The particle collection that fancied itself a physicist

I haven’t done a “Physicists say the darndest things” post in a while.  People usually ask me to write one up every time a Lawrence Krauss, Sean Carroll, or Stephen Hawking (well, lately not Hawking) publishes a new “gee whiz” pop philosophy book masquerading as a pop science book.  I find the genre extremely boring.  It’s always the same dreary, sophomoric PBS-level stuff: We’re all just heaps of particles, but golly this really increases rather than decreases the wonder of it all, and here’s some half-baked amateur metaphysics and life lessons that even hardcore materialist philosophers would regard as fallacious and banal.  The only variable is whether the crap philosophy in these $30 time wasters is coupled with clueless arrogance (cough, Krauss) or at least presented with some humility. 

In Brian Greene’s case we have someone who seems a pleasant enough fellow.  But his new book Until the End of Time nevertheless exhibits the usual foibles of the genre.  I’ll focus here on what he says about the place of the human mind in the physical universe (the topic of chapter 5).  The basic metaphysical assumption is a crude reductionism: All that really exists, we are assured, are basic particles governed by mathematical laws.  Hence consciousness, free will, etc. must somehow either be reduced without remainder to these, or eliminated from our picture of reality.  The problem Greene wants to solve in the chapter is to explain how this program can most plausibly be carried out.

Physics ain’t all that

There are three main difficulties with Greene’s solution to the problem.  First, the solution is a non-starter, because second, he doesn’t understand the problem in the first place.  But third, it doesn’t matter, because the reductionistic assumption that creates the problem isn’t true anyway.

Let’s start with that last point.  Greene insists that the “evidence” supports his basic reductionist assumption.  In fact, the “evidence” does no such thing, and the assumption is false.  Greene himself inadvertently hints at the reason why.  For one thing, he writes:

The art of science, of which Newton was the master, lies in making judicious simplifications that render problems tractable while retaining enough of their essence to ensure that the conclusions drawn are relevant.  The challenge is that simplifications effective for one class of problems can be less so for others.  Model the planets as solid balls and you can work out their trajectories with ease and precision.  Model your head as a solid ball and the insights into the nature of mind will be less enlightening.  But to jettison unproductive approximations and lay bare the inner workings of a system containing as many particles as the brain – a laudable goal – would require mastering a level of complexity fantastically beyond the reach of today's most sophisticated mathematical and computational methods. (p. 117)

Very true.  The problem, though, is that Greene seems to think that if we did have a complete mathematical description of the particles that make up a brain, then we would have captured all there is to the brain.  What he fails to see is that such a description would itself be just as much a simplifying abstraction as modeling your head as a solid ball would be.  True, it would be a model that captures much more of the mathematical structure of the brain, but the point is that it would still be capturing only mathematical structure and nothing else. 

But is there anything more to matter than its mathematical structure?  Of course there is, because there is no such thing as mathematical structure without some concrete reality that has the structure.  Mathematical structure by itself is a mere abstraction from concrete reality.  A description of the brain in terms of nothing more than particles governed by mathematical laws, no matter how complex this description, can no more give you a complete description of the brain than spherical geometry can give you a complete description of a planet or a basketball.  And Greene himself inadvertently admits this too.  He writes:

I don’t know what mass is.  I don't know what electric charge is.  What I do know is that mass produces and responds to a gravitational force, and electric charge produces and responds to an electromagnetic force.  So while I can’t tell you what these features of particles are, I can tell you what these features do… For gravitational and electromagnetic influences, any concern that substituting action and response for an intrinsic definition amounts to an intellectual sleight of hand is, for most researchers, alleviated by the spectacularly accurate predictions we can extract from our mathematical theories of these two forces. (p. 133)

What Greene is acknowledging here is that the methods of physics don’t capture the intrinsic nature of phenomena, but only those relations between phenomena susceptible of mathematically precise description.  Hence physics simply doesn’t tell us everything there is to know even about the material world (let alone anything beyond the material world).  As I have noted many times, this is a point that used to be often commented upon by scientists and philosophers (Poincaré, Duhem, Russell, Eddington, et al.) and has in recent years been getting renewed attention in academic philosophy. 

What Greene doesn’t see is that the point completely undermines his basic reductionist assumption.  Why should we assume that what is real must be reducible to physics’ mathematical description of basic particles, if we already know that that description doesn’t capture every aspect of reality in the first place?

