Is the universe conscious?
A conversation with philosopher Philip Goff about “cosmopsychism” and his new book
The last decade or so has been a golden age for the philosophy of mind as well as the scientific study of the mind and brain. The major new development in this time frame has been a growing acceptance that the prevailing view of mind and consciousness — what we can label “materialism” or “physicalism” as shorthand — is increasingly seen as inadequate for providing answers to the big questions about the nature of consciousness.
It seems, nevertheless, that some version of materialism may still be the prevailing view among philosophers and scientists. But alternative views are growing fast. One of the major alternatives is panpsychism, the view that all matter has some associated consciousness, albeit in a very limited manner for most matter.
Philip Goff, who was until recently associate professor in philosophy at the Central European University in Budapest but has just taken up a position at Durham University in the UK, is a pioneer in elaborating on the various versions of panpsychism and developing in some detail his preferred version known as “cosmopsychism.” This is the view that the entire cosmos is one consciousness and that all other conscious entities, humans and everything else, are grounded in this higher level of consciousness. This is the opposite of the more common notion of grounding: grounding in the smaller constituents of the mind such as neurons, molecules, etc.
He fleshes these ideas out in his 2017 book (his first), Consciousness and Fundamental Reality, published by Oxford University Press. Goff’s book is very well done, a rare example of cutting edge philosophy that tackles viciously complex issues with clarity, modesty and humor.
For the record, I outed myself long ago as a panpsychist, but I’m of the more traditional (it seems funny to use that term in the context of panpsychism even now) school of constitutive panpsychism. I find Goff’s arguments for cosmopsychism interesting but not entirely convincing, as we discuss below. I approached Goff about an interview with the expressed goal of exploring his ideas and how they connect with Alfred North Whitehead’s process philosophy, which is a version of panpsychism I’ve explored and advocated in modified form.
We conducted this interview by email in mid-2018.
Why do all the “isms” discussed in your book matter to anyone outside of academic philosophy?
I have a love/hate relationship with “isms.” There’s a danger of getting lost in the jungle of all the possible views philosophers have dreamt up, so that you end up forgetting about the issue you were supposing to be dealing with in the first place. But the fact is there are many different theories of what reality is like. And what I’m interested in is having our best guess at which one is true. So you need at least some “isms” to capture what these different theories are.
What got you into philosophy as a career?
I’ve always been obsessed with philosophy. My parents tell me that when I was four I asked “Why are we here?” (although that might have been because we’d just moved house…). I’ve always been interested in how different bits of our worldview fit together. How do moral truths fit with scientific truths, or mathematical truths? How does consciousness fit in with the physical facts about the body and brain? How does it all “hang together”?
Actually, in terms of career, philosophy was plan B. I wanted to drop out of school and become a rock star. That didn’t work out though, so I had to find some other way of paying the bills. Actually, my band’s just got back together, so maybe we’ll make it this time and I can quit the day job.
What is panpsychism? And what is cosmopsychism?
We’ve spent many decades now trying to explain consciousness in terms of utterly non-conscious processes in the brain and got precisely nowhere, at least on the central problem of explaining consciousness. Panpsychists propose an alternative research programme. Rather than trying to explain consciousness in terms of non-consciousness, panpsychists hope to explain the complex consciousness of human and animal brains in terms of simple forms of consciousness, simple forms of consciousness that are then postulated to exist as basic features of matter, perhaps basic properties of fundamental particles. The view sounds crazy, but you should judge a view not on its cultural associations but on its explanatory power. What panpsychism offers is a way of integrating consciousness into our scientific picture of the world, a way that avoids the deep difficulties that plague its more conventional rivals, such as materialism and dualism. An increasing number of philosophers and neuroscientists are coming around to the idea that it may be our best hope for solving the problem of consciousness.
Cosmopsychism is a form of panpsychism. Regular panpsychism starts from the commonly held assumption that fundamental things exist at the micro-level. I call this the “Lego brick” view of the world: there are loads of little things, and when you put them together you get big things. I’m inclined to think the Lego brick view doesn’t fit very well with contemporary physics, in particular with the phenomenon of quantum entanglement.
