Article in Science Daily here, which claims that a lot of new evidence supports Roger Penrose's old conjectures about the the way that quantum physics is implicated in consciousness. If any philosophers of mind feel like explaining this to the rest of us, that would be very cool.
10 responses to “Was Penrose right all along?”
-
Here’s the original article text, a review by some of the theory’s proponents:
http://www.sciencedirect.com/science/article/pii/S1571064513001188LikeLike
-
Deepak Chopra, really?
LikeLike
-
Surely others know more about this than I do, but I will kick things off: One objection to Penrose’s idea was that the the brain was too hot and noisy for anything interestingly quantum to function, since decoherence would happen much too fast. This objection now pretty clearly seems to be wrong. We have evidence that various biological system “genuinely use” QM to do various things.
But this is a long way, I would think, from actually showing that Penrose was right about all of his elaborate ideas about wave collapse and consciousness. It only shows that one can no longer object to his view for one particular in-principle reason.LikeLike
-
Yeah, what does one say? I thought Penrose was pretty polite.
LikeLike
-
very gentlemanly, what is with the editors there, maybe Gary Zukav was too busy dancing wu li with Oprah?
LikeLike
-
Ha!
That’s exactly what happened, though I think Fritjof Capra must have been involved in some manner as well.LikeLike
-
I think the bigger problem here is not with the explanans but the explanandum. Is “qualia” even a coherent concept latching onto a determinate phenomena amenable for targeted physical theorizing? People use terms like “consciousness” in such multifarious and ambiguous ways applying it to everything and nothing that I am extremely skeptical that these competing theorizers of consciousness are even trying to explain the same thing, or even have a good idea themselves of what needs explaining. If you ask them to “point out” what they are trying to explain, they all point in wildly different directions. Orch-OR seems to be no more or less scientific than similar physics-based “proto-consciousness is everywhere” theories e.g. Integrated Information. How can Orch-OR disprove integrated information theory and vice versa?
LikeLike
-
In the absence of a philosopher of mind, let’s try a philosopher of physics with a mind AOC…
To be clear: Penrose (and I take it Hameroff, though I’ve studied this exclusively through Penrose’s work) is not claiming that quantum-mechanical effects per se are relevant to consciousness. The idea is rather that certain deviations from the predictions of orthodox quantum mechanics play a role in consciousness.
Here’s a sketch of the argument:
(1) Certain cognitive tasks performable by humans are not computer-simulable (because of Lucas-style arguments that no computer can understand the truth of Gödel’s theorem but that humans do)
(2) Both classical mechanics and standard quantum mechanics are computer-simulable.
(3) So some of the cognitive functions of the brain must rely on physics other than classical or quantum mechanics. (From 1,2)
(4) Because of the quantum measurement problem, we know [says Penrose!] that quantum mechanics must be modified to include a new collapse mechanism. Since the physics of this process is completely unknown it is possible that it is not computer-simulable.
(5) Again because the physics of this process is completely unknown it is possible that it plays a role in the cognitive functioning of the brain.
(6) There are no other plausible sources of physical processes that are not computer-simulable and that could play a role in the cognitive functioning of the brain.
(7) So the quantum-collapse process must both be non-computer-simulable and play a part in the cognitive functioning of the brain. (From 3,4,5,6).
To do the argument credit, it’s more careful than the usual quantum-mechanics-of-consciousness arguments, which might be caricatured as: conscious=mystery, quantum=mystery, therefore conscious=quantum. (And it doesn’t rely in any way on the supposed phenomenal features of consciousness, i.e., qualia.) But by that same token, it loses its force completely if its premises are rejected: in particular, if (as I do) you find Lucas’s argument from Gödel’s theorem wholly unpersuasive, then there is no reason to take Penrose’s and Hameroff’s theory seriously. (There is to the best of my knowledge no other reason to suppose that the quantum measurement problem requires non-computer-simulable physics to be solved, and no extant mathematically formulated variation of quantum mechanics that is non-computer-simulable.)