Greene also acknowledges that we need what he calls “nested stories” about different “layers of reality” – not just the story about what is going on at the level of fundamental particles governed by mathematical laws, but also “higher-level accounts” couched in language about learning, creativity, thinking, deliberating, and other concepts that have no applicability at the level of laws and particles (pp. 154-55).  Readers of Daniel Dennett will recognize in this a warmed over variation on his theory of the different “stances” we might take toward phenomena (namely the physical stance, the design stance, and the intentional stance).

All the same, Greene insists on giving the lower-level story about particles and laws a privileged status, and treating any part of a higher-level story that cannot be reformulated in lower-level terms as merely a useful fiction.  But why should we agree with that?  For one thing, the thesis that the higher-level stories are just convenient fictions faces the same problem that all versions of anti-realism do, namely that it is hard to see how these “stories” could work so well, and indeed be as practically indispensable as they are, if they weren’t true.  (This is known as a “no miracles” argument in the philosophy of science literature.)

For another thing, why should we not turn the tables and hold instead that it is the higher-level story that tells us the truth about the world, whereas the lower-level story is merely a simplifying abstraction that is useful for certain purposes (such as technological ones) but that leaves out much of concrete reality and thus is not strictly true?  This is essentially the view taken by anti-reductionist philosophers of science like Nancy Cartwright.

Greene claims that the “evidence” provided by the successful predictions made using the laws of physics supports his reductionist position, but it does nothing of the kind.  After all, as Greene himself happily acknowledges, there are no laws that allow us rigorously to predict the behavior of systems conceived of as dogs, cats, basketballs, dollar bills, human beings, etc.  We have to abstract out all that is distinctive of these things qua biological, cultural, economic, etc. phenomena and describe them instead in the simplifying terms of physics, and then we will get rigorous predictions (though only of those aspects of their behavior that are reflected in the simplifying description). 

So, all that Greene is entitled to say is that mathematically precise laws accurately describe the behavior of systems modeled at a high enough level of abstraction to be characterizable in terms of mathematically precise laws.  Which, though not entirely unimpressive (since it could have turned out that the laws failed no matter how abstractly we modeled the phenomena) is still not nearly as impressive as Greene needs it to be.  In particular, it hardly shows that there is no more to physical reality than is captured by the laws and abstract models.

Greene’s fallacy is like that of someone who says that, since a map is enormously useful for getting around a certain bit of terrain, predicting what you’ll see when you reach this or that part of it, etc., it follows that there is nothing more to the terrain that what is captured by the map.  As Alfred Korzybski once said, “the map is not the territory.”  If only more physicists were capable of seeing what a crackpot linguist could!

Anyway, whether you agree with me or with Greene, here’s the thing: The dispute is not a scientific one but a philosophical one.  As I have argued at length in Aristotle’s Revenge and elsewhere, the compelling arguments are all on the anti-reductionist side.  But even if we anti-reductionists were wrong about that, Greene has said nothing to show that we are.  Greene thinks he has solidly established a metaphysical result by drawing it out of physics, but all he has actually done is to read a dubious and unsupported metaphysical claim into physics. 

Physics cannot solve the problem, because it creates the problem

Let’s move on to the second difficulty, which is that Greene does not understand the problem he is trying to solve.  To be fair, he does at least see that there is a problem facing anyone who wants to insist on the kind of reductionism he favors while also affirming the reality of conscious experience.  He appeals to arguments like Thomas Nagel’s “bat argument” and Frank Jackson’s “knowledge argument” to illustrate the problem, and he realizes that they can’t be waved away after the fashion of the village Reddit materialist. 

All the same, he fails to see the depth of the problem, and in particular fails to see that the methods of physics are precisely what generate the problem in the first place, so that it is clueless to think (as Greene does) that the problem can be resolved by further application of those methods.  Moreover, some of the writers Greene cites make this point themselves.

