Further Reflections on Analytic Philosophy

My last blog post garnered some interesting discussion which I thought I would collate here. The impetus for this is that the long-suffering Preston Stovall kindly emailed with a detailed response. We both agreed it would be nice for him to be able to publish the response on my blog post, so it’s below. He did this in a timely manner after I published the blog post, so naturally I’ve sat on it for a month as I deal with my disorganisation and psychodrama. Sorry Preston!

Before getting to that, I thought I would link to some of the other discussions so one can get a span of opinions. First, the ever reliable Daily Nous ended up hosting quite a detailed discussion in its comments. I’d say the vibe is most people think I am overly pessimistic, but check it out for yourself! 

One style of response that I found quite common in my own interactions was very well spelled out in this blog post - here the thought is that while it is true that philosophers generally cannot plausibly believe they will achieve rational consensus, this is not such a bad thing. The mistake was ever hoping for that in the first place, and once we have gotten over that hangup we can enjoy the sort of pluralistic free play of ideas that comes with a taste for dissensus. For my own part, while I think I share some similar sensibilities, I am less confident this is really a way to avoid the worry. To me this is something like a paradox of happiness situation - it’s true that happiness is a valuable thing indeed, but if you set out consciously with it in mind “ok I am going to do things so I am happy” you are probably going about it wrong. Happiness is attained as a byproduct through doing that which brings you intrinsic satisfaction. My suspicion is the value of this pluralistic free play of ideas in a similar way depends on being a byproduct of people seeking (and failing to attain) the sort of rational consensus that I am so pessimistic about.

Another response was more in line with my own pessimistic tendencies. This blog post explored more of the cultural and institutional factors that led to the bad situation I outlined. It describes a dialectic which culminated thus: “[e]ventually analytic philosophy worked itself pure: a cohort of charismatic but somewhat abrasive professors debated arcane topics with one another, isolated from most of the greater academic community, their leisure and keynote travel supported by an ever expanding cast of adjuncts, visiting professors and graduate assistants whose career prospects were vanishing before their eyes.” This, it is argued, was a not stable equilibrium, and the signs of decay and decline I pointed to are really just people fighting about what comes next. Maybe so - but I didn’t expect to get outflanked on the pessimism front! Just as with the above and below, it’s an interesting read and I recommend you check it out.

Ok those were the three sets of responses I wanted to highlight. So without further ado, see below for Dr. Stovall’s defence of philosophy! He’s a more optimistic sort than I am, so it’s the sort of pleasant situation wherein the more plausible it seems I am wrong in light of his arguments, the happier I shall be to learn of my error!

---------------------------------------------------------------------------------------------------------------------------------------------------

In Defense of Philosophy: Analytic, Pragmatist, and Transcendental

Preston Stovall

June 4^th, 2021

I

I found myself nodding along, at one step removed, to some of the claims Liam makes in his essay. For instance, though he describes a “pervasive pessimistic skepticism” among young analytic philosophers that is no doubt pervasive in some areas, pessimistic skepticism has not been my impression of the state of things in the circles I move. And while the “grand march to Kripke” narrative has a sound sociological basis in the development of 20^th century philosophical logic, I find it rather unconvincing as a claim about the formal foundations for much of the work we do (I realize Liam’s not endorsing that narrative himself, and I appreciate his call to engage more with historians of philosophy). 

In fact, I think there’d be much less pessimistic skepticism about the future of analytic philosophy if people were more familiar with some of the reasons to reject (or at least substantially qualify) that narrative. In response to Liam’s essay, then, I’ll lay out a case for optimism about the state of analytic philosophy today (or about the state of a descendant of what used to be called “analytic philosophy”). In the rest of part I, I’ll begin by sketching a critique of the “march to Kripke” narrative, before using that critique to highlight some of the work that gives us reason for optimism (or so I claim). In part II, I’ll broaden the lens and consider the question of whether anything unifies the study of philosophy. In the end, I want to suggest that our reasons for optimism about contemporary analytically influenced philosophy are also reasons to think we are taking part in a shared activity that has been ongoing since almost the first use of the term “philosophy”.

