Book review: Giulio Tononi's Phi: A Voyage from the Brain to the Soul

Phi is easily the most unusual book on consciousness I have read in awhile. It’s hard to describe, but Tononi makes his case for “integrated information” using poetry, art, metaphor, and fiction. Each chapter is a fictional vignette or dialogue between characters inspired by famous scientists like Galileo, Darwin, or Francis Crick. At the end of every chapter is a “note” written in normal academic language explaining the context of the stories. On just about every page there are huge full-color glossy pictures of famous art. The book is simply beautiful as a physical object in an attempt, I suspect, to convince qualiaphiles that Tononi is “one of them”.

The theory of integrated information itself, however, is less appealing.  Here is how integrated information is defined:

Integrated information measures how much can be distinguished by the whole above and beyond its parts, and Phi is its symbol. A complex is where  Phi reaches its maximum, and therein lives one consciousness- a single entity of experience.

And with that Tononi hopes the “hard” problem of consciousness is solved. However, the intellectual weight of Phi  rests on a thought experiment involving a photodiode. A photodiode discriminates between light and no light. But does the photodiode see the light? Does it experience the light? Most people would think no. But the photodiode does integrate information (1 bit to be precise) and therefore, according to the theory of integrated information, has some experience, however dim. The theory of integrated information is therefore a modern form of panpsychism based on the informational axiom of “it from bit”. For obvious reasons Tononi downplays the panpsychist implications of his theory, but he does admit it. Consider this quote:

“Compared to [a camera], even a photodiode is richer, it owns a wisp of consciousness, the dimmest of experiences, one bit, because each of its states is one of two, not one of trillions” (p. 162)

The reason the camera is not rich is because it can be broken down into a million individual photodiodes. According to Tononi, the reason why the camera has a low level of  Phi compared to a brain is that the brain integrates information between all its specialized processors and the camera does not. But nevertheless, each photodiode has a “wisp of consciousness”.

Tononi also uses a thought experiment involving a “qualiascope”, a hypothetical device that measures integrated information and can therefore be used to detect consciousness in the world around us. In the vignettes, Tononi writes that when you use the qualiascope:

“‘You’ll look in vain at rocks and rivers, clouds and mountains,’ said the old woman. ‘The highest peak is small when you compare it to the tiny moth'” (p. 222).

This is how he downplays his panpsychism. Notice how he doesn’t say that rocks and clouds  altogether lack consciousness. It’s just that their “highest peak” of  Phi is low compared to a moth. The important part however is that the  Phi of rocks and clouds is low but not nonexistent.

Why is this important? Because Tononi wants to have his cake and eat it too. To see why just look at some of his chapter subtitles:

Chapter 3 “In which is shown that the corticothalamic system generates consciousness”
Chapter 4 “In which is shown that the cerebellum, while having more neurons than the cerebrum, does not generate consciousness.”

 This is because Tononi admires the Neural Correlates of Consciousness methodology founded by none other than Francis Crick, who has a strong intellectual presence throughout the book. According to most NCC approaches, consciousness seems to depend on “corticothamalic” loops and not just specialized processors alone (like the cerebellum).This finding comes from research correlating behavioral reports of consciousness with activity of the brain. When most people report being conscious, higher-order system loops are activated. And in monkey experiments the “report” is a judgement about whether they see a stimulus, which can be made by pressing a lever. What they find in the NCC approach is that consciousness seems to depend on more than just specialized processors operating alone. It requires a kind of globalized network of communicating modules to “generate” consciousness.

It should now be plain as day why Tononi is inconsistent in trying to have his cake and eat it too. If a lowly inorganic photodiode has a “wisp of consciousness”, then clearly, by any standard, a single neuron also has a wisp of consciousness, as well as the entire cerebellum. Tononi acknowledges this:

“Perhaps a whiff of consciousness still breathes inside your sleeping brain but is so feeble that with discretion it makes itself unnoticed. Perhaps inside your brain asleep the repertoire is so reduced that it’s no richer than in a waking ant, or not by much. Your sleeping  Phi would be much less than when your brain is fast awake, but still not nil” (p. 275).

