Quote for the Day – The Fractal Nature of Scientific Knowledge

I believe that scientific knowledge has fractal properties; that no matter how much we learn; whatever is left, however small it may seem, is just as infinitely complex as the whole was to start with. That, I think, is the secret of the Universe.

~Isaac Asimov, I, Asimov: A Memoir

A Critical Review of Thomas Nagel’s latest book Mind and Cosmos

Thomas Nagel’s book Mind and Cosmos has garnered a lot of attention since its publication, with the reaction being intense, heated, and of largely mixed opinion (For a nice overview and response to negative views, see here). Not surprisingly, those sympathetic to naturalism have been harsh in their assessment, and those not so sympathetic have been full of praise. My assessment is ultimately negative, but I think both the defenders and critics of Nagel’s book have relied on a set of shared assumptions that are deeply problematic and masked by ambiguity. By exposing the hidden assumptions framing the debate between “reductionism” and “non-reductionism”, I show that Nagel’s criticism of “reductionism” is sound, but this is no reason to reject physicalism, only an outdated and confused version of physicalism.

1. The Problem of Hierarchy: Two Views of Reduction

The controversial subtitle of Mind and Cosmos is “Why the Materialist Neo-Darwinian Conception of Nature Is Almost Certainly False”. Quite the claim! But what exactly is “Materialist Neo-Darwinism” and who really holds this view? More precisely, Nagel’s target for criticism is what he calls:

“A particular naturalistic Weltanshauung that postulates a hierarchical relation among the subjects of those sciences.”

I have a lot to say about this notion of a “hiearchical relation”, so bear with me! I propose Nagel’s invocation of a “hierarchy” is ambiguous between two senses of the term. First, there is a hierarchy of scale that is compositional e.g. Society is composed of individuals, which are composed of cells, which are composed of molecules, which are composed of atoms, etc. I take compositional hierarchy to be non-controversial.

The second sense of “hierarchy” is what John Heil calls a “hierarchy of being” or “hierarchy of reality”. On this view, you have lower-level “realizer bases” out of which higher-order properties “emerge”, in the sense of “strong” emergence where the whole is non-compositionally greater than the sum of the parts (whatever that means). For example, on the hierarchical view of reality, the mind “emerges”, “arises”, or is “generated” as a higher-order property “out of” the lower realizer bases, and is said to “supervene” on that lower base. Nagel is perfectly right to think this is a mysterious process, but that’s a good reason for physicalists to reject the entire hierarchical view of reality, not give up on physicalism.

The alternative picture is that reality is flat. There is one level of reality: it’s called reality! Within that level we can find hierarchies of scale and composition, but ultimately there are not two different levels or types of “being”, one physical and the other mental, with the latter “emerging” or “arising” out of the former. Nagel seems to think that physicalism is committed to this view, but ironically the problems he locates with reductionist physicalism stem from his own dualist intuitions about the existence of “higher” levels of reality. However, an ontologically serious physicalism has no room for different “levels” of reality, and thus it is not a “problem” to try and reduce the mental “level” to the physical “level” because that view of reduction relies on the problematic notion of there being a hierarchy of reality in the first place. In his book From an Ontological Point of View, John Heil recommends that

 …[W]e abandon the notion that reality is hierarchical. We can accept levels of organization, levels of complexity, levels of description, and levels of explanation, without commitment to levels of reality in the sense embraced by many self-proclaimed anti-reductionist philosophers today.The upshot is a conception of the world and our representations of it that is ontologically,but not analytically,reductive.

Agreed! This result generalizes to the entirety of Nagel’s critique of physicalist reductionism. While Nagel is right to argue that reductionism is problematic, the entire problematic of reducing higher levels of reality to lower levels of reality is a product of the dualist worldview, not the physicalist one. It bears repeating that on the physicalist view, there is no problem of “reducing” the mental to the physical because setting the problem up like that buys into the traditional dualist framework of reality as hierarchically layered. In a nutshell, Nagel is criticizing a dualistic view of physicalism, and thus his critique of physicalism says more about how Nagel views the world than it does about how physicalists do.

