FW: [FOM] Re: Contrasting methodologies

tom holden thomas.holden at balliol.ox.ac.uk
Fri Oct 3 19:23:04 EDT 2003

The following message was originally a private response to a private e-mail
Friedman sent me; having now posted his message to the list, he suggested I
do the same. As such, please do not get overly "up in arms" over my comments
on Godel, Tarski etc., they were intended as examples, not as detailed
summaries of the current philosophical thinking. Also with this in mind, I
apologise for the briefness of some of the quotes, please refer to
Friedman's original mail if their context is not clear:
Thank you for your reply.

I must say I'm still puzzled by your notion of "great permanent value" and
would be inclined to classify many of the great philosophical texts
(including some modern ones) as of considerably more "permanent value" than
even the development of ZFC. Lack of popular consensus on a work does not
mean it is of no value. But we shall agree to differ.

As for Godel, yes I am perfectly aware of his work and its place in the
literature, but the impression I get is there is nowadays a distinct
movement against attributing any great philosophical significance to it.
After all there are still formalists and Maddy/Yablo-esque

>I take it that the philosophical question you are concerned about is
>*what does it mean to follow a trivial rule like: given a finite string of
>bits, append 1 at the end*

Not quite. More "can we learn to follow a trivial rule like:... and in what
sense are we 'justified' in the continuation we choose," maybe this is the
root of some of our misunderstandings.

>My main f.o.m. like suggestions are by no means appropriately characterized
>in terms of "formal complexity measures". E.g., I was talking about the
>deep intellectual work needed to
>*get a computer to follow simple rules when appropriately prompted by a
>human to do so*

I know. I was merely using complexity measures as an example. With my more
closely specified "philosophical question", I hope you can see my point that
as computers are made "in their creators [linguistic] image" at the
ASM/program level they do not answer the "philosophical question." And at
higher levels, "neural nets" suffer the same problems of finite learning
sets we do, and "evolutionary algorithms" often include complexity measures
as one of their optimization methods.

>I suspect that 
>***the deep 'cart before the horse' issues that had to be fleshed out in
>order to build a real (even very low level) computer system based on the
>stored program idea***

I don't, simply because as no one actually worries about whether they're
adding 1 right, it's unlikely computer scientists would. But as we're both
working on the level of suspicions, again I propose we agree to differ.

>Now do something with that idea. Provide a calculus for determining
>which ones concur with linguistic use and which ones do not. Go well beyond

Use is transient, formalisation in a calculus defeats the purpose. It may be
interesting to formalise a snap shot, (just as people have done many times
before with other areas e.g. the many logics in existence) but it's not
going to answer any of my questions.
>I know of nothing invulnerable to formal treatment.

Yes nothing is invulnerable to formal treatment, but many things are
invulnerable to formal treatment in a way which preserves the answers to
philosophical issues. E.g. What is truth?: Philosophers: Errr. Logicians:
Easy, formalise it, Tarski condition is all there is. Few philosophers (e.g.
Davidson): Ahh so that's all there is to truth. Many more: No however
correct Tarksi may be, it does not answer the philosophical issue.

>An informal form of some symmetry argument...

But with implications for the possibility of ever successfully applying any
predicate, even "=100+1"

>And how does this relate to the fact that we have no problems now with
>computers misbehaving on trivial tasks?

As I said in my first message, reliance on computer behaviour comes down to
a combination of finite testing of individual ops. against our conception of
the "right" way of applying such ops. combined with scientific knowledge.

It has been interesting having this discussion and I thank you for taking
the time to reply in depth. I look forward to more such over the coming

Tom Holden

More information about the FOM mailing list