“Early on, an embryo’s consciousness – the value of its  Phi – may be less than a fly’s. The shapes of its qualia will be less formed than its unformed body, and less human than that: featureless, undistinguished, undifferentiated lumps that do not bear the shape of sight and sound and smell” (p. 281)

” Phi may be low for individual neurons” (p. 344)

But if a single neuron has a wisp of consciousness, then clearly consciousness is not “generated” by the corticothalamic system. It is instead a fundamental property of matter itself. It from bit. What Tononi means to say with his chapter subtitles is that “The corticothalamic system generates the right amount of  Phi to make consciousness interesting and precious to humans”. The difference between the photodiode and the corticothalamic system is a difference of degree. The corticothalamic system has a high enough level  Phi such that it makes an interesting difference to human experience such that we can report or notice it, distinguishing coma patients (very low  Phi) from awake alert adults (very high  Phi).

But now there is an interesting tension in Tononi’s theory. If there is a low but nonnegligible amount  of  Phi in a human embryo, Tononi’s theory must now figure out how to make a cut-off point between the lowest amount of  Phi we actually care about so we can figure out when to stop giving people abortions. Until Tononi answers that question, his “solution” to the hard problem of consciousness is fairly disappointing. He came up with this notion of integrated information to explain qualia, but now we are faced with the difficult question of “How much  Phi is necessary for us to care about?” Clearly no one really cares about the “wisp of consciousness” in a photodiode. So having solved the “hard” problem of qualia, Tononi just creates an equally difficult problem: how to figure out the amount of  Phi worth caring about from a moral perspective. And he plainly admits he hasn’t solved these problems.

But for me this is a huge problem. You can’t have your cake and eat it to if you are a panpsychist. You can’t say that photodiodes are conscious but then say the only interesting consciousness is that of corticothalamic systems. This seems rather ad hoc to me; a solution meant to fit into prexisting research trends. If you are a panpsychist you should embrace the radical conclusion. According to  Phi theory, Consciousness is everywhere. It is not “generated” in the brain. It only reaches a high level of  Phi in the brain. And if that’s the case, then the entire methodology of NCC is mistaken. NCC is not a true NCC but rather the “Neural Correlates of the Amount of Consciousness Humans Actually Care About”.

Overall conclusion: Phi is an interesting book and worth borrowing from the library. But I wouldn’t say it adequately solves the hard problem of consciousness. Not even close. What it does is arbitrarily stipulate criteria for pointing out consciousness in nonhuman entities. But Tononi never makes a real argument beyond appeals to intuition for why we should accept a definition of consciousness such that the ascriptions come out with photodiodes having a “wisp” of consciousness. I think most people will want to define stipulation criteria such that the ascriptions come out with only biological creatures having consciousness. Panpsychism is just too radical for most. So while I applaud Tononi for exploring this ancient idea from a modern perspective, I ultimately think that when people truly understand that Tononi is a panpsychist they will be less attracted to it despite its close relationship to Francis Crick and the wildly popular NCC approach.



Filed under Consciousness, Philosophy, Psychology

17 responses to “Book review: Giulio Tononi's Phi: A Voyage from the Brain to the Soul

  1. You beat me to the punch, Gary! I’ve been planning a review of my own, but more focussed on the rhetorical tactics used in the book. Schwitzgebel recently tackled Tononi for his series of Is America Conscious Given Theory x reductios, and focussed on what he termed the ‘nesting problem’: the problem is that phi seems to raise the possibility of consciousnesses *within* consciousnesses (Tononi recognizes this in his most recent papers and calls it the superposition problem). It is a prickly pear indeed, and more problematic for Tononi because he is committed to it.

    As far as the panpsychism goes, I’m not convinced that Tononi is waffling on the issue or just plain vague because he’s genuinely agnostic on whether some kind of neurochauvinistic emergence is involved. He is a scientist, after all, and so I think should be forgiven for not checking all the metaphysical boxes. Imagine if Bohrs (especially Bohrs!) or Einstein had metaphysical hangups!

  2. Check out Koch’s recent portrayal of Tononi’s view in his own 2012 book, if you haven’t already. In some ways, it’s clearer and more consistent about the implications of Tononi than Tononi himself is — though Koch pulls back from fully endorsing the view. Notable: Koch does *not* discuss Tononi’s recent exclusion/anti-nesting principle.

  3. Eric Schwitzgebel

    P.S. — Whoops, Gary, I see you have discussed exactly this in an earlier post. Sorry I missed that!

  4. Nick Mann

    This is the most comprehensive and knowledgeable review of Tononi’s opus I’ve read so far. Which may not sound like the compliment it’s meant to be.

    He doesn’t get into the issues of emergence from specific physics versus multiple realizability and stuff like that, preferring to remain agnostic. (In the online video of a lecture he gave about 3-4 years ago he says, regarding in silico true intelligence, that any substrate that can generate phi can, well, generate phi. Don’t make enemies you don’t need.

