Friday, March 23, 2012

Why Tononi Should Think That the United States Is Conscious

The is the fourth and probably last in a series of posts on why several major theorists of consciousness should attribute literal "phenomenal" conscious experience to the United States, considered as a concrete but spatially distributed entity at least partly composed by citizens and residents. Previous posts treated Dennett, Dretske, and Humphrey. Humphrey and I have an extended exchange in the comments field of my post on his work, and I have offered general considerations supporting the view that if materialism is true the United States is probably conscious here, here, and here (page 18 ff). A full-length paper is in the works but not yet in circulatable shape.

I chose Dennett, Dretske, Humphrey, and Tononi as my sample theorists for two reasons: First, they represent a diverse range of very prominent materialist theories of conscious. And second, they are theoretically ambitious, trying to explain consciousness in general in any possible organism (and not just human consciousness or consciousness as it appears on Earth, like most scientific and neural accounts), covering the metaphysics from top to bottom (and not, say, resting upon a relatively unanalyzed notion of "representation" on which it would be unclear whether the United States literally has the right sort of representations).

Of our four theorists, neuroscientist Giulio Tononi’s view (2004, 2008; Balduzzi and Tononi 2009) enables the quickest argument to the consciousness of the United States. Tononi equates consciousness with “integrated information”. “Information”, in Tononi’s sense, is abundant in the universe – present everywhere or almost everywhere there is causation. And information is integrated, at least in a tiny degree, whenever there are contingent causal connections within a system with a bit a structure – a system that is not collapsed into maximum entropy. Since integrated information is pervasive, so also, Tononi says, is consciousness. He says that “even a binary photodiode is not completely unconscious, but rather enjoys exactly 1 bit of consciousness” (2008, p. 236; cf. Chalmers 1996 on thermostats). Likewise, Tononi attributes “qualia” (that is, consciousness) to simple logical AND and OR gates (Balduzzi and Tononi 2009). On Tononi’s view, what distinguishes human consciousness from photodiode consciousness, OR-gate consciousness, and speck-of-dust consciousness is its richness of detail: The brain is massively informationally complex and integrated, and thus enjoys consciousness orders of magnitude more complex than that of simple systems.

Before we saddle Tononi straightaway with commitment to the consciousness of the United States, though, there is one issue to address: Despite the liberality of his view, Tononi does not regard every putative system as an “entity” that could be the locus of consciousness. If a putative system contains no causal, that is, informational, connections between its parts, then it is not an entity in the relevant sense; it is not, he says, a “complex”. Also, a putative system is not a conscious entity or complex if a larger, more informationally integrated system entirely subsumes it. For example, two disparate nodes do not constitute a conscious complex if a third node lies between them creating a more informationally integrated network. This restriction on the possible loci of consciousness is still extremely liberal by commonsense standards: Complexes can nest and overlap, for example, within the brain, where tightly integrated subsystems interact within larger less-integrated systems.

It seems straightforward that residents of the United States also form multiple overlapping, causally connected complexes. Despite Tononi’s general caveat about what can legitimately count as an entity or a complex, there seem to be no Tononian grounds for denying that the United States is such an entity or complex and thus a locus of consciousness. Its subsystems are informationally connected, and it doesn’t appear to be subsumed within any more tightly informationally integrated system. (I’m assuming the world community and the Earth as a whole are not more tightly informationally integrated than is the U.S., but doesn’t matter for my ultimate argument if we relax this assumption and grant that on Tononi’s view it would be the world community or planet as a whole that is conscious, rather than the United States.) This conclusion seems especially evident given Tononi’s assertion that conscious complexes exist “at multiple spatial and temporal scales” “in most natural (and artificial) systems” (2004, p. 19). Choose the right temporal and spatial scale and Tononi’s view will deliver group consciousness.

