Random review All Reviews Rating Form Contact

Mind and Cosmos by Thomas Nagel

Mind and Cosmos was not well received. Written by the eminent analytic philosopher Thomas Nagel and published in September 2012 by Oxford University Press, it arrived at the tail end of a long, vehement debate between biologists and creationists about whether ‘intelligent design’ was a credible scientific alternative to evolutionary theory. Nagel subtitled his book Why the Materialist Neo-Darwinian Conception of Nature is Almost Certainly False, and in the introduction he announces that he’s surveyed both sides of the controversy and ruled in favour of the creationists. He doesn’t agree with everything they say, nor does he believe in God. He just thinks that in this fight they’re less wrong than the biologists, declaring that it is ‘highly implausible that life as we know it is the result of a sequence of physical accidents together with the mechanism of natural selection.’

This went down as well as you'd imagine (‘the shoddy reasoning of a once great thinker’ - Steven Pinker; ‘the most despised science book of 2012’ - The Guardian). Most of the rage focused on the pro-creationist statements in the introductory chapter. But this introduction is also where Nagel explains that Mind and Cosmos isn’t really about evolution. It’s about the mind and the cosmos, and his real target is a philosophical worldview that encompasses evolution, along with all of biology, physics and chemistry. By the first chapter he’s conceded that evolutionary theory is - probably - mostly correct. It’s just incomplete. His goal is to point towards a new framework that preserves the discoveries of the natural sciences while encompassing the things they leave out.

~

What is the philosophical worldview that Nagel is critiquing? In 1944 Erwin Schrödinger gave a series of lectures at Trinity Dublin that were later published as the book What is Life? It begins by asking: 'how can the events in space and time which take place within the spatial boundary of a living organism be accounted for by physics and chemistry?’

This is a perfect encapsulation of Nagel’s target, which he refers to as reductive materialism. It is the belief that the universe and everything in it can be understood in terms of material objects interacting physically, the nature of which can be studied and understood by a hierarchy of scientific disciplines. Physics is at the base; it explains all of chemistry; chemistry explains all of biology, biology explains all forms of animal behaviour. There might be emergent properties or phenomena that appear at certain scales, arrangements or levels of complexity - like the surface tension of a drop of liquid water - but there’s no mystical energy field that controls our destiny. It’s tangibles all the way down.

Explaining the universe through reductive materialism has been an extremely successful project. It’s the logic Schrödinger followed when What is Life suggested that cells must contain an ‘heredity molecule’ for transmitting information between generations; that the information might take the form of a combinatorial code, and so changes in molecular structure could account for mutation and variance. Watson and Crick acknowledged his influence when they discovered the double-helical nature of DNA. Centuries of similarly astonishing scientific discoveries all resolve back to the materialist model of existence. Why does Nagel think it’s wrong, and what could he possibly hope to replace it with?

One of Nagel’s most influential essays is about what it’s like to be a bat. To have the subjective experience of sleeping upside down, flying around, eating bugs, navigating via echolocation etc. It must be like something, just as it’s like something to be a human in pain or to see the colour red. But, he argues, this form of knowledge is off limits to the scientific method as it currently stands. You can have a map of the bat brain, model all the neural correlates of bat consciousness, be able to predict bat behaviour with total accuracy, and you still won’t know what it is like to be a bat. Or any other creature subjectively experiencing reality. He believes this limitation is attributable to methodological decisions made during the early modern period when the philosophical tenets of scientific rationalism were laid down. Decisions that the thinkers of the era saw as arbitrary conveniences gradually transformed into unexamined metaphysical truths.

I want to spend more time with this idea than Nagel does (i.e. more than two sentences: Mind and Cosmos is a slim but dense book, so much of this review consists of elaborations of its ideas rather than summaries or rebuttals). Here he’s making an abbreviated version of an argument developed by Martin Heidegger back in the 1920s and 30s. Heidegger’s favourite metaphor describes existence as a dark forest. The philosopher’s role is to find paths through the darkness, ‘to be a walker in unlit places’. He told us that during the early modern period thinkers like Francis Bacon, René Descartes and Galileo devised a new method for natural philosophy; a new path through the dark. They resolved to begin with their own self-evident existence as thinking beings then use mathematics, logic and empirical inquiry to build up a new ‘system of the world’: an objective model of reality.