It’s true that, sociologically, the discipline glommed on to possible-worlds semantics and representational theories of meaning in the second half of the 20^th century. Given the development of philosophical logic in the first half of that century, there were sound reasons for doing so. The work of people like Ruth Barcan Marcus, Saul Kripke, and Georg Henrik von Wright showed that the alethic (and other) modalities could be tolerably modelled in terms of possible worlds, by showing that the same kinds of recursive semantic evaluations as had already been given for predicate logic and the Boolean operators could be given for sentences falling under these modalities. What’s more, the axiom systems botanized by C.I. Lewis at the beginning of the 20^th century could be accounted for in terms of mathematical properties of the accessibility relations between worlds, offering (what appear to be) substantive answers to questions about whether, e.g., something could be necessary without being necessarily necessary.

Equipped with this extension of the old logic to models including possible worlds, and as W.V.O. Quine’s animus against modal logic receded into the background of philosophical consciousness, philosophers began to put these models to use in their work. But as Adam Tamas Tuboly argues in working through the literature of that period, Quine’s concerns weren’t so much addressed as ignored (Robert Brandom has made this point in a couple of places as well). There’s good research being done by people like Greg Frost-Arnold (who makes a showing in the comments in Liam’s post), Kevin J. Harrelson, Joel Katzav, and Adam Tuboly in questioning conventional narratives about the development of analytic philosophy (and Katzav’s work, along with Krist Vaesen, on the apparent journal-capture by analytic philosophers taking place during the interwar through post-WWII periods, is pretty eye-opening). 

The development of model-theoretic possible-worlds semantics has done much to shape areas like epistemology, metaphysics, the philosophy of mind, artificial intelligence, and the philosophy of perception over the last few decades, to say nothing of the influence it’s had in linguistics and computer programming. But there are a number of undercurrents to the development of philosophical logic since the pioneering work of Gottlob Frege and C.S. Peirce on extensional first-order languages, and many of these currents flow through channels that do not lead to Kripke’s work. The existence of a century’s worth of vibrant research programs in proof-theoretic semantics, for instance, shows that the recent dominance of model-theoretic representationalism is a historical contingency. And if more philosophers were familiar with the proof-theoretic foundations of contemporary logic and linguistics, it would be clear that there are at least two directions that philosophy has been “marching” since the middle of the last century. 

Familiarity with the proof-theoretic strand in the history of analytic philosophy also helps clear up some of the conceptual oddities that lie at the base of much of the work that has been done in metaphysics since the development of possible-worlds semantics. The interest in so-called “hyperintensional” semantics, and the much-lauded “hyperintensional revolution”, for instance, are artifacts of Rudolf Carnap’s decision to replace Fregean senses with intensions as functions from state descriptions to extensions in Meaning and Necessity. The fact that “intensional semantics” is now, in many idiolects, coextensional with “possible-worlds semantics” is evidence enough that the very need for a hyperintensional revolution has been self-imposed as a byproduct of an impoverished notion of meaning.

Carnap was clear that his notion of intension wasn’t meant to do the work that older notions of intension (sense, comprehension, connotation, content, etc.) were able to do, however, and his purposes were served by interpreting intensions in this way. But important developments in 20^th century philosophy take on a new significance once this historical picture is brought into view. Kit Fine’s work on essences in the 1990s and early 2000s, for instance, is predicated on the observation that possible worlds cannot by themselves distinguish the asymmetric ontological dependence (so the thought runs) of the singleton set {Socrates} on the human being Socrates. Each exist at exactly the same worlds, and so the ontological dependence of a set on its members cannot be modelled in those terms.  Consequently, models restricted to possible worlds cannot explain the different truth-conditional meanings of “Socrates is essential to {Socrates}” and “{Socrates} is essential to Socrates”. 

In response, Fine developed a model-theoretic representational metalanguage that uses talk of essence to interpret object-language talk of essence. What this establishes is not that there’s a realm of metaphysical essence that the philosophical logician can divine the logical contours of, but rather that the notion of intension at work on possible-worlds semantics is inadequate as a complement to extensional notions of meaning (as Carnap himself knew).

Today, model-theoretic and realist representationalism about metaphysics (employing metalanguages of possible worlds, essences, etc.) is widespread, and the field of inquiry into essences has been substantially shaped by Fine’s work. But if we go proof-theoretic, we can interpret object-language talk of essence in terms of a metalanguage of explanatory inferences, where metaphysical language is mentioned in a metalanguage that appeals to features of the proof system. It is then possible to use ordinary reflection on our pre-existing habits of explanation, adverting to material facts about the domains in question, to settle on object-language interpretations for essentialist talk. In this regard, and to adopt a Kantian turn of phrase, the study of metaphysics becomes a kind of reflection on the conditions under which it is possible to so much as think, understand, or make judgments about things like human beings and sets.