The only question that would appear to remain is whether the United States is informationally integrated enough to have a rich stream of conscious experience, or whether its consciousness is substantially impoverished compared to that of a normal human being. This matter is somewhat difficult to assess, but given the massive informational transfer between people and the highly sensitive complex contingencies in human interaction, including in large-group interactions over longish time frames, I would think a plausible first guess from Tononi’s perspective should be that the United States (or world community), when assessed at the appropriate time scale, has at least as rich a stream of conscious experience as does a small mammal.

Update April 3:
In the comments section, Scott Bakker has kindly pointed me toward a new paper by Tononi. This paper seems to reflect a substantial change in Tononi's position with respect to the issues above. While I think the view above accurately captures Tononi's view through at least 2009, it will require substantial modification in light of his most recent remarks.

Update June 6:
See here for my reaction to Tononi's updated position.


Scott Bakker said...

Tononi's most recent account can be found here:

The references you cite have always made me curious since Tononi's therapeutic aim has always been to find some way to localize and quantify consciousness in the brain to help determine 'levels of consciousness' in those suffering traumatic brain events. So in his 2005 "Consciousness, information integration, and the brain" (collected in Progress In Brain Research 150) his interest is to show how the thalamocortical structural features capture the kinds of informatic intergration requisite for consciousness, whereas the structural features belonging to the cerebellum do not. So I've always just assumed that he's talking about a 'sliding scale' possessing some steep slopes. In which case, he could shrug his shoulders and say something like, 'Sure, Eric, but that isn't the kind of consciousness anyone is interested in. America? That's background radiation: it's the plasma that's interesting because that's where we're likely to figure it all out (because that's where we can see it in action).'

But I'm just guessing.

That hasn't stopped me from cribbing the idea of a 'consciousness Goldilocks zone' for my next SF novel!

Neil said...

Most neurobiological accounts of consciousness are motivated and justified by evidence regarding report. If they are accounts of consciousness, it is access consciousness they explain. It is nothing more than a fond hope that these accounts explain or ever describe phenomenal consciousness. But the US probably *is* access conscious.

Baron P said...

Is the US aware that it's conscious?

Eric Schwitzgebel said...

Sorry for the slow reply, folks! I had a feverish weekend of flu. (Still not wholly recovered.)

Scott: I agree that Tononi could quite reasonably say that as a practical matter people who aren't theorists of consciousness shouldn't care too much about the simple consciousness of simple or poorly integrated systems. You're right that this would fit with his clinical interests. At the same time, Tononi does himself seem to be interested in the simple consciousness of simple systems. Those are the sorts of examples he uses in developing his own theory in the papers cited, so I don't think he could quite make the "plasma" move. (Besides, CBR turns out to be hugely important and interesting to cosmology, after all!)

Crib away! If you end up using these ideas in a novel, I'd be interested to see.

Eric Schwitzgebel said...

Neil: Ah, the old A-consciousness/P-consciousness two-step! By "consciousness" I always mean what Block means by "phenomenal consciousness". If you're a skeptic about our capacity to figure out phenomenal consciousness, I wouldn't disagree!

Baron P: Not to judge by linguistic reports. But neither, then, are rabbits.

Baron P said...

I believe that rabbits are aware that they are consciously aware. Also they have the awareness of anticipation. The argument however that the US is conscious because it's aware would require that it also be alive to be aware of consciousness.

Scott Bakker said...

Your CBR point is taken. But my guess is that Tononi would say that other informational sciences have the CBR version of 'consciousness' well in hand. It's not that they need change anything because they're missing some 'special' ingredient.

Remember he has a specific pedagogic purpose in mind when using the diode and the camera: to illustrate the incremental, yet VAST, nature of the difference between the these and (highly integrated) human processors.

So in a sense, I think he would complain that, even though you are reading his examples correctly, you are misconstuing their significance.

To identify consciousness with information integration allows you to tar any system possessing differences making differences with the 'consciousness brush,' and to so license applications of the concept far beyond the comfort range of most everyone except panpsychics, thus putting skeptical pressure on the identification.

But it depends which way you're pressing, doesn't it? When you make the noo-informatic identification you're either suggesting that information is really something special, or that consciousness really isn't all that special after all.