  5. Matt Sigl

    Tononi makes no qualms about his panpsychism. His papers make it clear: ANY system that generates information over and above its parts, over a certain timescale, is conscious. In his lecture posted on YouTube he admits he is a panpsychist, and Christof Koch, in his book, also is clear that the IIT is a panpsychist theory. So, if that’s not something obvious in PHI, I doubt it’s anything like dodge on Tononi’s part; he just wanted to make his point about the brain’s relationship with consciousness clear. There is no question that he considers consciousness “a fundamental property.” Even in the book Tononi claims that shapes in qualia space are the “only thing that’s really real.” I think the fundamental nature of consciousness is something the IIT is quite up front about.

    As for the nested consciousness problem, after reading the book I still think the issue is a murky one for the theory, but nonetheless clearer. I think if Tononi’s view is right it would have to be the case that each mechanism of a system can only be part of one consciousness, since information cannot be counted twice. So, instead of an onion, as the metaphor goes in the book, it’s only the onion’s core, one specific consciousness, that remains. Another way to say this is that every causal chain of yes/no questions that a series of mechanisms “asks” can only ever lead to one consciousness: the qualia shape that is generated when the yes/no questions come to an end and there remain no more mechanisms available to break the symmetry. An instance of consciousness is the guess the system makes about how the world is after its very long game of “20 Questions.” Every mechanism only plays the game in one series; the series that generates the highest value for phi.

    FWIW I think the book is a masterpiece of thought and expression. But, I’ve been in the tank for the IIT for some time now so, maybe that was a foregone conclusion. Still, I notice there are other similar panpsychist ideas out there, bubbling up in scientific thought. This paper on “conscious realism” by UCIrvine professor Donald Hoffman was wholly convincing to me and works well within an IIT paradigm:

    Anyway, glad to see a substantial review of the book. They’ve been few and far between so far.

    • Matt: I actually don’t see anything in IITC (the theory, apart from the theorist)that implies panpsychism. It’s not information that it identifies with consciousness, but a rather strict constraint on information (integration), the same way life requires strict constraints on matter. I must be missing something here.

      Also, I found Hoffman’s piece very interesting, but not nearly so radical as he seems to think, which again leads me to think I’m missing something important. One almost thinks he reads the term ‘representation’ as meaning the same thing (HFD) wherever he encounters it, when many theorists implicitly follow, say Clark’s usage, which strikes me as very similar to what Hoffman is driving at with MUI. Maybe I’ve just become accustomed to reading the term in a non-isomorphic, heuristic sense. Mechanistic explanation in cognitive science generally adopts the ‘representation – computation’ model as a facon de parler – even Dennett uses it!

      But again this could be an artifact of my non-institutional study of these things. I was under the impression that the work on ecological rationality (Gigarenzer especially) had pretty much swept the table. (If not, then it will).

      It did get me wondering, however, if anyone in PoM has taken systems-theory approaches to the problem of consciousness. I found Hoffman’s discussion reminiscent of Luhmann at a couple of turns, and the notion that autopoietic systems manage environmental complexity via adaptive complexity reduction. Which is just to say that if Hoffman’s notion is that its heuristics all the way down, I’m not sure what the big whup is. Again, I have the feeling that I’m missing something basic.

      Otherwise the larger position, Conscious Realism, strikes me as a terminological twist on idealism, which I just have to reject on cognitive psychological grounds. Characterizing the ‘all’ on the basis of the ‘for me’ is a ubiquitous human conceit. For the man with a hammer, as the saying goes, every problem is a nail – or an *icon,* as the case might be.

      • Matt Sigl


        The paper is radical because it argues that ALL our perceptions are best seen, not as representations of truth, but merely representations that facilitate useful behavior. These representations, as he points out in his paper, can be as different from the real thing as a operating system like Windows is to the internal causal dynamics of the circuit board. If that’s not a radical shift in the way we view our grasp of “reality” I don’t know what is. It’s not just that our minds represent the world in an incomplete way – we know that already – it’s that the world may look NOTHING like our perceptions at all.

        For what it’s worth I think Hoffman’s “conscious realism” is absolutely a modern version of idealism. But I see no points against it for that. (I never can quite get Bishop Berekely to stop yammering in my head.)

