Bas van Fraassen
Daniel Hoffman
who happens to very interesting to many of us who love technology.
Allow me to provide a paragraph of tecnical history, for context.
We didn't need to know the details of the valence physics of doped
silicon to make semiconductors. We don't need any theory more
sophisticated than that of the valence theory of electrons to
develop photoetching, and probably could have done the same without
that theory. The discovery of semiconductors was a product of
technology, not scientific theory, which was used to explain the
phenomena discovered when an oscilliscope was used to test the
voltage drop across a silicon wafer. John Bardeen and William Shockley,
who subsequently got into hot water for advancing observations not
unlike those of Murray and Hernstein, had to find a theory to expalain
the phenomena of a pulsating signal riding upon the dc going through the
crystal. It turned out to be caused by variations in the light passing
through an electric fan. Once they knew that light altered conductivity,
they knew that they could optically couple an electronic signal. They
had the results in hand before they applied a theory of physics. From
what they had discovered by observing the effects of light, the
early silicon based transistors were field-effect, where a
non-conducting junction changed the resistance of the material to which
it was attatched. The effect of light on a junction between doped
silicon materials was known before the Heisenburg's theory of matrix
mechanics was applied, and that effect gave rise to the invention of the
PNP and NPN transistors. The point of all this is that tremendous
progress was made in technology without knowledge of the "true" nature
of the processes involved. Kant notwithstanding, technology can get by
just fine, thank you very much, without a secure foundation.
What's Bas van Fraasen have to do with all this?
In "To Save the Phenomena" (Journal of Philsophy, 73, no 18, (Oct 21,
1976 and reprinted in "The Philsophy of Science, Richard Boyd, ed.,
MIT Press, 1991 [a book well worth purchasing, BTW]), Van Fraassen
says:
The aim of science is to give us `a literally true story of what
the world is like;' and the proper form of acceptance of a
theory is to believe that it is true. This is the statement
of scientific realism: "To have good reason to accept a theory
is to have good reason to believe that the entities it
postulates are real," as Wilfrid Sellars has expressed it.
Accordingly, all intirealism is a position according to which
the aims of science can be well served without giving such a
literally true story, and acceptance of a theory may properly
involve something less (or other) than belief that it is true.
("Philosophy of Science," p.187)
The example of the semiconductor is meant to show that the belief
in the truth of the valence theory, with Heisenburg's "forbidden
zones" and all that, was not needed to advance semiconductor
technology. However, the history of that technology was advanced
by the use of that theory. When van Fraassen says "The phenomena
are saved when they are exhibited as fragments of a larger unity,"
(ibid, p.191) he sounds a lot like Alasdair MacIntyre, who said that
we cannot understand particular events and actions except as part of
larger wholes. We need a working context into which we can situate
experiences. The question of whether that context has to be
"correct" in order to situate experiences in such a way that progress
in science can be made is the point of dispute between realists and
anti-realists. To see where van Fraassen stands on this:
The idea of a literally true account has two aspects: the
language is to be literally construed; and, so constructed, the
account is true. This divides the antirealists into two sorts.
The first sort holds that science is or aims to be true,
properly (but not literally) construed. The second holds that
the language of science should be literally construed but that
its theories need not be true to be good. The antirealism I
advocate belongs to the second sort. (ibid, p.187)
Those of us who read and accept Berkeley's conclusions that we, to
anachronize a tad, understand in terms of the "brains in vats"
hypothesis, might also fall into the second sort. It doesn't
matter if the data I get from my senses reflect the true world
or are fed via a computer. If my mind sorts the given phenomena
into a larger unity that is able to meaningfully incorporate new
phenomena. If a "vatworlder" has a theory that adequately represents
the given world of empirical data, whether or not it adequately
represents the "true world" is a problem the vatworlder cannot answer.
It would be nice to be able to answer that latter question, but
most of us are getting by ok without being able to do so. I am
not certain that van Fraassen's second sort of anti-realist fits this
description, but will leave it as an open question for the moment.