For someone like myself, who thinks they can 'explain away' most all the things that make consciousness seem 'special,' I really don't feel that pressure. 'Sure,' I might say, 'it's a sliding scale, but if THAT'S the way you're going with it, we probably should just chuck 'consciousness' altogether, and talk varieties of information processing (culminating in the one you're living this-moment-now).'

And for Tononi, I suspect, it all comes down to the differences in 'phi' value. If the cerebellum can't pull together what it takes, I certainly can't see a system as NOISY and unintegrated (as Tononi defines it) as a bunch of god-fearing Americans passing muster!

(I'm afraid the Republican Primaries have put a substantial dent in your case!)

One of my long time readers is actually studying with him. I'll drop him a line and see if he won't steer Guilio this way...

Neil said...

Hi Eric,

My point is that theories like Tononi's - though I suspect he takes it to be about p consciousness - are always constructed on the basis of evidence about a consciousness. It takes several further inferential leaps (while closing ones eyes and hoping like hell) to get from there to p consciousness.

Arnold Trehub said...

The retinoid theory of consciousness exhibits integrated information (as does a Google server center), but, contra Tononi, its explanation of phenomenal consciousness depends on a putative neuronal mechanism that realizes a transparent brain representation of the volumetric world from a privileged egocentric perspective. In this model of consciousness, one could have a very low level of integrated information in one's brain and yet be conscious. In fact, consciousness is a necessary precondition for having any level of integrated *phenomenal* information. This minimal level of consciousness would correspond to C1 in retinoid theory. See "Space, self, and the theater of consciousness", here:

Eric Schwitzgebel said...

Arnold: How would your theory apply to blind group-mind aliens? (Or would it apply?)

Arnold Trehub said...

Hi Eric,

I have no idea how the group mind/brain of blind aliens might work. I would say, however, that if the brain of an alien is conscious, then he/she/it must have subjectivity -- a transparent brain representation of the world in which it exists from a fixed perspectival spatio-temporal origin.

The retinoid theory of consciousness is a claim about creatures in this world. It proposes that consciousness did not exist before the evolutionary emergence of the retinoid system, and that autaptic-cell activity in a brain's retinoid space is necessary and sufficient for consciousness to exist. For example, see here:

Anonymous said...

Eric, if the US is indeed conscious and you proceed to convince a lot of people of this idead - would that entail that you just have made the US conscious of it's own conciousness? Well done sir!

Matt Sigl said...

Is the US conscious according to the IIT? I don't think we know enough about what sorts of causal structures are necessary and sufficient to make a system integrated to know. Tononi certainly can't rule it out though.

In his lecture available on YouTube Tononi asks and then answers the question about Ned Block's thought experiment of whether a billion Chinese people on cell phones can be a single conscious entity. Tononi claims that he can't give a certain answer but stresses that getting a high value for phi requires a very unique and special causal structure. Not a lot of physical systems have this feature, or so it seems. There's a reason the brain is the most complicated matter that we know of.

I don't think a system having "multiple overlapping, causally connected complexes" is a complete enough criterion to warrant attaching a high phi value to it. Unless, that is, you already have established that "causally connected" implies integration, but that would be tautological as it would be proving America to have something (a phi value) that it already has by definition (causal connectedness). But, assuming that America IS complex and integrated enough to produce phi it would again be a danger to ascribe a human-like mentality to this system. As Tononi points out, the phenomenal aspects of a system, the actual content of its qualia, is specified by the geometry of the informational space over a certain timescale. Phenomenal experience is an abstract shape in multi-dimension qualia space. If conscious, a system like America is bound to have a very alien qualia space compared to a normal human mind. The shapes produced in this space are probably totally incomprehensible to us, even if the phi value is high. I would be reticent to ascribe to it the kind of intentionality we experience. Still, it's true that the IIT admits the possibility of such bizarre superordinate minds.