But this method excluded subjective experience: ‘what it is like’ to be someone or something. Partly because subjectivity is messy and complex, but mostly because they thought they already understood it. They believed that humans experience the world subjectively because we have an immortal soul, placed there by the supernatural being who created everything. That’s also why we have free will, a sense of good and evil and the faculty of reason. And even if you doubted the contemporary explanations for these phenomena and wanted to figure out how any of them worked, the religious authorities discouraged that even more strenuously than they discouraged questioning whether the sun really moved around the earth.

We’ve spent centuries exploring the paths the early moderns laid out for us, and we’ve learned many things. And in the light of those discoveries we no longer share their assumptions about God or the soul as explanations for subjective experience. ‘We who used to think we understood it have now become perplexed’. But at the same time, we’ve convinced ourselves that their technique - the empirical, mathematical study of the physical world - is the only way we can ever truly know anything. Instead of finding new paths through existence, philosophy becomes ‘the handmaiden of technology and the sciences’, insistent that the present path is the only path. We can’t see any other paths and we can’t see that we can’t see them. The result is the ‘reductive materialist’ framework.

Nagel’s goal is: ‘to extend the boundaries of what is now regarded as unthinkable, in light of how little we really understand about the world.’ He’s setting out into the dark forest. And he’ll start by flipping Schrödinger’s logic back on itself. If he can find something ‘within the spatial boundary of a living organism’ that can’t be accounted for by physics and chemistry then reductive materialism fails. Meaning there must be something else to the universe other than physical objects, and our assumptions about existence must change. He’s got three candidates: consciousness, cognition and values.

Something bad happened at the end of the Ediacaran era, about five hundred and forty million years ago. There was a mass extinction event and most of the biosphere got wiped out. Some of the survivors were complex multicellular organisms and their descendents spent the next twenty-five million years radiating out into all the ecological niches they could find. And some of them had nervous systems not unlike our own. I like this paragraph from a 2021 Topics in Cognitive Science paper:

Comparative anatomy indicates that major subdivisions, tracts, and histological features of our nervous system were present in the first vertebrates. Human brains appear to be essentially gigantic fish brains specialized for legged locomotion, dexterous manipulation, complex dynamical visual scene analysis, and talking about ourselves.

Were any of the billions of creatures alive during the early Cambrian conscious? Was it ‘like’ anything to be a soft bodied, bottom feeding fish drifting through the warm oceans half a billion years ago? Let’s say it wasn’t; that they were all just complex aggregations of chemicals with no consciousness at all. Heidegger would say that this world is closed off from itself; concealed. Unlit.

But eventually there’s a mutation or an epigenetic event, an adaptation in a gene or gene regulator leading to physical changes in the brain, and something flickers into awareness and feels pain when it's attacked by a predator. Or it feels desire when it detects a mate. Maybe it sees a flash of light when something moves instead of just registering the information as electrochemical changes in its nervous system. What’s the basis of this process? What is pain? What’s happening in terms of the biochemistry?

There are various theories about the functional role of consciousness. Maybe it's a global workspace for unconscious processes? Maybe it’s a product of ‘metacognition’ in which information in the nervous system becomes aware of itself? But none of those are physical theories. Mental events must have physical correlates for reductive materialism to be true. So what were they? If no one can say then it must be false, because:

subjective consciousness, if it is not reducible to something physical, … would be left completely unexplained by physical evolution—even if the physical evolution of such organisms is in fact a causally necessary and sufficient condition for consciousness.

The argument against this line of reasoning goes: ‘We don’t know how consciousness works, or how it evolved. Yet. But one hundred years ago we didn’t know how heritability worked. There wasn’t a biochemical model of DNA until the 1950s. So people like Nagel could - and did - say: ‘Oh oh! Looks like science can’t explain everything! Quick! We need to look for radical alternatives to materialism!’ But now we do know how heritability works, and surprise surprise, it’s reductive and materialist. The same thing will almost certainly happen with consciousness. It’s just more complex so we’re not there yet. And which is more likely? That a complicated thing hasn’t been understood in the very short time we’ve been studying it, or that our extremely successful model of the universe is wrong?’