        Can you explain more about why you reject idealism on “cognitive psychological grounds.” Seems to me that reason alone is the way in which we can break out of our “conceit” and model the way things are in themselves, or at least attempt to. You seem to subscribe to a radical Kantian division between phenomenal and noumenal, the latter of which, in principle, we can have no access to. At ALL. After all, being generated by our brains, won’t ALL attempts to articulate how things “really are” fall into this “fallacy”? I don’t agree but, it’s a serious POV. Is this an adequate representation of your criticism?

      • It could just be what I’ve been reading, a consensus fallacy thing. I just don’t think verisimilitude has much to do with ‘representation’ any more with a majority of philosophers of mind. I thought it was the logical relationships anyone was really interested in with orthodox representationalists. I think most have problems with the very notion of the ‘world looking like for real.’

        They’ve been heuristic cartoons, recapitulating structure more like DNA and proteins for me for quite some time. For me the question of access, is just that, access. No division into the ‘phenomenal’ and the ‘noumenal,’ just a brain churning information in an environment. Mysteries lurch in all directions, IN as much as OUT, but with a three pound brain, go figure.

        If you want some kind *intentional* relationship with the world, it’s hard not to find yourself continually sealed in. Between Hume and Wittgenstein, subjective and normative constraints, objectivity becomes a difficult thing to understand. But if you just look at it in mechanistic terms, we are plainly continuous with our environments. People can argue where to impose the boundaries, in the skull or five feet away, but the assumption of continuity with a absolutely dwarfing environment is going to be mighty hard to rattle… In addition to explaining consciousness.

        In cognitive psych terms: as an outsider confronted with Hoffman’s claims regarding the metaphysical centrality of consciousness, I simply look to the history of science and see the slow exorcism of anthropomorphisms from the natural world. It seems clears to me that we’re powerfully predisposed to do this, assert the centrality of what seems most immediate, so there are grounds to at least suspect Hoffman in this regard. The converse notion seems to me something hard won.

  6. Gary-

    It sounds like I would not like this book. I am a panpsychist, but I have no patience for “it from bit”, which I consider Platonism run amok. What is “pure information” but an abstraction, a figment of our intellect? What is integrated information? Information, as I have seen it in the wild, is always something real, actually happening: a semiconductor discharging, a ball rolling, etc. Any “integration” must be some physical causal network, playing out over time. From your posting, Phi seems like soft-focus, squishy panpsychism.

    Nevertheless, you say “Panpsychism is just too radical for most” – as a philosopher, do you really want to voice that sentiment? The same could be said, in their eras, of heliocentrism or natural selection. As philosophers, as I heard one elderly professor say once, we have nothing to gain by being conservative. I would argue that when examined closely, all the other competing theories are, at least in some sense, more radical than panpsychism.

    But mostly, I don’t think you are being quite fair to Tononi in taking him to task for not having worked out the ethical and social implications of his ideas. The universe does not owe us moral comfort. Sometimes we find out truths that knock us for an ethical loop. If Tononi thinks he has figured something out about how the world is put together that opens up some sticky questions about whether or when abortions are ethically problematic, the fact that he does not solve those problems about abortion does not have any bearing on the truth of the things he thinks he has figured something out about how the world is put together. Don’t shoot the messenger. The moral problems are for us, as a society, to work out over the course of time.

    -John Gregg

    • Gary Williams

      Hi John,

      The “it from bit” stuff doesn’t make a huge appearance in the book, but he does emphasis the “irreducibility” of integrated information in a way that might be to “soft” for you.

      As for its radical nature, you say “The same could be said, in their eras, of heliocentrism or natural selection.” However, there is a crucial difference between panpsychism and heliocentrism or natural selection. Galileo could look through his telescope and see the truth of heliocentrism. Darwin could look at Finch beaks and see the truth of natural selection. But it’s impossible to look at nature from the third-person perspective and “see” that there is phenomenal consciousness there because by its nature it is private and subjective.

      Darwin presented compelling reasons for believing in natural selection. Once you look at all the evidence, you can’t help but believe in evolution. But I see no such analogy with panpsychism. Obviously brain science itself doesn’t directly support the idea that a lowly photodiode experiences the world. No science does. It requires a leap of faith. A leap based on intuition. But my intuition screams against such a possibility.