Why do we balk at this? I think it's challenging for us to a imagine a mind with no clear causal power. The "American mind" would presumably have no such power. Where would it be? Yet the IIT doesn't assume that OUR minds have causal power either. The relation of minds to action or physical causation is not an issue that Tononi discusses. His view could be seen as a kind of epiphenomenalism though Tononi would probably want to stay agnostic on the question. A monist view is hinted at but not explicit in the IIT as is an attenuated idealism, though it's not explicit or necessary for Tononi argument to go through.

Personally, I find the idea of an American consciousness (or "world consciousness") beautiful. Just as we peer deeper and deeper into the material nature of reality only to discover an infinite, expanding, inflating and splitting multiverse, it would be equally beautiful and awe-inspiring to discover that the reality of mind in the universe is equally baroque. An even deeper, potentially theologic question is whether the whole universe is itself an integrated system of information. Certain discoveries in physics—I'm thinking Bell's theorem and "spooky action at-a-distance"—may imply just such a structure. Tononi has a fascinating footnote about the quantum world in his IIT Manifesto.

I couldn't tell if this blog entry was attempting to imply that the possibility of an "American" consciousness makes the IIT a theory worth rejecting out-of-hand or if the idea, while a bit extreme, nonetheless deserves serious attention. My vote is for the latter.

Baron P said...

fficeya someMatt, best comment yet. Keep it up!

Baron P said...

I hate that anti robot device. It's consciously inhuman.

Scott Bakker said...

Wouldn't Tononi be in the same boat as Edelman with his 'phenomenal transform,' Matt?

Since information IS causality for Tononi, and consciousness is a function of information integration, how could consciousness be epiphenomenal?

The real question would be whether it was causal in the way that it assumed it would be.

Matt Sigl said...

I don't know about Edleman's "phenomenal tansform" so I can't comment on that, but your second point is an interesting one Scott.

I actually agree that Tononi's theory can be seen to imply an identity thesis between causation and information (or changes in information states over time perhaps) but I don't want to put those words in his mouth. Tononi isn't all that interested in fundamental metaphysical speculation. He is clear that he sees "consciousness" as a fundamental property in the same way that mass or charge are fundamental properties. He does not claim, however, that objective mass or charge can be themselves reduced even further into an information paradigm. (Ironically enough with the holographic principle it seems contemporary physics might be starting to do just that.)

Whatever the case for fundamental ontology, all the IIT needs to demonstrate is that consciousness is fundamental and, at the very least, corresponds to the causal flux of the world in certain lawful ways. I don't think Tononi would want to take the leap that I would endorse, which is that consciousness is what grounds the reality of all causation itself.

So, I don't think Tononi would agree with your characterization and claim that for the IIT to be true, causation would have to be identical with information. The IIT is neutral about the ontology of information other than to say that wherever information occurs there is a corresponding instance of consciousness whose character is lawfully determined by the structure of its qualia-space. Mind may be ubiquitous in the IIT but it may still be epiphenomenal.

Eric Schwitzgebel said...

Thanks for the continuing comments, folks!

@ Arnold: I figured you might say something like that. By not aiming for full metaphysical generality, your theory doesn't fall into the class of target theories I am currently evaluating.

@ Anon: That would be a funny result, if true.

@ Matt: Thanks for the detailed reflections! I agree that Tononi might not be committed to a *high* phi value for the U.S. (or other similar group entity), just a non-zero phi value, which is all that is necessary for consciousness on his theory. High phi seems to me the most natural interpretation, given our complex informational integration, but evaluating that isn't entirely straightforward!

@ Scott/Matt: I'm inclined to agree with Matt that Tononi's theory might be open about such issues, although I'd be happy to see textual evidence that he's committed. I wouldn't leap quickly to the conclusion that the consciousness of the U.S. is epiphenomenal just from the fact that U.S. behavior fully depends on the behavior of individuals. Note that a parallel argument could be made for the epiphenomenality of a human being's consciousness, since a human being's behavior (presumably) fully depends on the behavior of her lower-level physical structures.

Arnold Trehub said...

Eric, must a theory aiming at full metaphysical generality be isolated from empirical constraints? I'm really interested in how you conceive of full metaphysical generality.

Scott Bakker said...

