiar with the literature, and not people with a surficial knowledge of the problems in question.

For instance, I am inclined to think some form of non-classical logic is true, but I don’t even come close to knowing enough about the literature to have a considered opinion on the matter, and so my answer on that survey should indicate nothing about whether there is a consensus among the relevant experts.

(c) The questions on the surveys are far from exhaustive (as they should be).

(d) The questions that are covered on the survey tend to be very broad (e.g. empiricism or rationalism). The truth to these broad questions hinges on countless more precise philosophical questions that are being worked out in the literature.

Were we to survey philosophers who are experts in the various literatures about the more narrow philosophical questions, I suspect there would be something much closer to a consensus about a number of them.

(e) I think there is often much more agreement among adherents to the various, seemingly contradictory, positions listed than is assumed. I even think some of these seemingly contradictory positions are compatible. For example, I think virtue ethics is consistent with various forms of consequentialism and deontology. (Don't believe me? Keep an eye out for Yishai Cohen's and my forthcoming paper "The Limits of Virtue Ethics" in Oxford Studies in Normative Ethics. Or email me for the penultimate draft.)

Third, you write that “there are at least two possibilities here: (1) philosophy has made real progress on free will, or (2) about 60% of people are psychologically inclined to compatibilism, and that's just a psychosocial fact about human beings, not a form of real philosophical progress.”

These possibilities don’t seem to be mutually exclusive or exhaustive. It might be true that 60% of philosophers are psychologically inclined to accept compatibilism, and their inclinations towards compatibilism might be unjustified insofar as they don’t know the relevant literature. At the same time, there could be (and I think has been) real substantive progress made on free will. This can be seen by reading the work of numerous people working in the literature right now.

I know I am moved by arguments for compatibilism, but certainly don’t know enough of the relevant literature to be very confident in my judgments. (Though, of course, that didn’t stop me from answering that question on the PhilPapers survey, nor should it have).

Fourth, I don’t see any tension in the surveys about naturalism and moral realism for the reason Mike already listed. In response to your response to his comment, I am not sure if natural moral realists are in the minority. But even if they are, I see no tension in the data since anti-realists and people who selected “other” could be naturalists too.

Fifth, here is a point I made on Facebook and which is a variant of a point found in Nathan Ballantyne’s absolutely fantastic paper “Knockdown Arguments.” (https://www.academia.edu/4279942/Knockdown_Arguments)

Progress in any non-philosophical field assumes certain contentious philosophical positions. So, if there is no progress in philosophy in the sense that answers to philosophical questions cannot be known, then there is no progress in any field in the sense that answers to any non-philosophical questions cannot be known either. If progress in other fields is understood as taking the form of showing that P is true given certain contentious assumptions, then there is that very same kind of progress in philosophy. I, of course, think that there is progress in philosophy in the sense that answers to contentious philosophical positions can come to be known from doing philosophy or reading and understanding sound or cogent philosophical arguments. But the point here is that such skepticism about philosophical progress will generalize to skepticism about progress in any field.

Sixth, related to that last note, we should be careful to make sure that the standards for “progress” that we apply to philosophy are also ones that we would apply to fields outside of philosophy. Anecdotally, I sometimes see philosophy-progress skeptics apply a notion of progress to philosophy that entails no progress is ever made in any field. The most interesting question, I think, is whether there is progress in philosophy according to the standard(s) by which there is progress in other fields.

Is there? I think the answer to that question is clearly “Yes.” I take solace in the fact that anyone who disagrees with me also has to believe that they can make no progress in showing I am wrong.

[Mike B]

Here is something I take to be a nice example of philosophical progress from my own field. Famously, in the first half of the 20th century most philosophers were sense-datum theorists. Motivated by the many mismatches between sensory appearance and reality, they thought that the direct objects of perceptual experience must be something distinct from, but which represented, the actual distal sensory stimuli. Thus, on their view, what we experience is a representation.

Now, the progress came when people realized that this approach was conceptually confused. Namely, we need to distinguish between representational vehicles and representational content. What the arguments for sense-data really showed was only that our perceptual experiences were representational vehicles; they did not show that *what* we experienced was a representation. In fact, the only coherent way to unpack the details is to say that what we experience is what those vehicles represent, i.e. their content. Since they represent distal sensory stimuli, we experience distal sensory stimuli.

The point is that philosophers in the first half of the 20th century were confused about an important concept, representation, and applied that concept to the evidence in the wrong way, yielding a silly theory of perceptual experience. Later the conceptual confusion was recognized, and theories reshuffled into modern forms of representationalism.

Did this shift get us closer to the truth? I'm not sure, because there are other reasons to be skeptical of modern representationalism (e.g., see all the people defending naive realist accounts). So it's not clear that this shift was a shift to the right theory. But it was a shift away from an incorrect theory. So it seems clear to me that it's a nice example of progress, in the sense that we learned about what was wrong. At the very, very least, we learned that a certain form of argument for a certain position was wrong. It's also a form of conceptual progress, where we enrich our concepts into more robust form, allowing more subtle and reality-conforming applications of those concepts.

I guess it's also worth noting that this philosophical progress hasn't received much uptake outside philosophers of perception, and even within philosophers of perception it's limited to certain traditions or working groups of researchers (albeit very central and influential ones in the anglophone world). For example, most contemporary psychology and neuroscience textbooks will, at some point, reproduce the fallacious reasoning of the sense-datum theorists and assert some crude form of indirect realism/sense-datum theory as the obvious truth about perceptual experience. My impression from reading a lot of research papers and pop articles is that most working perception scientists still hold the older view as well. But I don't think this state of affairs speaks against either the veracity of contemporary representationalism or the progress that's been made. It's not as if all these sources have carefully considered the arguments and still stuck with the old theory; they're simply unaware of the conceptual moves. Perhaps someone needs to write a short book targeted at perception scientists in psychology and neuroscience explaining the issue.

[Brad Cokelet]

Hi Marcus,

I think there is progress in philosophy but that it is more fine grained. My view overlaps with Travis' and Justin's to some extent, but I want to look at a couple of examples.

Before Marquis wrote his paper on abortion philosophers thinking about the morality of abortion were not thinking about what makes it wrong to kill adult human beings. Marquis saw that we can think about that topic bracketed from questions about status and that it does seem to bear on the morality of abortion. This has spawned lots of interesting work on what makes killing wrong.

Another example would be the reasons literature as it moves from Williams/Korsgaard/McDowell, to Dancy's second book, to Schroeder's slaves of the passions. All along this path people are interested in the relation between reasons and desires but over time we find more carefully distinguished questions which people then discuss in detail. Williams was interested in desires as a logical condition on reasons, then we get Dancy raising questions about the ontology of reasons, and then we get Schroeder helping us to notice how the pragmatic meaning of "reason" can mislead our thinking about cases. Obviously these are just some of the discoveries and advances in the reasons-literature swamp.

In these examples we see progress of various kinds, and it goes beyond discovering new interesting questions:

Progress in distinguishing and identifying questions
Progress in identifying and evaluating competing answers to the questions
Progress in identifying *justification relations* between topics or questions
Progress in identifying improved *methods* for answering questions

I have not followed the phil science debate but when I was in grad school I took a class on it and found the pessimistic meta-induction argument pretty interesting. It drove home how much change there has been in science over time. One line of response was to look for structural features that have been carried over across time as science makes progress. I was not convinced that John Worrell's version of this worked to support realism, but the point here is that I think we can make an analogy with science and philosophy.

Philosophy makes progress in converging on the conceptual and justificatory structure of the intellectual terrain and on improved methods for making judgments and theories about it. It is a matter of coming to agree that it is important to distinguish various questions, that they bear on each other in various ways, etc. For example even those who reject Marquis' argument now have chapters in their books on abortion on the wrongness of killing and its relation to the morality of abortion. The idea that you can have a well justified view on that issue without thinking about the wrongness of killing is recognized by all as false.

[Brad Cokelet]

Hi Marcus,

I have always meant to study the book Gutting wrote on progress, but this article makes it look like a worked out view of the sort I favor. If you know of good responses to his work I would love to know about them!

https://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199668779.001.0001/oxfordhb-9780199668779-e-2

[Marcus Arvan]

Hey Travis: Thanks for weighing in, and for suggesting Nathan's paper! I think I saw a version of it at the APA several years ago. I remember thinking it was a good paper then, and I agree that it's an excellent paper now.

That being said, when Nathan deals with differences between science and philosophy, I worry that Nathan never really gets to the heart of the fact that there are obvious differences between science and philosophy, and why those differences matter. On the one hand, he says, "While there *undeniably* is a difference between philosophy and other fields, it is hard to say exactly what it amounts to." Then he adds, "There are certainly ways in which philosophy differs from the sciences. It is sometimes suggested that the subject matter of philosophy has to do in part with very general and abstract concepts like knowledge,causation, luck, and justice.the methodology of philosophy depends critically on ‘‘rational insight’’ or ‘‘intuition’’ in exploring its subject matter. Or it may be that the critical distinction between science and philosophy concerns agreement about what constitutes a proper test or experiment of a thesis: in science there is (arguably/some) agreement, in philosophy there is not.
agreement about what constitutes a proper test or experiment of a thesis: in science there is (arguably/some) agreement, in philosophy there is not."

This is what my comments (and published work) have emphasized. So, let us focus on it a bit more. Here are a few striking facts about science:

≈100% of experts in chemistry agree that hydrogen is a fundamental element.

≈100% of experts in biology believe that plants and animals are made out of cells.

≈100% of experts in particle physics believe that quantum mechanics is our best theory of fundamental particle interactions.

>99% of experts in physical cosmology believe that relativity is our best physical theory of gravitation (there are a few outliers, such as proponents of Modified Newtonian Dynamics, but each known alternative is thought to have problems that relativity does not).

Okay, so there are some pretty stunning convergences in science about (A) basic facts, and (B) true theories. Now, on top of that stunning convergence, is there disagreement? Yes, absolutely! In particle physics, there are string theorists, brane theorists, and so on and so forth. There's half a dozen or so interpretations of quantum mechanics, and so on. But--and this is crucial--these are recognized to be theories on the 'speculative edge' of the discipline. These speculative theories on the boundary of those disciplines are still predicated on vast *agreement* about basic facts.

Now turn to philosophy. Simply put, none of this is true. If we look at moral philosophy for example, there is profoundly little agreement about what the basic facts are or which theories, if any, are actually correct. Now, you might say, there's pretty wide agreement that there are moral facts, such as the wrongness of murder. But the problem with this is twofold. First, there are plenty of human beings that don't even seem to recognize that much: warlords, mafia kingpins, violent criminals, and so on. When we look at Plato's dialogues, character after character Socrates interacts with (Thrasymachus, Glaucon, Polus, etc.) simply appears to work with an instrumental conception of normativity--even going so far as to argue that if you can murder and get away with, you should go for it! Or, to take a contemporary case, my wife taught in a business school last year. She reported student after student saying things like, "If you can get away with it, what's wrong with it?" She had no good answers--and, for my part, I still don't think moral philosophy has given a good answer (but rather tended to simply evade the question by trying to distinguish 'moral reasons' from prudential ones). Second, to the extent that anyone already thinks murder is wrong, we don't need philosophy to tell them so. We need philosophy to give *theories* of why it's wrong.

The point I want to make then is this. In 'Epistemology Naturalized', Quine writes the following: "Turning back then to our task of defining observation sentences, we get this: an observation sentence is one which all speakers of the language give the same verdict when given the same concurrent stimulation...This formulation accords perfectly with the role of the observation sentence as the court of appeal of scientific theories."

The point then is this: in science, there are genuine observation sentences. Anyone with vision who grasps the meaning of the word 'cell'--that is, anyone who has taken high school biology--can look through a microscope and see a cell with their own two eyes. By a similar token, anyone who wants to test Einstein's theory of relativity can set up an experiment and see if the data match those predicted (hint: they have, in experiment after experiment). Now, at this point Nathan may say, "Aha, gotcha! Not everyone in the language will agree with that because there are philosophers like van Fraasen who argue that the things we see in microscopes aren't observables."

While this is very nice and clever, I think it is an evasion. Whatever we want to say about those deep philosophical matters (about which I think there can probably be no progress, since disagreements about them will come down to differences in intuitions), it is still the case that any biologist will say, "Look, call it what you want: an observable, observable, whatever. We call those things 'cells', and we know how they are structured and function." Now, of course, Nathan might want to go all the way down the rabbit hole and philosophically challenge those claims--but now I think we are just playing games. There is clearly a difference between philosophy and science, which is that setting all of these deep philosophical questions aside, *every* darn biologist agrees there are cells.

