Functionalism

1
Functionalism

• Albert Parker
• May 3, 2001

2
Two Camps

• Dualist
• Mind is formal and abstract
• Substance two radically different kinds of
  substances in the universe material objects and
  immaterial minds (Descartes)
• Property - two radically different kinds of
  properties in the universe material properties
  (like weight) and immaterial properties (like
  pain)

• Materialist
• Mind is the wet and slimy stuff in our heads
• Behaviorism mind reduces behavior or
  dispositions to behavior
• Physicalism mental states are brain states
• Functionalism mental states are defined by
  causal relations
• Strong AI minds are programs implemented in
  computers

• (Searle MLS, p46-7 and SA 31)

3
How do we observe the mind?

• Dualists Problem Metaphysical Gap
• The mind can be observed through introspection.
  It is just the minds of others that cannot be
  observed.
• Materialists Problem Leibniz Gap
• It must be confessed, moreover, that perception,
  and that which depends on it, are inexplicable by
  mechanical causes, that is, by figures and
  motions. And supposing that there were a
  mechanism so constructed as to think, feel and
  have perception, we might enter into it as into a
  mill. And this granted, we should find only on
  visiting it, pieces which push one against
  another, but never anything by which to explain a
  perception. This must be sought in the simple
  substance, and not in the composite or in the
  machine (CMB, Cummins 4)

Analysis of consciousness bogged down for lack of
analytical tools! (Psychology crippled and
behaviorism was spawned by taking the mind out of
psychology) (6-7)
4
What to do?

• For the materialist, Functionalism provides a
  bridge over the Leibniz gap Mental states are
  defined in terms of their abstract causal roles
  within the wider information processing system.
  A given mental state is characterized in terms of
  its abstract causal relations to environmental
  input, to other internal states and to output
  (Churchland, NP, 351).
• Example 1 Pump is a functional kind, being
  implemented by hearts, propellor and case,
  vibrator and one-way valve, centrifuges, piston
  and sleeve arrangements (MBC, Cummmins 7).
• Example 2 Mouse Trap is a functional kind, being
  implemented by spring traps, cage traps, a sack
  of grain attached to a trip wire, a cat or
  specially bred killer rat
• Example 3 Being in pain is a state carried out
  characterized by its causal relations to
• behavior wincing and crying out
• external input skin being burned
• other internal states the desire to make the
  pain go away and belief of what will bring
  relief.
• Functional kinds are specified by their roles and
• not by the material in which they are
  instantiated. (Churchland, NP, 351).

5
How does Functionalism Deal with the Leibniz Gap?

• Functionalism mental concepts are revealed
  through function not in terms of intrinsic
  features. That is, deal with function, to heck
  with the form.
• Leibniz Gap occurs because function cant be read
  from form
• Hence, Functionalism is a bridge over the Gap.
• trying to understand perception by studying
  only neurons is like trying to understand bird
  flight by studying only feathers (Marr 27)
• once we have explained the causal basis of
  consciousness in terms of the firing of neurons
  in the various cortical layers, it seems we still
  have a phenomenon left over (Searle MLS, p55)

6
Whose Idea was this?

• Emile Durkheim created the functional orientation
  by codifing the distinction of structure and
  function in the social sciences. He published
  his functional analysis in a number of empirical
  studies
• The Division of Labor in Society (1893)
• The Elementary Forms of Religious Life (1912)
• Durkheim recognized the potential problems
  of functionalism that critics like John Searle
  and Patricia Churchland still espouse today in
  reaction to neurophilosophical functionalism
• To discover the need that a structure
  functions to meet does not necessarily reveal its
  cause - the sequence of events that created the
  structure in the first place. To assume it does
  puts the cart before the horse. (Turner 17)
• J. Dewey The Reflex Arc Concept in Psychology
  (1896) is the opening shot of Functionalism in
  Psychology (Bergmann 678).
• H. Putnam Minds and Machines in Dimensions of
  the Mind (1960) was the first to argue that minds
  are things that we can conceive solely in terms
  of input, output, and various functional
  relations. (Hardcastle 2-3)

7
Some Functionalist Achievements

• von Neumann architecture (1943-1957) Still used
  in todays desk-tops, lap-tops, calculators and
  palm pilots that grace every office and half the
  homes in America (von Neumann xii)
• Barlows First Dogma (1972) A description of
  that activity of a single nerve cell which is
  transmitted to and influences other nerve cells
  and of a nerve cells response to such influences
  from other cells, is a complete enough
  description for a functional understanding of the
  nervous system (Barlow 380)
• Theory of the Cellebar Cortex (1969) This
  regular
• cortical structure is interpreted as a
  simple but powerful memorizing
• device for learning motor skills (Marr
  14)

8
something was going wrong

• I was myself caught up in this excitement (of
  the 60s). Truth, I also believed, was
  basically neutral, and the central aim of all
  research was a thorough functional analysis of
  the structure of the central nervous system the
  way seemed clear but somewhere underneath,
  something was going wrong the cerebellar study
  suggested that one could hope to understand
  cortical structure in functional terms but at
  the same time it did not much enlighten one
  about the motor system - it did not, for example,
  tell one how to go about programming a mechanical
  arm (Marr 14-15)
• Functionalism looks nice on paper to those who
  know nothing about how brains work. (Freeman
  115).
• Functionalism is merely the last gasp of
  classical psychology. (Bergmann 678)
• The functionalist strategem is a smoke screen
  for the preservation of error and confusion It
  needs to be revealed for the shortsighted and
  reactionary position it is. (MBC, Churchland
  506-7)
• If you are tempted to functionalism, I believe
  you do not need refutation, you need help.
  (Searle, TRM, 9)