Daniel Hoffman
Sysop, Republic 518 BBS
(215) 487-0863
rp51...@republic.netaxs.com
* OLX 2.1 * You're not in the traffic jam: You ARE the traffic jam!
Aaron Boyden
> In "To Save the Phenomena" (Journal of Philsophy, 73, no 18, (Oct 21,
> 1976 and reprinted in "The Philsophy of Science, Richard Boyd, ed.,
> MIT Press, 1991 [a book well worth purchasing, BTW]), Van Fraassen
> says:
Just thought I'd second your book recommendation here. I also found
Boyd's own essays quite interesting; is it just me, or does have have an
uncanny gift for making his foes' views seem more plausible than his own?
> From van Fraassen:
> The idea of a literally true account has two aspects: the
> language is to be literally construed; and, so constructed, the
> account is true. This divides the antirealists into two sorts.
> The first sort holds that science is or aims to be true,
> properly (but not literally) construed. The second holds that
> the language of science should be literally construed but that
> its theories need not be true to be good. The antirealism I
> advocate belongs to the second sort. (ibid, p.187)
>
> Those of us who read and accept Berkeley's conclusions that we, to
> anachronize a tad, understand in terms of the "brains in vats"
> hypothesis, might also fall into the second sort.
I would have thought that a Berkeleyan would fall into the first class of
anti-realists. Didn't Berkeley think that, for example, "chair" referred
to a complex of actual and potential perceptions? From the above, it
would seem that on van Fraassen's view "chair" refers to no such odd
logical construction. It purports to refer to chairs, just as the
realist would claim. Van Fraassen's break with the realist is in being
indifferent as to whether it actually succeeds in referring, prefering to
instead consider whether it's useful to suppose that it does. Sort of
like the contrast between Mackie's and Stevenson's subjectivism;
Stevenson thought that moral statements actually referred to people's
attitudes, while Mackie thought that they purported to refer to
extra-mental moral facts. Of course, Mackie went a little farther with
respect to morality than van Fraassen does with respect to science;
Mackie thought that those extra-mental moral facts definitely didn't
exist to be referred to.
Personally, I find myself more sympathetic to the first sort of
anti-realism; if the actual existence of objects is shorn from everything
which is relevant to determining whether supposing their existence is
useful or not, I get terribly confused about what it means to say an
object actually exists. It begins to strike me as an unintelligible
question, rather than just an unimportant one. I suppose that that is
not incompatible with the second view; if one says no more about the ding
an sich than to explain that nothing can be said about it, I can't say
that any great mistake has been committed. However, I think it creates
less potential for misunderstanding if you don't give a name to that
about which you cannot talk, particularly a name (that which exists)
which so obviously invites confusion with something that actually matters
to us (that which we suppose to exist).
Still, the distinction is not really of any great significance. Both
classes of anti-realism separate knowledge from the ding an sich,
connecting it instead to pragmatic factors; where precisely the
separation occurs is of much less importance than that fundamental insight.
Calvin Bruce Ostrum
Aaron Boyden <650...@ucsbuxa.ucsb.edu> wrote:
| > In "To Save the Phenomena" (Journal of Philsophy, 73, no 18, (Oct 21,
| > 1976 and reprinted in "The Philsophy of Science, Richard Boyd, ed.,
| > MIT Press, 1991 [a book well worth purchasing, BTW]), Van Fraassen
| > says:
|
| Just thought I'd second your book recommendation here. I also found
| Boyd's own essays quite interesting; is it just me, or does have have an
| uncanny gift for making his foes' views seem more plausible than his own?
I would also recommend this collection. As for Boyd, I think he does an
admirable job of fairly presenting the views of his opponents, but I also
find his own views immensly more satisfying (does that mean I find them
more "plausible"? I'm not sure. I think there still may be room for some
kind of pragmatic anti-realism that parallels his inference-to-the-best-
explanation-reflective-equilibrium-situated-tracking-of-natural-kind type
realism much better than it does what he identifies as constructivism and
empiricism).