‘But,’ Nagel could protest (and again, this is my interpretation of Nagel). ‘Surely if this model of reality holds true you can talk about what the biochemistry of consciousness might look like in really broad terms, the way Schrödinger did with his heredity molecule? And I accept that human consciousness must be complex. But if reductive materialism is correct then it started out as something simple. So just give me a guess about the nature of those very early forms. And if you can’t even do that . . . Isn’t that kind of a problem for your theory? Consciousness is the central feature of our existence and you can’t even come close to an explanation?’

Next up: cognition, by which Nagel means reason. Let’s go back to Schrödinger for a moment (and we’ll return to him again later, but relax: Mind and Cosmos doesn’t suggest that consciousness is ‘quantum’: it’s more inventive than that). In 1926 Schrödinger published his seminal paper on wave mechanics presenting what came to be known as the Schrödinger equation, given here in its general form:

It describes the distribution of probability waves governing the motion of small particles. He validated it by predicting the properties of a hydrogen atom. It’s one of the most accurate formulae in all of mathematical physics.  

We could ask a few interesting questions here. Firstly, if the universe is strictly material then what is the physical nature of the object the equation describes? Secondly: what’s the relationship between mathematical objects and the material world? If physicalism is true then how does a complex number like i interact with reality? But Nagel wants us to consider a simpler, deeper problem: how can Schrödinger do this? If he - and the rest of our species - evolved through natural selection then why is the universe intelligible to us in this way?

Remember: reason wasn’t a problem for the early moderns. When Newton developed differential calculus he thought reason was an aspect of his immortal soul, and that enabled him to discover ‘the language of God’. It’s not until the late 19th century that Nietzsche asks the awkward question: if we’re an evolved species, and there’s no rational God creating us and imbuing us with reason, how could we access any form of objective truth about existence? And how could we ever know that we couldn’t know?

Darwin had similar reservations, writing in 1881 in a now-famous letter to a friend (much beloved by intelligent design advocates): ‘with me the horrid doubt always arises whether the convictions of man's mind, which has been developed from the mind of the lower animals, are of any value or at all trustworthy. Would any one trust in the convictions of a monkey's mind, if there are any convictions in such a mind?’ There’s a recursive trap here which Nietzsche delighted in pointing out. Is it true to say there’s no such thing as truth?

But Thomas Nagel is not postmodern. He thinks we have access to objective truth. So again he turns the question around. How does evolutionary theory explain our ability to accurately model the probability waves of subatomic particles? How did the prehistory of our species select for this trait, and how do we avoid the recursive trap?

It is not enough to be able to think that if there are logical truths, natural selection might very well have given me the capacity to recognize them. That cannot be my ground for trusting my reason, because even that thought implicitly relies on reason in a prior way.

Here’s the electromagnetic spectrum: it shows the wavelengths we’ve evolved to see and, as you’d expect, it is only a tiny sliver of reality. We perceive the light it was advantageous for our ancestors to detect, and which was available via the path-dependent branches of our genetic history as hypertrophied fish brains. But reason lets us ‘see’ all of it. There is a difference between the world as it appears to us and the world as it truly is, Nagel argues, and our ability to perceive the second is an astonishing fact that - like consciousness - we’ve forgotten to be astonished by.

He believes that evolved species should think using heuristics, and our ability to go beyond that and reason deeply and successfully about the true nature of existence requires an explanation. He’s aware that there’s a just-so story that explains the emergence of reason:

The capacity to generalise from experience and to allow those generalisations, or general expectations, to be confirmed or disconfirmed by subsequent experience is also adaptive . . . With language we can hold in our minds and share with others alternative possibilities, and decide among them on the basis of their consistency or inconsistency with further observations. Complex scientific theories that entail empirical predictions are therefore extensions of the highly adaptive capacity to learn from experience - our own and that of others. This story depends heavily on the supposition of biological origin of the capacity for non-perceptual representation through language, resulting in the ability to grasp logically complex abstract structures.

But even if this story can be proved true, and a similar just-so story can be demonstrated for the evolution of consciousness, there’s no reductive materialist explanation for the existence of those complex abstract structures, let alone their availability to the nervous systems of evolved beings. In his commentary on Eugene Wigner’s well known paper regarding the mysterious efficacy of mathematics in the natural sciences, the computer scientist Richard Hamming cites a number of celebrated scientists who made profound breakthroughs not by experimentation or observation, but by logical and/or mathematical reasoning about the deep nature of the universe. And Nagel finds it significant that these two immaterial phenomena - consciousness and reason - interact with one another and with moral values, the third and final strand of his argument.