The supposed asymmetric ontological dependence of {Socrates} on Socrates, for instance, can be explained by recourse to the prosaic way we identify and individuate sets and human beings. For we show that two sets are identical by showing that they include exactly the same members (in the formal mode: we write up two lists and verify that exactly the same names occur on each list), but we do not show that two human beings are identical by comparing the sets containing them. Instead, we trace the lives of human beings. Object-language talk of essence goes over into a metalanguage of explanation, where the latter is read off of existing material commitments about the things we’re talking about. To the extent that there’s a metaphysical project here, it’s one that proceeds by reflection on the way we habitually reason about things like organisms and mathematical objects. 

Rather than try to build a notion of intension inside model theory, then, it seems more promising to treat model theory and proof theory as two formalisms for reconstructing the old extension/intension distinction concerning complementary notions of meaning – one ontological, and concerned with word-world (and, as I’ll note in a moment, world-word) relations, and the other deontic, concerned with word-word relations. And there are interesting philosophical discoveries to be made when we do so.

Although mainstream work in analytic philosophy tends to be informed almost entirely by the model-theoretic realist and representational picture of meaning exemplified in possible-worlds semantics, proof theory has remained a viable research program in philosophical logic and linguistics since the work of Gerhard Gentzen in the 1930s (there’s a story to be told here about the considerations that led Tarski and Gödel to convince Carnap in the 1930s to adopt a metalanguage of truth, and of the subsequent shift from Carnap’s so-called “syntactic” to his “semantic” period – or better, from his implicitly proof-theoretic semantic period to his explicitly model-theoretic semantic period). More analytic philosophers are coming around to proof-theoretic notions of meaning today, and I suspect that proof-theoretic methods for recording the structure of derivations will prove useful in modelling modal content in AI systems meant to represent complex domains like human agency, weather patterns, and contamination propagation (this is a conjecture on my part).

Nevertheless, scientifically informed model theory may still do productive work in philosophy and, perhaps, metaphysics. On that front, analytic philosophers working in cognition seem to be another counterpoint to the pervasive pessimistic skepticism Liam talks about. There’s productive and scientifically informed philosophy being done on dialectical processes of reasoning (e.g. with the work of Catarina Dutilh Novaes and Ladislav Koreň), on perception and knowledge (Bence Nanay and Tyler Burge), on epistemology (Hilary Kornblith and Josef Perner), on broadly inferential (and Sellarsian) theories of meaning (Cathy Legg, Jaroslav Peregrin, Glenda Satne, and Michael Tomasello), and on the logic and phenomenology of shared intentionality and deontic cognition (Margaret Gilbert, Elisabeth Pacherie, Raimo Tuomela, and Dan Zahavi come to mind). I don’t mean these to be exhaustive or representative lists (this is a blog post, after all). But even a passing familiarity with this work establishes that scientists and philosophers are productively interacting with one another in these fields, many of them employing more-or-less overtly model-theoretic representational semantic paradigms, and none of them look anything like a degenerate research program.

II

Regarding the idea that there’s no unifying research project animating our work today, I’ve always found this suggestion puzzling. The discipline we call “philosophy” traces its roots to a tradition that grew out of interest in (among other things) beauty, truth, and goodness. While not a definition of philosophy, and though the people working in analytic philosophy today may not give much thought to these transcendental ideas together, they remain the focus of much of what we do. Furthermore, this has taken place within an intellectual lineage – beginning with direct tutelage in the relations between Socrates, Plato, and Aristotle – that successive generations have participated in since roughly the coining of the term “philosophy”.

That’s nearly two and half thousand years of more-or-less continuous effort on the part of a historical tradition of people who called their study “philosophy”, and who generally engage with the work of earlier and contemporary figures in this tradition. I don’t see what more we need as a unification than the fact that we trace our work to, and continue to engage with, a millennia-spanning tradition – using the very term “philosophy” – that addresses foundational questions of, inter alia, beauty, truth, and goodness. And notice that both the thematic and the historical characterization permit expanding the canon of philosophy so as to include work that developed outside this historical network (my thanks to Kevin Harrelson and Bharath Vallabha for conversation over this issue).

