Nagel and his critics, Part VI

We’ve been looking at the critics of Thomas Nagel’s recent book Mind and Cosmos.  Having examined the objections raised by Brian Leiter and Michael Weisberg, Elliott Sober, Alva Noë, and John Dupré, I want to turn now to some interesting remarks made by Eric Schliesser in a series of posts on Nagel over at the New APPS blog.  Schliesser’s comments concern, first, the way the scientific revolution is portrayed by Nagel’s critics, and second, the role the Principle of Sufficient Reason plays in Nagel’s book.  Most recently, in response to my own series of posts, Schliesser has also commented on the status of naturalism in contemporary philosophy.  Let’s look at each of these sets of remarks in turn.

Aristotle and the scientific revolution

Schliesser notes several respects in which the scientific revolution was more complicated than Nagel’s critics seem to imply.  In particular, it was not an immediate and complete transition to what Bernard Williams has called the “Absolute Conception of Reality” -- a world of colorless, odorless, teleology-free material elements whose behavior can be exhaustively described in terms of mathematical laws -- and it was not a transition from a completely failed Aristotelian science that was a mere systematization of common sense devoid of predictive successes.  In fact, Schliesser says, Aristotelian science, whatever its failings, was “a majestic achievement” and in fact “it is an open question if teleology has ever been fully eliminated from physics (and if so, when).”  (I hasten to add that while these quotes might make Schliesser sound like… well, like me… he is certainly not putting forward the sort of neo-Aristotelian-Scholastic position I would defend.  He just wants to make it clear that the conceptual territory is more complicated than the standard Heroic Age of Science narrative would have it.  Read the whole post to get a sense of his views.)

Now if I understand Schliesser correctly, he thinks that Leiter, Weisberg, and Noë more or less identify the “Absolute Conception of Reality” with science itself.  They differ in that Leiter and Weisberg essentially think that the Absolute Conception (and thus science) probably captures more or less all of reality (or at least that Nagel hasn’t shown otherwise); whereas Noë thinks that the Absolute Conception (and thus science) is inadequate as its stands and needs to be supplemented.  Schliesser, by contrast, thinks it is a mistake to identify science with the Absolute Conception in the first place.  The Absolute Conception is rather something that has gotten tacked on to modern science but could be separated from it, whatever we might want to replace it with.  (Again, if I understand Schliesser’s position correctly -- I welcome correction if I have misinterpreted him.)

My own view, of course, is that the so-called Absolute Conception most certainly does not capture all of reality.  The question is whether it is Noë’s characterization of the Absolute Conception’s relation to science, or Schliesser’s, that is correct.  I am not sure that the dispute is more than terminological, and the reason is that it seems to me that terms like “science” and even “physics” are not really as well-defined as is often supposed.  Certainly they are contested terms, as I think the dispute at hand itself indicates.  Schliesser writes, vis-à-vis teleology and physics:

Leibniz promoted least action principles, and variational principles have remained popular through Hertz, Hamilton, into quantum mechanics. This is not the place to decide whether such principles must be interpreted as teleological (and in what sense 'teleological' has been transformed along the way); all I claim is that they have been understood as teleological by prominent philosophers-scientists.

I would imagine that some readers will say “That’s true, so maybe physics could in principle include teleology (whether or not it should),” while others would say “That’s true, but since physics can’t include teleology, what you’re talking about are really scientists who were doing metaphysics (whether or not that metaphysics is correct).”  

Similarly, when a J. Scott Turner, Marjorie Grene, or André Ariew suggests that immanent teleology (i.e. the Aristotelian kind rather than the Paleyan kind) might have a legitimate place in our understanding of developmental processes in biology (even if it is not necessary to understanding adaptation), some readers might say “OK, maybe a science like biology can affirm teleology after all,” whereas others might say “Biology is a science, and thus cannot affirm teleology; so, even if there is reason to attribute teleology to developmental processes, that would be a metaphysical point rather than a scientific point.”

Some Aristotelian-Thomistic thinkers of the early to mid twentieth century argued that whereas physics characterizes nature in an essentially mathematical way, Aristotelian notions like formal and final causality, act and potency, and the like are still essential to an adequate philosophy of nature, which addresses deeper questions about the nature of the material world than physics addresses.  (I have put things this way myself many times, such as here.)  On the other hand, NancyCartwright says that though “the empiricists of the scientific revolution wanted to oust Aristotle entirely from the new learning,” in fact “they did no such thing” and physics itself does indeed reveal the Aristotelian “inner natures” of things.  Similarly, the physicist Stephen Barr has objected to the way some Thomists characterize physics, and insists that physics really does uncover the “intelligible structures” of things (and thus, in a sense, their “forms”).

