Reponse to the 'review' of brain symbols

The message below was sent by email to the editor of 'Minds and Machines' on 5Oct96.

related texts

--------------------------------------------------------------------------

Dear James H. Fetzer,

re: "The neurons in the brain cannot implement symbolic systems"

I just received your letter, rejecting the above article, and the reviewer
comments. I don't think I can describe these comments without being
extremely rude. The ignorance of the reviewer is so deep, that he does
not even know that the patterns of neurons in the human brain are static. 

Below I attached the full text of the review, with my comments
(indented and start with *). These demonstrate that the reviewer does
not have even an elementary understanding of neurons in the brain and
of symbolic systems theories and models, and failed to read substantial
portions of the text .  This review is a shame for 'Minds and Machines'.

I request that the article will be given to  reviewers which have
at least an elementary understanding of neurons and symbolic systems
theory,  are open to new ideas, and are capable of reading a
relatively short article without skipping large portions of it. 


Yehouda Harpaz


************************************************************************

The author is not recommended to seek publication of this paper: it
can only damage his reputation.

   * Start with a paternalistic comment.

The paper describes the characteristics of 'symbol systems' from such
sources a Newell's book on SOAR; it gives an account of neurons as
culled from text-books, stressing their stochastic nature. Its key
argument is as follows (from P. 8):

'Since it must be possible to store symbol tokens in arbitrary
structures during computation (in other words, they are dynamic), they
cannot be implemented by static features. This means that symbol
tokens cannot be implemented by patterns of neurons and the
connections between them.' There is a logical gulf between the first
proposition and the second proposition. We are not told why if symbol
tokens must be dynamic, they cannot be implemented by patterns of
neurons.

    *  The 'logical gulf' is the ignorance of the reviewer. Everybody 
    *  with  an elementary knowledge of neurobiology knows that patterns
    *  of neurons are static. In addition, the text explicitly list
    *  this as part of the properties of neurons in section 3 (p. 6,
    *  top), showing that the reviewer did not read this section. 

The paper is poorly written, and contains many grammatical mistakes
and slips of the pen.

   * There are probably 'many grammatical mistakes and slips of the
   * pen'. 'Poorly written', however, is not obvious, and probably
   * means that the development of ideas is not done in the way the
   * reviewer expect. 

My best advice to the author is to discuss his ideas with a colleague
who knows about neurons, symbols and the theory of computability.

   * This 'advice', with its implication that I don't know much about
   * "neurons, symbols and the theory of computability", is amazing, 
   * considering that the reviewer could not find anything that points
   * to any lack of understanding on my side. The only base for it is
   * the arrogance of the reviewer. 

There is currently much work that models the properties of real
neurons computationally. 

   * How is that relevant to the discussion?

And indeed computers can readily model stochastic and
non-deterministic systems.

   * And how is that relevant to the discussion? The discussion is
   * about what neurons can do, not what computers can do. Either 
   * the reviewer failed to understand what the claim of the article
   * is, or maybe he is just babbling randomly. 

Finally, whatever may be meant by a 'symbol', surely written
communications are symbolic?

   * Here the reviewer exposes his lack of understanding of the
   * symbolic  systems theory.  The symbols of written communications 
   * have static meaning, so written communications is not a symbolic
   * system, according to symbolic systems theory. This is discussed,
   * including references, in section 2, and  systems which do not 
   * require dynamic meaning are  explicitly excluded from the
   * discussion (p. 4, end of first paragraph). Thus the reviewer did
   * not read this section, either. 

And surely writing and reading depend on brain mechanisms?

   * Not only the brain. At least the eyes and several other
   * supporting systems are essential, too. This is 
   * an important point, see in the discussion below. 

So, why does the author write papers postulating that the brain cannot
implement symbol system? He may as well write papers -- as do
deconstructionists -- postulating that written communication is
impossible

  
   * The reviewer did not bother to present his argument properly, so
   * we need to guess what the argument is. My guess is that it is: 
   * "A person can perform symbolic operations, so neurons must be able
   * to perform symbolic operations". that is broken logic (Even if we
   * ignore the fact that the reviewer did not give an example of a
   * symbolic operation). 
   *
   * The person as a whole is different qualitatively from the components
   * of the brain in at least two fundamental properties:

   * 1) In addition to the brain, a person has also sensory systems. 
   *   These differ from neurons in the brain in that their direct
   *   source of input (e.g the objects that emit the photons that the eyes 
   *   receive) is very variable. In the case of the eyes, the direct
   *   source of input changes in time scales of several 10ms, both
   *   because the eyes move, and because objects in the world (or
   *   images of objects in TV) are moving. In other senses the
   *   changes are slower, but the even the slowest changes (in taste
   *   and smell) happen many times in each  day.

   *   In contrast, the direct source of information to any neuron, or
   *   a group of neurons, in the human CNS, is almost static (with very few
   *   exceptions). Once the development of the brain stops in the first
   *   year of life, the set of neurons that deliver information to
   *   any group of neurons in the brain is essentially constant for the
   *   rest of the life of this person. Major changes are rare, and do
   *   not happen during normal computation.

   *   Thus any argument that relies on static direct input is
   *   applicable to components of the brain, but not to the person
   *   as  a whole.

   * 2) The brain implement a learning and thinking system. We don't
   *   know much about the details of this system, but we know it works.
   *   This system is capable of learning new skills,  by
   *   trial and error learning, by analyzing situations and deciding on the
   *   right actions, and by receiving communication from other people.
   *   These new skills are not limited to the capabilities that are
   *   built-in in the genetic makeup of the person, and written
   *   communication  is an example of such a new skill. 

   *   It cannot be assumed that component in the brain also have this
   *   learning capabilities, so they are limited in a way that the
   *   whole person isn't. The possibility of learning how to deal
   *   with symbol tokens inside the brain is discussed in my article
   *   (section 5. P.10  second paragraph). 
   *
   * Since components of the brain do not have sensory input and the 
   * learning capabilities of the whole person, there are many tasks
   * that the whole person can do that components cannot do. Thus,
   * that the person can perform some task (e.g written communication, 
   * symbolic operations) does not prove that components of the brain 
   * can do it.