This is because in science, there are universally recognized true observation sentences. The problem in philosophy is that for the most part there just isn't. What one person takes to be a datum/premise for an argument, another philosopher will simply deny. So, for example, when a Kantian claims that X is a constitutive feature of agency, non-Kantians are like, "yeah, I don't buy that premise." And so on, and so forth. And so, unlike in science, where vast numbers of theorists converge on the same basic answers (viz. ≈100% of biologists believing in cells), here's what we get in moral philosophy:

Moral-realists
Moral-antirealists
Moral skeptics
Moral error-theorists
Cognitivists
Noncognitivists
Expressivists
Moral rationalists
Moral sentimentalists
Kantians
Aristotelians
Utilitarians
Pluralists
Care ethicists
Particularists
Contractarians
Contractualists
Etc.

And, when you actually look at these views, there is hardly agreement on *anything*. Sure, do some of the views agree murder is wrong? Yes, but all of the non-criminals/non-psychopaths already believed that--and the theories that do say murder is wrong say that it is wrong for mutually inconsistent reasons.

Now, let me be clear, this isn't to say that there is no hope for philosophy, or that some of the above theories cannot be reconciled. For my part, I've recently tried (like Yishai Cohen) to reconcile several leading moral frameworks. But a large part of my project here has been to argue that we should change how we do moral philosophy--specifically, by making it more scientific. In brief, my aim is to base moral theory on true observation sentences as Quine understands them: namely, propositions that virtually all competent language users can recognize to express facts--which I argue is best exemplified by empirical moral psychology and the virtually universal recognition of instrumental, means-end rationality (see https://www.marcusarvan.net/neurofunctional-prudence-and-morali ). Of course, my approach could be misguided. But, at least for now, I am holding out some hope that this approach may help put moral philosophy on firmer footing by actually requiring moral philosophy to answer to empirical science (and yes, I do defend this approach against the charge it commits the naturalistic fallacy and violates the is-ought gap).

In any case, to (finally) address some of your other concerns, I don't think this type of argument is self-defeating. It doesn't run into the peer-disagreement problem you mentioned because it aims to resolve peer-disagreement by reference to observation sentences: sentences that epistemic *peers* recognize to be true together. This, I think, is the genius of science. It gives us a common method for resolving peer-disagreement that participants in inquiry recognize to be truth-apt, rather than differing intuitions about premises (which I think is the problem with armchair philosophy). Consequently, by the same token, I think I absolutely am "[making] sure that the standards for “progress” that we apply to philosophy are also ones that we would apply to fields outside of philosophy." My standard for progress isn't mere facts of convergence. It is that science converges for particular reasons: namely, its insistence upon basing theories on observations that virtually *everyone* recognizes as basic facts. And my claim then is that we have very good reasons for taking this kind of convergence to be a kind of progress that philosophy, for the most part, cannot stake the same kind of claim to.

Finally, I'm not sure you should "take solace in the fact that anyone who disagrees with me also has to believe that they can make no progress in showing I am wrong." I never said philosophy makes no progress. I've only argued that it makes less progress than optimists tend to claim.

[Marcus Arvan]

Hey Brad: Thanks for weighing in as well. I think your comment has helped me to clarify what I really want to say.

Like I said in response to Justin's comment, to the extent that I think philosophy does make progress, his general conception of progress is the one that I think I'm inclined to find the most attractive--and I think your proposal is broadly in line with his. Let me explain why before laying out what my residual concern is.

To take your first example, Marquis as you note raised very good questions about what makes killing adult human beings (normally) wrong. That has indeed spawned a lot of interesting work--on competing answers, 'evaluating' them, and so on and so forth. The same is true of your second example, the reasons literature.

So, I'm happy to accept that philosophy progresses in the four ways you list. But what I want to say in response is two things:

(1) Proponents of philosophical progress typically go much further than that, claiming that philosophy progresses by giving truth-apt *answers* to philosophical problems.

(2) The four ways that you list philosophy as progressing are nevertheless epistemically deficient in a way that optimistic proponents of philosophical progress often do not seem to recognize.

Let me begin with (1). So, in the interview I linked to above, Tim Maudlin comes right and says that he thinks philosophy makes progress by giving answers that he, Tim Maudlin, knows to be correct. Indeed, he says explicitly that he thinks philosophy has achieved "convergence on the truth." He gives atheism and compatibilism as main examples here. This kind of answer bugs me for many reasons, not the least of which is that I think theism might be true and I think compatibilism is either false or true but trivial (I give a disjunctive answer here because I don't think philosophical methods suffice to clearly support one disjunct over the other). [Note: I choose to believe that theism is true as a matter of religious faith, but that is a different story]

That brings me to my deeper concern, which (following my previous comments in this thread) is that I simply don't think (most) philosophical methods have any good claim to be truth-apt. Consequently, when people like Maudlin say they think philosophy has progressed by leading to the truth in cases like compatibilism, I think the answer is doubly wrong: a main example he's giving as progress is either false or true but trivial, and the methods that he's implying led to the truth aren't truth-conducive.

Maudlin seems to me far from alone in making these kinds of claims. If I recall correctly (and please do correct me if I'm wrong), Scott Soames' history of analytic philosophy book makes a number of claims about certain philosophical discoveries being true (such as some Kripkean claims in philosophy of language modality). And my general sense is that philosophers in general think their arguments bear some relation the truth.

So, I guess that's my first concern. I'm apt to agree with you that philosophy can progress in the four ways you list, but that proponents of philosophical progress don't in general stop there: they think that when 'evaluating' rival theories or theses, philosophy makes progress by getting closer to the truth (note: your conception of progress is strictly ambiguous between 'evaluation' being truth-apt and not. Alchemists and astrologers certainly 'evaluated' each other's claims, as did ancient Greek cosmologists. But none of their evaluations bore the mark of truth-aptness).

This brings me to (2). I guess my real position, the more that I think about it, isn't the "philosophy makes little progress." It is that the *kind* of progress it makes is deeply epistemically deficient in a way that optimistic proponents of progress don't generally seem to want to recognize. Another way to put my concern is, "Sure, philosophy may make progress. But is it truth-apt progress or Dennettian chmess progress? If it's the latter, then sure philosophy may make progress--but it's not the kind of progress we should be very happy with."

That's my real concern--whether we should be happy with status quo methods and how we understand 'progress.' In my case, for example, early drafts of my first book--Rightness as Fairness--initially defended a constitutivist theory of categorical reasons. I thought it was interesting, and so did some reviewers. But one thing a reviewer said to me was (to paraphrase), "Sure, Kantians will buy this. But why should anyone else?" That comment hit me like a ton of bricks. Yes, Kantians have been giving constituvist theories for ages--but, given that there are others out there that just don't find the premises behind those theories very plausible, what grounds did I as a Kantian theorist have for thinking they actual bear some important relation to the truth?

My worry here, fundamentally, was about the kind of peer-disagreement Travis alludes to. It occurred to me (as it sort of had implicitly for a very long time), that philosophy lacks any good method for resolving basic disagreements over premises--and that insofar as different philosophers disagree radically over which are plausible (let alone true!), there is a very deep problem here for philosophy, and by extension, for the claim that philosophy makes much (if any) *truth-apt* progress: the kind of progress that (again) philosophers like Maudlin repeatedly assert in interviews or imply in their work.

This post, and the comments in it, as well as the (very different) approach I ultimately ended up pursuing in my two books are all based on the thought that philosophy needs to fess up to this problem and try to do better. How can we do better? Well, I think in two ways. First, I think we *can* keep doing speculative philosophy, particularly in areas that science has not progressed far enough in to help us. But, when we do that kind of philosophy, I think we should fess up to the fact that our speculative methods are epistemically problematic, possessing little claim to truth-aptness. Second, I think we do non-speculative, naturalistic philosophy whenever possible. Rather than asserting (as people like Scanlon have) that empirical psychology can have nothing fruitful to say about moral philosophy, we should (or so I argue) think that it has the *most* fruitful things to say, since empirical moral psychology can tell us how people actually do cognize and reason about moral matters (something which I've argued can indeed be very fruitful normatively speaking).

So, ultimately, I'm actually very "bullish" on philosophy. I think speculative and naturalistic philosophy should *both* be done, but that our profession should be much more skeptical of the truth-aptness of the former and much more open to the latter (as there are still, to this day, many philosophers who seem to think that naturalistic philosophy either isn't philosophy at all or can have nothing fruitful to say about philosophical questions--both of which I think are false).

[Chris]

I'm not saying this is relevant to defending Maudlin or others' defenses of progress on the "big questions", but to me it is clear that philosophy makes progress in giving "answers" to philosophical questions (in your sense, (1), Marcus) - just narrower philosophical questions than are typically examined in the broad phil papers, survey. Maybe you think it is Chmess, but when a philosopher such as David Lewis proves his triviality results that show the problem of identifying conditional probability with the probability of a conditional event - that's progress! Its a proof, that means high standards. Or when philosophers of science prove things, or formal epistemologists, there is consensus (at least in the cases where their proof isn't flawed).

I don't know about ethics or political philosophy, but in philosophy of science, sometimes progress is made by making simple distinctions - pointing out that biologists (or philosophers) are confusing tokens with types, or necessary with sufficient conditions. There is work in my field, that, for those who care about it, has uncontroversial results - at least as uncontroversial as the science, at any rate.

(Incidentally, for a critique of van Fraassen that doesn't just rely on controversial "intuitions" see Sober "Epistemology for Empiricists" - the argument there is pretty clear. IF you accept certain basic ideas about how confirmation works, then van Fraassen is wrong." Of course, someone might reject these assumptions, but it is still progress if the argument is valid - it still shows that if you accept X, then Y follows. (It provides a truth apt answer to the question: If I accept this feature of how evidence supports a hypothesis, then is van Fraassen's view mistaken?)

(Maybe you don't disagree with any of this - I think that those who defend progress in philosophy should focus on these smaller problems where consensus is achieved (analogous to that in science/math) but not on the "big questions".)

[Mark Walker]

The recent collection of papers on this topic, Philosophy’s Future: The Problem of Philosophical Progress, includes contributions by Frank Jackson and Timothy Williamson. My paper in the collection argues that we should think that many of our philosophical views are probably wrong. That is, we should disbelieve many of our philosophical views. If this view were adopted, we could increase philosophical progress at least in one sense: the accuracy of our philosophical beliefs would jump dramatically. It is our epistemic hubris that gets in the way of this. My paper, “Between Gods and Apes: On the Lack of Philosophical and Scientific Progress”, can be found here: https://drive.google.com/file/d/1JrFMDZbOWHrhLBPRdGhJJrmN3ds4e_j0/view . Admittedly, I hold a minority position: all the other papers are more optimistic, some quite so.

[Marcus Arvan]

Hi Mike: so I think a lot of what you said is right, and a good example of genuine (though modest) philosophical progress using standard philosophical methods.

First, although I haven't thought about the issue in a very long time (though I did study philosophy of perception seriously in graduate school), I'm willing to accept--based on what I recall and what you say--that sense-datum theory has been refuted. One quick note, though. I'm not entirely sure what you mean when you assert: "Since they represent distal sensory stimuli, we experience distal sensory stimuli." But on at least one reading (say, an adverbialist theory of perception), the latter doesn't seem to me to follow. But maybe I'm misunderstanding what you meant there.

In any case, I think there are occasionally fairly (if not always completely) convincing arguments in philosophy to rule out particular views. Arguments against sense-data are plausibly one case; and in moral philosophy, my sense is that a lot of philosophers tend to give the Euthyphro dilemma against divine command theory as an example.

I think this kind of progress is indeed important, as it does rule out at least some false positions. But I think there are three problems here:

(1) Cases like this seem very few and far-between. My experience is that when you ask philosophers for further examples of this sort, there aren't many further examples to give--and oftentimes, when someone does give one, other philosophers will chime in and say, "Actually, I think that view you all think has been refuted is still perfectly defensible." And they'll typically have arguments claiming to show it. (Example: not too long ago, I made an argument on the Cocoon that ordinary language philosophy had been effectively ruled out--and I had some present-day defenders of ordinary language philosophy contend that their work shows that all of the arguments against ordinary language philosophy are bad and its still an important methodology!)

(2) As you note, the rejection of one false view doesn't, in itself, drive us any closer to the truth--not if all of the other views we are considering (or taking seriously) are also false.

(3) If I recall (I can't read paywalled articles at home during COVID), Jason Brennan indicates in the article I linked to in the OP that philosophy may actually increase the probability of false belief. This is because prior to doing philosophy, your commonsense view(s) about a philosophical matter may already be true and philosophy lead you to erroneously take seriously all kinds of false options--since, as we see above, philosophy generates vastly different incompatible theories (e.g. of morality).

Long story short, I think you're right that philosophy can rule out some false views (on the basis of internal incoherence). I also think you're right that in some cases, it can be important to make scientists aware of it (though, in other cases, I'm not too sure--an example: logical positivism is widely rejected in philosophy, but positivist beliefs led Einstein to relativity by convincing him that any conception of 'space' that is physically meaningful must be measurable--directly contrary to the Newtonians who preceded him. So, in some cases, false philosophical views may actually be instrumentally useful in spurring on scientific progress). In any case, while I agree with you on all this, I'm still inclined to think it is pretty rare for philosophical views to be ruled out like sense-datum theory. I also think that even in cases where theories are ruled out as incoherent, we still face the problem of whether traditional philosophical methods provide any truth-apt grounds for deciding among the remaining theories (e.g. adverbialism about perceptions, the theory of appearing, direct perception theory, etc.).