Here’s the basic idea.  The founders of modern physics put at the center of the scientific conception of the world the idea that matter should be characterized in terms of quantifiable primary qualities alone – size, shape, motion, position, etc.  Irreducibly qualitative features like color, sound, heat, cold, and the like were to be treated as secondary qualities, reflecting the way we experience the world, but not how the world really is in itself.  To be more precise, for purposes of physics, colors were to be redefined in terms of surface reflectance properties, sounds in terms of compression waves, temperature in terms of molecular motion, etc.  Hence, if by “red” we mean such-and-such a reflectance property, then we can say that a certain apple is red; but if instead we mean by “red” the way red looks to everyday experience, then that exists only in the conscious observer, and not in the apple itself.  If by “heat” we mean such-and-such a pattern of molecular motion, then we can say that the water in a certain bathtub is hot; but if instead we mean by “heat” the way heat feels in everyday experience, then that too exists only in the consciousness of the observer.  And so on for other secondary qualities (as Greene himself recounts at p. 139).

Though the details of the story have changed over the centuries, what has persisted to the present day is a tendency to treat so-called secondary qualities as merely the qualia of our conscious experience of the material world, rather than anything to be found in the material world itself.  They are simply not the kind of thing that can be captured by the purely quantitative, mathematical language to which physics confines its description of matter.  And the problem is that this conception of matter entails a kind of dualism.  For if these qualities do not exist in the material world, then they must not exist in the brain, which is part of the material world.  Yet if they do exist in the mind, then the mind must not be identical with the brain or with any other part of the material world.

Like so many other superficial materialists, Greene thinks the problem merely has to do with its being intuitively difficult to see how conscious experiences could be material.  No, the problem is much deeper than that – it is that modern physics essentially defines the physical world in a way that entails that consciousness is non-physical.   The problem has less to do with consciousness than with matter as physics conceives of it.

Descartes and his followers saw this implication, and that (rather than intuition, religious prejudice, or some other anticlimactic rationale) is why they judged consciousness to be immaterial.  Indeed, the basic problem was recognized by the ancient atomist Democritus, who is, ironically, quoted by Greene himself.  In particular, Greene cites the Intellect’s side in an exchange Democritus imagined the Intellect having with the Senses:

Intellect: Color is by convention, sweet by convention, bitter by convention; in truth there are but atoms and the void.

What Greene does not quote is the retort that Democritus put into the mouth of the Senses:

Senses: Wretched mind, from us you are taking the evidence by which you would overthrow us?  Your victory is your own fall.

Democritus’s point is that if the atomist says both that atoms are all that exist and that color, sweetness, etc. and the other qualities of conscious experience are not to be found in the atoms, then we have a paradox.  For conscious experience is what provides the empirical evidence on which the atomist account is itself based!  The atomist thus seems unable to fit the very evidence his theory relies on into the picture of the world the theory describes.  Democritus was intellectually honest enough to take note of this problem, though we don’t know how he tried to resolve it, if he did.

Erwin Schrödinger noted that the same problem afflicts modern physics, which takes for granted a conception of matter that is in the relevant respect like that of the ancient atomists (though of course in other respects it is very different).  And Nagel’s argument in “What Is It Like to Be a Bat?” makes the same point.  The way Nagel puts it is that since physics works with a conception of matter as essentially objective (in the sense of being independent of any particular observer’s point of view), it cannot incorporate into its picture of reality the subjectivity of conscious experience (which is precisely tied to the point of view of the observer).

For this reason, Nagel and other contemporary philosophers of mind like David Chalmers have argued that consciousness cannot be explained in physical terms unless physics revises its conception of matter.  Greene considers Chalmers’ version of this idea, but replies that there is no “convincing evidence” for such a thesis (p. 135).  But the reason he doesn’t see the evidence is, as Orwell would say, because it’s right in front of his nose.  It is there in physics’ own conception of matter, which excludes consciousness from the material world precisely by allowing into that world only what can be described in the language of mathematics.

Nobody in here but us particles

Greene is keen on saying that we are all just “collections of particles,” and goes on at length about how he is himself just a collection of particles (pp. 156-57).  That we seem to be more than that is, he suggests, just an illusion.  Here again he borrows from Dennett, by way of neuroscientist Michael Graziano.  Just as, according to the primary/secondary quality story, we project onto external reality properties that aren’t really there (such as the red of a Ferrari, in Greene’s example), so too do we project onto the internal world of the brain a stream of conscious thoughts and experiences that aren’t really there.  Greene writes:

You continuously create a schematic mental representation of your own state of mind.  If you are looking at the red Ferrari, not only do you create a schematic representation of the car, you also create a schematic representation of your Ferrari-focused attention.  All the features you bind together to represent the Ferrari are augmented by an additional quality summarizing your own mental focus…

[This] is the heart of why conscious experience seems to float unmoored in the mind.  When the brain's penchant for simplified schematic representations is applied to itself, to its own attention, the resulting description ignores the very physical processes responsible for that attention.  That is why thoughts and sensations seem ethereal, as if they come from nowhere, as if they hover in our heads. (pp. 140-1)

Now, I certainly understand the attractions of this “higher-order representation” sort of view.  I once defended a version of it myself, in my doctoral dissertation during my naturalist days.  But it’s hopeless.  Here are some of the problems with it.