On an alternative view, it is the universe as a whole that is fundamental, and the existence of everything else is derived from the existence of the universe. If you combine that view with panpsychism you get cosmopsychism: the view that the conscious universe is the one fundamental entity. It sounds a bit mystical. But you needn’t think of the conscious universe as anything like God. In the book I suggest that the consciousness of the universe is probably just a kind of mess.
Why, in a nutshell, is materialism false?
Materialism can’t account for the reality of consciousness, which I think is a datum in its own right. We know that consciousness is real and so it has to fit into our worldview somehow. If there’s a supposedly complete theory of reality that can account for all of the data of observation and experiment but that can’t account for the reality of consciousness, then that theory cannot be true.
The problem is that physical science works with a purely quantitative vocabulary, whereas consciousness is an essentially qualitative phenomenon. In saying that consciousness is “qualitative” I simply mean that it involves qualities. Think about the redness of a red experience, or the sweet smell of flowers, or the taste of coffee. You simply can’t capture these kinds of qualities in the purely quantitative language of physical science. This was well understood by the founder of physical science, Galileo. Galileo only ever intended physical science to be a partial description of reality. He hoped that it could deal with the mathematical properties of reality, but he never dreamt that it could capture the qualities of experience, which Galileo believed resided in the soul. Indeed, I argue in the book that the reason physical science has done so well is that Galileo kicked things off by placing consciousness outside of its domain of enquiry, thereby giving physical scientists a more manageable task.
This historical fact is important, because it undermines a common argument for materialism. Although the problem of consciousness is taken very seriously, many argue that the astonishing track record of physical science in explaining more and more of our universe ought to give us confidence that, if we just plug away with our standard methods of investigating the brain, we’ll one day crack the mystery of consciousness. But this argument is not supported by the history. As a panpsychist, I would argue that physical science has done so well precisely because it was never designed to deal with consciousness.
After concluding that materalism can’t solve the mind/body problem, you argue that idealism and dualism have overwhelming problems of their own; therefore, some version of panpsychism must be true since it’s the last candidate standing. But then you knock down most versions of panpsychism based on what you call the subject irreducibility problem. Finally, you argue for cosmopsychism, according to which all things are mind-like but they are grounded not in fundamental “particles” of consciousness (downward grounding “by analysis”), but in the cosmos as a whole (upward grounding “by subsumption”). Let’s unpack a couple of these ideas, starting with subject irreducibility. The core of your argument that subjects (like human consciousness) are irreducible is that the grounding by analysis relation can’t explain how micro-level facts ground macro-level facts about consciousness. I admit to having a hard time following this argument, which you describe as an “extremely powerful challenge to constitutive Russellian monism.” Could you sum up this argument here?
It is a complicated argument, but I’m personally convinced there’s something right about it. The basic idea is that you can’t analyse facts about conscious subjects into facts about more fundamental things. Contrast with the case of a party. All it is for a party to exist is (roughly) for people to be gathered together having a good time. In this sense, we can analyse what it is for a party to exist in terms of something more fundamental: people. I don’t think you can do the same with my conscious mind. You can’t analyse what it is for me to exist in terms of more basic entities. You couldn’t say, for example, “All it is for Philip to be conscious is for lots of little conscious things to be grouped closely together in the right way.” That’s just not what you mean when you say “Philip is conscious.” When we talk about “parties” this is really just a disguised way of talking about something else, namely people. But when we say “Philip is conscious” we are making a basic claim about reality that can’t be analysed in more fundamental terms.
There’s a lot more to the argument, but that’s the basic idea. And if it’s right, we have to find some other way of fitting conscious subjects into reality. That’s where “grounding by subsumption” comes in; it’s an alternative model of how conscious subjects fit into our overall picture of reality. The idea is that even if we can’t fit conscious subjects into reality via analysis, we might be able to fit them in by subsuming them in a more expansive entity, such as the universe.
I wish I knew a way to put all this stuff more simply… I don’t even try to talk about it in my writing aimed at a general audience ….
If the universe as a whole is conscious in some manner where does its consciousness reside if not in and between its constituents? If the universe as a whole is conscious and the speed limit for any causal influence is the speed of light, is it then the case that the universe’s moments of experience are incredibly slow compared to ours?