To be fair, many physicalists have unwittingly bought into the same “levels of being” view that is problematic, but insist that the “reduction” of the highest levels to the lowest is tractable. I agree entirely with Nagel that this project of reduction is doomed to failure, but not for the same reason Nagel does. My problem with the project of reduction is that it sets up the problem wrong, whereas Nagel think it reveals the truth of dualism (or at least the falsity of physicalism).

Nagel claims that his book is

“…[N]ot just about acknowledging the limits of what is actually understood but of trying to recognize what can and cannot in principle be understood by certain existing methods.”

Having cleared up the ambiguity of hierarchies, we can now analyze Nagel’s conclusions about what can and cannot be understood by the physical sciences. If the project is to say it is impossible for the physical sciences to “reduce” the “mental level” to the “physical level” then I agree entirely. But not because I doubt the truth of physicalism. The mental level cannot be reduced to the physical level because the whole idea of there being different levels is a product of the traditional dualist worldview where reality is divided into base terrestrial materials and higher spiritual realms.

Thus, when Nagel criticizes the project of “materialist reductionism,” he kind of has a point, but it’s impossible to see this unless we clear up ambiguities about how physicalists can think about reductionism. I propose there are essentially two types of materialist reductionism: call them Reductionism and Eliminativism. Reductionism is the confused view we have been discussing, where it’s acknowledged at the start there are two levels of being, and the problem is to reduce the mental level to the physical level or provide psycho-physical “bridge laws” between the two levels. In contrast, Eliminativism says all that exists is the physical level. Rather than starting with two levels, Eliminativism starts with one level and all phenomena as well as explanations of phenomena are framed in terms of this level of reality.

As I see it, Nagel is right to criticize Reductionism, but his arguments don’t apply to Eliminativism. Unfortunately, Nagel doesn’t make this distinction clear and doesn’t engage with the alternative one-level view of reality, which is the true competitor to the traditional dualist worldview. I can imagine what Nagel would say if he heard about this distinction. He might say it’s completely counter-intuitive because it denies the reality of “first person subjective experience”. But again, this talk of “denying reality” is ambiguous between denying that psychological phenomena “live” in the higher-realm of a two-level reality or denying that it exists at all. Physicalists don’t have to deny that psychological phenomena exist; they will just say that, whatever they are, they “are” just like everything else: existing in the complex, flat ontology that is our physical universe. Thus, physicalists only deny the reality of mental phenomena if it is assumed that mental phenomena “live” in a higher plane of existence in the two-level picture of reality.

Of course, given the complexity of the universe, physicalists have no expectation that humans will be cognitively comfortable (or even capable) of accounting for everything they want to explain in terms of the lowest compositional scales (atoms, quarks, etc). Rather, we abstract to higher compositional scales to deal with complex phenomena like human civilization. But all this abstraction is abstracted within a single level of reality. The psychological necessity of parsing the world into higher and lower scales is not a metaphysical proof of separate planes of reality. In fact, a physicalist understanding of human cognition would predict that humans would not be capable of handling reams of data without compressing it with language, graphs, equations, simulations, models, etc.

Thus, the following statement by Nagel is actually a prediction of the naturalistic concept of human cognition:

“For a long time I have found the materialist account of how we and our fellow organisms came to exist hard to believe.”

Of course such a massively complicated proposition is “hard to believe”. A lot of things are hard to believe because we are forced to use our puny primate brains to understand a large, scarily complex physical universe with weird things like black holes, quarks, and gravity. Why would we expect it to be easy to imagine abiogenesis? Essentially, Nagel’s argument that not-P is: “I don’t understand P, therefore not-P.” But if P is physicalism, and physicalism predicts that humans will have a hard time understanding complicated things, our lack of understanding (especially of nonscientists) is surely not a refutation of physicalism, merely a recognition of its cognitive complexity.