Who is right?  In all these cases I think it depends on whether you think modern physics and modern science more generally are defined by the early moderns’ move away from immanent, Aristotelian formal and final causes.  If you think that something like immanent formal and final causes are real but allow (at least for the sake of argument) that physics and science are non-teleological, then you will say that physics and science more generally need to be supplemented by an Aristotelian philosophy of nature.  If you think that something like immanent formal and final causes are real and do not think the anti-Aristotelian move of the early moderns has any essential connection with the practice of modern science as such, then you will be happy to allow that formal and final causes are properly part of physics and science more generally.  But it does not seem to me that there is much of a substantive dispute here.  Or at least, both sides essentially agree about what exists in nature itself; they just disagree over which parts of the study of nature belong to the sphere of physics and which to the sphere of philosophy.  (I’ve talked with Steve Barr about this subject at length, and I am convinced that the differences between him and the Thomists he is criticizing over how to interpret physics are largely verbal.)  

Schliesser himself, I realize, is not necessarily committed to all or even any of the Aristotelian apparatus, and does not have a dog in any intra-Aristotelian fight over how modern physics is related to notions like teleology, formal causes, etc.  As far as I can tell he is just making the point that Williams’ “Absolute Conception” should not be confused with science itself, whether or not one wants to reject the Absolute Conception, and (if one does reject it) whether one wants to replace it with some kind of neo-Aristotelian position or with something else entirely.  

But I think something like the point I’ve been making applies to the relationship between Schliesser’s position and Noë’s, and Nagel’s position and mine too for that matter.  All of us essentially agree that the Absolute Conception is not something that science has shown to be a complete account of reality.  Where we differ is (perhaps) in our degree of confidence in the judgment that the Absolute Conception is incomplete, in our views about whether the Absolute Conception should be identified with science itself, and in our views about what the Absolute Conception should be supplemented with.  Those are significant disagreements, but the point of agreement is not insignificant.

(Chomsky once remarked that “as soon as we come to understand anything, we call it ‘physical.’”   Something similar might be said about the word “science.”  Take Aristotelian-Scholastic philosophy of nature, metaphysics, and natural theology.  These have been worked out in systematic detail -- the Neo-Scholastic manualist tradition essentially presented them the way modern physics, chemistry, or biology would be presented in a textbook, as a worked out body of thought -- and we old-fashioned Thomist types regard them as about as secure and clearly right at least in outline as anything else we know.  Hence we are happy to regard them as “sciences,” in the Aristotelian sense.  Naturalists, of course, would reject all of that.  But many of them are so confident of the truth of an essentially materialist metaphysics and a broadly empiricist epistemology that they more or less regard those positions as “scientific.”)

Before moving on I want briefly to remark on something Schliesser says about Aristotle.  He writes:

I am no Aristotle scholar. But I read Aristotle as a relentless naturalist with an unmoved mover uninterested in the goings on in the world. (I am open to correction from more learned readers.) His final causes are rooted in the worldly natures of worldly things. Such natures and material stuff are constraints on any teleological explanation; I bet that even the most hard-nosed Darwinian would be impressed by the extraordinary detail and empirical richness of Aristole's biological writings.

End quote.  A commenter took Schliesser to task for these remarks, which led to an exchange in his combox (in which my work was cited).  I certainly wouldn’t call Aristotle a “naturalist,” though in fairness to Schliesser it should be noted both that the term “naturalism” is itself somewhat elastic, and that twentieth-century Catholic critics of Neo-Scholasticism such as Henri de Lubac liked to suggest that Aristotelianism tended toward “naturalism.”  

But Aristotle was clearly not a “naturalist” in the contemporary sense.  The most obvious reasons are that he affirmed the existence of an Unmoved Mover who is pure actuality (and thus immaterial), and affirmed also the immateriality of the intellect (even if the precise interpretation of Aristotle’s views on this subject has of course been controversial for centuries).  These were by no means incidental features of his philosophy, any more than atheism and materialism are incidental features of contemporary naturalism.  Moreover, while Nagel and others might want to marry something like Aristotle’s immanent teleology to naturalism, that will be a shotgun wedding if it happens at all.  For however we want to define “science,” naturalism, at least as that has been understood in contemporary academic philosophy since Quine, is surely wedded to the “Absolute Conception” or something like it, and the Absolute Conception is inherently anti-Aristotelian.  (That, I might note, is why I alluded to Cartwright in an earlier post as someone whose work -- neo-Aristotelian as it is -- poses a challenge to naturalism.  New APPS contributor Mark Lance took issue with this, but I was of course not claiming that Cartwright is in general agreement with more self-consciously anti-naturalist philosophers like analytical Thomists.)

