Dear Bill,

Here is my take on the Putnam article we were discussing.
[Cognition 2(1) pp.136-140]

Putnam was the (principal) originator of the "functionalist"
approach to the mind-brain problem: he proposed that a 
psychological state, such as a pain, was identical to
some functionally characterized activity. According to
this doctrine, having a certain kind of pain WAS THE SAME 
as being engaged in a certain kind of FUNCTIONALLY 
CHARACTERIZED activity. 

Putnam's aim in section 1 is to explain why he is 
retreating on his earlier the reductionist position 
that the laws of higher level sciences REDUCE to the
laws of lower level ones, and eventually to the laws of 
particle physics. 

A key point, here, is that he is focussing on LAWS: the laws 
of the various scientific disciplines. He argues that
the LAWS OF THE VARIOUS SCIENCES are basically autonomous
logical structures designed to deal with different levels
of phenomena, and that there is therefore no rational
requirement that they have any strict logical or mathematical
connections to each other: the laws that are useful in 
EXPLAINING connection among phenomena at some level need not,
for that purpose, be logically or conceptually linked onto 
lower level laws in any exact way. Indeed, an optimal
high-level law ought not be encumbered with a sub-structure
that is not needed to do the job at hand, which is to 
explain connections between phenomena at that level---
for example at the psychological level. 

He gives examples in which the microstructure is basically
irrelevant to the connections that need explaining, which 
depend on macroscopic geometric rigidity, not on the cause
of that rigidity. And he gives other examples where 
connections to still higher level structures are important
to an adequate explanation.

This all boils down to the possibility of effective "autonomy" 
of the LAWS of the different sciences, or at least to a 
non-need for the higher-level explanatory laws to "reduce" to
lower-level ones.

In section 2 on the Turing machine model of the "whole
human being" he wants to explain why his earlier argument
in favor machine functionalism was incorrect. 
His earlier position had, by his own account, two parts: 
1) A whole human being IS a Turing machine,
and 2) The psychological states of a human being ARE Turing
machine states or disjunctions of Turing machine states.

Notice that whereas in section 1 he was speaking 
of LAWS that EXPLAIN PHENOMENA in useful ways, he is now
using the words IS and ARE: he is addressing the ontological 
question of what a human being IS, and what that beings
psychological states ARE.

Previously (in section 1) he was focussing on the nature of
the LAWS BY WHICH WE EXPLAIN, in useful ways, connection
between phenomena in some field or discipline. For that
sort of discussion the issues `what something IS' was not 
essential, and, indeed, his whole line of argument was based 
on this fact that explanation need NOT be based on ontology.

But in section 2 he brings in ontology. 

He says (p.136, final paragraph: his emphasis marked by *x*;
mine by capital letters)

  " These characteristics already establish that *no* 
psychological state in any customary sense can BE a Turing
machine state. Take a particular kind of pain to be a
`psychological' state. If I *am* a Turing machine, then my
present `state' must determine not only whether or not I am
having that particular kind of pain but also whether or not
I am about to say `three', whether or not I am hearing a 
shrill whine, etc. So the psychological state in question 
(the pain) IS NOT THE SAME AS my `state' in the sense of 
*machine state*, although it may still be possible that 
the machine state *determines* my psychological state."

Putman is talking here about our real psychological
states, such as `pains', and he is denying that they ARE
IDENTICAL TO machine states, but is admitting they may 
BE DETERMINED BY machine states. But this notion that
the psychological realities are DETERMINED BY the
machine state realities betrays some shifting away from
the notion of autonomous LAWS OF EXPLANATION to the idea
of some basic reality, or realities, or existing aspects 
of nature that lie behind these perhaps imperfect man-made
idealizations. The reference to these DETERMINATIONS 
betrays a belief that there is some sort of beingness 
that ties these differently described aspects together. 
Ontology concerns that underlying structure that ties
together the aspects of nature that we describe differently.

Putnam is rejecting here a certain kind of identity theory: 
the pain IS NOT THE VERY SAME THING as a Turing machine state. 

THUS A PAIN IS SOMETHING ELSE! 

This conclusion appears to be moving him toward a 
DUALIST position! 

(p.137, l.5) "My description *qua* Turing machine 
(machine table) and my description *qua* human being 
(via a psychological theory) are descriptions at two 
totally different levels of organization."


The structure that the psychological state `is different from' 
is narrowly defined here in terms Turing machine states. 
However, he goes on to broaden his net, and in the end 
concludes that nothing is really achieved, in terms of 
explanatory power, by bringing in the Turing machine concepts,
in place of those of physics and chemistry.

So he does seem to be moving toward a dualist position:
`being in a psychological state (having a pain)' is 
NOT THE SAME AS
`being in a certain functionally characterized state.'

However, his argument refers back to different levels of
EXPLANATORY LAWS (as is emphasized by his parenthetical 
remark on p. 137 l.6 that the human being is described 
"*via* a psychological theory.") Thus the question 
of the connection between these two kinds of states---
psychological versus functional---if posed at the 
deeper ontological level (involving perhaps causal 
determination) is skirted: although the question seems to be
posed at the ontological level, is resolved dualistically
only at the level of explanatory laws.


A little later (p.138 l.4) he gets closer to
ontology. He says: "the *exact* law has to be compatible 
with ... the laws of physics, if materialism is correct."

(l. 25) "It is sometimes believed that a non-idealized 
description, an `exact' description, is possible 
`in principle' at the level of physics; be that as
it may, there is not the slightest reason to believe
that it is possible at the level of psychology..."