[Travis Timmerman]

Thanks for your reply Marcus. Here are my initial thoughts.

While I’m sure it’s true that 100% of experts in biology believe that plants and animals are made out of cells, for instance, they disagree about all sorts of other broader claims. Expert evolutionary biologists, for example, disagree about whether evolution takes place at the level of genes, cells, organism, groups, species, or multiple levels. Likewise, a 100% of experts agree that modus tonens (excluding cases of conditional obligation, and McGree-type counterexamples) is a valid argument, that events that result in the permanent cessation of one’s existence are prudentially relevant insofar as they result in one having a total lower well-being than they otherwise would have, that early versions of actualism violate ought distributes through conjunction, while versions that build in a control condition don’t, and the list goes on and on. Though there will be disagreement about the truth of broader more difficult philosophical questions. In all areas of intellectual inquiry, I expect there to be 100% agreement among experts about a number of claims, and disagreement about broader more contentious issues. Philosophy seems no different here.

I am not sure what you mean by the distinction between “basic facts” and “theories”, but I think there is lots agreement about what I take to be basic facts in any subfield of philosophy, and disagreement about theories. The supposed counterexamples you provide don’t seem like counterexamples at all since they all concern disagreement between novices and experts. You seemed to have moved the goalposts for philosophy. While it may be true that a mob boss might not think murder is wrong, that does no more to threaten convergence among experts in ethics than the fact that professional athlete Kyrie Irving thinks the earth is flat threatens convergence among geoscientists.

As an aside, there’s plenty of good answers to the question your wife’s student raised. I would start with David Enoch’s “Taking Morality Seriously” and Doug Portmore’s work on moral rationalism. But I digress. The main point is that the fact that uniformed novices (undergraduate students with no background in the field, mob bosses, and fictionalized characters from dialogues) don’t all accept that it’s wrong to murder presents no reason to think that there hasn’t been progress in ethics, much less philosophy. Moreover, there is plenty of agreement among experts about basic facts in ethics. I listed one above. There are plenty of examples, though, such as sentience is sufficient more moral considerability, and so is personhood, for every non-consequentialist normative ethical theory that renders deontic verdicts for acts, there is an analogue consequentialist normative ethical theory that renders the same deontic verdicts, some of Sidgwick’s axioms (not the moral theory he tries to derive from it), and the list goes on and on.

I don’t think you’re fully appreciating a point I raised earlier, and that is in the Ballantyne article too. It’s not an evasion to point out that knowledge claims about even basic facts in science depend on us being able to make knowledge claims about contentious philosophical issues that can only be settled by a priori reasoning. If you think a priori philosophical inquiry is an unreliable methodology (or, as you put it, “has no good claim to truth-aptness”), then you must think no claim that depends on the truth of claims that can only be assessed using a priori reasoning have a claim to truth-aptness. That includes not only every claim about “basic facts” in science, but every claim ever. This problem cannot be defined away, or ignored simply because all scientists agree that there are cells. Scientists are not thinking about the contentious philosophical issues that their empiricism presupposes, nor should they be. But assuming their claims are true (or we can know they’re true) requires presupposing a number of contentious philosophical claims are true (or that we can know they’re true) too and if we can’t know the truth about those philosophical issues from using philosophical methodology, then we can’t know the truth about any scientific issues that presuppose the truth of said philosophical issues.

Your own reply to me hinges on all sorts of contentious philosophical issues concerning epistemology and the philosophy of science. You couldn’t know if the claims you are making are true unless you also knew the truth of the philosophical positions they presuppose. Maybe you can know the truth about these contentious philosophical issues, but then it looks like philosophical methodology applied to those questions would have to have a claim to truth-aptness by your own lights.

I don’t think a lot of work in philosophy could be made more scientific, nor should it be made more scientific, as that wouldn’t in any way help us answer the philosophical questions. Just as one example, none of the work I do in ethics or the philosophy of death seems like it could be bolstered even the slightest bit by empirical science. But maybe some issues in your new book could be. I’d have to read it to know for sure.

My comment about taking solace with people who disagree with me was meant to be humorous, and was directed at the people who are complete philosophical skeptics, not anyone (like you) who grants that philosophy does make at least some progress. I am glad we agree it does.

[Marcus Arvan]

Hey Chris: Those are all good points. I think you're right that formal proofs like Lewis's are progress. But it's interesting that you give two types of examples:

(A) Formal proofs
(B) Work in the philosophy of science (correcting conceptual errors biologists may make, viz. tokens vs types)

What's interesting about these cases is that type (A) formal work is akin akin to mathematics, and (B)-type work draws on empirical science.

Here's why I think this is important: both of these types of work conform to what, following Quine, I want to call true observation sentences. *Everyone* providing and evaluating a formal proof in mathematics or modal logic (for example) can agree upon the axioms and inference-rules involved (viz. IF we assume S5 *and* X is the case, then we can derive Z through the following rules). There can then of course be further philosophical debates about the rules (e.g. intuitionistic logics, paraconsistent logics, etc.)--which I think can be hard to ever settle for the reasons I identify in this thread--but the point is that when giving a formal proof there tends to be universal agreement about what an axiom is, how other things can be derived from particular axioms given other rules, and so on.

For these reasons, I'm apt to agree that formal proofs can lead to definitive progress--for precisely the same reasons that math proofs do. But, I will add, formal proofs like these are very niche area of philosophy! There's not the vast majority of what philosophers do, and they're not the kinds of things that proponents of progress typically single out as progress (since, as we see in the case of Maudlin and others, defenders of progress typically want to claim *substantive* rather than merely formal progress in answering philosophical questions).

That brings us to type-(B) work: philosophy that grapples directly with empirical work (such as philosophy of biology, physics, etc.). Here, I'm inclined to think there can be some genuine progress because, unlike in other areas of philosophy, the philosophical work is being driven by empirical research: research based upon true observation sentences (thus affording subsequent critical reflection some real claim to truth-aptness).

Long story short, I'm inclined to agree that both (A)-type work and (B)-type work can give rise to truth-apt progress. But I'm also inclined to say that this isn't what most philosophers mean when they suggest that philosophy makes progress. For again, in interviews and in historical books, the more usual cases are substantive philosophical claims (such as that compatibilism is true, theism false, Kripke's claims about morality and reference true, and so on). It's those kinds of claims about progress that my concerns are primarily about.

[Marcus Arvan]

Hey Travis: Thanks for your reply. I'll have a longer reply later. But one quick thing for now.

You write: "The supposed counterexamples you provide don’t seem like counterexamples at all since they all concern disagreement between novices and experts...As an aside, there’s plenty of good answers to the question your wife’s student raised. I would start with David Enoch’s “Taking Morality Seriously” and Doug Portmore’s work on moral rationalism. But I digress. The main point is that the fact that uniformed novices (undergraduate students with no background in the field, mob bosses, and fictionalized characters from dialogues) don’t all accept that it’s wrong to murder presents no reason to think that there hasn’t been progress in ethics, much less philosophy. Moreover, there is plenty of agreement among experts about basic facts in ethics. I listed one above."

I simply disagree with most of this. First, as an expert, I share my students' persistent concerns about morality (and Thrasymachus', Polus', etc.). This is because, from a methodological/metaphilosophical perspective, I think their concerns are good ones (since what my students--and Thrasymachus, etc.--are doing is appealing to an instrumental conception of normativity, which everyday life and the empirical science of moral cognition indicate to the dominant and most universal conception of normative rationality and deliberation across ordinary human agents ranging from children to adolescents, adults, criminals, etc.). Second, there are other experts--both in moral philosophy and empirical moral psychology--who share these concerns (Richard Joyce, Hobbesian contractarians, and Daniel Batson, to name just a few--see e.g. https://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199355549.001.0001/acprof-9780199355549 ). Third, although there are experts out there who find existing answers to these concerns convincing, there are other experts who do not (though I think we can all agree that many answers are interesting!). In any case, these kinds of disagreements--including disagreements among experts on whether the concerns of novices are good ones--are precisely the problem I am trying to raise. When you claim "there is plenty of agreement" among experts in ethics, I recognize that. But what I do not agree that it is the kind of agreement we should take to be truth-apt (above and beyond appealing psychologically to the normative standpoint of a certain kind of person), since again there are plenty of experts who disagree. In this regard, I think ethics as a discipline is similar to the free will literature. Sure, there may be a fair amount of agreement in favor of compatibilism (e.g. 60%)--but, for all that, there are quite a lot of experts who think compatibilism is deeply mistaken. Ethics is similar.

[Malcolm]

Is this a question about Anglo-analytic philosophical progress? Or philosophy as a global human pursuit?

If the latter, I'd say progress involves also becoming aware of the different questions we should be asking, which requires looking beyond our own textual traditions. If the former, then I'd still say that without some awareness of whether that tradition is asking good questions, as Justin Weinberg is suggesting, it's hard to measure--and looking beyond is one approach to do that.

All of this suggests some reasons to think history of philosophy (writ large) is helpful for "first-order" philosophical progress, that is, getting better questions and better answers to them. And, insofar as there is more intercultural work happening, I'd suggest that itself is part of philosophical progress.

[Travis Timmerman]

Thanks Marcus,

I'd be interested to hear where you think Enoch and Portmore go wrong.

I still don't think your analogy is a good one, as you're giving your students and an imagined mob boss too much credit. If you want to change the analogy to focus on people like Richard Joyce (whose arguments for nihilism are very different from you wife's student who suggests something can't be wrong if you can get away with it), then that's fine. But even granting that there is disagreement among experts about the correct metaethical view doesn't, as far as I can tell, give me reason to reject any of the claims I made in the above post.

There is plenty of universal agreement about seemingly "basic" facts and disagreement about broad "theories" in most or all disciplines. Though, even if this weren't true, and there wasn't even universal agreement about so-called "basic" facts in ethics, I wouldn't think that would give us reason to doubt that ethics makes real progress.

Why would the fact that experts disagree about broad questions give us reason to think their agreement about other issues is not "the kind of agreement we should take to be truth-apt" (whatever you mean by truth-apt)? That seems unmotivated to me, and raises concerns that your position is self-defeating.

Also, I meant to write "modus ponens" in my last comment. Please excuse that and any other typos. I wrote my responses quickly during breaks from some editing duties I am trying to finish.

[Mike B]

Hi Marcus, I think we're mostly in agreement. I just wanted to expand a bit more on the role I see for philosophy (at least the philosophy I do). The problem with the sense-datum theory isn't so much its incoherence. I don't think that it entails any contradictions. The problem is the really ham-fisted way in which it handles the concept of "representation". For example, it fails to see that something can appear F without being F (the "sense-datum fallacy"), and it fails to distinguish between representational vehicles and representational content. So sense-datum theories (and early modern philosophers before them, and many perception scientists today) applied this impoverished concept of representation to the empirical data (namely, that often appearance mismatches reality) and got a silly theory out as the result.

Take, for instance, your cell example. You get 100% agreement on those observation sentences because the concepts in play are straightforward observational concepts, and fairly simple. But many important concepts (and the phenomena they denote) aren't. "Consciousness", "perceptual experience", etc., aren't simple observational concepts.

Certainly it's not the case that any interesting question about consciousness or perceptual experience could be answered by a single observation. (Unlike questions about whether living organisms are constituted by cells, which can be answered in a single experiment.) If you bungle these concepts, or related ones like "representation", then you're going to misinterpret the empirical data you can collect (e.g., data about neuron responses to stimuli).

So I see my role as a philosopher working in cognitive science as interpreting empirical results using sufficiently sophisticated concepts, concepts which are subtle enough to better fit the contours of reality. I also think there's a role for this kind of work under the normal division of labor: just as I'd be no good at designing and running (say) neural recording experiments, and such work takes years of training, most scientists are no good at doing conceptual work, and such work takes years of careful training.

So while I agree that there are probably few philosophical theories that we can rule out using purely philosophical methods, I think the application of philosophical conceptual work to empirical work in the sciences can result in real advances, and help push the conversation forward in constructive ways. A great example of this is how Ned Block has actually shaped discussion in cognitive science over the last 25 years via his distinction between access and phenomenal consciousness, or how David Chalmers' writings on the neural correlates of consciousness have have shaped discussion of neural recording studies.

[Benjamin LS Nelson]

I've said this before in other venues, but it has always seemed to me that, to ask "has philosophy made progress?", is about as sensible as asking, "who is the leader of traffic?" It misunderstands the nature and aims of the thing it's asking about. There is no such thing as a leader of traffic, because not everybody has the same start or finish line, either in absolute or relative terms. And there is no such thing as progress in philosophy, because the aims are as various as the starting-points.

Granted, in both cases, you can point to instances where there is a local leader -- e.g., in a caravan, or parade. But these aren't objections to the complaint. By the same token, advancements in modal logic or potentially temporary consensus on some particular issue do not count as objections to the thesis that there is no progress in philosophy.