A very nice recently published book that deals with naturalised epistemology
Boyd-style is Hilary Kornblith's "Inductive inference and its natural
ground" and I'd highly recommend it also.
| like the contrast between Mackie's and Stevenson's subjectivism;
| Stevenson thought that moral statements actually referred to people's
| attitudes, while Mackie thought that they purported to refer to
| extra-mental moral facts. Of course, Mackie went a little farther with
| respect to morality than van Fraassen does with respect to science;
| Mackie thought that those extra-mental moral facts definitely didn't
| exist to be referred to.
The "error theory" versus the "shrug theory". It's funny you should bring
morality into it, after mentioning Boyd. Are you familiar with his
"How to be a moral realist"? In this paper he attempts to show that
the same arguments he uses for scientific realism can also be used to make
moral realism more plausible. Unfortunately I'm afraid this cuts both
ways: someone who can't see how moral realism could possibly work, upon
finding Boyd's arguments allegedly support it, is likely to have his
confidence in scientific realism undermined also (and I think there are
quite a few scientific realists who are moral anti-realists).
| Personally, I find myself more sympathetic to the first sort of
| anti-realism; if the actual existence of objects is shorn from everything
| which is relevant to determining whether supposing their existence is
| useful or not, I get terribly confused about what it means to say an
| object actually exists. It begins to strike me as an unintelligible
| question, rather than just an unimportant one. I suppose that that is
You appear to be suggesting that "unintelligible" implies "unimportant".
Is this not perhaps rather presumptuous? What do you think of, for
example, Nagel's argument contra Davidson, and "Davidson Junior"? People
like Nagel and Fodor seem quite correct to say that there are many other
organisms where "unintelligible" does *not* imply "unimportant", but the
analogy really doesn't seem to go through, does it. It's like once you
hit a certain spot, you can form a thought ultimately general enough that
it succeeds in referring to "everything that is" and "everything that
matters", even though very little else can be attributed to its referent.
But is this enough to make anti-realism true?
---------------------------------------------------------------------------
Calvin Ostrum c...@cs.toronto.edu
---------------------------------------------------------------------------
It is a far, far better thing to have a firm anchor in nonsense than
to put out on the troubled seas of thought. -- John Kenneth Galbraith
---------------------------------------------------------------------------
Aaron Boyden
> | Personally, I find myself more sympathetic to the first sort of
> | anti-realism; if the actual existence of objects is shorn from everything
> | which is relevant to determining whether supposing their existence is
> | useful or not, I get terribly confused about what it means to say an
> | object actually exists. It begins to strike me as an unintelligible
> | question, rather than just an unimportant one. I suppose that that is
>
> You appear to be suggesting that "unintelligible" implies "unimportant".
I'm not sure how you got that impression. I derived "unimportant" from
van Fraassen's arguments, and considered "unintelligible" to be something
separate which van Fraassen would, so far as I can tell, not endorse.
However, now that you mention it, I think I have an argument to the
effect that unintelligible entails unimportant. Importance is relative;
something can only be important for some purpose. If a given X is
important for purpose P, then it is at least slightly intelligible;
however great our ignorance about other aspects of X, we at least know
something about it, namely that it's important for P. Thus, something
which is totally unintelligible could not be important, at least not for
us (and who else have we the right to speak for?) This is as opposed to
the only partially unintelligible, which can of course be very important,
but I wasn't meaning to suggest unintelligibility in any but the
strongest sense above.
Nancy McGough
>On Sat, 18 Mar 1995, Daniel Hoffman wrote:
>> From van Fraassen:
>> The idea of a literally true account has two aspects: the
>> language is to be literally construed; and, so constructed, the
>> account is true. This divides the antirealists into two sorts.