(p.140 l.13) "Psychological states are not
machine states..... But what *is* the nature
of psychological states."

Putnam's argument, although leading to a dualist position
within his framework of LAWS OF EXPLANATION at each level,
evades the issue of dualism at a deeper ontological level; 
i.e., the issues of 
(1) what a psychological reality, such as a pain, 
actually IS, if it is not just a feeling or experience
that occurs in a stream of conscious experiences, and
(2) how this existing reality is connected to the aspects 
of nature that we described in our basic physical theory.

The difficulty that besets Putnam, and all philosophers
who demands compatibility of philosophy with science,
but who have not recognized the extreme relevance of
quantum theory to these considerations) is that  
ontology is straight-jacketed in their thinking by the
incorrect belief that basic science entails materialism,
or something similar. Restriction to classical physics (CP)
gives them only one model of physical reality, materialism. 
And that model does not have the capacity to provide a  
satisfactory ontological place for psychological realities 
such as pains. Hence these philosophers are forced to
turn to other non-ontological considerations---such as 
the EXPLANATORY theories operating in the different 
sciences---to find a place for psychological 
realities, such as pains. The ideal of the Unity of Science 
was sacrificed by Putnam, as he says himself (p. 131 l. 5), 
in order to make a place for what is most real to us, our 
pains and sorrows and joys and delights, and the visual, 
auditory, and tactile sensations that connect us to certain 
aspects of nature that are not us.

Can the dualistic solution suggested by Putnam's
conclusion be carried over to the ontological level?

Within the framework provided by classical physics (CP) 
a dualistic ontology would render the psychological 
realities epiphenomenal: the course of physical
events would be determined in principle without ever
acknowledging the existence of psychological realities.

What is the problem with the epiphenomenalism entailed
by dualistically construed classical physics?

One problem is simply that it seems absurd and unnatural
that nature would create a separate level of reality
that attends so closely to features of the environment
that are relevant to one's decisions about how to one
should act, but then give that level no power to affect
the course of physical events. 

If one accepts that the laws and principles of classical 
physics are in principle adequate for the study of the 
mind-brain system (quantum effects being allowed in 
principle at the microlevel, but not in a way that is 
conceptually essential) then one is led to the idea of 
the "neural correlate of a conscious experience: 
a pattern of neurological (or more generally, brain) 
activity that is the brain `counterpart' 
of the conscious experience.

Staying within the classical-physics framework we can ask:
What is the relation between the psychological reality
(e.g., the pain) and its brain counterpart.

It is important to recognize here that complex physical 
systems can develop properties that simpler physical systems 
lack. There is the "wheelness" of a wheel, the "automobileness"
of an automobile, the "temperature" and "pressure"
and "wind velocity" of a weather system, etc., and
these higher-level physical properties are both real
and useful in explaining the behaviour of these systems.
Thus an automobile is "not JUST a collection of atoms
and molecules": it is a collection of atoms and molecles
organized in a particular way, and in a way that causes
or allows it to behave in particular ways.

On the other hand, and automobile is, in another sense,
"nothing but" a certain sort of configuration of atoms
and molecules. In order for an entity or structure to BE 
an automobile it should be a collection of atoms and molecules
with certain structural and functional properties, and those
properties should, according to the materialistic stance
of classical physical theory, be, in principle, consequences 
of the laws and principles of classical physical theory plus
the (initial) conditions on the locations and velocities
of those particles (plus the EM and Grav fields).

The BRAIN STATE "associated" with a psychological reality
(e.g., a pain) is such a physical structure, if (classical)
materialism is true. But what is the connection between
that brain state and the associated psychological reality?

The essential verity here is that although the relevant
properties of the BRAIN STATE are analogous to the
properties of "wheelness", and "automobileness", and
"temperature", and "pressure", and "mitosis", and
"digestion", and "solidity", etc., the properties of
conscious experiences are not. There is an essential
difference, which eliminates the possibility of drawing
pertinent analogies between "feelings" (which include,
if broadly construed, all conscious experience) and these
other properties. For these other properties the physical 
implementation of the concept is achieved by restrictions
on the motions ultimately of atomic particles and molecules,
and the existence of a physical system that is behaving
in conformity to these restrictions rationally entails 
in principle the existence within nature of a system
that exhibits the defining characteristics of the concepts.

But there is no way to deduce from the laws and principles
of CP the existence of "feelings": the laws and principles 
of CP provide no basis for deducing that some
physically characterized system "feels" anything at all.
Some extra postulates would be needed, in order to bring
feelings in. But CP is dynamically complete without any 
such appendages.

Adding such appendages to a logically complete dynamical
system is gratuitous and unnatural. And once the essential
logical distinction between `feelings' and the other
mentioned properties is made clear it becomes, I believe,
irrational to claim, within the CP framework, that a conscious
feeling is THE SAME AS a physically describable property
of the brain.

There is, apart from its senselessness and unnaturalness, 
a more decisive argument against epiphenomenal consciousness.
I take it as given that consciousness co-evolved
with the brain during the development of homo sapiens.
But epiphenomenal consciousness has no physical
consequences, and hence can have no survival value.
Evolutionary psychology needs a physical theory in 
which our conscious thoughts and feeling are efficacious 
in the physical world. I see this as a compelling reason 
to turn to quantum physics for the foundations of the physics
of the mind-brain connection. Within that truer physics
our conscious thoughts and feelings are naturally
causally efficacious.