2. Flubbing the Facts

Ontology aside, I do think Nagel makes some empirical gaffes exposing his ignorance of mainstream biological science, especially when discussing the state-of-the-art materialist conception of abiogenesis. He says, incredibly:

“The more details we learn about the chemical basis of life and the intricacy of the genetic code, the more unbelievable the standard historical account becomes”

If history has taught us anything, believability by non-scientists is not a good metric for the truth of scientific claims. For many non-scientists, Einsten’s theory of relativity is unbelievable, but that obviously has no bearing on its truth. Nagel continues:

“[Abiogenesis] flies in the face of common sense…

“….It is prima facie highly implausible that life as we know it is the result of a sequence of physical accidents together with the mechanism of natural selection

“…When we go back far enough, to the origin of life – of self-replicating systems capable of supporting evolution by natural selection- those actually engaged in research in the subject recognize that they are very far from even formulating a viable explanatory hypothesis of the traditional materialist kind”

It would behoove Nagel to google the phrase “hydrothermal origins of life” to find proposed models of abiogenesis that start from recent discoveries of deep-sea heat vents. A quick search on Wikipedia and you can find the “iron-sulfur world theory” of abiogenesis. There are probably other rival hydrothermal theories, and I have no idea whether any of them are true, but the point is that they seem, to me, like a “viable explanatory hypothesis of the traditional materialist kind”.

Perhaps there is an ambiguity in Nagel’s use of “formulation”. Maybe he means “nobody has formulated a complete explanation established to be almost certainly true”. This would be true, but that’s simply an argument from ignorance. You can’t infer from the fact that we haven’t yet formulated a “complete” explanation that there isn’t a complete explanation.

However, if Nagel means “nobody has even formulated a viable educated guess” about the origins of life, this is almost certainly false, unless he can show why hydrothermal models aren’t “viable” as possible explanations.

3. Good Darwinians vs Bad Darwinians

Another thing that confused me is who Nagel considers to be a “bad Darwinian materialist” and who is a “good” materialist. Right after harshly criticizing Richard Dawkins for explaining the evolution of the eye “merely as a result of chemical accident, without the operation of some other factors determining and restricting the forms of genetic variation”, Nagel gives an example in a footnote of presumably “good”, non-reductionist Darwinians, “who insist that the evidence calls for a more restricted account of the sources of variation in the genetic material.” Surprisingly, Nagel cites Stuart Kauffman has a “good” non-reductionist, saying in that same footnote “Stuart Kauffman suggests in several books that variation is not due to chance, and that principles of spontaneous self-organization play a more important role than natural selection in evolutionary history.”

This is confusing on multiple levels. Is Nagel implying that Kauffman is “on his side” merely because he denies that all variation is due to chance alone and not chance in addition to other factors like the material property of spontaneous self-organization? If big, bad “Darwinian materialism” is supposed to be the claim that chance alone is the full story, then Nagel is making a stink about a view that few biologists actually hold. It might be convenient to treat chance has the primary operator, but this is because dealing with the entire material evolution of cellular matter is absurdly complicated, and would require a ridiculous amount of computing power to simulate.

I can’t speak for Kauffman’s own ontological views, but Kauffman-esque “non-reductionism” is ambiguous between materialist Reductionism and materialist Eliminativism. It’s conceptually possible that all the “spontaneous” self-organization of matter is a property of the single level of reality that is the physical universe. This might be wildly complex, but why shouldn’t it be? It’s the universe itself we’re talking about here, not a toy model of it. But nothing about wild and strange self-organizing properties of matter is inconsistent with materialist Eliminativism. Thus, Nagel’s appeal to Kauffman has a counter-example to reductionism is ineffective as a genuine criticism of the physicalist worldview.

Conclusion

As a card-carrying “Darwinian materialist” who wears his naturalistic credentials proudly, I am sympathetic to the received criticisms of Nagel’s book. However, unlike other naturalists who have criticized Nagel’s book, I think Nagel’s criticism of Reductionism is sound: it is impossible to reduce the mental level to the physical level. Nagel’s critics have mistakenly limited their options to either joining Nagel in attacking Reductionism or accepting Reductionism as formulated by Nagel.

I reject both options in favor of Eliminativism. The reason why it’s impossible to reduce the mental level to the physical level is that this very way of setting up the problem belies the traditional view of reality as composed of “levels of being”, with the mental level “emerging out of” or “arising from” the physical level. For eliminativists, there is a single level: physical reality, and humans exist only on this level. But we certainly don’t talk or write this way. For good reason: we have puny primate brains with finite memory and processing power. It’d be futile to try to understand WWII in terms of Schroedinger’s equation, but this cognitive inability does not entail the falsity of quantum mechanics. Likewise with psychology.

