[ Pobierz całość w formacie PDF ]
relation between the content of experiences and the content of the concepts they
eventuate in locking to is so rarely arbitrary.
end p.128
Well, maybe. But, of course, that's cold comfort if what you want is a non-nativist
version of SIA. You can only trigger a concept that's there, genetically specified, waiting
to be triggered. So the Darwinian/ethological story about concept acquisition does no
better than the old-fashioned hypothesis-testing story at making DOORKNOB not be
innate. Out of one frying pan but into another; ethologists are nativists by definition.
And, anyhow, even if the doorknob/DOORKNOB relation is selected for by evolution,
what, if not inductive learning, could be the mechanism by which it is implemented? If
concept acquisition isn't inductive, then how does Mother Nature contrive to insure that it
is instances of F-ness (and not of G-ness) that trigger the concept F in the course of
ontogeny? After all, if Mother N wants to select for the doorknob/DOORKNOB type of
relation between concepts and their experiential causes, she has to do so by selecting a
mechanism that produces that relation between one's concepts and their causes. This is a
special case of the entirely general truth that whenever Mother N wants to select for any
phenotypic property she has to do so by selecting a proximal mechanism that produces it.
The obvious candidate to select if one wants to ensure that concept acquisition exhibits
the d/D relation is inductive learning. But we have it on independent grounds that
primitive concepts can't be learned inductively. There may be a way for a conceptual
atomist to get out of this dilemma, but waving his hands about Darwin certainly isn't it.
The preliminary moral, anyhow, is that radical nativism is very hard for a conceptual
atomist to avoid. If he starts out thinking about concept acquisition the way Empiricists
do as a kind of hypothesis testing radical concept nativism follows; and if he starts
out thinking about concept acquisition the way that ethologists do as a kind of
triggering radical concept nativism still follows. It looks like a conceptual atomist ends
up being a radical concept nativist pretty much however he starts out thinking about
concept acquisition. So maybe conceptual atomism is just false.
Or maybe radical concept nativism is true, despite its wide unpopularity in the
philosophical community. Speaking just as a private citizen, I've always sort of thought it
wouldn't be all that surprising if radical concept nativism did turn out to be true. So it
didn't much embarrass me that all the roads from concept atomism seemed to lead there.
It is, after all, God and not philosophers who gets to decide what creatures have
genotypically built in. That is surely much the best arrangement from the creature's point
of view.
So, in any case, it seemed to me in 1975 or so. But maybe this relaxed stance won't do
after all. The problem with the theory that the primitive concepts are learned inductively
was that it's circular. But now we seem to
end p.129
have an apparently respectable argument that they must be learned inductively: nothing
else appears likely to account for the content relation between the concept that's acquired
and the experience that mediates its acquisition. But look, it can't be that inductivism
about the acquisition of primitive concepts is both circular and mandatory.
Please note that, though this is an embarrassment for those of us who are inclined towards
atomism, it is also an embarrassment for those of you who aren't. For, whatever you may
think about the size of the primitive conceptual basis and, in particular, about whether
DOORKNOB is in it on any version of RTM some concepts are going to have to be
primitive. And, on the one hand, SA does seem to show that primitive concepts can't be
acquired inductively. And, on the other hand, whatever the primitive concepts are, their
acquisition is pretty sure to exhibit the familiar d/D relation between the content of the
concept and the content of the experience that occasions it. Of what concept does the
acquisition not?9
In fact, it's the concepts that have traditionally been practically everybody's favourite
candidates for being primitive that exhibit the doorknob/DOORKNOB effect most
clearly. Like RED, for example. To be sure, philosophers of both the Cartesian and the
Empiricist persuasion have often stressed the arbitrariness of the relation between the
content of sensory concepts and the character of their causes. It's bumping into photons
(or whatever) that causes RED; but RED and PHOTON couldn't be less alike in content.
(According to Descartes, this shows that not even sensory concepts can come from
experience. According to Locke, it shows that secondary qualities are mind-dependent.)
Well, if the relation between sensory concepts and their causes really is arbitrary, then
there can be no d/D problem about sensory concepts. In which case, if Empiricists are
right and only sensory concepts are primitive, everything turns out OK. Sensory concepts
don't have to be learned inductively, so they can be innate; just as the Standard Argument
requires, and just as Empiricists and Rationalists have both always supposed them to be.
Empiricism would be cheap at the price if it shows the way out of a foundational paradox
about RTM.
But, on second thought, no such luck. The thing to keep your eye on, pace Locke and
Descartes both, is that the relation between the content of a sensory concept and the
character of its cause is not arbitrary when the cause is intentionally described. The thing
to keep your eye on is that we typically get the concept RED from (or, anyhow, on the
occasion of) experiencing things as red.