Schliesser and his commenter also discuss some other issues, such as whether God, as Aristotle understands him, is an efficient cause as well as a final cause, and whether the existence of immanent final causes must ultimately be explained in terms of God’s directing natural substances toward their ends.  Aquinas and other Scholastic thinkers would of course answer in the affirmative in both cases -- quite rightly, in my view -- but I agree that whether Aristotle himself would have gone in these directions is controversial, to say the least.

PSR and teleology

In a later post, Schliesser discusses the role the Principle of Sufficient Reason (PSR) plays in Nagel’s book.  He begins as follows:

Analytical philosophy has made great progress over the last century. But its original, necessary biases did some harm, too. In particular, detailed working knowledge of the history of philosophy and metaphysics was banished for several generations. While metaphysics is thriving again, we still lack (despite the brilliance of David Lewis' modular approach) complete systems of thought that can rival in depth and interlocking breadth the past masters (say, Suarez, Leibniz, etc.). The damage has also been more narrow. For example, one of the most obvious so-called ‘Kuhn Losses’ is our relative ignorance of the nature and implications of the Principle of Sufficient Reason (PSR). This is no surprise because analytical philosophy was founded in the act of rejecting PSR. Our forefathers’ attempt to balance between common sense and the truths of science meant -- as science and the PSR parted ways -- the willing submission to brute, ultimate facts…

End quote.  I essentially agree with this, and warmly (though I imagine Schliesser and I would disagree about the nature and extent of the “progress” analytic philosophy has purportedly made, and about how “necessary” its original biases were).  Schliesser also says that “unlike some of my Darwinian friends I think we should keep philosophy a welcoming place to folk that want to explore possible future sciences in opposition to present science.”  

However, Schliesser goes on to criticize Nagel for going beyond what PSR would allow in appealing to teleology.  This is not because of any hostility to teleology as such on Schliesser’s part, but rather because he thinks Nagel simply does not do enough to show how the specific sort of teleological principles he considers meet the demands of “any PSR worth having.”  Because Nagel’s views are so inchoate, Schliesser certainly can make that case; although, precisely because Nagel’s views are so inchoate, there may be a way to reconstruct them so that Schliesser’s criticisms are rebutted.

I’m not inclined to try to defend Nagel on this score, however, because I do not think PSR -- or more precisely, PSR as it is understood in rationalist philosophy -- has the importance Nagel, Schliesser, and many others (whether pro-PSR or anti-PSR) think it does.  Here we come to one of the many crucial differences between classical (and especially Aristotelian-Scholastic) philosophy on the one hand, and the varieties of modern philosophy on the other, on which I so often harp.  (What follows is in the way of a digression of sorts, but one that is relevant to the issues Schliesser raises.)

For the Aristotelian-Scholastic philosopher, the work writers like Leibniz or Spinoza would do with PSR is properly done instead by what is sometimes called the principle of causality.  That principle can be formulated in different ways, but the most fundamental formulation states that no potency can actualize itself, but must be actualized by something already actual.

Notice a few things about the relationship between this principle and PSR.  First, the principle of causality is grounded in the theory of act and potency, which is the foundation of Aristotelian philosophy of nature (and indeed of the entire Aristotelian-Scholastic system), first developed in response to Parmenides’ and Zeno’s denial of the reality of change and multiplicity.  The Aristotelian argues that change and multiplicity can be real features of the world only if potency -- that which is neither fully actual nor sheer non-being or nothingness, but an irreducible middle ground -- is also a real feature of the world.  By contrast, PSR, as it is understood in the modern rationalist tradition, was developed precisely in an intellectual context which had essentially jettisoned the theory of act and potency.  (To be sure, the notions of act and potency are not entirely absent in the early moderns, but they have nothing like the centrality they have in Scholasticism, they are not considered essential to understanding the natural world, and they would soon disappear almost entirely within “mainstream” Western philosophy -- or at least they did until contemporary analytic metaphysicians started to take seriously the idea that there is a distinction to be drawn between “categorical” and “dispositional” properties.)  PSR lacks the wider theoretical context of the Scholastic principle of causality, standing alone as a supposed “law of thought.”