This reading has the virtue of making sense of the pursuit of philosophy as the study of what Peirce called the “normative sciences” of aesthetics, logic, and ethics. And just as the normative sciences map onto these transcendental ideas, so do they map onto the moments of the reflex arc. This opens up into a view on which one task for philosophy today is to help construct a set of categories through which to understand the natural evolution of more complex sensory, central, and motor neural structures as a process that is continuous with the socio-historical development of the ideas of beauty, truth, and goodness; and where the study of philosophy is the self-conscious motor for the socio-historical side of that natural-cum-spiritual process. In this regard, the birth of philosophy in ancient Greece is a birth into self-consciousness of who we are as rational animals.

We don’t have those categories, but it seems clear that we’ve been implicitly constructing them at least since Kant thought to wonder what we may hope, what we can know, and what we ought do (which, as Kant notes in his logic, raises a fourth question: what is a person?). Again, though I move in different circles from Liam, my sense is that many analytically trained philosophers working today understand themselves to be taking part in this project, or something close enough to it to warrant the comparison. And it is perhaps noteworthy that a model theory using possible worlds for interpreting descriptive sentences having a word-world intentionality, and plans of action for interpreting intentional and prescriptive sentences having a world-word intentionality, offers the means of interpreting dimensions of meaning that correspond to the sensory and the motor moments of the reflex arc. With a proof theory using rules of inference as a basis for interpreting word-word relations as constituting the correlate contents of acts of central neural processing, this reinforces the impression that model theory and proof theory need to be understood as two complementary notions of meaning.

At any rate, the claim that “the game the original leading lights thought they were playing has long ago been ceded and no one dares think they are going to do better” doesn’t track the line of intellectual development that runs through the analytic philosophers I’m most familiar with. Perhaps the lesson is that conventional notions about who is (or should be) a “leading light” need to be revaluated.

None of this is to take away from the fact that Liam has no doubt articulated a view that many share in analytic philosophy today. Nor is it to minimize his remarks about the sociological problems the discipline faces in terms of recruitment, job security, job openings, or about the increasing turn to try to make analytic philosophy relevant in some applied capacity. And this remark from Dave Atenasio’s response to Liam’s essay cuts close to the bone:

“Eventually analytic philosophy worked itself pure: a cohort of charismatic but somewhat abrasive professors debated arcane topics with one another, isolated from most of the greater academic community, their leisure and keynote travel supported by an ever expanding cast of adjuncts, visiting professors and graduate assistants whose career prospects were vanishing before their eyes.”

But my sense is that the job market problems aren’t so much a fault of analytic philosophy as part of a crisis in the humanities and a general shakeup in higher education in the developed world, while the shift to more public-facing and interdisciplinary work is in many ways a good thing for the profession. In terms of the influence that American philosophy has historically had on American culture, for instance, it’s a shame (at least in the American context) that there’s really nothing like a “John Dewey track” through the discipline today. Given Katzav and Vaesen’s research into the journal capture by analytic philosophers that took place in the middle of the 20^th century, this state of affairs is perhaps unsurprising. It is nevertheless a loss for the Republic, I think.

I’ll close by bringing this back around to Quine’s animus against modal logic. If Katzav and Vaesen are correct about the displacement of classically pragmatist inflected American philosophy (among others) at venues like The Journal of Philosophy, Mind, and The Philosophical Review in the middle of the 20^th century, in favor of the ascendant analytical school of philosophy, then it is perhaps an irony of our situation that Quine’s reception on this front would take the shape that it did. For that reception brings to mind John Dewey’s assessment of intellectual progress at the end of “The Influence of Darwin on Philosophy”, first published in Popular Science Monthly in 1909:

“Old ideas give way slowly; for they are more than abstract logical forms and categories. They are habits, predispositions, deeply engrained attitudes of aversion and preference. Moreover, the conviction persists – though history shows it to be a hallucination – that all the questions that the human mind has asked are questions that can be answered in terms of the alternatives that the questions themselves present. But in fact intellectual progress usually occurs through sheer abandonment of questions together with both of the alternatives they assume, an abandonment that results from their decreasing vitality and a change of urgent interest. We do not solve them: we get over them.”