[Marcus Arvan]

Hi Travis: Sure, no problem. My initial constitutivist defense of Rightness as Fairness (in early drafts) was not at all that different than Enoch's indispensability argument: roughly, that when we're deliberating, we're looking for the (objectively) right answer as to to how to act. Enoch argues that it follows from the indispensability of this to deliberation that we are committed to objective reasons, from which it is inferred that there are objective reasons, etc. In my initial draft of Rightness as Fairness, I defended Rightness as Fairness on similar grounds, arguing that indispensable elements of deliberation normatively constitute reasons to conform to the particular moral theory I went on to defend (Four Principles of Fairness).

Now, I still think there may be something to this kind of argument (and I actually plan to return to my own version of it in the future). But I ultimately don't think it is an adequate defense of morality against important forms of skepticism or a good way to establish moral philosophy to be truth-apt for a variety of related reasons. First, there are well-known places in the argument for skeptical people (including me and my students) to resist. See e.g. https://ndpr.nd.edu/news/taking-morality-seriously-a-defense-of-robust-realism/ . I think there are reasons to doubt the indispensability premise (since, as I explain below, actual human deliberation appears to be primarily instrumental, only attempting to arrive at right-relative-to-our-own-ends answers, thus not generating objective moral reasons but at most instrumental reasons relative to our ends). I also think there are grounds for doubting the inference from indispensability of X in deliberation (belief in objective reasons) to the actuality of X (objective reasons)--as X could just be an indispensable fiction. And I think there are other issues as well (related to the metaphysics and epistemology) of primitive reasons.

Second, the empirical psychology of human moral cognition and motivation shows a variety of things that actually cohere with the mob-boss, business student, etc., perspective--which is why I think it is so vital to address on its own terms. The empirical psychology shows that:

1. Human deliberation is primarily instrumental (people only tend to behave morally when it satisfies their ends, lying nearly always and rationalizing it as 'the right thing to do' when it suits their interests, etc.--which of course we see constantly on a daily basis in real life where people flout their 'moral principles', make excuses, rationalize what they did, etc., which is why our world in general is such a disaster!).

2. Moral belief specifically appears to be largely instrumental, such that when people (such as psychopaths) lack any desire to behave morally they "see nothing wrong whatsoever" with behaving that way (a similar phenomenon occurs when ordinary individuals gain power over others--hence, why administrators, mid-level managers, CEOs, etc. can seem so amoral in their judgments and behavior).

3. Moral belief and motivation both systematically involve personal-risk calculation, mental time-travel (retrospective and prospective), concern for one's own possible future selves, regret-avoidance, modal imagination, areas of the brain involving narrative representation, and a variety of other things.

Long story short, the empirical literature indicates that most ordinary people aren't all *that* different than the mob boss or criminal: most of what passes for 'morality' and moral beliefs and motivation are instrumental, means-end deliberation (though it turns out that a very particular kind of long-term instrumental deliberation leads us to cognize moral principles and use them to govern our behavior, making them our ends--which is what my book is about). Consequently, in my view arguments like Enoch's simply don't appropriately engage with (A) actual human moral cognition, or (B) the actual normative standpoint that most human agents adopt the vast majority of the time when we act, including when we think we act 'morally'. I have similar methodological and empirical concerns about Portmore's view and other similar views. This isn't to say that I don't think these theories are interesting or well-argued (I do!)--but it is to say that I have deep methodological problems with their approaches (corresponding to the concerns I've been exploring in this thread and elsewhere).

My own view is that the only kind of adequate answer to these problems is for moral philosophy to be deeply and systematically rooted in empirical moral psychology. This is the approach I adopt in my new book, where I argue the empirical science supports a unified normative and descriptive theory of prudence and morality: a theory that I argue explains a wide variety of empirical phenomena and normative phenomena better than alternatives. I also try to give a better answer to where the mob boss, criminal, and so on, go wrong in deliberation--giving a long developmental story on how we learn not to be like them but instead engage in the forms of risk-aversion, mental time-travel, modal imagination, and narrative representation that the science shows underlie and comprise moral beliefs and motivation.

But this--the centrality of empirical psychology to ethics--is precisely what you and many others (including Scanlon, etc.) seem to think is misguided when you write, "none of the work I do in ethics or the philosophy of death seems like it could be bolstered even the slightest bit by empirical science." I respectfully disagree. I think nearly all work in ethics would benefit from empirical moral and social psychology. And I've tried to make the case for this in detail in my most recent work. I may be wrong of course, but I'll be frank: I'm placing my bet that the 21st century will be the one where philosophers recognize just how important empirical psychology (and science more generally) is to what we do.

[Dan]

Maudlin's claim that 'since Locke and Hume there's been no interesting debate about free will' betrays an ignorance of subsequent discussion of free will/moral responsibility. Frankfurt's 1969 challenge to the principle of alternative possibilities (the claim that the ability to do otherwise is necessary for free will/moral responsibility)-- widely accepted/presupposed before Frankfurt, including by Locke and Hume-- marked a large shift of dialectical tide and generated interesting debate, by any standard of 'interesting' met by extant contributions to philosophy.

[Travis Timmerman]

Thanks for the detailed reply Marcus. That’s helpful, and I’m glad we agree that there is at least something to Enoch’s sort of argument. I disagree with your claim that this sort of argument, combined with this others, isn’t an adequate defense of morality against skeptical people simply because “there are well-known places” for skeptics to resist. That seems true of almost any argument, and so can’t be the bar by which the “adequacy” of arguments is measured.

That would be interesting if the empirical moral psychology of human cognition and motivation shows that they deliberate and form beliefs in the way you describe. It seems like I should read your new book to learn more about this!

Prior to doing so, however, I don’t see how these claims about how folk supposedly do deliberate and form moral beliefs has any bearing on what the moral truth is, or truths about how rational ought to deliberate unless you’re also arguing that the empirical literature shows they can’t do otherwise. I know you mentioned you address is-ought concerns in the book, so maybe this concern is addressed.
The empirical literature might show that people are naturally inclined to think arguments that affirm the consequent are truth preserving. But that doesn’t mean the correct account of logic should take this into account in any way, except perhaps to point out this common error is, indeed, an error. Likewise, if people are inclined to “behave morally” only when it suits their interests, that doesn’t mean the correct account of morality should take this into account in any way, except perhaps to point out this supposedly common error is, indeed, an error.

I have to read your book to know for sure, but I am antecedently skeptical that metaethical arguments like Enoch’s are failing to take (A) and (B) into account in an objectionable sense. I have no doubt that some work in ethics (moral psychology itself counts as ethics, and applied ethics, and questions about moral responsibility) can benefit enormously from work in empirical moral and social psychology. So we agree about those cases, though respectfully disagree about the scope of ethics in which this would be beneficial.

I have no idea how work in moral or social psychology could bear on the actualism/possibilism debate in ethics, or the specific axiological issues in death that interest me. Take one of my most recent papers on the topic (A Dilemma for Epicureanism in Philosophical Studies). I show that Epicureanism is either involved in a mere verbal dispute with deprivationists or is committed to making obviously false claims about prudence, claims which no actual Epicurean in the literature accepts. It’s hard to see what, even in principle, empirical work in moral psychology could be relevant to whether my argument is sound. But maybe this is just due to a lack of imagination on my part, and maybe I’ll read your book and become convinced.

[Marcus Arvan]

Thanks, Travis. Again, I'll give a longer response tomorrow, but here's a quick response to some of your concerns.

"I disagree with your claim that this sort of argument, combined with this others, isn’t an adequate defense of morality against skeptical people simply because “there are well-known places” for skeptics to resist. That seems true of almost any argument, and so can’t be the bar by which the “adequacy” of arguments is measured."

The issue, as I see it, is a bit deeper: namely, it's the methodological issues we've been discussing here. The basic problem that I have with these kinds of arguments (and many arguments in philosophy) is that they are based on really controversial premises--the very issue that I've been arguing raises a serious problem for philosophical progress (in contrast to science). So, for example, like I said earlier, the earlier versions of my first book initially aimed to defend something similar to Enoch's argument. But when a couple of reviewers pointed out that *only* Kantians were likely to buy the premises I was appealing to, that seemed to me a very serious problem. I want to know what the truth is, not what Kantians-are-disposed-to-think-the-truth-is. By a similar token, I don't buy the premises of any standard arguments for mind-independent moral realism. At best, they seem to me to represent what-those-antecedently-attracted-to-moral-realism-are-apt-to-think-the-truth-is. This all seems to me very similar (as I point out in the first chapter of my first book) to debates in early 20th century psychology between Freudians, Jungians, Humanists, and so on. They all thought their premises were true and their opponents' premises false, and yet...future psychological science showed that, for the most part, none of them were based on true premises. Which again is why I think, in general, we need better, more scientific methods. Could some of the philosophical theories I mentioned above have true premises (such as what is indispensable to deliberation)? Sure, but I think empirical science can help us figure that out by studying how people actually do deliberate.

So, this is what I think ethics needs to do: namely, move away from controversial premise-mongering to instead arguments based on Quinean observation sentences. And what accomplishes that? Well, or so I've argued, two things: (1) instrumentalism, and (2) empirical moral psychology. We can discuss why I think this further. But anyway, I argue that the empirical science strongly supports this picture.

This brings me to your concern about the relevance of actual human cognition: "I don’t see how these claims about how folk supposedly do deliberate and form moral beliefs has any bearing on what the moral truth is, or truths about how rational ought to deliberate..."

Well, the entire point of both of my books is that once we understand how people actually deliberate, we can recover and explain most everything we want to explain: why selfishness is tempting (and hence why people behave that way so much), why it is also generally irrational, why we should instead act in ways we can justify to others, why we tend to think moral reasons are categorical, how the moral psychology this involves generates four principles of fairness that reconcile the competing insights of major moral frameworks and unify moral theory with social and political theory--and how all of these things are unified by a normative theory of prudence and general theory of prudential and moral cognition that engages with people's actual motives.

Long story short: I argue that a proper understanding of how moral and prudential cognition actually work doesn't end up reifying bad reasoning but instead enables us to provide a better, more unified, more explanatorily powerful account of *good* moral reasoning. I'm not the only philosopher pursuing this type of account (see Eleonora Vigano, Josh May, Jeanette Kennett, etc.), and my account may or may not be on the right track. But I do think this is a promising track with methodological, epistemic, and normative advantages.

[Marcus Arvan]

Hi Travis: Thanks again for the great discussion!

Let me begin with some of your questions about how empirical science might bolster your research. You write: "Take one of my most recent papers on the topic (A Dilemma for Epicureanism in Philosophical Studies). I show that Epicureanism is either involved in a mere verbal dispute with deprivationists or is committed to making obviously false claims about prudence, claims which no actual Epicurean in the literature accepts."

One general problem I have with a bunch of different literatures in philosophy is that oftentimes researchers make claims about what it "obviously" true or false that don't seem that way to me at all. See https://philosopherscocoon.typepad.com/blog/2015/12/when-you-have-the-wrong-intuitions.html . My worry here, as I've expressed many times before, is that researchers may falsely generalize from what they or their interlocutors take to be obviously true or false to inferences about what is in fact obviously true or false. Further, in at least some cases, empirical research has ended up challenging the claims that researchers previously took to be obvious from the armchair. For example, there is an entire literature premised on the claim that it's clear that ordinary folk are objectivists about morality. Yet this claim has recently been challenged by empirical work: https://philpapers.org/rec/BEETEC. Or, to take another recent case, a well-known researcher recently wrote a post on social media expressing puzzlement that a couple of reviewers asked them to defend a 'platitude' about morality. It then turned out in the following social media thread that almost no one took it to be a platitude except for them and other moral realists. There's also empirical work challenging the premises of influential thought experiments about linguistic meaning and reference. And so on.

In your case, here's how I think empirical research could help. I argue in my book that claims about prudence are ultimately empirical claims--since they are claims about what can be expected to make a life go well as a whole. I also argue that we're not entitled to simply defer to commonsense judgments about prudence, as it may turn out empirically that there are counterexamples--individuals who can live prudentially good lives in ways that don't conform to our armchair expectations.

In any case, in your Epicureanism paper you assert two premises as an Essential Desideratum related to prudence and death:

Preferring Life (PL): Any person P has pro tanto self-regarding reason to prefer (and ensure, if possible) continued life at time t if P’s total well-being
would be higher if P does not die at t than if she does die at t.

Preferring Death (PD): Any person P has pro tanto self-regarding reason to prefer (and ensure, if possible) death at time t if P’s total well-being would be higher if P dies at t than if she does not die at t.