This brings us to a second point, which is that whereas the principle of causality is a principle of metaphysics and philosophy of nature, PSR is essentially a logical-cum-epistemological principle.  The principle of causality purports to tell us something about the world itself.  PSR purports to tell us something about how we have to understand the world, or think about it, or make it intelligible to us.  That opens it to Humean, Kantian, and naturalist objections to the effect that there is no reason to think that objective reality has to correspond to our demands for intelligibility.  Like the frantic Steve Rogers (a.k.a. Captain America) in the comic book panel above, we crave explanations.  But perhaps (so this sort of objection goes, and to shift pop culture references) we are in the epistemic position of Inigo Montoya, and had better get used to disappointment.

Which brings us to the third point, which is that if the Scholastic position is correct, then (a version of) PSR is itself grounded in the principle of causality and the theory of act and potency more generally.  It is because any potency can be actualized only by what is already actual that the actualization of any potency -- and thus the coming into existence of a thing or the occurrence of any event -- will have an explanation.  It is because God is pure actuality and thus without any potency that could be actualized in the first place that he does not have (indeed could not have had) a cause, and is “self-explanatory” in a way nothing else could be.  Etc.  The theory of causality grounds the theory of explanation, rather than the other way around.  While Humean, Kantian, and naturalist objections to PSR may have force against rationalist versions of the principle -- which, again, present it as if it were just an a priori principle or something that merely reflects the structure of cognition -- they misfire against the Scholastic approach to PSR, which grounds it in something objective and metaphysically deeper.

I would say that this approach is invisible to most modern philosophers because they are used to looking at the issue in terms of the range of options that have come down to us from the rationalists, empiricists, and Kant -- a range of options the Scholastic regards as too limited.  For the Aristotelian-Scholastic tradition, our concepts are all grounded in experience but nevertheless outstrip experience.  I form the concept of a triangle or a tree only because I have experienced these things -- the concepts are not innate or a priori -- but the concept triangularity or treeness nevertheless cannot in principle be identified with a sensation, mental image, computational symbol structure, neural structure or any other “inner representation” of the sort the empiricist or naturalist would identify it with.  (See my article “Kripke, Ross, and the Immaterial Aspects of Thought,” forthcoming in American Catholic Philosophical Quarterly, for a detailed defense of the claim that a concept “cannot in principle” be identified with any such thing.  The reasons have to do with the fact that concepts and the thoughts they enter into inherently have a universal reference that no sensation, mental image, computational symbol structure, neural structure, or the like can have, and a determinacy of content that no sensation, mental image, computational symbol structure, neural structure, or the like can have.  I’ve addressed this issue before, but the new paper sets the argument out at greater length than I have previously and responds to every objection I know of.)  

Now what we grasp in grasping a concept is, for the Aristotelian, the form, nature, or essence of the thing the concept is a concept of.  The same form, nature, or essence that makes the tree a tree makes my cognition of a tree a cognition of a tree; it’s just that in the latter case the form exists “intentionally” rather than in mind-independent reality.  It is abstracted by the mind from the concrete, mind-independent tree and considered by itself.  And when I grasp necessary truths about a tree, the necessity I find in thought (conceptual necessity) is merely an echo or reflection of the necessity that already exists in a mind-independent way in the tree itself given its essence.  But I know the tree only via experience, from which experience I’ve abstracted the form or essence.  In that sense, knowledge of the essences of things and thus of necessary truths about them is grounded in experience rather than a priori.  It isn’t mere knowledge of concepts, whether innate or intuited as residing in some Platonic third realm.  It is knowledge of the mind-independent and experienced things themselves, through the abstracted concepts.  