First, it just keeps kicking the problem back a stage, ad infinitum.  Again, the view starts with the primary/secondary quality thesis that redness, heat, etc. as common sense understands such qualities don’t exist in the external material world but only in our representations of it, as the qualia of conscious experience.  This opens up the problem that if these qualities don’t exist in external material things like the Ferrari, then they can’t exist in the brain either, since it too is material.  Greene’s answer to this problem is to say: “OK, then not only do they not exist in the Ferrari, but they don’t exist in the brain’s representation of the Ferrari; instead they exist in the brain’s higher-order representation of the brain’s representation of the Ferrari!”  And of course, that just moves the pea under another shell, because higher-order representations in the brain are just as material as first-order representations in the brain.  And if Greene deals with this problem by appealing to some third-order representation, then the problem will just pop up again there, like the proverbial whack-a-mole (if I can introduce yet another analogy).

Second, it’s actually worse than that, because the notion of “representation” is itself a mental notion which, like consciousness, cannot be assimilated to the conception of matter physics has inherited from the early moderns.  On that conception, matter is just colorless, odorless, soundless, tasteless, and meaningless particles in motion.  Matter on its own does not stand for, point to, refer to, or represent anything; it lacks intentionality or “directedness toward” any object beyond itself (to put the point using the standard technical philosophical jargon).  The early moderns’ conception of matter took intentionality no less than qualia to exist only in the mind’s awareness of the physical world and not in the physical world itself – which, again, is why they took the modern conception of matter to entail the immateriality of the mind.

So, if Greene is beholden to this conception of matter, he owes us an explanation of how intentionality or representational content can arise in a material world thus understood.  But his “explanation” essentially amounts to saying: “Representation arises in a world that doesn’t already include it once the brain starts representing itself as the sort of thing that has representations in it.”  That’s like Feynman’s example of the painter who claimed he could make yellow paint by mixing only red and white paint, as long as he also added some yellow paint to the mix somewhere along the way.  The painter can make yellow paint where it doesn’t yet exist – but only if there’s already some yellow paint lying around.  And the brain on Greene’s account can make mental representations where they don’t already exist – but only if there are some mental representations lying around.  If you don’t see the fallacy here, then you might be qualified to write a pop science bestseller.

Third, Greene’s position entails a self-defeating skepticism.  Not only do we have no genuine access to the external world – but only to our inner representations of it – it turns out we don’t really have access even to those inner representations of the external world either, but only to representations of them.  And in fact (if we follow this out consistently), we don’t have access even to those, but only to yet higher-order representations ad infinitum.  So how do we know that anything is real, including Greene’s own account of what is really going on? 

Even professional philosophers like Dennett who peddle these sorts of views are unable to solve the problems facing them.  Poor Greene isn’t even aware that the problems exist.  And yet, though less obnoxious than a blowhard like Krauss, he is no less confident in his absurd conclusions. 

It can be charming when a child pretends that he is a cloud, or a boulder, or a lion, or even – I suppose – a collection of particles.  It’s considerably less charming when a grown man does it, and when he is a grown man with a Ph.D. and a tenured position at Columbia, it’s downright embarrassing.  But you need only turn on the news to see that otherwise intelligent people believing ever more ridiculous things on the basis of ever more convoluted sophistries is the story of our age.  There is a crucial but widely overlooked lesson here.  When your basic assumptions are unsound, greater intelligence by no means guarantees that you will come to see this.  On the contrary, sometimes you will end up only more hardened in error than a less intelligent person would be, because you will be able to come up with subtler fallacies and cleverer self-deceptions.

Related posts:

Greene on Nozick on nothing

Why are (some) physicists so bad at philosophy?

Not understanding nothing

A gigantic book royalty check from nothing

Scientists should tell Lawrence Krauss to shut up already

Carroll on laws and causation

Mad scientists