You claim these are modest, seemingly axiomatic principles. (p. 243) But, as it happens, I think both of them are false. Why? Because I think the normative reasons a person has fundamentally depends, among other things, on their epistemic situation--that is, about what they have *evidence* about regarding their life as a whole. The two principles you take to be axiomatic have nothing to do about a person's evidence but instead make reasons claims merely on the basis of objective facts about the world and about what would be the case if someone were to die at a particular time. I don't think objective facts give anyone pro tanto reasons to prefer anything unless those people also have some evidence of what those facts are or are likely to be. So, here's what seems to me to be the case: you take these to be modest principles that are obviously true because you assume an objective theory of reasons. I do not. Hence, you and I disagree over basic premises on methodological and substantive grounds. I claim we need the method of empirical sciences to help us resolve the disagreement on non-question-begging grounds. For example, I argue in my recent work that our total evidence (both normative and empirical) best supports a theory of mind-and-epistemic-position-dependent reasons like my own theories positing mind-independent reasons. Although I know you doubt that empirical science can help answer these kinds of questions, there are those of us who think empirical science can help that might be more persuaded by such arguments if they had the kind of empirical backing we take to be important.

Now turn to your very good paper on Singer and drowning children: https://philpapers.org/rec/TIMSTI. Here too I think empirical science might bolster your case. Why? Because, in brief, I think the empirical science of moral motivation broadly coheres with the claims you make in the paper. I don't think this is the place to go into that in detail--but assuming this is correct, why wouldn't it help your case to be able to show, "Oh, and by the way, the empirical science of moral judgment and motivation coheres with my account in ways X, Y, and Z far better than Singer's?" I think by all means it would, as it would help support the longstanding worry that Singer's account is too demanding and appeal to moral theorists who think that moral theories and applied ethical arguments should at least be broadly consistent with realistic facts about human cognition and motivation.

Anyway, these are just some quick thoughts. I'll try to address your metaphilosophical concerns (about science, philosophy, and Nathan's argument) later today!

[Marcus Arvan]

Hey Mike B: I think you and me are on the same page. That’s the kind of intersection between philosophy and empirical science that I strongly favor!

[Marcus Arvan]

Benjamin LS Nelson: here’s why I don’t think the question misrepresents the aim of what’s being talked about when we do philosophy. Philosophers make assertions: those assertions typically claim on the surface to express truths: truths about things like grounding, the nature of mind, morality, and so on. So, it seems—both in the language philosophers typically use and they way pitch or describe the results of philosophical inquiry (see Maudlin’s claims about convergence to truth)—a central aim of philosophy is to ascertain the truth or what we are justified in believing the truth to be. But progress in that core aim manifestly can be evaluated. It’s what we do in every other area of inquiry that purports to seek the truth: namely, evaluate whether we are getting closer, further away, or have no idea which. So, I don’t think the progress question is akin to asking who’s the leader in traffic. It’s more akin to asking who’s leading a race where the finish line is the truth. Now, maybe this isn’t how some philosophers conceive the core aims of philosophy. But it seems to me than many clearly do and that in any case philosophy do purport to make assertions that can be true or false and hence evaluated as such in evaluating progress toward that goal.

[Benjamin LS Nelson]

Hi Marcus, thanks for the reply. Please, call me Ben!

Not every philosophical assertion is taken seriously as truth-bearing in productive philosophical discussion, given the multiplicity of contexts that define what counts as 'truth' for the purposes of answering the question at hand. e.g., for Moore, all our Cartesian demons were dismissed through the verbal equivalent of an eyeroll.

Maybe we share the same types of ends -- e.g., truth-seeking in the broadest sense, directed towards the resolution of questions of wide scope (or related to such questions). That seems right, to me. But how much work does that do? Going back to the analogy, every car in traffic means to go somewhere -- namely, its destination. That is, each car has its own context that defines what counts as a 'destination' for that car. That doesn't mean that when I let some dude merge ahead of me on the freeway that he's closer to his destination than I am.

I don't mean to suggest, incidentally, that contextualism is correct. It may very well not be -- actually, I think it cannot be quite right. But it seems to me that it does greater service to the surface experience of doing philosophy, and that taking the appearances seriously means that analogies to 'crossing the finish line', racing, etc., are going to have to be made far more complicated, if not abandoned.

[Marcus Arvan]

Hi Ben: That's a very interesting perspective, and I really appreciate the opportunity to think about it.

I guess I have real concerns about truth-contextualism, and by extension about respecting the surface experience of doing philosophy. As I explain below, I think the surface experience of doing philosophy is ambiguous between two things: (1) truth-simpliciter, and (2) some other end(s). My concern then is that in that case the surface experience of doing philosophy runs into a very serious dilemma--one represented somewhat in this great article by Abe Graber, "Creating truths by winning arguments: the problem of methodological artifacts in philosophy" - https://philpapers.org/rec/GRACTB-4

In brief, the dilemma is this. On the one hand, philosophers sometimes seem to think their arguments get at truths *simpliciter*. In my experience interacting with philosophers, this seems to me to be what they generally take themselves and their theories to be doing. They aim at true theories (for example, I think mind-independent moral realists think their theories are either true or more likely to be true than alternatives). However, to the extent that this is the surface experience of doing philosophy, then I think philosophers typically use bad methods for meeting that goal (viz. the kinds of concerns I've aired here and elsewhere). In other words, when philosophers do philosophy with truth simpliciter as their goal (which it seems they often do), I think the analogy of a race with a single finish line is absolutely correct and we have little or no reasons to think anyone is getting close to the finish line they think they are.

On the other hand, if we adopt a serious form of truth-contextualism--or any other standard for doing philosophy aside from truth simpliciter--then I think we run into serious problems with 'creating apparent truths out of thin air' that really aren't truths at all but rather something else: specifically, representations of what some subclass of people 'find plausible', and then simply identifying that with 'truth in a context.' But in that case I think we're doing something *very* dubious that I explained here: https://philosopherscocoon.typepad.com/blog/2020/03/neurofunctional-prudence-and-morality-some-intellectual-history-part-1.html . Specifically, what I think we're doing then is dressing up something probably not worth doing--namely, investigations that we have no reason to take to have any relation to truth--as though they do have some relation to truth.

To see what I think the problem is, consider some of the examples I give early in my first book and have mentioned here: Freudian, Jungian, Humanist, etc., theories of psychology. People in the early 20th Century took these theories very seriously, and many of their proponents were hardly any less convinced that their theories were true or well-justified than many philosophical practitioners today. A lot of times, proponents of the theories vastly oversold the evidence in favor of them, prescribing treatments to people (such as psychoanalysis), and so on. And finally, when modern psychological science came along, the evidence turned out to mostly put the lie to those theories.

I have similar concerns about philosophy. If we enter into classrooms teaching moral theories, claiming they're well justified by arguments, and so on, we need to ask: by what standard? If it's an only an internal-to-the-practice standard (viz. the intuitions of professional philosophers), then I fear that we work with a non-truth-apt method akin to what Freudians, Jungians, and so on, were doing. But I take it we don't want to be doing *that*, as we now know in retrospect that those theories were mostly nonsense. When we present ourselves to our students, I take it that we tend to present ourselves as something like guardians of careful thinking, and people who are pursuing the truth about philosophical questions.

And so the dilemma is this: either (1) we are doing that (seeking truth simpliciter) but often not using truth-apt methods, or (2) we're not seeking truth simpliciter, in which case there are very serious epistemic and practical questions about the value of what we're doing.

[Marcus Arvan]

Hi Travis: Thanks again for the great discussion.

Allow me try to address your (and Nathan's) very challenging points about science, philosophy, and truth.

You write: "I don’t think you’re fully appreciating a point I raised earlier, and that is in the Ballantyne article too. It’s not an evasion to point out that knowledge claims about even basic facts in science depend on us being able to make knowledge claims about contentious philosophical issues that can only be settled by a priori reasoning. If you think a priori philosophical inquiry is an unreliable methodology (or, as you put it, “has no good claim to truth-aptness”), then you must think no claim that depends on the truth of claims that can only be assessed using a priori reasoning have a claim to truth-aptness. That includes not only every claim about “basic facts” in science, but every claim ever. This problem cannot be defined away, or ignored simply because all scientists agree that there are cells. Scientists are not thinking about the contentious philosophical issues that their empiricism presupposes, nor should they be. But assuming their claims are true (or we can know they’re true) requires presupposing a number of contentious philosophical claims are true (or that we can know they’re true) too and if we can’t know the truth about those philosophical issues from using philosophical methodology, then we can’t know the truth about any scientific issues that presuppose the truth of said philosophical issues."

I fundamentally disagree with the following claim, which is a linchpin in your argument: "even basic facts in science depend on us being able to make knowledge claims about contentious philosophical issues that can only be settled by a priori reasoning."

I don't think we can substantively settle *anything* a priori (except proofs of logic, mathematics, etc., which are not substantive but rather formal in nature). This is for broadly Quinean 'Two Dogma' and later-Wittgensteinian grounds. First, like Quine, I doubt there is a viable analytic/synthetic distinction--and, by extension, I doubt that there is a viable a priori/a posteriori distinction. Rather, I think most (if not all) 'a priori' philosophy is predicated on surreptitious a posteriori assumptions about what the facts in the world are. For example, if it is claimed on 'a priori' grounds that moral language is cognitivist, I think when we trace out the argument we will see that the 'a priori' argument depends on a posteriori facts about how moral language is used (this is a general point Quine makes about analyticity generally: that whenever we try to define it, we either find ourselves in a circular regress or make synthetic/a posteriori assumptions about language use). Second, like later-Wittgenstein, I think substantive a priori philosophy is predicated on misconceived picture of the nature of language: the supposition that there are determinate facts of the matter about what our concepts mean that can be revealed by conceptual analysis rather than through investigating a posteriori facts about how words are used ('meaning as use'). For both of these reasons, I regard nearly all substantive 'a priori' endeavors in philosophy as misguided. And so I'm inclined to think that philosophy instead should be thoroughly naturalistic (note: I don't think this is self-defeating, because I don't think Quine's and Wittgenstein's arguments are a priori ones).

Bearing this in mind, I think as an a posteriori matter that the core ordinary use of the word 'true' (in ordinary language) is to pick out things that people universally converge upon viz. basic Quinean observation sentences. I think this is the core use because when we use the term 'true' in everyday life, we mean that the truth is not what you think it is, or what I think it is, but we can *see* it to be together (which is how Quine understands basic observation sentences, and which is the backbone of all science).

I take it a posteriori that this is the ordinary meaning of the word 'true', and that there are good practical reasons to have it as a concept and primary use in our vocabulary (to distinguish mere subjective option from intersubjective fact). Given that this is the ordinary meaning of the word--which is the one I think we're all interested in, including philosophers (see below)--I think it is safe and correct to say that the findings of science are true (because they are based upon Quinean observation sentences in turn generating theoretical convergence by making correct predictions), but that it is not correct to say that the findings of most armchair philosophical arguments are truth-apt because (in general) philosophical arguments tend to be based on controversial premises that some investigators *think* are true but other investigators don't.

Now, I expect there will be many objections here (as of course there always are in philosophy)--such as that I am not warranted in simply deferring to the ordinary usage of the word 'true', and so on. At which point I will contend a posteriori that what philosophers seem to be really looking for is truth in the ordinary sense of the term--since in moral philosophy, for instance, moral realists don't just think it's 'true for them' that reasons are objective but rather true simpliciter that reasons are objective (in the sense that they think that's the view that other investigators should converge upon). So, ultimately, I will try to make the claim that philosophers are ultimately just out to provide true theories in the ordinary sense of the term--which is simply what science is trying to do as well. I will then maintain that science has proven to be the best methodology of arriving at that (true predictions), whereas countless other forms of inquiry (astrology, alchemy, etc.) don't. And then I will point out the things I have in this thread: that as an a posteriori matter science clearly gets at the truth in *that* sense better that traditional philosophical methods as well (which I think is hard to deny, since we continually debate whether philosophy makes any progress in that sense whereas we don't debate whether science does).

At this point, of course, we may have arrived at a fundamental disagreement--since I expect you want to say that all of the above need to be settled a priori rather than a posteriori and I will demur. At which point, unless we can find some common ground to continue the debate (maybe we can!), we may have to concede that we fundamentally disagree in philosophical outlook at have to leave it to posterity to determine which outlook to prefer. However, I will point to the history of science supplanting what were once considered a priori philosophical fields (ancient cosmology, a priori arguments space must be absolute, philosophy of perception, etc.) and then providing good a posteriori answers (modern physics, relativity, etc.) that are now widely accepted and practically useful (since they enable us to make true predictions about the world). And I will will express the belief and hope that philosophy will do better this time around by embracing empirically-informed approaches rather than rejecting them--as I fear that if (say) in moral philosophy we largely stick to the armchair, then the empirical psychologists will come take the game away in a similar manner to scientists in other areas in centuries past. Better to work with science, I think, than to fight or neglect it (not just for practical reasons but again, or so I think, for epistemic reasons).

