Friday, November 27, 2015

On the Indeterminacy of the Physical

James Ross has argued that thought must be at least partly non-physical, int his variant of the Argument from Reason, which he outlines in his paper Immaterial Aspects of Thought.  This argument is quite similar to CS Lewis' AFR, since they both claim that human intellect has properties that cannot be accounted for by any purely naturalistic explanation.  Both of these arguments make claims that are epistemically unjustified and inconsistent with a scientific understanding of cognition.  Victor Reppert's defense of the AFR does this too, and I have argued this point with him, and even showed him some reading material that would help him understand the scientific perspective, but he continues to refuse to learn any relevant science. 

The argument made by Ross is that thinking of a certain type has the property of 'determinacy', which no physical process can have.  In my previous post, I addressed the issue that Ross does not clearly define what is meant by the term 'determinacy'.  He seems to use the term in two distinctly different ways.  But if the word is understood in either of those ways, his claims about the determinacy of the intellect and the indeterminacy of physical processes deny the reality that is clearly observable.  I suspect that like Reppert, Ross must have a similar anti-scientific bent.  I will show that his argument is false.

Here is Ross' argument stated as a syllogism:
(1) All formal thinking is determinate, but
(2) No physical process is determinate, so
(3) No formal thinking is a physical process.
Formal thinking, according to Ross, makes use of mathematics or logical rules such as modus ponens or conjunction.  He says formal thinking is characterized by understanding the logic involved in the thinking process.  So in the case of adding, for example, the human intellect can perform "genuine adding", as opposed to a physical device that only simulates adding by following some mechanistic process or an algorithm.
Adding - genuinely adding, not estimating - is a sum-giving thought form for any suitable array of numbers. ...  There is a great difference between adding incorrectly and doing something else, like guessing, estimating, or following a routine or algorithm. The adding I am talking about, like conjoining, is a form of understanding.
Ross appears to be saying that logical processes involved in formal thinking are not accomplished in the same way as any physical process, and that is why they are determinate.  Aside from the problem that it is unclear what Ross really means by "determinate" (see the discussion in my earlier post), Ross claims that the process of formal thinking is different from an algorithmic process that a computer would perform.  But that's not really true.

When we solve a problem of arithmetic, mathematics, or logic, we actually use algorithms.  We go through a step-by-step process, something like this: Eight plus seven is fifteen.  Write down the five and carry the one ...  That's how we actually do our formal thinking, and it's indistinguishable from a mechanistic algorithm that a computer might do.  The process of adding base ten numbers requires using 55 memorized facts, which are the sums of the digits, and following a mechanistic procedure.  In fact, all logical thinking involves a combination of basic facts and procedures.  On what basis, then, does Ross claim that our thinking is different from a physical process?  The only real difference is that a person performing this process is aware of what he's doing.  But that's irrelevant to the procedure used, or the outcome.  The claim of determinacy, based on formal thinking being somehow different from a physical process, appears to be completely unjustified. 

But Ross goes beyond the claim that the thinking process itself if different.  He claims that the outcome of the process is different, too.  Because formal thinking supposedly involves a true realization of the Platonic form of some process, and no physical process can fully realize that form, the output of the process must be at best an approximation of the ideal result.  Ross makes the analogy of a door being an approximation of the Platonic form of a rectangle.  So does that imply that a machine adding 2 and 2 will arrive at a result that is not exactly 4, but an approximation of it?  Of course not.  That's not how discreet arithmetic works, even if it is done by a machine.  The answer is exactly 4.  And that's the answer that comes out every single time, as long as the machine is functioning properly.  So what is there about this physical process that is indeterminate?  And if the outcome of the process is indistinguishable from formal thinking, on what basis does Ross make the claim that one is determinate and the other one isn't?

Ross also speaks of a different kind of indeterminacy in a physical process.  He says that we can never be absolutely certain what function may actually be implemented by a machine.  We may see that the machine's output appears to be consistent with some function, such as  x + y, but we can't be sure that at some future time, it won't produce x + y + 1 as the output.  For that reason, Ross thinks that any physical process is indeterminate in a way that formal thinking isn't.  But this kind of indeterminacy only applies when the process by which the machine produces these results is hidden from our examination.  If we know that a computer's program specifies x + y, then there is nothing at all about this physical process that is indeterminate.  We can be sure that it will never produce x + y + 1, as long as the computer is functioning properly.  So once again, the claim of indeterminacy made by Ross on this basis, is lacking justification.

Ross' claims about the determinacy, as they relate to formal thinking and physical processes, are not only epistemically unjustified, but they contradict what we know, both about thinking and about other physical processes.  Like Lewis' AFR, this argument is based on premises that are unscientific and unfounded.


  1. Nice analysis. Germane to how the human brain deals with numbers, I recommend "The Number Sense" by Stanislas Deheane. (His other books are good as well.) He doesn't deal with philosophical implications just the neurosceince but it is good stuff.

    1. Thanks. I just got a copy of it. Looks interesting.