Thoughts on the fundamental problem of representation

I’ve been thinking a lot about the so-called “fundamental problem of representation”: what is a representation and how does it work as a representation? How do representations represent? We need to first answer what a representation is. Many philosophers seem to agree that it has something to do with “standing in”. The obvious example is a photograph. A photograph of a cat is a representation of a cat because the photograph “stands in” for the real cat. How does this work? Well, it seems to need an interpreter to interpret the photograph as a representation of a cat. But if we want to explain how the brain represents something, it obviously won’t do to posit an interpreter, for this is just a homuncular explanation.

So the photograph example is kind of a nonstarter when it comes to understanding how the brain represents something. Many philosophers believe that in order for the brain to perceive the world, it must form a representation of the world. In this way, perceiving the world is seen as forming a model of the world, which is used to compute action plans. But is this really a scientific explanation? When a brain perceives a cat, what does it mean for brain activity to “stand in” for that cat? This “standing in” function is obscure. For this reason, Eric Dietrich prefers to talk about representations as “mediators”. A representation is a mediator between a stimulus and behavior.

This makes sense to me, for I can imagine what it means for something to mediate between a physical perturbation and physical behavior. But doesn’t a thermometer also mediate between a stimulus and a behavior? What makes neurons different from thermometers? Aren’t neurons just complex bio-machines? And machines are machines. But I’m convinced there is a difference between a thermometer and a brain. I think representations in the brain are genuinely mental whereas I do not think this is true of the thermometer. Why? I haven’t quite worked this out, but I think the difference is that the mechanisms of mediation in the brain are responding to meaningful information, whereas the mechanisms of mediation in the thermometer are not responding to meaning at all. Meaning is mental. Mental mediations are mediations in response to meaning. We can thus make a fundamental distinction between nonmental representation and mental representation.

But what is meaningful information? How can we understand meaning ontologically? I think the concept of affordances is useful here. Let’s start simple. Think of sucrose molecules. Now imagine a thermometer like machine that had sensors designed to respond to sucrose, mechanisms of mediation (“processing”), and then an output behavior (turning on a green light). We have no reason to think of this machine as instantiating any truly mental mediations. Its mediations are purely responding to the sucrose causally. But now imagine a bacteria. It too has biochemical sensors designed to discriminate sucrose, mechanisms of mediation for processing it, and output behaviors. Some people think I’m crazy for holding this view, but I genuinely think that in the case of the bacteria, the mechanisms of mediation are mental. Why? Because the sucrose affords something to the bacteria, namely, nutrition. The sucrose is thus meaningful to the bacteria, whereas sucrose is not meaningful to the machine. There is an affective valence even at the level of the bacteria, it is just hard to imagine. But put yourself in the “mind” of a bacteria. Its whole world has a valence. It is attracted/repelled by physical perturbations. But unlike the machine, the bacteria perceives these perturbations as genuinely meaningful, for the sucrose affords the possibility of an opportunity for helping maintain a norm (the norm of survival). I think this emphasis on survival and affective valence is important, because I think of it as a means to solve the frame problem. Having a norm which regulates behavior enables the mechanisms of mediation to be responsive to more than just brute causal information. In enables the perception of affordances. The norm of survival is the Ur-desire, the spark of mentation. Arguably, the category between life and nonlife is fuzzy, so it’s not quite clear where to draw the line, but that there is a line is undeniable. I don’t doubt that robots could in principle instantiate their own norms to solve the frame problem, but I imagine they will look similar to biological norms.

But what about neurons? Many contemporary philosophers of mind think that “mental stuff” only happens in sufficiently complex brains. I think this is a mistake, for any association of representation with strictly neural processes will fail to answer the fundamental problem of representation. I think Dietrich is right that the concept of mediation is the right way to understand representation. But I think that neural mediation is just one form of mediation. Evolutionarily speaking, neural mediation was highly adaptive for it allowed organisms to increase the complexity of the mediation between stimulus and behavior. More complexity in mediation leads to “deeper” processing i.e. more complex behavior. Neural processing allowed for the mediations to become “abstract”. By this, I mean that the mechanisms of response become sensitive to more “global” features of a stimulus profile. This is called an increase in invariance, for differences in low-level stimulus detail will make the “higher-order” circuits fire steady. Think of perceiving a chair. We can recognize a chair from almost any angle of viewing. As we change angles, the lower-level mechanisms of response fire only in response to very specific and low-level features of the chair. At higher levels of processing, the response is steady regardless of the lower-level features. In this way, we say that the representations in the brain have become more “abstract”. Language is the ultimate in abstract mediation, for linguistic “tagging” of the world enables us to respond to very abstract kinds of information, particularly in respect to social cognition and our collapse of human behavior into abstract folk psychological categories. The vocabulary term “mind” is one of these ultimate abstractions, for it abstracts over all physical behavior and gives us a new category of response: person. Such linguistic representations are meta-representational insofar as they allow organisms to represent representations, to mediate mediations. Many theorists, including myself, think that it is meta-representation which separates the mental life of humans from that of other animals.