Cool question! Although I think I’d reply by saying this question only arises when one’s still thinking in the “lego-brick” mindset. For the cosmopsychist, the fundamental level consists of just the universe, and it has consciousness itself as a single, unified whole. The universe comes first; the parts come later (obviously, I’m using temporal vocabulary slightly metaphorically here).
You describe an ostensible infinite regress problem in terms of the combination of smaller entities being comprised of yet smaller entities, etc. But why couldn’t there be a fundamental level of micropsyches just as there is (as far as we know) a fundamental level for microphysical entities? This would seem to make a lot of sense in the panpsychist worldview since matter and mind are generally conceived by panpsychists to be two sides of the same coin.
Good point. At the stage of the book you’re referring to (p. 212), I’m discussing what we mean when we say, e.g. “Tam is conscious”, and I’m trying to argue that we don’t mean “There are lots of little conscious things arranged in Tam in a certain way.” In the bit of the argument you’re focusing on, the thought is that if that was what I meant, then we’d want to apply the same analysis to the talk of conscious things in the second sentence. So “There are lots of little conscious things arranged in Tam in a certain way” really means “There are lots of little conscious things inside the little things inside Tam.” But then we’d have to do the same again, and again, and again, and again, ad infinitum. I don’t think it’s plausible that that’s what I mean when I say “Tam is conscious.” So to repeat: in the first instance it’s just a claim about meaning not ontology.
I actually have no problem with my consciousness being described as the combination of smaller conscious entities arranged in certain ways (shared resonance leads to combination, in my preferred solution to the combination problem). And under the mind/matter dual aspect nature of panpsychism, if there is a fundamental physical level there would be an accompanying fundamental mental level, halting the infinite regress at that level. So I guess my question is why should consciousness not be subject to normal part/whole relations when it seems that everything else in the universe is?
But notice that my argument involves a claim about meaning. Do you think it’s plausible that what someone means when they say “Tam is conscious” is “There are lots of little conscious things arranged in Tam in a certain way”? Ok, so you might wonder what why I’m talking about meaning when what we’re really interested in is reality. But my account of analysis depends on there being such connections. This is the most complicated part of my work, and I’m not sure I can do the argument full justice here. Hope that doesn’t sound like too much of a cop out…
I’m sympathetic to the notion that all that exists is causal structure (cause and effect at many different scales of nature), but this is a view you argue fervently against in your critiques of materialism. You argue that physics has been successful because it has ignored the “concrete categorical nature” of things by focusing only on causal structure. This is a tendency that I’ve called “absent-minded science” in my work and I agree with the fact that physics adopted this stance and then basically forgot that it had ignored half of nature. But what if the solution here is something like Rosenberg (A Place for Consciousness: Probing the Deep Structure of the Natural World) has advocated: viewing nature as nothing but causal structure but that causal structure has both internal and external aspects (dual aspectism), in a view that we could call “causal panpsychism” for lack of an adopted term for this species of panpsychism? In this view, the “concrete categorical nature” of all things is just causal structure, but that causal structure is both matter/energy (external) and mind (internal).
I think there may be just a difference in terminology here. When I reject the view that all that exists is causal structure, I mean the view that the nature of the physical world can be entirely grasped in causal terms. I take it that’s not Rosenberg’s view; on his view, if you just knew the causal stuff you’d be missing out on something, namely the consciousness stuff. Rosenberg is an important pioneer of Russellian monism [the general approach to consciousness that Goff develops in his book, which is comprised of various types of panpsychism], so I hope we’re on the same side! Incidentally, David Chalmers told me that it was Rosenberg who persuaded him of the merits of what later became known as “Russellian monism.”
Rosenberg does argue that the universe should be viewed as nothing but causal structure, but all causes have dual aspects, both mental and physical. So mind/consciousness is built into the fundamental fabric of the universe in a way that makes it impossible for it to be an epiphenomenon (a meaningless side effect of some other more important phenomenon). This is also the case with Whitehead’s system. I guess my follow up question here is whether you see this dual aspect notion of the nature of causation as satisfactory in terms of incorporating causation into the mind-body discussion?
I’m broadly sympathetic to this kind of picture. I’d probably want to phrase things slightly differently, but it’s just a question of details. I defend what I call the ‘consciousness+ hypothesis’, according to which each conscious state is an aspect of a more expansive state which has mental and non-mental aspects. My view is that it is the whole consciousness+ state that is causally efficacious, and that the conscious state gets to be causal as part of that state. I certainly reject epiphenomenalism.