Peter Singer: Do you think it’s wrong to allow a child to suffer?

Almost Everyone: Sure. Why do you ask?

Peter Singer: Because if we think rationally we see that by failing to donate to effective charities in the developing world we-

Frederick Nietzsche: Wait. Why is it wrong for children to suffer? What do we even mean when we say that?

Almost Everyone: (Appalled, awkward silence)

This is the famous problem of values. It should sound familiar. Moral values are an aspect of subjective experience: we feel that it is wrong to allow a child to suffer. But there’s nothing objective we can point to in the world to validate this feeling. Yet again this wasn’t a problem for the early moderns who believed values were objectively real because they were God’s values. For them moral statements about good and evil could be true or false in the same way that mathematical statements can be true or false. Morality was hardcoded into the universe by its creator. But if ‘God is dead’, and we stop believing in that creator then that moral framework loses its coherency. When we say a thing is ‘good’ or ‘bad’, we’re using words that no longer mean anything, other than that we personally like or dislike that thing. For Nietzsche this places us in an age of incomplete nihilism in which our values are meaningless but we keep referring to them anyway. We’re like cartoon characters that have run off a cliff but haven’t looked down yet.

Nagel disagrees with all of this. He’s a moral realist, meaning he thinks statements about moral values can be objectively true or false. But he doesn’t think this is because God decreed it that way. Remember: he’s also an atheist. Instead he thinks statements about moral values can be irreducibly true - that is, they’re just true because they’re true. Not all moral statements function like this. Instead he thinks they can be reduced to primitive types - ‘pain is bad’, maybe - and then built up into complex combinations of objectively true statements. And this is yet another problem for materialism and evolutionary theory.

The logic goes like this. We might have evolved intuitions about appropriate behaviour - kin selection, reciprocal altruism, various taboos - that we’ve acquired through natural or even cultural selection. But they can’t be moral, any more than having fingers and a spleen is moral. They’re just advantageous traits (or spandrels, products of genetic drift etc). But we can subjectively apprehend values that have no apparent evolutionary advantage, like the notion that animal cruelty is wrong. So our ability to perceive such values is a problem in the same way that our use of reason is a problem. Actually more-so, because there’s a just-so story about the evolution of reason, as explained above, but no comparable story about the evolution of our ability to apprehend moral values.

Nagel concedes that readers might not be persuaded by his argument for moral realism. But he wants us to bundle it in with consciousness and reason AND free will, because it’s important for his project: trying to understand how we can consciously reason about right and wrong, then take action in the world based on those judgements. This is where the cosmos enters the picture. Mind and Cosmos wants to map out a system that is compatible with discoveries in the natural sciences but also accounts for the existence of consciousness, reason and values.  

~

Let’s start with the universe. What could it be made of? Wikipedia has this useful diagram:

Physicalism is Nagel’s primary target. He discounts dualism for the same reason as everyone else: the problem of causal interaction - how could purely physical and purely mental substances relate to each other? And he doesn’t like idealism because it doesn’t really explain anything: we’re all just ideas in the mind of God. That leaves neutral monism which postulates that the universe is made of a single substance that has both mental and physical properties.  (There’s a vast, VAST literature on this and it also takes in panpsychism - the notion that consciousness is ubiquitous throughout the universe. Nagel has written about these subjects in other books and papers but we don’t need to get caught up in any of the jargon or intricacies of these debates because Mind and Cosmos seems to use the terms panpsychism and neutral monism interchangeably.)

Neutral monism was developed by William James and Bertrand Russell as an attempt to solve the same problems Nagel is interested in: what is consciousness and how could it evolve? And it begins with a critique of physics. When someone like Erwin Schrödinger calculates the momentum or spin of an electron, Russell notes that they’re describing the external qualities of the particle (or wave or whatever) but that the operation can’t tell us what the entity is:

All that physics gives us is certain equations giving abstract properties of their changes. But as to what it is that changes, and what it changes from and to—as to this, physics is silent.