9
Whats wrong with Functionalism?

• ABSURDITIES CAN BE DEFENDED for example,
  alchemy, in the face of elemental chemistry, can
  be defended using the functionalist stratagem,
  thusly
• Ensouled in mercury or sulphur is an
  abstract functional state. Mercury, for
  example, actually refers to the disposition to
  reflect light,to liquefy under heat, to unite
  with other matter, etc. (these are functional
  kinds). It is the total syndrome of occurent and
  causal properties of a metal or substance that
  matters, not the corpuscularian details of the
  substrate (function matters, not the
  implementation). Alchemy, it is concluded,
  comprehends a level of organization in reality
  that is distinct from, and irreducible to, the
  organization found at the level of corpuscularian
  chemistry. (MBC Churchland 507).
• NEED DETAILS! - the success of implementing
  Neural Networks illustrates that an elementary
  understanding of brain microstructure funds a
  fertile conception of what cognition really is.
  (MBC Churchland 198)
• QUALIA can not be reduced to something else,
  because if you could they would be something
  else, and they are not something else. (Searle,
  RM 51)

10
Lets Patch This Up!

• Oceans of Ink have been spilled in the debate
  over functionalism (Hasker 29)
• Valerie Hardcastle Functionalism is neither
  very strong nor very controversial.
• Jaegwon Kim global reductionism is not possible,
  but some local reductions are qualia are
  intrinsic properties if anything is, and to
  functionalize them is to eliminate them as
  intrinsic properties. (Hardcastle 29).
• Elliot Sober Functionalism got off on the wrong
  foot. The problem is that function is ambiguous
  (97)

11
New Paradigm or Paradigm Shift?

• John Searle Most of the recently fashioned
  materialist concepts of the mind such as
  behaviorism, functionalism and physicalism end
  up denying that there are any such things as
  minds as we ordinarily think of them Now, why
  do they do that? (MBS 15). (We need to stop
  using) the antique and obsolete vocabulary of
  mental and physical, mind and body (MLS
  47). Any satisfactory account of the mind must
  take into account consciousness, intentionality,
  subjectivity, and mental causation (MBS 17).
• Reactions Many of them, especially
  the younger
• generation, agree with me, but I am amazed
  at the
• number and vehemence of the defenders (SA
  29)
• David Marr Almost never can a complex system of
  any
• kind be understood as a simple
  extrapolation from the
• properties of the individual components
  (there are three)
• levels at which an information processing
  device must be understood (24-5)

Computational Theory What is goal of computation
and logic of the strategy?
Representation and algorithm What is
representation of the input and output and the
algorithm map?
Hardware Implementation How are the
representation and the algorithm realized
physically?
12
My Take

• Cummins states that if a theory is any good, it
  must be explanative in and of itself. Some
  functionalist defenses to attacks sound like the
  non-intuitive hook-hook argument illustrated on
  page 3 of MBC handout.
• Functionalism has problems in its pure form
  (everything is function, defined in terms of
  inputs and outputs). NOTE The new paradigms
  presented here (at least Marrs) have
  functionalist components.
• Searles paradigm may be a new one. Marrs is
  a shift.

13
References

• H. B. Barlow. Single snits and sensation a
  neural doctrine for perceptual psycology?
  Perception 1.
• G. Bergmann. The Contribution of John B.
  Watson. J. M. Scher editor. Theories of the
  Mind. The Free Press. New York, 1962.
• P. S. Churchland. Neurophilosophy. The MIT
  Press. Cambridge, 1986.
• R. Cummins and D. D. Cummins. Minds Brains and
  Computers The Foundations of Cognitive Science.
  Blackwell Publishers. Malden, MA, 2000.
• W. J. Freeman and C. A. Skarda. Mind/Brain
  Science Neuroscience on Philosophy of Mind. E.
  Lepore and R. V. Gulick editors. John Searle and
  His Critics. Basil Blackwell. Cambridge, 1991.
• V. Hardcastle. How to Build a Theory in
  Cognitive Science. State University of New York
  Press. New York, 1996.
• W. Hasker. The Emergent Self. Cornell
  University Press. Ithaca, 1999.
• D. Marr. Vision. W.H. Freeman and Company. New
  York, 1982.
• J. von Neumann. The Computer and the Brain.
  Yale University Press. New Haven, 2000.
• J. Searle. Is the Brains Mind a Computer
  Program?. Scientific American. Jan 1990, p
  118.
• J. Searle. Minds, Brains and Science. Harvard
  University Press. Cambridge, 1984.
• J. Searle. Mind, Language, and Society. Basic
  Books. New York, 1998.
• J. Searle. Rediscovery of the Mind.. The MIT
  Press. Cambridge, 1992.
• E. Sober. Putting the Function Back into
  Functionalism. W. G. Lycan editor. Mind and
  Cognition. Basil Blackwell. Cambridge, 1990.
• J. H. Turner and A. Maryanski. Functionalism.
  The Benjamin/Cummings Publishing Company. Menlo
  Park, CA, 1979.