One of my goals in this dialogue is to connect some dots between Russellian monism and Whitehead’s process philosophy — what he called “the philosophy of organism.” This connection is more than a little poetic because as you know Russell and Whitehead worked together for over a decade on their magisterial multi-volume Principia Mathematica, which tried to ground mathematics in logic — before that effort was blown up by Gödel some years later. Their collaboration ended before they both focused separately on metaphysics in a number of later works. Both were arguably panpsychists in some manner. So here’s my question: have you examined Whitehead’s metaphysical system, based on “actual entities” and “eternal objects” as the fundamental atoms of reality, with actual entities oscillating in each moment between mental and physical “poles,” as a species of Russellian monism as you frame it in your work?
I’d like at some point to write something on the connections between Whitehead and Russellian monism. I do admit this was something lacking in the book. My student Marta Santuccio is currently working on this topic, so maybe I’ll just copy what she says … It’s possible I’m dodging the issue because I disappoint a lot of panpsychists when I say the following: I’m actually a fan of substance ontology… please, don’t hate me…
I don’t really have any arguments one way or the other, but, when I reflect on my conscious experience, it just seems to me more natural to describe it in substance/attribute terms (i.e. as an enduring thing with changing characteristics) rather than as a process. And I don’t get the arguments for thinking of things in terms of process. Of course, that’s not to deny that there are events and processes, but I don’t see why we can’t analyse them in terms of facts about individuals and their changing properties. Maybe you can help me see where I’m going wrong…
Well, Whitehead would accuse you of being a “substantivalist” and overlooking the manifest passage of time in our lives. But you’re in good company because most physicists today are substantivalists … The alternative that Whitehead and others have fleshed out is an event-based ontology. Events are happenings in time — processes — and thus the “process philosophy” designation. Under this view, properties exist but they attach to events rather than static things (which don’t actually exist, since there is always a passage of time). And that’s why the event-based ontology is proposed: to incorporate the fact that there is always a passage of time in our consciousness and the universe more generally. Anyway, this leads to my next question, about essentialism. You discuss essences and essentialism many times in your book, but aren’t these notions only in our minds (a la Hirsch, who you discuss at p. 261)? Doesn’t this use of the notion that things have essences go against the strong evolutionary trend in philosophy and science, in its recognition that all things are always changing (here we get process again), and there can be no essences in things/processes that are always changing? Under this view, any identified essences are linguistic conventions, useful sometimes but not ontologically real.
But why do we have to think of substances as static? Why can’t they be constantly changing? That’s what I don’t get.
I’m glad you’ve brought up the topic of essence, as it would be useful to clarify my views on this. The notion of “essence” I employ is very thin. I think properties have essences, but all I really mean by that is that the nature of the property can be grasped by the intellect. So, for example, you can understand what it is for something to be a triangle: it means something is a three-sided figure. It’s really just a commitment to the intelligibility of reality. And it’s completely consistent with properties being messy and vague, and having historical definitions, a view which I find plausible in the case of biological species. So my commitment to essence doesn’t mean I think everything has nice, tidy, eternally unchanging definitions.
You worry that under the perdurantist “stage theory” view, that we are best conceived as 4-dimensional wormlike beings rather than persisting in some manner through time, then “my conscious mind, that thing I know with certainty to exist, won’t be around in a couple of seconds’ time. It will be replaced by some other conscious mind, which will be very similar to it, which share its memories, but which won’t be me… This is precisely the content of the fear of imminent death: the fear that I won’t exist in the near future. … I am dying every second. … [This] is a radically sceptical scenario, different in degree but not in kind to solipsism or the hypothesis that I am a brain in a vat.” This “perpetual perishing” of each actual entity, in each moment, is at the heart of Whitehead’s process philosophy and I personally have no problem with the idea of my perishing in each moment, partly because I’m also reborn in each next moment (as are all actual entities) until bodily death occurs. But here’s my question: in what sense do you endure in each present moment under your preferred presentist view? Isn’t literally every physical and mental aspect of you changing in each moment in some manner (except your name…), even in a traditional materialist worldview? This is also a Buddhist view (“no-self”) and I personally don’t see any way around it except to argue for some personal essence/soul that doesn’t ever change, and that seems a big lift indeed given everything else we know about the universe.