Of course, this is just my philosophical outlook and hope for the future. I know not everyone shares them (not by a long shot!), and that it may well be dead wrong (as I am well aware of my own fallibility, strong views notwithstanding!). But I love philosophy for these very reasons--because, despite my own views about metaphilosophy and methodology, there are so many of us with different sensibilities making the best arguments and theories we can, challenging each other argumentatively and methodologically, etc. However deep our philosophical disagreements may be, I deeply respect you, Nathan, and others as thinkers and the work you do--and I want to reiterate that I don't think that armchair theorizing has no good use or role to play. If you look at my publication list (and indeed, at significant parts of my books), you'll find that I've done plenty of it myself! It is merely my belief that philosophers shouldn't be satisfied with those methods alone, and that we should supplement those methods with empirical science whenever possible. Maybe this is wrong. I dunno. But it's my current view, and this is why I'm so interested in having these kinds of conversations!

[Benjamin LS Nelson]

Hi Marcus, thanks again -- very interesting.

I am with you in thinking that philosophers often prioritize the use of methods that appropriate during inquiry guided towards truth-simpliciter. Be that as it may, this is not going to confront the analogy or its underlying point. After all, plenty of cars never reach their destinations, either, due to accidents and so on. We know what makes philosophy go bad, just as we know how cars fail to reach their destination. But the presence of a uniform standard for failure does not guarantee a uniform standard for success, especially not the success of the peculiar and demanding kind connoted by the notion of "progress".

Incidentally, I do not accept your dilemma. It seems to me that "truth simpliciter" is an elliptical shorthand for a truth-definition that is both valid inside and across contexts. I think it is unlikely that anything quite like that exists, though I also don't think that that makes philosophy unproductive. There is a middle position: I acknowledge that some contexts of evaluation are comparatively better than others, but only up to a point. By the same token, a busted old Model T with its hood on fire (say, creationism) is relatively unlikely to reach its destination, and we will have achieved progress of a kind insofar as we make cars and systems of traffic that are less likely to fail to get to their various destinations. But that does not tell us who is the leader in traffic, who has crossed the finish line.

(Aside: I don't think either Freud or Jung were nonsensical, so much as often false, e.g., with the Oedipal complex. But at times their sort of baroque and imaginative inquiry helped to improve our understanding of an area of life that is difficult to come to terms with. Example: suppose we ask, "Is there a death instinct?" The question doesn't really occur to analytic philosophers or normal, non-spooky social scientists. But to me, that sounds like an interesting question -- I mean, as we speak, there are plenty of "black pill" or "Jokerified" Americans in the streets who sure sound like they want to die. Now, on this matter, from what I understand, Jung says no, there is no such thing, and Freud says yes there is. I'm inclined to agree with Jung. And I'm also inclined to say that it's an important question.)

[Marcus Arvan]

Hey Ben: I'm inclined to think there is a truth-definition that is valid inside and across contexts, and come to think of it I think I want to try to write a paper on this. But in any case, this is very helpful in pinning down where our disagreement seems to lie. If you're right that there is no such definition, then indeed, there is no dilemma. But I'm inclined to think there is such a definition--and that once we focus on that one, we can see that philosophical cars (as it were) are crashing en masse and we need to construct them differently (viz. updating our methods).

Anyway, these are deep issues (the nature of truth) that we can't settle here--but it's been a great discussion for pinning these issues down!

In terms of Freudianism, Jungianism, etc., I agree the theories aren't literally nonsense. As I mentioned in an earlier comment, I merely meant to say that they are 'nonsense' in the colloquial use of the term (making a bunch of false claims that science later showed to be such). However, I do have qualms with the idea that in some ways they improved us our understanding of human life. Having lived through the era when these views were still popular (and used to justify expensive treatments with no scientific support), they were treated by many with the utter seriousness of a religious cult--in ways that I think were profoundly harmful to people, particularly patients but also to basic standards of evidence and critical thinking (seriously, the way in which many people expounded Freudian ideas seemed to me akin to talking about the secret powers of healing crystals). I think early 20th century psychology was just another step in a long line of poor thinking dressed up with utter seriousness of Truth (e.g. astrology, alchemy, bloodletting, etc.).

This is why I've been pressing these concerns about philosophy so hard. I love philosophy. I wouldn't do it if I didn't. But I think there are serious issues with standard philosophical methodologies that really aren't that different than the issues with pseudoscience. Further, quite a lot of times (e.g. Marxism), I think pseudoscientific philosophical theorizing can no less profoundly harmful than pseudoscience (in brief, substantive philosophical problems aside, Marx's overall theory is so out of line with what is known in empirical psychology that it is no surprise at all that Communist dicatorships have been systematically plagued with persecution and mass murder). My aims in pushing these things so hard is not to condemn philosophy but rather to enjoin us to do better: to not shun empirically-informed philosophy (as a considerable number of philosophers do), but instead realize that empirical methods really are the one place where we can be confident that the philosophical 'rubber hits the road' (i.e. theories tracking truth).

All that being said, I'm apt to agree with you that it's an interesting question whether some people have a death wish. But I don't think it take bad psychological theories to recognize that as a question worth investigating!

[Benjamin LS Nelson]

Hi Marcus,

One last thought on truth, just to offer a pre-emptive defense before you write that paper! From a critical perspective, I do think there are cross-contextual *constraints* on an appropriate truth-definition (where critical means, insofar as we are trying to fix our wayward truth-talking). But such critical constraints on a truth-definition do not act as a definition for truth, because I can still make sense of the misguided ways that people speak about truth. An ameliorative truth-definition would potentially make such (truth-pretending) expressions unintelligible, when all a language-fixer ought to want to say is that wayward truth-talk only amounts to the commission of understandable mistakes.

(If I am reading you right, then by 'nonsense', you probably just meant 'bullshit'. That's fair, and captures the gist of one colloquial sense of the term. And it does apply to some -- but not all -- of the spooky humanistic literature from last century -- but only some. That said, I think it a digression.)

Cheers!

[Marcus Arvan]

Thanks, Ben - that's super helpful! I'm not sure that an ameliorative definition would have to make truth-pretending expressions unintelligible. Given my later-Wittgensteinian proclivities, I'm more inclined to say that (A) there's a core use of the truth-predicate (viz. Quinean observation sentences), (B) there are good practical reasons for that core use and for using its attendant truth-definition as the one that guides inquiry, whereas (C) there are non-core uses (and hence, alternative definitions) of the truth-predicate that are possible and not nonsensical but nevertheless *defective* for guiding inquiry (because, unlike the core use, they are unable to reliably distinguish subjective BS-ing from the kind of intersubjective agreement we normally take to definitive of truth on in every other area of life, including science).

[William Peden]

My main areas of interest are formal epistemology, philosophy of science, and logic, where it's hard not to feel that philosophy has made a lot of progress since, say, 1920. I don't think that we know the answers to a lot of the big questions in these areas (How should we reason with probabilities? Can one rationally withhold belief in our best-tested scientific theories? Is classical logic sufficient for reasoning about the world we live in? etc.) but I think we've made a lot of progress on smaller questions, which is generally how intellectual progress works. I am also confident that, in 2120, philosophers in these areas will know lots of things that I don't, which is extremely exciting.

[Bryan Frances]

I published a paper defending the existence of philosophical progress. Here is the abstract:

I first argue that there is plenty of agreement among philosophers on philosophically
substantive claims, which fall into three categories. This agreement suggests that there
is important philosophical progress. I then argue that although it’s easy to list several
potential kinds of philosophical progress, it is much harder to determine whether the
potential is actual. Then I attempt to articulate the truth that the deniers of
philosophical progress are latching on to. Finally, I comment on the significance of the
agreement and (potential) progress.

You can read it here:

http://bryanfrances.weebly.com/uploads/4/6/6/2/46624483/agreement_and_progress_metaphilosophy.pdf

Cheers!

[Marcus Arvan]

Thanks for chiming in, and for the link to your paper Bryan. I just gave it a quick look and may post a response in due course!

[John Keller]

Travis, I don't think Ballantyne's argument works, for reasons outlined here: http://link.springer.com/article/10.1007/s10670-014-9720-z

(Or here if you don't have access: http://johnakeller.org/OKA.pdf )

Very briefly, I argue that since we are asking different questions in philosophy than we are asking in other domains, arguments that would be circular in philosophy are not circular in those other domains.

I know a podcast by Chalmers is linked to above, but this paper of his is good too: http://www.consc.net/papers/progress.pdf

As is van Inwagen't reply, but I think you need to get the book for that.

[Untenured Ethicist]

Here are three claims I'm confident are true:

(1) There is no largest prime number.
(2) It is wrong to cheat a customer to get money to pay for a vacation.
(3) I have two hands.

Though I feel confident about all three, I believe that my justification for believing the first two is stronger. For (1), there is a proof. For (2), there is an a priori argument (a Kantian one) that seems to me sound. To believe (3), I must trust my senses, which involves a leap of faith.

There is a list of things moral philosophers know in Richard Sharvy's magnificent polemic "Who's to Say What's Right or Wrong? People Who Have Ph.D.s in Philosophy, That's Who."

http://www.luvnpeas.org/rsharvy/whostosay.html

Some of the claims on Sharvy's list are overstated. For instance, instead of "The most valuable things in life are useless," I would have said, "The value of the most valuable things does not consist in their usefulness." This would leave open the question whether the most valuable things include things that have both instrumental and final value. But I think Sharvy is right to think that moral philosophy has contributed to moral knowledge, even if it leaves some important questions unsettled.

[Marcus Arvan]

Untenured Ethicist: I’m confident they’re all true too, but I don't think takes studying philosophy for someone to know them. First, even before Newton gave a formal proof of an infinite series of primes, it is fairly intuitive that insofar as mathematical number lines are infinite (something school children generally get), basic patterns of math almost certainly repeat at an infinite level. Sure, Newton's proof proved it--but that was a formal proof (which I haven't questioned can be used to make progress).

In terms of ethics, most ordinary people learn it's wrong to cheat even without having done philosophy. Yes, many ordinary people are garden variety relativists—but I actually think they are onto something important: namely, that there *are* good reasons to doubt whether moral objectivism is anything more than an article of faith (see Thrasymachus, Glaucon, Adeimantus, Polus, and Nietzsche’s famous claim that of course the sheep will condemn birds of prey, but for all that birds of prey have no reason to care!). I've argued in my newest book that moral philosophers have never adequately resolved this problem, and that it is ultimately an empirical question whether moral norms objectively apply to everyone or whether there are outliers to whom morality doesn't apply at all (in a sense making it 'relative'). Now, of course, I know moral philosophers generally despise this notion (as the dominant view is that moral norms must objectively apply to everyone to *be* moral norms. However, I argue there are very good reasons for taking it seriously, despite its unwelcome implications.

Finally, Kant’s a priori arguments in moral philosophy, while very interesting, are notoriously problematic. Philosophers to this very day debate what his arguments are exactly, and whether any of core ones he makes in moral philosophy are successful! For my part, I don’t think any of them are (though, as an ex-Kantian, I find them fascinating nonetheless!).

[Marcus Arvan]

Hi Bryan: I read your paper with interest (thanks for sharing it!). While I think you provide a pretty good case that there is some real progress in philosophy (which I accept), I have to disagree with some details as well as more with how you understand progress (as I still think there is something deficient at a basic level with the *kind* of progress philosophy tends to make when it does, at least insofar as we use standard armchair methods).

First, a couple of points on details. Consider the first two examples of progress you give to start the paper:

1. Epistemicism faces a serious objection regarding how sharp meanings are fixed.

2. There are good reasons to think that truths about causation are closely connected to certain
counterfactual truths and/or truths about laws of nature.

Although you say on p. 3, "it can take months of
exposure to many readings and lectures before a student can even grasp virtually any of (1)-(10)", I think this is clearly false.

Consider epistemicism. Sure, the average person has never heard of the theory. But suppose you told them that some philosophers have defended this theory that exactly *1* hair makes the difference between a balding person and a non-balding person, and that *1* grain of rice makes the difference between a heap and a non-heap--because according to their philosophical theory, all of our words are perfectly precise like this. Suppose you then added that philosophers who defend this theory will admit that *we can never know* which hair or grain of rice that is.

Quite frankly, here is what will probably happen. First, the person will give you an incredulous stare, expressing disbelief that philosophers could defend such an absurd view seriously. Then they would probably say, "Yes, but we don't *use* words like that. We don't use 'bald' to refer to precisely how many hairs a person has or 'heap' to refer to an exact number of grains of rice. So, how in the world do your philosophers think one hair can make a difference?" I should know. I used to work on vagueness, and whenever I talked about it with non-philosophers, this is the kind of stuff they would say. Such people are clearly demonstrating an implicit understanding of (1): that there is an obvious problem given how ordinary language is used for how in the world sharp meanings are fixed given how we use language. And there's a good reason for that: it's an *obvious* problem right on the face of it.

Now turn to (2). Does one *really* need to do philosophy in order to appreciate that causation has something to do with counterfactuals or laws of nature? If you were to ask someone why hitting a cue ball on a billards table causes the other balls it hits to ricochet around, if they are a halfway intelligent person, they would say something like, "Well, that's just how the laws of physics work" or "If you hadn't hit the cue ball, the other balls wouldn't have moved."