Eric and Matt: his most recent article
actually lays out his position regarding causality and information.

Eric Schwitzgebel said...

Thanks for the reference, Scott (again)! I hadn't seen it before. His section on "complexes" puts a spin on his work that makes it less straightforward to reconcile what he says with the consciousness of groups. I'll have to reflect on that more!

Matt Sigl said...


Thanks for directing me to Tononi’s new paper. Lots of food for thought. Once again, Tononi impresses me greatly with his philosophical sophistication. Though he rarely explores his ideas to their furthest philosophic conclusions, it's not his job to.

His new explanation of the relationship between information and causation is a rich treasure trove of ideas. Tononi tends to put some of his juiciest material in footnotes, as is the case here. Footnote 33 is worth reading and re-reading; its content has deep philosophic implications for the problem of emergence and the ontology of causation and the mind. It's worth quoting a little bit. He writes:

“However, things are different from the intrinsic perspective - that of the complex itself. If a set of macro-causes accounts for the complex’ behavior better that a set of micro-causes, i.e. if the micro-level is causally less complete than the macrolevel, then the exclusion principle (Occam’s razor) implies that the behavior of the system is determined by the macro-causes only, and the micro-causes cannot be double-counted.”

If I read him right, he is claiming that since integrated information is always ONE THING, not many smaller things, it's prudent to view the "macro-state" of the complex as more causally efficacious than the entirety of the micro-elements considered independently. He doesn't, it's important to note, claim that the causal flow of the micro-elements is incomplete, only that, when a complex is active, it, as a whole, is a better, more informative, more accurate causal story than mere micro-elements. This is somewhat hard to conceptualize, and there may even be an argument to be made that it doesn't make total sense. Briefly, if the micro-elements entail any and all macroscopic effects then in what sense can the macro integrated state be said to be more primary than it's micro-physical parts?

I think the answer has to do with information's identity (and not just correlation) with causation. Since causation IS information, (and microphysical elements like quarks and electrons are best perhaps conceived of as elements in causal processes, not isolated entities), then the ontology of the world admits a causal role for consciousness that truly is above the sum of its parts, even if the "parts" entail the conscious state.

Later in the footnote Tononi says, "In terms of dynamical systems, this means that microvariables are ‘enslaved’ by macro-variables." This strikes me as a conceptual revolution over how we understand the causal flow of the world. It presents a reality in which causal processes build upon themselves, creating larger and larger "units" of causal efficacy. Furthermore, these "units" are always necessarily conscious, reaching their apogee, so far as we know, with the human mind.

Tononi's theory smells like the truth to me. Given the impoverished state of the science of consciousness (as opposed to the philosophy of consciousness, which has been a robust field of study since at least early 90's) this is itself a triumph. Until now, even sounding plausible was beyond the pay grade of most scientific theories of consciousness. Whether or not the IIT is in fact the case should be known soon enough as rapidly advancing experimental correlation could quickly give the theory serious empirical support. It could also show it to be fundamentally mistaken. What the IIT means however is a topic that, even if the IIT is accepted scientifically, will take a long time for philosophers to unpack and correctly conceptualize.

Scott Bakker said...

It was a gem to find for me as well. He's been under some critical pressure from Edelman and Seth (here's a link - it was openly available a couple of months back, but no longer )which might explain a couple of the new twists. Edelman now thinks that it could take several different analytic approachs to causation and information to home in on consciousness.

I've had my chips on Tononi (or someone from the Edelman 'bratpack') for quite some time. He was the first theorist I encountered who self-consciously took the 'intraneural perspective' that destroyed my dissertation (on phenomenological ontology) way back when.

I actually think it'll prove a disaster for philosophy as we understand it if any of these guys wins the empirical day. But then science has always been in the business of popping human balloons, not matter how sophisticated.

One of the immediate issues that the 'intraneural perspective' and IITC raises, is what might be called the 'Positioning Problem.' Information integration means information horizons, points at which integrative availability peters out. The cognitive sufficiency of ANY feature of conscious cognition would depend entirely on the informatic topography of integration. IITC raises the possibility that 'conscious cognition' could be a thoroughgoing 'intraneural perspectival artifact,' not so different than the one informing the Ptolemaic conception of the universe.