Maybe it’s just that I haven’t been enlightened yet, hence why perpetual perishing still gets me down a bit… When it comes to individuals, I basically think there are primitive facts about identity and difference. Here’s a thought experiment. Suppose the universe is symmetrical, so that everything happening in this half is perfectly mirrored by what’s happening in the other half. So there are twin versions of us right out on the other side of the cosmos having exactly the same discussion as we’re having here. You have all the same properties as Twin Tam (your twin on the other side of the universe) but you’re not the same person. Why not? I don’t think there’s any answer to that question. You are you, and Twin Tam is Twin Tam, and that’s all there is to it. So I don’t think there’s anything that makes you you, beyond the fact that you are you. Maybe you and I could swap all of our physical and psychological properties, though a series of operations and memory implants. But I would still be me and you would still be you. Why? Simply because I’m me and you’re you.
I should add, my slightly old-fashioned views about essence, substance and identity are not an essential part of Russellian monism. In fact, probably most Russellian monists would be more on your side on these issues. Russellian monism is a broad framework for taking consciousness seriously without collapsing into dualism, and there are many different ways of spelling out the details.
Staying with the more spiritual implications of your work, let’s look at the nature of the new kind of philosophy and science you advocate. You write: “[T]rue post-Galilean metaphysics hasn’t yet begun. It might not work, but it’s worth a try.” Your key suggestion for this new approach to metaphysics is to allow “carefully considered intuitions concerning the nature of phenomenal consciousness” to be included in our considerations, along with parsimony, common sense, empirical data, etc. I agree entirely with this suggestion and it is the core of what Wilber has called “deep science.” Others (Richard Smith) have called this way of obtaining knowledge “introception,” referring to the broader universe of intuition, including spiritual revelation. Do you intend your carefully considered intuitions to include various kinds of spiritual revelation and intuition or are you limiting your suggestion in this regard to more everyday domains?
No, I wouldn’t say this was spiritual at all. I’m not hostile to spirituality, and in fact I’d like to write something on this topic at some point, but I don’t think the claims of this book have any connection to spirituality. I think of the reality of consciousness as a cold-blooded scientific datum, just one that is known through introspection rather than observation and experiments.
Similarly, what are the spiritual implications of cosmopsychism for you? I also have to ask if you’ve dabbled in psychedelics at all since, as someone who has tried most of the psychedelics available, I smell a whiff of them in your work. Forgive me if I’m just projecting…
Again, I don’t think there are any spiritual implications. Of course, there is a cultural association between certain spiritual views and the idea that the universe is conscious. But I argue for cosmopsychism simply as the best explanation of consciousness. Sorry that’s a bit of a boring answer!
I think it’s important to emphasize this. Many panpsychists and cosmopsychists are complete atheists who are just trying to explain consciousness. If, on independent grounds, one is motivated to defend the universe having a spiritual aspect, then perhaps panpsychism/cosmopsychism has certain advantages. But that would be a further step, and I don’t think any of the arguments I make in my book point in that direction.
Having said that, I think perhaps panpsychism can make us conscious beings feel a little more at home in the universe, and a little bit closer to nature. I suggest in my new book that panpsychism might help us deal better with the climate crisis.
Last, how does your cosmopsychism differ from notions of idealism like Berkeley’s or (more recently) Kastrup?
Berkeley thinks the fundamental constituents of the world are human and divine minds (and perhaps animal ones too…I’m not sure what Berkeley thought of animal minds). The table exists because it is an idea in the minds of humans or of God. The problem with this kind of idealism, at least once you take God out the picture, is that it’s hard to explain the commonalities in our experience. Why is it that two people in the same room both see a table? The panpsychist can give exactly the same answer as the materialist: It’s because there is a table, and light bounces off it and goes into their eyes. The only little detail I add is that that the table is made up of things that are minimally conscious.
I find Kastrup’s view really interesting, but I guess I find it hard to make sense of the view that we’re all literally the same mind. I have a paper called ‘There Is More Than One Thing’ in which I argue against the coherence of this kind of radical monism according to which there is literally only one mind. I’d rather say there’s only one fundamental mind, which subsumes all the other minds as aspects.