The Hameroff-Penrose argument cited in Science Daily strikes me as rather opportunistic. Here’s the scientific discovery that’s underpinning it: It used to be thought that distinctively quantum-mechanical effects require rather delicate and controlled (and, in particular, cold and isolated) conditions to manifest. Warm, open systems like biological systems therefore seem to be places where we can ignore quantum phenomena and just use classical physics. But the last few years have provided a lot of very exciting evidence that quantum effects are being exploited all over biological systems. But because (to repeat) these quantum effects are computer-simulable, they don’t in any way offer the kind of physics that Penrose wants to appeal to to solve the Gödel’s Theorem problem. (Quantum computers cannot be effectively simulated by classical computers, which is to say that they can’t be simulated in any reasonable amount of time – but the Lucas argument doesn’t in any way rely on timescale.)
The most charitable interpretation of Hameroff’s and Penrose’s recent arguments (so far as I can see: I haven’t studied them closely) is that the same features of the brain that were supposed to block orthodox quantum phenomena from having any interesting cognitive effects might plausibly be thought also to block exotic quantum-state-collapse physics from having those effects. But it’s very hard to know for sure because we don’t know anything at all about what this physics is supposed to look like. Their actual paper seems to me (on quick reading) to be a bit of a bait-and-switch, equivocating between the idea that quantum computation plays a role in cognition and the idea that Penrose’s supposed quantum state reduction plays that role.
I find this story kind of depressing, to be honest: the discovery that quantum physics is playing an important role in lots of places in biology is (I’d say) one of the most exciting empirical discoveries in physics in the last decade, in many ways more so than finding the Higgs boson because it was completely unexpected, but it’s a shame that it gets press time here only through the connection to Penrose’s frankly fringe set of ideas.LikeLike
-
Thanks David. That’s really helpful. I note another problem with the argument: it doesn’t mention consciousness. So even if you accept the premises, it doesn’t support the claim that the quantum-collapse process plays a role on consciousness (relatively little of cognition is conscious, after all).
LikeLike
-
Just to reiterate Neil’s thanks.
A couple of years ago me and Jason MeGill built on some of Crispin Wright’s and my earlier arguments to produce a new version of the Lucas-Penrose argument that is valid, and (we show: http://link.springer.com/article/10.1007%2Fs11023-010-9203-1#page-1 ) not subject to any of the traditional criticisms.
But doing so requires making explicit two suppressed premises in the original: (1) that the only alternative to epistemically problematic platonism about mathematics is a kind of constructivism that satisfies three Dummettian requirements about mathematics and/or higher order logic (in the paper we show that Robert Brandom is also committed to these requirements), and (2) that Chomsky’s account of the performance-competence distinction is correct (finite humans must instantiate the same processes, subject to performance limitations, that yield a potentially infinite output). With these premises, nobody needs to recognize their own Goedel sentence (whatever that was supposed to mean all along).*
But the defender of the computational theory of mind (which I don’t find plausible for other reasons) can produce independent arguments against (1) or (2), and so it’s not crazy to see Lucas-Penrose as a reductio of one or both of those views rather than a reductio of computationalism.
In particular, Goedel’s theorem causes serious problems that Dummettians and Brandomians don’t take seriously enough. The person who equates mathematical or (higher order!) logical truth with provability now must consider the provability relation in question to be non-axiomatizable (showing this involves Craig’s Theorem, which is constructively kosher). But then what’s the Benaccerafian gain? Supposedly constructivism had a bad semantics but a better epistemology. But does it really? Is behavioral manifestation of an “open textured” (that is infinite and non-recursive) proof relation really any epistemically better than Goedelian insight into Plato’s heaven? You still have the bad kind of infinity constructivism was supposed to save us from. This is why Megill and I called our paper “Are Turing Machines Platonists?”
For me, the only nice upshot about Lucas-Penrose is that they motivate this question with respect to Benacerraf’s dilemma. From your description, it doesn’t seem like the cool quantum physics in biology gets us anything more than that.
[Notes:
*The literature is vague on this excrutiatingly simple issue! Is it the Goedel number of an axiomatization of the Turing machine that you yourself are? Or is it the Goedel number of the axiomatization of the mathematics that you the Turing machine in some sense implicitly knows via the ability to solve mathematics? If it’s the former, there’s no reason at all to think people can discover their own Goedel number in this sense. If it’s the latter, a Turing machine can discover this Goedel number, because Goedel’s procedure is constructive! Obviously the procedure iterates to infinity, but that doesn’t say anything about computationalism. We consider this in our paper.]LikeLike
New APPS 
Leave a comment