Sure, philosophers have developed all kinds of intricate theories about (1) and (2)--but the point is simply that you don't need to do a lick of philosophy to know that (1) and (2) are obviously true.

I think something like this is true of a number of others claims you list (such as that some beliefs are true and others false, false beliefs cannot be knowledge, a belief can be reasonable but false, and so on).

The problem then, as I see it, is that once we excise all of the intuitively obvious cases that one doesn't need to be a professional philosopher to know, you have to admit things like this claim you make on p. 3: "I’m not saying that all or even any of (1)-(10) is true. Second, there is of course no universal agreement about any of the ten claims, even restricting the pool of philosophers to those with the relevant expertise. But that is true for just about any field of study that has at least a thousand thoroughly trained practitioners."

As I mentioned earlier in this thread (and John Keller's response to Nathan Ballantyne's paper shows), this claim about other fields is false. In physics, chemistry, biology, and other scientific fields, ≈100% of researchers accept some *particular* leading theory (i.e. quantum mechanics, relativity, organic chemistry). This simply isn't the case in any area of philosophy. The closest we get is probably Atheism, where something like 80% of philosophers agree. Take any other area of philosophy you like, there is *no* single leading theory that ≈100% of theorists accept. This is categorically different in kind than what we see in the sciences--and I've tried to diagnose what the difference reduces to: truth-apt methods in the case of science (viz. Quinean observation sentences) versus non-truth apt methods in philosophy (viz. differences in intuitions over premises, which as we see in the history of human inquiry from ancient Greek cosmology through astrology through early-20th Century psychology has a consistently poor pedigree across different areas of inquiry).

This, I think, is the real issue. Yes, there are areas in philosophy where significant areas of agreement arise--but there are deep reasons to doubt whether philosophical agreement in general is truth-apt, given especially that (again) in just about every field, there is vast dissensus over premises and theories--something again, that is categorically different than science and ordinary-everyday standards for judging something to be factual/true.

I also don't think conditional knowledge (pp. 5-6) should satisfy us, because it's one thing to do that *if* (e.g.) consequentialism is true, then Y. Sure, that may be interesting to know--but if we have no further truth-apt methods for determining whether consequentialism is in fact true, then knowing the conditional doesn't take us very far (in terms of giving us first-order truths about morality, which is what consequentialism is of course a theory of).

[Bryan Frances]

Hi Marcus,

Some replies:

Regarding claims such as “Epistemicism faces a serious objection regarding how sharp meanings are fixed”, I think we’re talking past one another. I meant that in order to have a *deep* understanding of it, it takes a great deal of time. For comparison, I was once at a talk by an excellent philosopher of physics, and some of his talk was on the traditional problem of induction. He thought he could solve it without too much difficulty. My colleague Peter Simons started off the Q&A with, “I don’t think you really understand the problem”. He was right, and that’s the type of thing I was gesturing at.

When it comes to claims like “false beliefs aren’t knowledge”, I insist that plenty of non-philosophers reject some of these “Basic” claims. It’s amazing but true. I remember discussing with a smart undergraduate, for at least 30 minutes, the possibility that one might suspend judgment on some claims. No matter how clear-cut my examples and commentary, he found this profoundly unthinkable. Many people have incredibly dim grasp of even simple claims about basic notions. The more I shut up and let undergraduates and non-philosophers talk, the more I see this.

I do agree with you that scientific fields have a great deal more agreement on substantive and esoteric matters compared to philosophy. No doubt about it. But I think what you say about relativity and other theories is false. Many scientists walk and talk as if they believe these theories but really just accept them for research purposes only.

I agree with you that conditional claims should not “satisfy” us, as you put it. For my own part, I find anything less than knowing the truth of the really big issues quite unsatisfying. As you would guess, that makes me think philosophy is very unsatisfying!

I’m not sure what you mean by ‘truth-apt’. (I'm sorry if you answered that question elsewhere.) Are you saying that our philosophical claims don’t have truth-value, at least typically? If so, is it due to semantic incompleteness or unresolved ambiguity or something similar? Or by ‘truth-apt’ did you mean something epistemic?
On the face of it, sometimes philosophers genuinely contradict each other. I may hold that all ordinary positive knowledge attributions in everyday life of the form ‘S knows that P’ are false. And you may say that some of them are true. It seems that one of us has a true position and one has a false position.

Bryan

[Untenured Ethicist]

Marcus, thanks for your reply. I wonder what you think about Hume's Law. Trying to test moral claims empirically seems to me a conceptual mistake. Facts about how people do behave or about how people think people should behave tell us nothing about how people should behave without a further value premise.

My reason for mentioning mathematics was to point out that mathematics is not an empirical subject and to suggest that moral philosophy may be likewise non-empirical. It is not from experience that we know there are infinitely many primes. We need Euclid's proof. It remains unknown whether there are infinitely many twin primes.

There are mathematical theorems that are counter-intuitive. We should be open to the possibility of moral theorems that prove widespread moral intuitions to be incorrect.

[Marcus Arvan]

Hi Untenured: You'd really have to read my most recent book for my full answer to the Hume's law issue (I also have an unpublished draft paper developing my response further). For an overview, see here: https://www.marcusarvan.net/neurofunctional-prudence-and-morali

But anyway, here's the brief story. I argue in my book that if we use strict standards of evidence (i.e. the same standards we use to evaluate truth in everyday life and science), we can see that the best foundation for moral philosophy is:

(1) Instrumental, means-end normative rationality.
(2) Empirical psychology.

In brief, the argument for (1) is that this is the only conception of normativity that *all* competent linguistic users commit themselves to on a day-to-day basis--including children, criminals, psychopaths, and ordinary adults. I argue that because all competent users share this normative commitment, we can give a naturalistic semantics (a Humean reduction) for instrumental normative propositions that makes them come out *objectively true simpliciter*, using the same standards of truth as science.

In contrast, I argue that all other purported normative foundations for morality (including 'objective moral reasons', Kantian a priori arguments, etc.) always end up appealing to premises that large bodies of competent language users reject--thus making premises for those kinds of arguments *not* truth-apt in any way remotely close to how truth is normally understood in everyday life or science.

My argument then is that when we examine empirical psychology, we indeed find that (1) instrumental deliberation is dominant across all populations (including those who believe they are acting for 'moral reasons'), and that (2) people only experience moral norms as normatively binding the extent that they have instrumental interest in them via prospective mental-time travel (note: there is a lot of empirical evidence for this picture given in the book, including but not limited to a wealth of evidence that psychopaths don't find moral norms binding for these reasons and ordinary adults not finding moral norms binding when they gain power over others--much as Glaucon famously argued in the Ring of Gyges case in Book II of The Republic).

In brief outline, I argue that once we think about evidential standards in moral philosophy and metaethics well, it turns out that morality itself--normatively and descriptively--reduces to prudence. But, I argue, prudence is *empirically tractable*, as we can study scientifically what in fact makes people's lives go better over the course of life as a whole, as well as how people cognize this in making moral decisions.

If this is right, then moral questions generally reduce to empirical questions about prudential and moral psychology, as well as empirical facts about the world around us. Further, I argue that those very facts support a normative moral theory (Rightness as Fairness) that claims to reconcile the major insights of utilitarianism, Kantianism, contractualism, and virtue ethics, providing a better (instrumental and moral) way for making applied moral decisions (see also https://www.marcusarvan.net/my-book ).

As for Hume's law, I make two claims in my newest book (pp. 126-7):

(1) My argument does not presuppose that Hume's Law is false, as I do not assume that a naturalistic reduction of moral semantics is correct. All I assume is that instrumental normativity is the only conception of *normativity* ('oughts') that satisfies appropriate evidential standards of truth-aptness.

(2) Nevertheless, I am optimistic that a naturalistic reduction of normativity is correct--as I argue that both the truth-conditions and truthmakers of instrumental normative propositions are simply descriptive facts about the world--making instrumental normative propositions true but reducible to descriptive empirical facts (thus bridging rather than violating the is-ought gap).

I also discuss these issues in my first book. Finally, in the conclusion of my newest book, I note that I believe that my theory is ultimately compatible with speculative Kantian foundations (as my initial draft of my first book defended Rightness as Fairness on Kantian a priori grounds).

My longest term project is to unify the Humean/naturalistic account of morality with a Kantian a priori one--because, as just noted, I believe both of these foundations converge on the same result: Rightness as Fairness. I simply chose the naturalistic/instrumental approach first for the reasons outlined in this thread: I think it has far better epistemic bona fides than a priori approaches to morality.

[Untenured Ethicist]

Marcus, thank you for your thoughtful and detailed reply. I will have to read your book to grasp the position fully.

I have a different view about the appropriate epistemic standards for practical reasoning. I think they are substantially the same as the appropriate standards for mathematical reasoning. Incomplete understanding of important concepts is common. That many intelligent adults reject the correct answer to the Monty Hall problem does not make it any less objectively correct. People are bound by the norms of sound reasoning about probability even if they do not recognize these norms.

[Marcus Arvan]

Hey Untenured: Thanks for your reply as well!

I guess I'm not entirely sure what you mean when you say "I have a different view about the appropriate epistemic standards for practical reasoning. I think they are substantially the same as the appropriate standards for mathematical reasoning."

In mathematics, when a correct proof is found, ≈100% of mathematicians will generally accept it--because the proof will be based on basic axioms that ≈100% of mathematicians accept as axiomatic or previous proofs based on those fundamental axioms. I haven't denied that purely formal areas of philosophy (such as mathematical logic or formal epistemology) can make truth-apt progress for these reasons.

What I do deny--and what Keller's paper linked to above also suggests--is that armchair philosophy in general conforms to anything like the same standards of evidence and practical reasoning as science or mathematics. Again, science, mathematics, and everyday understanding of evidence are all based on what Quine calls "observation sentences": basic observations that enjoy ≈100% of intersubjective agreement. For example, any scientist can run a test to confirm the predictions of Einsteinian relativity with clear, intersubjective data (readings on clocks, apparent position of stars near the Sun, etc.). Yes, there are underdetermination issues with respect to theory-confirmation (the same observational data being consistent with multiple theories)--but the point is that the basic observational data themselves are clear and intersubjectively verifiable to ≈100% of observers.

In contrast, substantive a priori arguments in philosophy--such as Kantian arguments for the categorical imperative, moral realist arguments for mind-independent 'moral facts'--are based on controversial premises that some people find attractive but other investigators reject. In this regarded (as noted above), armchair philosophy methods on substantive matters appear to be much more akin to speculative psychology in the early 20th century (which wasn't truth-apt), astrology (which isn't truth-apt either), and ancient Greek physical cosmology (which wasn't truth apt either).

None of this is to say that I think we should give up philosophical speculation altogether--as it can be interesting, particularly when it concerns questions that science cannot address yet. It *is* to say that we should be much more skeptical, however, of whether philosophical speculation is truth-apt--and it is certainly to say that philosophical speculation (as standardly practiced) in no way conforms to the same epistemic standards as science or abstract mathematics.

[nawzad]

I think the comments and feedback's are more fruitful than the article in itself. What I mean by this, is discussion, critics and deliberation are the "only" way of progress in philosophy.

[Marcus Arvan]

Hi Bryan: Thanks for keeping the conversation going, and sorry for the delay in responding. It was a busy end of the week! Here are some of my thoughts in reply.

You write: "Regarding claims such as “Epistemicism faces a serious objection regarding how sharp meanings are fixed”, I think we’re talking past one another. I meant that in order to have a *deep* understanding of it, it takes a great deal of time."

Here's the problem. This assumes that there is anything deep about epistemicism to genuinely understand--beyond what the ordinary person already knows (that is absurd for the very reasons most philosophers who study vagueness already reject it). Sure, philosophers can come up with all kind of armchair theories to "better understand it." But here my worry is that they are engaging in the equivalent of what Dennett calls "Higher-order truths about Chmess." To go back to an analogy I've given throughout this thread (and I think it's an apt analogy), in the early 20th Century Freudians, Jungians, etc. *thought* they were getting at a deeper understanding of psychology. But, as later scientific investigation, they weren't. By and large, they were just BS-ing. We cannot assume that armchair philosophical methods get at any kind of genuine understanding unless we have at least some reason to think they bear some relation to the truth (as opposed to simply misguided intuitions).

This brings me to: "I’m not sure what you mean by ‘truth-apt’. (I'm sorry if you answered that question elsewhere.) Are you saying that our philosophical claims don’t have truth-value, at least typically? If so, is it due to semantic incompleteness or unresolved ambiguity or something similar? Or by ‘truth-apt’ did you mean something epistemic?"

