Feser and Ross reprised

Ed Feser  responds to a post by physicist adversary Robert Oerter, who has picked up James Ross's Immaterial Aspects of Thought article, here, here, and here.  Oerter seems to have misunderstood Ross somewhat.  Not difficult given Ross's rather opaque writing style.  The only author I know to use the word incompossible.  Several times per page.  Ross tries to put clear blue water between the mental and the physical.  Some formal thought is determinate he says, physical processes never are.  Ne'er the twain shall meet.  This would be a nice coup if you could pull it off. But alas, human thinking is less determinate than Ross would have us believe, and physical behaviour is more.

1. Formal thinking is an abstraction from actual thinking.  Actual thinking is rather less determinate.  Dutifully calculating 1857 times 253 by long multiplication my mind wanders to Auntie Flo's imminent birthday.  On the other hand, water flows determinately downhill.  And there is determinately one hole in a torus.  There is no absolute determinacy.  One needs to ask, Determinate in what respect?

2. Ed says,

For another thing, even to deny the premise All formal thinking is determinate requires that one determinately grasp precisely the patterns one denies we have a determinate grasp of.

Machines can recognise patterns.  Patterns are a syntactic phenomenon.  The more one tries to bring out the determinacy of thought the more recourse one makes to purely syntactic accounts.  And machines can do syntax.  

3.  Ross and Feser enlist Quine to show that there is no fact of the matter as to what a physical device, a calculator, say, means by its behaviour.  But, of course, there is a fact of the matter as to what I mean by the rows and columns of digits in my multiplication.  Except perhaps when I break off to sketch what a generous art critic might see as an old lady wearing a hat.  Why the asymmetry of treatment?  Oerter picks up on this in a later post.  Ross allows the human calculator to look inside himself to find his meaning.  His argument, in so far as it concerns meaning (and Feser also emphasises meaning), is the argument from intentionality in disguise.

4.  But meaning is a red herring.  What's at stake is the determinacy or otherwise as to what is going on.  Now Kripke is wheeled on to show that there is no determinate fact as to what is going on in the calculator inferable from its past behaviour (or indeed its whole lifetime of behaviour).  Sure.  But why treat the machine as a black box?  Look carefully inside and you see that at some level of abstraction, what it's doing is multiplying.  We have been here before.  Likewise, abstracting away the temporary excursion into millinery, that is what I am doing with my rows and columns of digits.

UPDATE Friday 18 October

Ed Feser responds to Oerter here.   He makes a clarification

Part of the problem here might be that Oerter is not carefully distinguishing the following two claims:

(1) There just is no fact of the matter, period, about what function a system is computing.
(2) The physical properties of a system by themselves don’t suffice to determine what function it is computing.

Oerter sometimes writes as if what Ross is claiming is (1), but that is not correct. Ross is not denying, for example, that your pocket calculator is really adding rather than “quadding” (to allude to Kripke’s example). He is saying that the physical facts about the machine by themselves do not suffice to determine this. Something more is needed (in this case, the intentions of the designers and users of the calculator).

In which case, I fail to see the point of the lengthy discussion regarding the underdetermination of the calculator's function from the history of its inputs and outputs.   But then Ed says

Ross’s and Kripke’s argument is completely unaffected even if we add not only all future, but all possible, physical behavior of the machine.

But this is giving too much away.  If we know all possible inputs and all the corresponding outputs then we know the domain and range and mappings of the function the machine calculates.

Let's return to point (2).  Nowhere in Ross's article or Feser's summaries can I find an argument for this.  That any finite history of inputs and outputs underdetermines the function the machine is calculating is accepted but irrelevant.  Ed once again goes through the motions of explaining that the meaning of a physical symbol is undetermined by its structure.  Again this is accepted but irrelevant.  The function calculated isn't the meaning of the calculator or its behaviour.  It's a description of the behaviour.  I fail to see why the intentions of the device's designers and users are relevant to a description.  Are Ross and Feser claiming that such a description leaves something out?  Well, of course.  Every description leaves something out.  No description can be complete and all descriptions are abstractions. Are they using function here more in the sense of purpose?  If so, human intentions would then come into the picture.  But they aren't.  They are clearly talking about the mathematical function the machine computes.