In summary, the fundamental problem of representation is to understand what a representation is and to answer how it works as a representation. Representations are stand ins for stimuli. A stand in for a stimuli is a mechanisms of mediation between stimuli and behavior. There are two fundamental types of mediation: mental and nonmental. Nonmental mediation is ubiquitous in the physical world, whereas mental mediation is rare. Mental mediation is mental because the mechanisms of mediation are sensitive to affordance information, which is grounded by norms, the most evolutionarily basic being the norm of survival. Mental representations thus form a continuum of possible abstraction, with neural representations only being a kind of mediation, enabling deeper abstraction through stimulus-invariance. There is thus nothing mysterious about representation. The term itself is a shorthand description of the complex mechanisms of mediation intrinsic to an entity.

Reductionism or Holism? MU!

This picture was drawn by Douglas Hofstadter for his classic dialogue “Prelude…Ant fugue” in his masterful Godel, Escher, and Bach. In the dialogue he asked a simple, yet highly illuminating question: what does the picture say?

On the top level, it says “MU”, alluding to the Zen parable of Joshu and the dog. In this parable, a monk asks Joshu “does a dog have Buddhe-nature?” Joshu, with immediate fervor, replies “MU!”, which basically means that Joshu denied the legitimacy of the question. He “unasked” the question. The relevance of the parable is made clear when one attempts to answer the above question: what does the picture say? One cannot simply assert that it says “MU”, although on the top and bottom levels it says as much.(You can’t see it in the above copy, but the individual letters are made out of tiny iterations of “mu”). That would be too simplistic of an answer, although it is trivially “true”.

The beauty of Hofstadter’s diagram is only apparent after you let the meaning of the parable sink in. Do all questions have definite answers? Can we really ask questions like “Is the key approach to the mind reductionist or holist?” and expect simple answers? Is it wiser to instead “unask” such questions? Perhaps that question should be unasked as well!

I love this diagram on multiple levels, which is fitting I think, so I thought I would share it with you all. Please spend the time to ponder on the question of what the picture reallysays, and see if you come to the same conclusion as Joshu.

:: :: :: :: :: :: :: :: :: ::

The Careenium

I am a big fan of Douglas Hofstadter, the author of the classic Godel, Escher, Bach and more recently, I Am A Strange Loop. Hofstadter is a master of metaphors and today I would like discuss one metaphor in particular, The Careenium.

Hofstadter asks you to imagine a frictionless billiards table with lots and lots of tiny, magnetic marbles, or “simms”(small interacting magnetic marble), bouncing around, careening endlessly. Because these simms are slightly magnetic, they are apt to stick together into clusters called “simmballs”(see where this is going yet?). These simmballs are more or less stable with simms transferring in and out endlessly. Furthermore, imagine that the walls of the billiard table are sensitive to the outside environment and for every force, the walls flex inwardly slightly. Naturally, this flexibility is reflected in the careening simms and ultimately in the large simmballs.

Thus the simmballs be be said to encode for the events in the environment and in principle, if someone was well-versed in Careenium mechanics, they could interpret the simmballs as being symbolic. In case you haven’t figured out the mappings of the metaphor yet, let me lay it out explicitly. The simms map onto neurons(small events) and the simmballs map onto patterns of neurons(larger events) and by virtue of encoding for the environment, the simmballs(symbols) have representational qualities.