      However, I believe the problem is deeper than just lack of empirical evidence. I am deeply skeptical about the coherence and scientific legitimacy of the term “phenomenal consciousness” or “qualia”, which is Tononi’s target explanation. As is widely recognized, it is impossible to noncircularly define phenomenal consciousness. Now, you might say, like Koch or Crick, that “Well what if we had tried to define ‘gene’ before we completed the science! We would have gotten it all wrong! Definitions come last, not first.” But this is a false analogy. For we could have formulated a good noncircular definition of “gene” prior to the science that discovered what it actually is: a gene is a hypothetical mechanism that has something to do with passing hereditary information. Nothing circular there. And it helps the science by getting us to focus our efforts in finding such a mechanism. Koch is confusing an attempt to define a gene as figuring out exactly what it IS, but we knew prior to investigating what it was in a more general sense: something that had to do with heredity. Nothing similar can be said with phenomenal consciousness. Whereas we knew a priori that genes had something to do with heredity, we have no a priori idea what phenomenal consciousness is, what it’s for, or how it could exist. We are clueless. I think this is an inevitable result of grounding the concept in a phrase as ambiguous as “what it is like” and then thumping the table at skeptics and saying “if you gotta ask you aint never going to know”.

      I have no problems with consciousness studies so long as we define consciousness as something like introspection or reflection, which can be defined noncircularly. But when we attempt to do science of phenomenality, I am deeply skeptical. And this is why I am ultimately unsatisfied with Tononi’s “solution” to the problem. The problem of what? I have no idea! The whole concept of qualia is mysterious to me. It signifies nothing real that can be studied by science. Panpychism removes the mystery only by saying the mystery is everywhere. Qualia seems mysterious in humans? Well, the mystery doesn’t go away if we say that qualia exist for photodiodes and rocks. That just confirms my suspicions that I have no idea what people mean when they use the term “qualia”.

      With that said, your point about the objectivity of truth vs tricky moral questions is problematic. With natural selection and evolution, the point is true: we shouldn’t let moral questions dictate the objective truth of evolution. However, I don’t think a theory of qualia is on the same “objective” footing as the theory of evolution. I think in the study of qualia the right metaphor is not one of “finding truth” but rather “stipulating truth”.

      • VicP

        What I don’t think you can dispute Gary is that as you move up the ladder of nature from physical particles, ionic bonding, covalent bonding, complex organic molecules and biological cells; with coherent complexity of repeatable structures also come slower structures that manifest at higher/slower dimensions in space and time. Cellular evolution brings more complex organisms with specialized systems for movement in time and space. All other body systems are characterized by some form of coherence of physical structure at the cellular level to perform some specialized function at an appropriate level of time and space for the organism, accompanied by physical coherence are functions like filtration (kidneys), pumping (cardiac muscle), chemical production (liver) etc. Until the development of microscopes and advanced biological science, the understanding of coherence at the cellular level of development did not causally connect at the physical level of time and space for these systems and functions.

        For neural systems however, the functional coherency at the cellular level leads to the metaphysical question of qualia, feelings, consciousness etc. The appearance or epiphenomalism may not square yet with the understanding of physical extension properties of the previously mentioned systems into space and time, until one realizes that for neural tissue, nature has perhaps extended inner cellular function not outside of cells (appeared dualism) but rather between cells. To operate at the dimensions of space and time required of early advanced biological systems (bugs, frogs, mice etc.) requires not as primary an appearance of time and space but the primary required detection of objects. If the brain is the primary organ of objective reality, the phlogeny of brain evolution extends from detection from a primary object which begin at the subjective. What’s this bicameralism all about? The question can be answered by development of complex motor systems which requires bicameral division of motor systems and environmental detection/motor control systems for balance and movement. What is this primary object of subjective reality? When I got out of bed this morning my feet did the same thing they’ve always done for my remembered adult life, rather than naturally walk on the walls or ceiling; my feet hit the floor (and my organ of objective reality got closer to the ceiling). We need to deal with the gravity of this situation.

  7. Matt Sigl


    Just to get an understanding of your thought process: What are your thoughts on the Zombie argument for substance dualism?

  8. Pingback: A Brick o’ Qualia: Tononi, Phi, and the Neural Armchair « Three Pound Brain

  9. Pingback: Phi: An Overview | Headbirths

  10. loz

    John Gregg: “Any “integration” must be some physical causal network, playing out over time.”

    This is EXACTLY what IIT posits.

  11. Reblogged this on janpilotti and commented:
    Very clear argumentation! I will definitely writings more. your.
    In the meantime allow me to give link to my latest poster presented onTowards a Science of counsciousness:
    Conscious Spacetime. A possible connection between phenomenal properties and six dimensional spacetime.
    Best regards
    Jan Pilotti M.D. B.Sc. (math, theoretical physics)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s