Zoltan Torey’s new book is, in measures, exciting, provocative and infuriating. It is exciting because, contrary to what this review might be taken to imply, it has some really good ideas, not least that of a perspectival oscillation in conceptually structured thought between the observer being drawn into the foreground and pushed into the background.  It is provocative because of the many strong claims that it makes, beginning with the claim that language is necessary not only for (self-) consciousness but for any meta-cognition.  And it is infuriating for its failure to back up many of those claims, either by adequate references or incremental argument.

Take for example the claim that hemispheric brain structure in all animals besides humans is strictly redundant: i.e., only in humans is there any division of functions between left and right hemispheres.  Torey takes this as a given, but I am unfamiliar with the research supporting this, and so I would like to be able to follow up — but I cannot, because no references are offered.

Or take that claim that language is necessary for meta-cognition, i.e., higher-level thoughts or “thoughts about thoughts”. I can find no argument for this in the book; rather, it seems to be taken as an article of faith, an axiom from which much of the rest of the book follows, including such tendentious claims as that all animals besides humans (which must also include pre-linguistic infants) engage in no offline simulation-type thinking, indeed lack either minds or thoughts, and are strictly stimulus-response systems:  creatures of instinct. It leads Torey (unfortunately in my view) essentially to equate language with concepts and concepts with percepts or at least assume a straightforward translation of each to the other. It leads him to make such to my mind naïve statements about language as this one: “…The spoken language is thus a kind of conveyor belt for the transportation from one individual to another of the [conceptual] material” (page 70).  Errors in translation are glossed over.

Predictably Torey is dismissive of the mirror test, which has been widely claimed to show self-recognition in a number of higher mammals. “No thought operations are involved or needed and the assumption that the ape’s behavior signifies reflective awareness of a ‘self’ is not justified” (page 90).

It does not help the charitableness of my reading that Torey is remorselessly scathing of viewpoints in disagreement to his own, typically describing them as “absurd”. He has no patience for machine consciousness, which he describes as being on a par with leprechauns and ghosts. “The asymmetry between the brain and the computer is complete, all comparisons are flawed and the idea of computer-generated consciousness is nonsense” (page 155).

As many physicalists do, he lumps all forms of dualism together, making no attempt to distinguish substance dualism from property dualism, David Papineau’s conceptual dualism, or any other form of dualism.  I have no disagreement with physical monism per se, only with the assumption that we have a clear and untendentious understanding of what “physical” is or what the physical/mental divide is supposed to be.

Or take his claim that “generally speaking, there are two ways of dealing with consciousness: reductionism and mysterianism” (page 5).  Torey, of course, is a reductionist, leading him to make what I take as a category error in his insistence on using “brain” where others would talk of “mind”: e.g., when talking about the brain’s understanding of itself or its beliefs or decisions. I, who am not reductionist, am on Torey’s account consigned to mysterianism.  To the contrary, I do not think that consciousness needs to be mysterious, though I do disagree with Torey’s faith that the conscious mind is capable of a complete and consistent account of it.

Along with Roger Penrose (THE EMPEROR’S NEW MIND, SHADOWS OF THE MIND), Torey believes that significant parts of human cognition cannot be described algorithmically and that this proves the superiority of human understanding over “purely” formal systems.  He writes, “the claim that formal systems are mind-like, or may be viewed as analogues of the mind, was finally settled by Gödel” (page 159), with no acknowledgement that others, including Douglas Hofstadter whom he quotes at regular intervals, take a very different lesson from Gödel, concluding that the human mind is just as bound by the logical incompatibility between consistency and completeness.

Indeed Torey seems at points to go beyond Penrose in toying with the idea that the human mind is not bound by logical constraints at all.  There seems to be nowhere it cannot go:  “…Since everything that exists is either already in the percept form or can be rendered so by the mind, it is possible to scan (think about) absolutely everything, including the mind itself, as even this is just a percept in and for the thinking process” (page 163, emphasis added).  Unfortunately, unlike Penrose so usefully does in SHADOWS OF THE MIND, Torey offers no step-by-step argument so that one might, as one might be tempted with Penrose, point to a particular step in the argument as the point of disagreement.

Ironically Torey is quite aware of the logical problems entailed by the mind (or the brain, as he prefers) arriving at a full account of itself.  “When the mind’s focus is the focusing mind, …problems arise.  The object and the instrument of the inquiry become one and logic is compromised” (page 15).  He explains the problem most eloquently in Chapter 5, and it is worth quoting at length:

…As thinking equipment the brain cannot get past regressive circularities about itself. …The regressive circularity that the brain’s self-investigation invariably entails is self-reference:  a relationship that characterizes the observer’s observation of himself or herself.  The difficulty has to do with the identity of whatever is the recipient of reception, as there is nothing but reception to receive itself.  Not only has the brain’s logic no answer to this paradox, but it is its source.  The thinking process — given the raw data of experience to think about — simply cannot get past this point. (page 96)

But then Torey assumes, without further argument, that this subjectivity can be set aside and the mind (or brain), by clever analysis, can escape these logical constraints and arrive at the singular and real truth of the matter.  How this is to happen seems, to me, not to be (as Torey seems to take it) straightforward application of neuroscience but instead shrouded in the very mystery Torey wishes to escape.

On some matters I think Torey is simply wrong, though in good company: e.g., his apparent equation of Turing machines with real-world computers and computer programs as formal systems with computer programs as instantiated on a real-world computer.  Every computer I am aware of is embodied in a particular physical form and embedded in a particular physical environment with which it interacts in non-trivial ways.  Computers, contrary to popular wisdom, do not only “do what they are told”.

Other matters I find to be distractions. At several points Torey is dismissive of religion as an unfortunate, if necessary, consequence of our species’ cognitive development, an attempt to palliate our fears with myths rather than facts.  The speculation in the final chapter that human consciousness is the universe’s solution to the question of its fate — entropic “heat death” on the one hand, collapse back to singularity on the other — strikes me as hopelessly anthropocentric.  Stuck with seeing the universe from our own perspective, we insist on giving ourselves a central role to play.

Torey’s central claim is that consciousness is a language-based, offline loop of the sensory-motor system, reducible to neuroscientific terms, that enables the human organism through reflective awareness to delay and modulate its response to its environment.  By the end, I do not feel that consciousness has been explained (or explained away) any more than Dennett claimed to have done in 1991. The value of this book is, pace Torey, not in the answers it attempts to provide but in the questions it forces us to address and the discussions that will doubtless ensue.

 

© 2010 Joel Parthemore

 

 

Joel Parthemore is finishing his thesis write-up as a visiting research student of the University of Lund in Sweden, and hopes to submit this summer.  His primary affiliation is as a member of the Philosophy of AI and Cognitive Science research group of the School of Informatics, University of Sussex, UK.  His thesis concerns enactive philosophy and theories of concepts, in particular Peter Gärdenfors’ conceptual spaces theory.  In his spare time, he plays with Linux systems.”