The point of Hofstadter’s metaphor is relatively simple. He wants you to imagine a scenario where the brain(Careenium) could be seen in two different perspectives. One perspective, which comes naturally to scientists, is reductionist. That is, one could in principle view all the activities of the Careenium in terms of the tiny simms bouncing around, acting in accordance with well-known laws of physics. On the other hand, one could take could the high-road, and view the system in terms of the larger simmballs and their macroscopic, representational properties.

In order to help you visualize the implications of the metaphor, Hofstadter asks you to imagine two perceptual shifts of the Careenium. The first shift is to speed everything up, so that the fast-moving simms become too fast to be seen by the naked eye and the larger, slower moving simmball clusters become more active, bouncing around in a lively fashion, interacting with each other. The second perceptual shift involves zooming out so that the the simmballs become the only thing one can attend to. With these two perceptual shifts in mind, Hofstadter asks the following question: who shoves whom around inside the Careenium?

On one hand, there is the view that the tiny, meaningless simms are the primary “shovers” and the simmballs are merely along for the ride. On the other perspective, zoomed out and sped up, the simmballs are the only interesting feature of the system, with a rich symbolic “logic” that corresponds to the environment being represented. Which perspective is the “truth”? Well, as Hofstadter says, it “all seems topsy-turvy.” I’ll leave you with a quote from the book:

From our higher-level macroscopic vantage point as we hover above the table, we can see ideas giving rise to other ideas, we can see one symbolic event reminding the system of another symbolic event, we can see elaborate patterns of simmballs coming together and forming even larger patterns that constitute analogies-in short, we can visually eavesdrop on the logic of a thinking mind taking place in the patterned dance of the simmballs. And in this latter view, it is the simmballs that shove each other about, at their own isolated symbolic level.

:: :: :: :: :: :: :: :: :: ::

The Supposed Drawbacks of Reductionism

Misreading the mind: If neuroscientists want to understand the mystery of consciousness, they’ll need new methods.

I thought this article was an interesting insight into the thought process of those who disregard the relevance of neuroscience for understanding the mind. The terminology Lehrer uses gives him away:

Even our sense of consciousness is explained away with references to some obscure property of the frontal cortex.
…
[According to reductionism]The mind, in other words, is just a particular trick of matter, reducible to the callous laws of physics.
…
You are simply an elaborate cognitive illusion, an “epiphenomenon” of the cortex. Our mystery is denied.

All of these quotes highlight a curiosity in Lehrer’s phrasing. Does he really think that if neuroscience succeeds in explaining cognitive phenomena in mechanistic terms, the mind will be “explained away”? Was heat “explained away” when we reduced it to the movement of molecules? Were the properties of water “explained away” when we reduced it to H2O? On the contrary, the phenomena of both are still with us and it is ridiculous to assume that if neuroscience is successful it will reduce the mind to “just a trick”. On the contrary, the mind will be seen as a complicated set of cognitive phenomena not just “reducible to” but explained by mechanisms in the brain/body system.

So, the question isn’t as Lehrer says whether or not neuroscience can move “beyond reductionism”, but rather, what can be successfully explained in mechanistic terms and what can’t? It is clear that there is useful phenomenological data to be had at the higher levels of abstractions that characterize our thoughts about the mind, but it should be said again that these abstractions aren’t “just” tricks, but rather, complicated phenomena in their own right that need explaining. Whether or not that explanation will be in the terms of neuroscience or at the higher level of cognitive psychology has yet to be determined, but it seems clear that the empirical method itself will give us a clearer picture of the mind.

This brings me to my last point, and that is whether or not neuroscience is capable, in principle, of explaining all cognitive phenomena. For me, the answer is a resolute yes, but I want to emphasize the term “in principle”, because explaining all cognitive phenomena at the molecular level may be pragmatically out of reach. We should be grateful that evolution has given us a language capable of discussing cognitive phenomena at a higher abstraction than that of science, but we should also learn to accept the fact that ultimately everything in the universe, including the mind, can be “reduced” to the physical motions of matter. It might seem like I am making a category mistake, but it seems intuitively plausible to me. I am not saying that all cognitive phenomena will be reduced to the physical level, but I think in principle, it can be. But I don’t think that is a very interesting idea. What is more interesting to me is the question of what will be explained in mechanistic terms and what won’t, and that is a pragmatic question of science that we will be continuously working on for what seems like an indefinite period of time.

:: :: :: :: :: :: :: :: :: ::