There is, I think, more than a hint of a muddle about this in Fodor 1981a, where the
following is a favourite line of argument: Look, everybody Empiricists and
Rationalists agrees that there is at least one psychological mechanism which effects a
non-rational, arbitrary relation between at least some primitive concepts and their distal
causes. In particular, everybody agrees that the sensorium works that way. [E]ven the [ Pobierz całość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl szamanka888.keep.pl
relation between the content of experiences and the content of the concepts they
eventuate in locking to is so rarely arbitrary.
end p.128
Well, maybe. But, of course, that's cold comfort if what you want is a non-nativist
version of SIA. You can only trigger a concept that's there, genetically specified, waiting
to be triggered. So the Darwinian/ethological story about concept acquisition does no
better than the old-fashioned hypothesis-testing story at making DOORKNOB not be
innate. Out of one frying pan but into another; ethologists are nativists by definition.
And, anyhow, even if the doorknob/DOORKNOB relation is selected for by evolution,
what, if not inductive learning, could be the mechanism by which it is implemented? If
concept acquisition isn't inductive, then how does Mother Nature contrive to insure that it
is instances of F-ness (and not of G-ness) that trigger the concept F in the course of
ontogeny? After all, if Mother N wants to select for the doorknob/DOORKNOB type of
relation between concepts and their experiential causes, she has to do so by selecting a
mechanism that produces that relation between one's concepts and their causes. This is a
special case of the entirely general truth that whenever Mother N wants to select for any
phenotypic property she has to do so by selecting a proximal mechanism that produces it.
The obvious candidate to select if one wants to ensure that concept acquisition exhibits
the d/D relation is inductive learning. But we have it on independent grounds that
primitive concepts can't be learned inductively. There may be a way for a conceptual
atomist to get out of this dilemma, but waving his hands about Darwin certainly isn't it.
The preliminary moral, anyhow, is that radical nativism is very hard for a conceptual
atomist to avoid. If he starts out thinking about concept acquisition the way Empiricists
do as a kind of hypothesis testing radical concept nativism follows; and if he starts
out thinking about concept acquisition the way that ethologists do as a kind of
triggering radical concept nativism still follows. It looks like a conceptual atomist ends
up being a radical concept nativist pretty much however he starts out thinking about
concept acquisition. So maybe conceptual atomism is just false.
Or maybe radical concept nativism is true, despite its wide unpopularity in the
philosophical community. Speaking just as a private citizen, I've always sort of thought it
wouldn't be all that surprising if radical concept nativism did turn out to be true. So it
didn't much embarrass me that all the roads from concept atomism seemed to lead there.
It is, after all, God and not philosophers who gets to decide what creatures have
genotypically built in. That is surely much the best arrangement from the creature's point
of view.
So, in any case, it seemed to me in 1975 or so. But maybe this relaxed stance won't do
after all. The problem with the theory that the primitive concepts are learned inductively
was that it's circular. But now we seem to
end p.129
have an apparently respectable argument that they must be learned inductively: nothing
else appears likely to account for the content relation between the concept that's acquired
and the experience that mediates its acquisition. But look, it can't be that inductivism
about the acquisition of primitive concepts is both circular and mandatory.
Please note that, though this is an embarrassment for those of us who are inclined towards
atomism, it is also an embarrassment for those of you who aren't. For, whatever you may
think about the size of the primitive conceptual basis and, in particular, about whether
DOORKNOB is in it on any version of RTM some concepts are going to have to be
primitive. And, on the one hand, SA does seem to show that primitive concepts can't be
acquired inductively. And, on the other hand, whatever the primitive concepts are, their
acquisition is pretty sure to exhibit the familiar d/D relation between the content of the
concept and the content of the experience that occasions it. Of what concept does the
acquisition not?9
In fact, it's the concepts that have traditionally been practically everybody's favourite
candidates for being primitive that exhibit the doorknob/DOORKNOB effect most
clearly. Like RED, for example. To be sure, philosophers of both the Cartesian and the
Empiricist persuasion have often stressed the arbitrariness of the relation between the
content of sensory concepts and the character of their causes. It's bumping into photons
(or whatever) that causes RED; but RED and PHOTON couldn't be less alike in content.
(According to Descartes, this shows that not even sensory concepts can come from
experience. According to Locke, it shows that secondary qualities are mind-dependent.)
Well, if the relation between sensory concepts and their causes really is arbitrary, then
there can be no d/D problem about sensory concepts. In which case, if Empiricists are
right and only sensory concepts are primitive, everything turns out OK. Sensory concepts
don't have to be learned inductively, so they can be innate; just as the Standard Argument
requires, and just as Empiricists and Rationalists have both always supposed them to be.
Empiricism would be cheap at the price if it shows the way out of a foundational paradox
about RTM.
But, on second thought, no such luck. The thing to keep your eye on, pace Locke and
Descartes both, is that the relation between the content of a sensory concept and the
character of its cause is not arbitrary when the cause is intentionally described. The thing
to keep your eye on is that we typically get the concept RED from (or, anyhow, on the
occasion of) experiencing things as red.
There is, I think, more than a hint of a muddle about this in Fodor 1981a, where the
following is a favourite line of argument: Look, everybody Empiricists and
Rationalists agrees that there is at least one psychological mechanism which effects a
non-rational, arbitrary relation between at least some primitive concepts and their distal
causes. In particular, everybody agrees that the sensorium works that way. [E]ven the [ Pobierz całość w formacie PDF ]