The idea is to look at consciousness, not as a problematic 'perspective' on the world mediated by a brain, but as a problematic perspective on a brain possessing a problematic perspective on the world. This is essentially what IITC and related theories surmise: consciousness is the result of accessing certain neural information in a certain way. Tononi's 'intrinsic perspective.'

And this is why I've been selling you so hard on my paper Eric! Not simply because I need professional help (which I do), but because your book can be read as an exploration of the integrative information horizons of consciousness from the inside (as opposed to Tononi's outside approach). And your conclusions, I think, show that the answer to the Positioning Problem ain't going to be pretty.

Consciousness is causal, but not in any way we should expect to be able to fathom given the information it gets.

Eric Schwitzgebel said...

Arnold: Sorry for missing your previous comment! No I don't think a theory aiming at full metaphysical generality will be isolated from empirical constraints. But I do think that someone aiming for a theory that encompasses weird aliens is playing a very different game from the person aiming for a theory that is silent about such hypothetical cases. To the extent a theory aims just to be about *human* consciousness, or *mammal* consciousness, or *consciousness in beings with brains*, it's unclear what implications it would have for the case of the U.S., since the U.S. is not human or mammalian or possessed of a single unified brain. That's the point of my remark about metaphysical generality.

Eric Schwitzgebel said...

Scott: Thanks again for the helpful reference! I don't seem to remember getting a paper from you, so please feel free to resend. I find your comments above interesting but a bit too compressed to fully evaluate.

Eric Schwitzgebel said...

Matt: I've always been attracted to the thought that privileging the small over the large causally and explanatorily is problematic. Rob Wilson calls privileging the small "smallism". Originally, some of John Dupre's work from the 1990s is what convinced me of this. I'm inclined to think that to the extent there's truth in smallism it's really not about size but rather about certain privileges the specifically detailed has over the coarse and approximate.

Marie said...

In one of the papers of Tononi (in 2010) he say that the consciousness could be one emergency. If Tononi is an emergentist (in the sense ontological) seems that the notions of causality are redefined. Maybe the phenomena of low level (for its organization and complexity) produce a new level, which can not be reduced to the inferior level. And I seems that could be one downward causation (no is a epiphenomenon).

Anonymous said...

To be clear, you have misread something in the theory. There are strong "Tononian grounds" to reject the consciousness of the US.

The United States would not be conscious because Tononi employs an "exclusion principle" to deal with subsets.

For instance, given a set of elements {ABC}, if the set {ABC} has a higher Phi value than any of its proper subsets {AB}, {BC}, {AC}. then the set {ABC} is the phenomenally real entity. Compare this to Kim's usage of an "exclusion argument" in philosophy of causality (different in details and usage, but they operate on the same principle).

This "exclusion principle" is stated throughout Tononi's papers. The same would apply for supervenient macro phenomena - since the US government has less phi than the set of its citizens (which it supervenes on), the US government is excluded.

It's essentially a problem of: once you've found the maxima, what do you do with the long tail of minima? A sensible answer is to exclude them.

Your confusion stems from a misreading of Tononi. You've stated that complexes have liberal boundaries, but this is false in all formulations of his theory. Complexes cannot overlap - for any set of system elements, each element belongs only to a single complex. He has always held this view, and it is in his 2004 paper as well.

Thus when you say there are no "Tononian grounds" for excluding the United States, that is incorrect. There may be reasons for excluding it that are not Tononian grounds, but under his grounds, it is strongly excluded.

Eric Schwitzgebel said...

Right! This was based on 2004-2008 Tononi. I know he now uses an exclusion principle to rule out such cases. But I think that leads him into other troubles:

csimmons said...

The idea that a photodiode or transistor has some minimal amount of consciousness but that the U.S. has no consciousness because it is an aggregate seems ... non-intuitive ... to me.