>> The first sort holds that science is or aims to be true,
>> properly (but not literally) construed. The second holds that
>> the language of science should be literally construed but that
>> its theories need not be true to be good. The antirealism I
>> advocate belongs to the second sort. (ibid, p.187)
>>
>> Those of us who read and accept Berkeley's conclusions that we, to
>> anachronize a tad, understand in terms of the "brains in vats"
>> hypothesis, might also fall into the second sort.
>
>I would have thought that a Berkeleyan would fall into the first class of
>anti-realists. Didn't Berkeley think that, for example, "chair" referred
>to a complex of actual and potential perceptions? From the above, it
>would seem that on van Fraassen's view "chair" refers to no such odd
>logical construction. It purports to refer to chairs, just as the
>realist would claim. Van Fraassen's break with the realist is in being
>indifferent as to whether it actually succeeds in referring, prefering to
>instead consider whether it's useful to suppose that it does.
I think that talking about an observable, like a chair, is a confusing
way to try to understand van Fraassen's flavor of antirealism. It's
better to replace "chair" above with something like "quark" and then
ask which sort of antirealist Berkeley is. I don't know much about
Berkeley but I'm interested in the comparing all the different types
of antirealism, including the "brains in a vat" antirealism. Isn't a
"brains in a vat" antirealism the antithesis of naive realism, where
by "naive realism" I mean just "the belief that there is a physical
reality independent of me that I can know something about"?
>Personally, I find myself more sympathetic to the first sort of
>anti-realism; if the actual existence of objects is shorn from everything
>which is relevant to determining whether supposing their existence is
>useful or not, I get terribly confused about what it means to say an
>object actually exists. It begins to strike me as an unintelligible
>question, rather than just an unimportant one. I suppose that that is
>not incompatible with the second view; if one says no more about the ding
>an sich than to explain that nothing can be said about it, I can't say
>that any great mistake has been committed. However, I think it creates
>less potential for misunderstanding if you don't give a name to that
>about which you cannot talk, particularly a name (that which exists)
>which so obviously invites confusion with something that actually matters
>to us (that which we suppose to exist).
You say:
[1] "that which exists" is "that about which you cannot talk"
[2] "that which we suppose to exist" is "that which actually matters"
To me [1] clearly makes you an antirealist and [2] puts you in the
first sort of antirealist that van Fraassen describes. Since I am
pretty much alligned with van Fraassen's antirealism (the second sort
in the BvF quote) I wouldn't say [2], rather I'd say:
[2'] We don't supppose any of our scientific entities exist, rather
they are useful "fictions" that are empirically adequate. (They
might in fact exist, but that's not relevant.)
>Still, the distinction is not really of any great significance. Both
>classes of anti-realism separate knowledge from the ding an sich,
>connecting it instead to pragmatic factors; where precisely the
>separation occurs is of much less importance than that fundamental insight.
I actually do think the distinction is significant. BvF's sort of
antirealism is much less constrained than the first sort. You can
easily move between different scientific models, e.g., different
interpretations of Quantum Mechanics, using whichever one works best
for the problem at hand. In the first sort you are always constrained
by the need to be working in a true model.
-Nancy
--
/\_/\ @..@ Vote for the humanities.misc /\_/\
( o.o ) Nancy McGough (----) newsgroup! The CFV is at ( o.o )
> ^ < Infinite Ink ( >__< ) http://www.jazzie.com/ii/ > ~ <
Aaron Boyden
> I actually do think the distinction is significant. BvF's sort of
> antirealism is much less constrained than the first sort. You can
> easily move between different scientific models, e.g., different
> interpretations of Quantum Mechanics, using whichever one works best
> for the problem at hand. In the first sort you are always constrained
> by the need to be working in a true model.
I think I understand this. If I'm correct, I would like to make a small
modification to my view; I favor being both kinds of anti-realist at the
same time. The two questions are whether we feel constrained by what we
purport to refer to, and what exactly it is that we purport to refer to.
It seems to me that if the first version of anti-realism is used to
answer the second question, it becomes more natural to answer "no" to the
first question, which seems to be the question addressed by the second
version of anti-realism. I'm not at all convinced that this argument
works, though.