Imagine molecular biology if you couldn’t see inside a cell, only observe its behaviour. That’s the situation physics finds itself in with respect to elementary particles. Russell theorised that in addition to the extrinsic properties measured by physics, such entities also possess intrinsic properties. He called them 'quiddities'; other philosophers call them ‘inscrutables’. We currently have no knowledge of, or access to quiddities/inscrutables, but he suggested that some of these intrinsic properties are related to consciousness. Remember: Nagel is interested in what it’s like to be a bat, and why this knowledge is unavailable to a reductive materialist conception of reality. So you can see why this approach appeals to him.

But . . . if we follow the logic here, does this mean Bertrand Russell and Thomas Nagel think that quarks and free electrons might be conscious? Well, maybe - but unless they’re part of a nervous system they have no access to memory or perception. Perhaps they have a degree of consciousness in the same way they have mass and charge: close to but not actually zero; some form of barely perceptible white noise. Perhaps it aggregates with scale and complexity via some form of mental chemistry as evolved beings acquire sensory systems and access to information processing. The neuroscientist Christof Koch observes that he’s often met with an incredulous stare when he endorses this direction of thinking. But the one thing we truly know about our own intrinsic nature is that we are conscious in a way that isn’t accessible to the contemporary physical sciences. Maybe everything is?      

~

Nagel doesn’t talk about art. Which is a shame: it’s another of the mysteries of subjective experience, and artworks are sites in which the subjective and physical worlds intersect. So let us briefly consider Vincent van Gogh’s Mulberry Tree, painted at an asylum in the final year of his life. Because neutral monism/panpsychism is not the most radical or controversial idea advanced in Mind and Cosmos: that would be the teleology. And van Gogh’s tree will (hopefully) help keep us grounded as we contemplate Nagel’s teleological universe.

Why was that tree in that place at that time, to be painted by that man? As moderns we answer this question in terms of prior causes. Someone planted it there; or maybe it grew by itself. It emerged from a seed from another mulberry tree, and this species evolved into its current form via natural selection. But in pre-modern philosophy it was common to reason in terms of final causes. To this way of thinking the tree caused the seed. Or, if you find van Gogh’s painting to be profound enough, the painting could be the final cause - the telos - of the tree, which was the cause of the seed and the species itself. Backwards runs causality until reels the mind.

Does Nagel really believe that the universe works like that? Kind of, but in an intriguing way. Remember the moral values that he argues are a fundamental aspect of reality? Values that are immanent, i.e. true because they’re true? He argues that they might be teleological. They might act as a final cause that can bring into existence beings that are conscious, moral and rational, so that we can perceive and act upon those values.

Teleology would have to be restrictive in what it makes likely, but without depending on intentions or motives. This would probably have to involve some conception of an increase in value through the expanded possibilities provided by the higher forms of organization toward which nature tends: not just any outcome could qualify as a telos. That would make value an explanatory end, but not one that is realized through the purposes or intentions of an agent. Teleology means that in addition to physical law of the familiar kind, there are other laws of nature that are "biased toward the marvelous"

And so it all comes together, Mind and Cosmos. All become one in Yog Sothoth:

Once there are beings who can respond to value, the rather different teleology of intentional action becomes part of the historical picture , resulting in the creation of new value. The universe has become not only conscious and aware of itself but capable in some respects of choosing its path into the future—though all three, the consciousness, the knowledge, and the choice, are dispersed over a vast crowd of beings, acting both individually and collectively.

~

A lot of people reviewed Mind and Cosmos back in 2012, and most of them fell into two camps. The first were scientists or science-adjacent philosophers outraged by the Intelligent-design friendly introduction; the second were Christian intellectuals who were delighted by that same introduction and then upset by the godlessness of the approach laid out in the first chapter. But the physicist Sean Carroll made it past the opening pages and he had questions about how Nagel’s grand teleological vision hangs together from a scientific perspective:

Imagine what it would entail to truly believe that consciousness is not accounted for by physics. It would entail, among other things, that the behavior of ordinary matter would occasionally deviate from that expected on the basis of physics alone, even in circumstances where consciousness was not involved in any obvious way. Several billion years ago there weren’t conscious creatures here on Earth. It was just atoms and particles, bumping into each other in accordance with the rules of physics and chemistry. Except, if mind is not physical, at some point they swerved away from those laws, since remaining in accordance with them would never have created consciousness. In effect, the particles understood that sticking to their physically prescribed behaviors would never accomplish the universe’s grand plan of producing conscious life. Teleology is as good a word for that as any.