Obviously all this raises lots of questions, but the point is to indicate how radically different the Aristotelian-Scholastic approach to these issues is from the usual modern approaches.  The early moderns who rejected the Aristotelian account of concept formation replaced it with either a rationalist or empiricist account, both of which essentially take half of the Aristotelian position while lopping off the other half.  The rationalist agrees with the Aristotelian that the necessity of conceptual truths reflects the necessity of mind-independent reality, but denies that these truths are in any way grounded in experience.  They thus come to seem to the critic of rationalism to float free of any objective epistemological foundation.  The empiricist agrees with the Aristotelian that these truths are grounded in experience (at least insofar as every idea must be traceable to some sensation), but denies that they reflect any mind-independent necessity.  The necessity instead reflects nothing more than our habits of thought, or linguistic practices, or our cultural or biological inheritance, or something equally anticlimactic.  Kant, rather than rejecting the rationalist and empiricist extremes and returning to the Aristotelian middle ground position -- mind-independent necessity, knowledge of which is grounded in experience -- for all practical purposes embraced the anti-Aristotelian aspects of both rationalism and empiricism and chucked out the remaining Aristotelian bits of each: Conceptual truths reflect no mind-independent necessity (at least as far as we can possibly know) and are not grounded in experience but are a priori.  

Against this rationalist/empiricist/Kantian background, it is only natural that if something isn’t empirical science, it will seem that the only thing left for it to be is “conceptual analysis” -- and “therefore” to be of dubious objectivity (reflecting mere biologically-based but still contingent predilections at best, and mere prejudice at worst).  And it is no surprise that when PSR is dismissed as dubious conceptual analysis, the “brute, ultimate facts” of empirical science (as Schliesser refers to them) will come to seem the only possible terminus of explanation.  

This is also one reason contemporary philosophers often so badly misunderstand the arguments of Thomists and others whose thinking is grounded in the classical and medieval traditions.  Such philosophers will assume, for example, that Aquinas’s proofs of God’s existence must either be dubious “god of the gaps” style empirical hypotheses or exercises in rationalist metaphysics.  They tend therefore to assimilate Aquinas’s Fifth Way to Paley’s “design argument” and the Third Way to Leibniz’s cosmological argument.  As I have argued elsewhere, these are egregious misreadings, and they are only the tip of the iceberg.

For the Aristotelian-Scholastic philosopher, then, modern metaphysics, epistemology, philosophy of religion, etc. are in the same shape that Alasdair MacIntyre famously argued modern ethics is in.  In the former cases no less than the latter, crucial philosophical notions have been ripped from the intellectual contexts that gave them their intelligibility and have become distorted as a result, and the range of theoretical options visible to the modern philosopher has shrunk drastically.  Nagel’s proposals are bound to seem odd and ill-motivated, not only because they are inchoate, but because fully to work out their implications would require a far more extensive rethinking of current orthodoxy than Nagel himself probably realizes.  (That is no doubt one reason why his ideas are inchoate.)  Questions about PSR, teleology, etc. cannot properly be understood if they are treated as mere add-ons to a basically naturalistic-cum-scientistic picture of knowledge and reality, which leave that picture essentially intact.  The picture as a whole needs to be rethought if any part of it is seriously to be rethought -- just as an understanding of Aristotelian-Scholastic arguments in philosophy of religion, ethics, etc. require an understanding of general Aristotelian-Scholastic metaphysics, philosophy of nature, and epistemology.

Naturalism in contemporary philosophy

This naturally brings us, finally, to Schliesser’s remarks on the status of naturalism in contemporary philosophy.  Schliesser quotes some of my own comments on the subject, wherein I note some anti-naturalistic trends in recent philosophy, and writes:

Now it possible that we live in a philosophical age of ferment…

Of course, an alternative possibility is that (analytical) philosophy has relatively low barriers to entry for people that wish to challenge any naturalistic consensus (we merely need to show some facility with various baby logics, know how to string arguments together, et voila). And we shouldn't ignore the fact that from Wittgenstein (and his students) onward, analytical philosophy has been welcoming to all kinds of anti-naturalists (so that Feser may just be noting that we are less naturalistic than talk of "consensus" warrants). For there is nothing in our collective (disunified) methods or teachings that would enforce/stabilize a "naturalistic consensus." 

End quote.  I don’t think these two possibilities are incompatible.  We could be living in a time of ferment precisely because analytic philosophy is of its nature open to anti-naturalist ideas.  And one could go back further than Wittgenstein to find examples -- for instance to Frege’s, or even Russell’s, Platonism.  Even logical positivism and Ayer’s phenomenalism are closer to idealism than to the realist materialism characteristic of contemporary naturalism.  The current dominance of materialist naturalism -- and “dominance” is a better word than “consensus” -- may turn out to be as much of an anomaly in the history of analytic philosophy as it is in the history of philosophy more generally.  

Either way, as Prof. Schliesser not only says, but demonstrates by his own fine example, “philosophy is not party-politics.”  I thank him for his comments.