I've explained what I mean by this throughout this thread, and elaborated upon it in the first chapter of my first book: https://www.marcusarvan.net/my-book . I don't think that most philosophical methods have any good to be truth-apt according to how truth is normally understood in everyday language or in science, which understands it along the lines of Quinean observation-sentences (viz. basic observations that enjoy virtually universal assent/convergence). For example, *anyone* with sight can look through a microscope and see what an expert biologist sees. The biologist can then explain *what* that person (indeed any person) sees using biological theory--a theory that makes testable predictions that, again, *anyone* can confirm or disconfirm via other Quinean-observation sentences. This is the beauty of science: it doesn't permit BS-ing, or substituting controversial intuitions for generally observable facts. In contrast, arguments in philosophy almost always based on deeply controversial intuitions--intuitions that some investigators share but others do not. Furthermore, to the extent that a priori philosophers reject empirical Quinean observation sentences as a final, independent court of appeals (which is also plainly what differentiates armchair philosophy from science), philosophy contains no reliable method for distinguishing what is actually true from what a given philosopher or their friends and interlocutors might *want* to be true or erroneously *think* to be true (when it's not).

In this regard, once again, armchair philosophy seems all too like early-20th century psychology. Freudians were absolutely convinced that Freudian psychoanalysis was true. They based their arguments on strong intuitions they had, and Freud developed a vast theory to systematize those intuitions. Jungians did something similar. And so on. And it was all...just...wishful thinking. Almost none of what those theories said bears any relation to the truth.

This is the problem: without conforming rigorously to the standard for truth normally appealed to in everyday life and science (Quinean observation sentences), armchair philosophy engages in a form of inquiry that has been repeatedly shown throughout history (when new areas of science replaced philosophical speculation) to be *non*-truth apt.

Finally, in terms of the occasional student you run into who rejects the idea that knowledge requires true believe, here's the problem: the occasional philosopher denies it for roughly the same reasons - see

Hazlett, Allan, 2010, “The Myth of Factive Verbs”, Philosophy and Phenomenological Research, 80(3): 497–522. doi:10.1111/j.1933-1592.2010.00338.x

Nolan, Daniel (2008). Non-Factivity About Knowledge: A Defensive Move. The Reasoner 2 (11):6-7.

So, here again, it doesn't look like armchair philosophy improves understanding. Rather, most of us are already inclined to think knowledge requires true belief before doing any philosophy--and so most philosophers defend that. But some of us are antecedently willing to deny that knowledge requires true belief before doing philosophy--and so some philosophers defend that. In both cases, philosophy looks like it just develops complicated theories of what people already believed--rather than, like science, providing any kind of independent test of who is right, who is wrong, etc., leaving us to all debate these questions (like vagueness and the sorites paradox) for thousands of years with few, if any, clear answers.

[adc]

Hi Marcus,

I think your judgment about psychoanalysis is too strong. Take Freud. Freud's corpus is vast. He wrote about dreams, religion, everyday foibles, sexuality, among many other things. The claim that "almost [none of it] bears any relation to the truth" would require a mountain of argumentation. There have been plenty of people throughout the 20th century who've tried to provide it, but as far as I can tell much of it ignores the fact that Freud is often (not always) engaged in a project that is interpretative in nature. He wan't "basing his theories on strong intuitions he had" but instead on the observation of many people and the attempt to make sense of their behavior. Some might argue that he's no more relying on intuition than any of us are when we interpret the actions of others in terms of what they think and desire. What's interesting about Freud is that he argued that we should treat a wide range of behavior that would normally be considered passive (e.g., slips of the tongue and dreams) as if they are expressions of agency, and by doing so we can make sense of them as things we do as opposed to things that merely happen to us. To evaluate those interpretations (and there are lots of them) would require going through the particularities of each one and seeing where Freud's interpretations are plausible and where they aren't. I understand why someone would think that Freud was wrong about some things. But to say that he was "just BS-ing" seems unfair. If he was just BS-ing, then what do you make of philosophers like Jonathan Lear who take him seriously? If he was just BS-ing and it's all nonsense, then why do literary critics borrow insights for interpreting texts from psychoanalysis and not from alchemy and astrology? Perhaps it's because what Freud was doing is quite different from those enterprises and ought not be dismissed out of hand so easily. Just to be clear: I'm not claiming that Freudian psychoanalysis provides effective treatments for mental illness. I am claiming, though, that the interpretative dimension of psychoanalysis makes it unclear that scientific investigation could show it to be nonsense anymore than scientific investigation could show that certain interpretations of literary texts are nonsense. Psychoanalysis is not empirical psychology.

[Marcus Arvan]

Hi abc: If anything, I have gone too easy on Freud in this thread.

First, Freud's research methods and the practice of psychoanalysis would almost certainly not pass an IRB ethics board today--since it basically experiments recklessly with human subjects without any sound research design.

Second, his theory and the treatment practice it supported (psychoanalysis) have been profoundly harmful. I used to work in psychology. People shelled out large amounts of money to psychotherapists who convinced them they wanted to kill their fathers and sleep with their mothers (the Oedipal complex), had them revisit their most traumatic experiences over and over again (in 'talk therapy'), etc.--all in a treatment protocol which has since been shown to not lead to any robust forms of psychological improvement.

Third, Freud claimed to be doing empirical psychology, in that he claimed to be giving causal explanations of human behavior. Your comment in turn shows that he was indeed basing it on nothing more than strong intuitions image had. As you write, he simply observed people and then came up with interpretations of their behavior that “seemed plausible to him.” That *just is* observing the world and coming up with armchair intuitions about what seems plausible, which is *exactly* what astrologers do with the stars: they observed the heavens and birthdates, and then came up with interpretations of the connections between the two (horoscopes) that seemed plausible to them. Moreover, if we used your standard for evaluating Freud (e.g. it would take “mountains of argumentation” to disprove every individual claim that astrologers make about cases!), then we should take astrology seriously as an interpretive endeavor as well. But this is all patently absurd. Recasting Freudianism as nonempirical but “interpretive in nature” simply makes it no no different than astrology—end of story.

Why do a few philosophers and a bunch of literary critics still use Freud but not astrology? Well, why do some philosophers still take Marx seriously even though his labor theory of value is deeply problematic (labor isn't the main causal factor in economic value) and Communism is blatantly inconsistent with even a cursory understanding of human psychology? My answer: because despite Freud and Marx saying a bunch of false things, a lot of people sorta liked what they had to say...and bad habits die hard.

[adc]

Hi Marcus,

I didn't say that Freud would pass an IRB board. And I don't deny that some patients have been harmed by therapy. I was merely responding to your claims that Freud's theorizing was "just BS-ing"; that most of what he said "bore no relation to the truth"; that it's just"wishful thinking"; that it's "nonsense". Such remarks strike me as hyperbolic. You've now said that some philosophers and other humanists take him seriously because they like what he said and "bad habits die hard". All of this also strikes me as uncharitable, not only to Freud but to those other philosophers and humanists. It's the lack of charity that concerns me.

I'll leave it to others to defend Marx. I'll simply say that insofar as there is a lack of charity there too (both to Marx and our colleagues), I have the same concern.

I don't know that much about astrology. The analogy that seems relevant to me is the literary interpretation of texts. The interpretation of literature is a nonempirical endeavor, but I don't think taking those efforts seriously is "patently absurd." (Is it absurd to try to understand what Hamlet is about?) I don't know what goes into casting horoscopes, but if it's like interpreting a text, then it should be evaluated according to the standards of good and bad interpretations and not on the standards we use to evaluate astronomical theories. (That's a conditional claim; again, I don't know much about astrology).

[Evan]

ADC: You make an interesting comment about interpretation. I think all philosophers, academics, and professionals of the past were TRYING to describe things and prescribed things. Did many of them turn out to be false? Yes. From a historical perspective, it was inevitable that they would make false claims. So I agree that I don’t think they were BSing. I think they truly thought what they wrote and theorized were true. I don’t think Marx and Freud and other theorists had the intention lie or make misleading claims. Perhaps some did. But those are probably little amount. I don’t blame a lot of people in the past for being false or ignorant about a lot of things. Ignorance then was just inevitable.

But now that we DO know they made a lot of false claims, we have to ask ourselves: Should we still be relying on their work? If so, why? And how much of their work should we rely on or use?

When it comes to interpretation of literary texts, it’s very difficult to know for sure what the author intended or meant since a lot of them are dead now. And nor did a lot of them provide background justifications and explanations for their stories. Some artists never tell their viewers what their art is about and so we’re left with our own interpretation(s) of it. All these interpretations exist because the original creator of that work often failed to provide reasons for even creating that work or academics failed to find any explanation from the author or he or she was just a bad writer to begin with. Fortunately for us in this day and age, interviews and press releases exist more than ever. There is little to no shortages in trying to understand what Harry Potter was about, why J.K. Rawling wrote it, why she chose the characters, themes, set, etc.

We don’t really need to interpret Harry Potter. Instead we can just ask Rawling or watch her interviews and other personal writings to fully know what she intended for the story and characters, etc. In terms of older literatures and philosophical works that have little to no explanation from the original author, we have no choice but to interpret their work if we truly care to understand it.

However, there are plenty of empirical and psychological studies of certain concepts that can be used to supplement or support the interpretation of a particular literature or work of art WITHOUT using or citing Freud. The fact that Freud is still being cited in these areas given better options we have today, is academic laziness. Why use Freud when there are plenty of modern day research on human behavior, minds, and emotions? Are Freud’s theories truly relevant in the interpretation of literature? If anything, I think modern day psychology would be more useful in aiding literary criticism since the language and semantics they utilize aren’t pretentious like Freud’s. Any methodology or tools used in any discipline must be adequately justified and compelling, lest we end up recycling false or bad academic works over and over again.

[adc]

Hi Evan,

I'm don't do literary theory, so I'm not sure I'm going to be able to adequately address your concerns. But my understanding is that whether all interpretive questions about a literary text could be answered by appeal to the author is controversial; the authority of authorial intent, as far as I know, is an unresolved matter. For example, take the question of why Hamlet hesitates to kill Claudius. Or the question of how to understand the relationship between Hamlet and Ophelia. Or the question of whether Hamlet is mad. Suppose Shakespeare were alive, and you asked him for answers to these questions. Is it obvious that his answers would be better than the other various interpretations that have been offered over the centuries? What about cases where an author doesn't have answers about the text? For instance, suppose Shakespeare couldn't tell you why Hamlet hesitates. Wouldn't we still be faced with the question?

I agree that there are other interpretive tools that are more informed by contemporary work. But I don't think those are necessarily in conflict with psychoanalytic tools because there can be more than one interpretation of a literary text. Again, I'm not a literary theorist, but my understanding is that there's no such thing as the one true interpretation of a literary text. Instead, there are relatively good and bad interpretations. Some interpretations are better than others, and sometimes there can be multiple distinct interpretations that are good. So whether some of Freud's ideas should be used as an aid in literary analysis is determined by how good the interpretations are that arise from the use of those ideas and not by the mere existence of alternative methods of interpretation inspired by modern empirical research. And those interpretations need to be evaluated on a case by case basis. (For example, maybe Freud wouldn't be useful in interpreting Hamlet but maybe he would be useful in interpreting certain horror movies. We'd have to look and see.) If we should understand psychoanalysis less as an empirical science and more as a hermeneutic enterprise, then the mere fact that modern psychology has moved past Freud doesn't seem relevant to whether Freud's ideas are useful in providing good interpretations.

[Evan]

ADC: If Shakespeare was alive and provided answers to our questions, I would argue that yes his interpretation would be better. Why? Because it’s fiction! In fiction, any explanation from the author would be “better.” The explanation does not have to be scientific or realistic in the realm of fiction. Critiquing a particular fictional work because it doesn’t correspond to the laws of science or plausibility is irrelevant since it is (as the name suggests) fictional. Shakespeare’s interpretation or explanation may not seem plausible to us, but in his fictional universe, it probably is.

In the realm of fiction, we just have to accept that many events are weird, implausible, or nonsensical. I don’t think we should expect too much plausibility from fiction. So if X author says this is what his or her work of art is about, then we should just accept it. After all, it’s fiction. What else should we expect from it?

In the event that an author can’t provide any explanation of his or her work, then it’s open to multiple interpretations and community standards of today. Some works are more fixed or open than others in terms of interpretation.

Second, I think there is some similarity between the way science interprets data and literary theorists interpret a text. Both usually provide explanations for something they are studying. For example, a scientist looks at a data that shows X amount of people are dying from cancer. But she still needs to provide an explanation and hence interpret the data for why X amount of people are dying from cancer. She may and should come up with multiple explanations since there can be multiple causes for cancer. But her explanations are bound by the laws of causality and plausibility of our reality..

Because when we truly DON’T know why or how something occurs in a literature, we have to be open to MULTIPLE interpretation. And here is one problem with interpretations or explanations: some or many people are essentialists. They often refuse to accept multiple explanations for a phenomenon. I see this both in modern science AND Freud etc.

Just read critiques of Germs, Guns, and Steel by Jared Diamond (a biologist) from multiple historians, psychologists, anthropologists and other academics. He overlooked the role of ethics, cultural norms, and human psychology in the shaping of human history, societies, and ways of life. Germs, guns, and steel are necessary in shaping human history today, but they are by no means sufficient. His interpretation or explanation of human history from those three aspects are not as accurate as he would suggest, which is such a shame because his work would have been more fruitful if he at least adequately acknowledged those areas of humanity he largely neglected.