Nagel does address this, albeit indirectly: ‘the possibility of principles of change over time tending toward certain types of outcome is coherent, in a world in which the nonteleological laws are not fully deterministic’. I assume he’s talking about the Schrödinger equation here; which tells us that at a fundamental level the universe is not deterministic but rather probabilistic. There’s a randomness at the heart of things that might not be so random, this implies, but rather weighted towards pathways leading to the emergence of conscious, intelligent life. If we decide that van Gogh’s tree is a final cause in this model it doesn’t have to go back in time and tamper with the rules of physics and chemistry to bring itself into being. Instead the tree functions like an attractor state in a chaotic system. The universe will, over time, naturally tend towards the existence of the mulberry tree at the St Paul asylum in Saint-Remy.

Is Nagel invoking quantum theory the same way new age mystics use and abuse it - as a form of magic disguised behind a veneer of scientific jargon? Maybe he’s sensitive to that critique and that’s why he doesn’t quite say what he means? On the other hand, most of Mind and Cosmos consists of mind-bending yet cursory assertions about the ultimate nature of reality. Maybe Nagel was just being true to his vision.  

And I think ‘vision’ is the right word here. Mind and Cosmos ultimately resolves itself into a spiritual text. Yes, it’s also a book of metaphysics and its explanations are naturalistic, and yes, Nagel regards himself as an atheist. But when you’re describing a universe with objective truth and immanent moral values that summon our species into being I think you’ve moved beyond what most of us mean by atheism.

I’m not saying that’s a bad thing. I find parts of Nagel’s doctrine attractive. It would be very useful if truth and morality turned out to be real. I think he’s wrong about the improbability of the emergence of life, complexity, and even intelligence (or, rather, that while these events are improbable, many improbable things can happen over a few billion years). But he’s probably right about consciousness being an impasse for evolutionary theory as it stands. I’m persuaded that the existence of a hidden order lying behind the phenomenal world, and that our ability to perceive that order is a deep and unsolved problem. And he’s convinced me that the materialist conception of reality is almost certainly false

Are fundamental particles conscious? Are moral values an innate property of the universe? Are they teleologically bringing intelligent life into being? It’s all possible, I suppose. These aren’t the central tenets of Nagel’s creed though - he’s merely exploring what the universe might look like. And he admits he’s likely to be proved wrong, predicting that reality will turn out to be stranger than he’s imagined.

~

In the epilogue of What is Life Schrödinger gets metaphysical:

Let us see whether we cannot draw the correct, non-contradictory conclusion from the following two premises: (i) My body functions as a pure mechanism according to the Laws of Nature. (ii) Yet I know, by incontrovertible direct experience, that I am directing its motions, of which I foresee the effects, that may be fateful and all-important, in which case I feel and take full responsibility for them. The only possible inference from these two facts is, I think, that I –I in the widest meaning of the word, that is to say, every conscious mind that has ever said or felt 'I' -am the person, if any, who controls the 'motion of the atoms' according to the Laws of Nature.

We are all God experiencing the world subjectively is his argument here, and this explains how we’re able to take action in a universe governed by physical laws. (He’d been reading Schopenhauer). And in some ways this resembles Nagel’s notion that we are the universe becoming aware of itself, or even Heidegger’s vision that we are openings in the darkness of being. But what Nagel really believes is that we don’t know what we are. We don’t know what the mind is, and we don’t know what the cosmos is. We understand a little more about these subjects than we did in Aristotle’s day - but not a lot. He attributes this lack of progress to a human addiction to final reckonings: philosophical systems that attribute unexplained phenomenon to a deity, or, in the case of reductive materialism, by ignoring them or confidently assuming they’ll be solved within the pre-existing framework someway further down the line.

The idea at the heart of Mind and Cosmos is that existence is both deeply mysterious yet, somehow, comprehensible. But we only deepen our comprehension if we continue to interrogate the boundaries and limitations of our current methodologies. The universe might ultimately prove to be unknowable, its final truths beyond the realms of our senses or cognitive capabilities. We can never know this, though, so there’s value in exploring the unlit paths - even though most of them will lead nowhere - rather than pretending that the darkness surrounding us doesn’t exist, or assuming that it merely conceals a god.