UPDATE Sunday 20 October

Ed Feser replies once more to Robert Oerter here.  After wasting six paragraphs on an embarrassing and unhelpful little story he gets down to further exegesis and clarification of James Ross.  He quotes the following paragraph from Ross, with emphasis added,

There is no doubt, then, as to what the machine is doing. It adds, calculates, recalls, etc., by simulation. What it does gets the name of what we do, because it reliably gets the results we do (perhaps even more reliably than we do) when we add by a distinct process. The machine adds the way puppets walk. The names are analogous. The machine attains enough reliability, stability, and economy of output to achieve realism without reality. A flight simulator has enough realism for flight training; you are really trained, but you were not really flying.

Feser then goes on to say,

End quote. So, Ross plainly does say that there is a sense in which the machine adds -- a sense that involves simulation, analogy, something that is “adding” in the way that what a puppet does is “walking.” How can that be given what he says in the passage Oerter quotes? The answer is obvious: The machine “adds” relative to the intentions of the designers and users, just as a puppet “walks” relative to the motions of the puppeteer. The puppet has no power to walk on its own and the machine has no power to do adding (as opposed to “quadding,” say) on its own. But something from outside the system -- the puppeteer in the one case, the designers and users in the other -- are also part of the larger context, and taken together with the physical properties of the system result in “walking” or “adding” of a sort.

In short, Ross says just what I said he says.

However, two paragraphs before the one that Feser quotes we find Ross saying,

For any outputs to be sums, the machine has to add. But the indeterminacy among incompossible functions is to be found in each single case, and therefore in every case. Thus, the machine never adds.

Those of us who find Ross's thought less than crystal clear are not helped by Feser's opening homily.  Let's continue, then, with the argument as Ed himself presents it.  He seems to be saying that a calculator is two steps away from genuine adding, multiplying, etc.  He is saying that the argument from physical indeterminacy of function shows that there is no fact of the matter as to what function the machine is calculating.  But if we impose, as it were, the intentions of its designers and users upon the device we get a sort of adding, multiplying, etc.  What we must do to get the genuine article Ed doesn't reveal.

Reading Ed's remarks I was transported back to my grammar school days in the 1960s.  Computers were just becoming widely known about.  We had a day at the National Physical Laboratory where we were taught the basics of Algol60.  I remember writing a tiny program to calculate e which I punched out onto paper tape and ran on a room-sized Elliott 4100 machine, I think it was.  What astonished me in those days was that an artificial device could be made that did all that complicated stuff that I had to do in my head and on paper in order to add, let alone multiply, a pair of numbers.  And you could tell them what to do in a restricted version of the kind of language we talked in maths classes---or so it seemed at the time.   It wasn't until a few years later, when I'd acquired a maths degree and further experience working with the damn machines and having to learn some computer science, that it became clear that what the things were doing came down to manipulating electronic representations of patterns.  That's all addition and multiplication and the rest were---just pattern juggling.  Quite sophisticated patterns, perhaps, and complicated juggling, but ultimately just that.  That's why long multiplication by hand was so dull and error-prone.  One had to force oneself to behave like a machine, something we are not well-adapted to do.  Of course, only we knew what adding and multiplying meant, when it was appropriate to add and when to multiply, and so on, in order to work something out.  But to do the actual addition or multiplication we had to go into cipher-mode and follow the recipe mechanically.

I think Feser and Ross get the puppet metaphor back to front.  When we add or multiply it is we who become the puppets, hobbled and jerked around on the strings of the puppeteering algorithm that we have to impose on ourselves in order to have some hope of getting the right answer.