[hist-analytic] Frrom AUNE: Analytic and A Priori
danny.frederick at tiscali.co.uk
Mon Mar 23 11:33:55 EDT 2009
Hi Roger and Steve,
The use of Popper's demarcation criterion for empirical knowledge to define
a notion of a priori knowledge is due to me. But I accept your point that
the way I have done this will not do as it stands. So let me amplify a
Objective knowledge is inter-subjectively testable. It is empirical if it is
testable against public observation statements. It is a priori if it is
testable in some other way, such as algorithms or derivations. Of course, we
need something in addition to algorithms and derivations, namely, intuitive
evidence of necessity. The latter can vary widely between people; but so can
observation reports of an event which they have just witnessed. In each case
we need there to be some statements (empirical or a priori) on which most
people can agree easily. Well-conducted and repeatable experiments in
science usually do this for empirical basic statements; and there are many
mathematical and logical 'basic statements' on which general agreement is
easily reached, such as '1 + 1 = 2' and 'if (p & q), then q.'
Subjective knowledge is not inter-subjectively testable. It is therefore not
a part of objective knowledge, not a part of shared human experience, and is
usually of interest only to the people who possess it (and perhaps some of
the people closest to them). But if we want to bother ourselves with
solipsistic epistemology, we can draw an a priori/empirical distinction for
it. In this case, I would distinguish statements prompted by sensory
experiences and statements prompted by reflection. And I would say that a
piece of solipsistic knowledge is empirical if it is testable against
statements prompted by sensory experience; but it is a priori if it is known
without being tested against statements prompted by sensory experiences.
Since my awareness of my thinking is not a sensory experience, 'I think' is
a piece of subjective knowledge which is a priori. And it is contingent.
As I have said before, I think I am following the traditional rationalist
philosophers in what I say about subjective knowledge; but I admit that it
is not a clear-cut matter because they do not seem explicitly to have
addressed the questions I am considering. But I am out of touch with the
history of philosophy, so if anyone knows better, I will be pleased to hear.
There is an important point I made a few messages back (in response to
Bruce) that will bear repeating. It is not the case that ALL subjective
reports of experience are debarred from science or from objective knowledge
in general. A psychologist or neuroscientist may well correlate behaviour,
bodily changes or brain states with psychological states by means of
subjects' reports of what they are experiencing under experimentally
produced circumstances. But such subject reports are essentially
inter-subjective in that the subject does duty for humankind in general: if
a repeat of the experiment with a different subject gave a different subject
report, this would debar the subject reports from science. So there can be a
science of psychological states; but it will recognise only those
psychological states that are generally reported by subjects in similar
situations (any idiosyncratic reports will be dismissed, or at least put on
one side for the time being, even though their truth may not be impugned).
Thus we can have an objective science of psychology; but any experiential
reports included in it will be ones that any similar subject in a similar
situation would have made, that is, they will be inter-subjectively
I hope that what I have said above answers some of your questions. Here
follow some answers to the others.
Although he officially rejects the notion of intellectual intuition, Kant
himself relies upon it. First, his account of analyticity needs it: how do
we know whether one concept is contained in another? We must be able to see
it somehow. But not by any sensory from of seeing; otherwise the analytic
would be empirical. Second, his transcendental deductions, which yield
synthetic a priori conclusions, require a form of intellectual intuition
that enables us to see, non-empirically, either the truth of synthetic
propositions which are premises of the arguments or, if the arguments have
no synthetic premises, the validity of non-analytic inferences. So, although
I said I was rehabilitating intellectual intuition after Kant's attack upon
it, I am really just exposing that Kant's attack was a sham.
If I perceive a red ball as a red ball, then the content of the intentional
object of my perception includes redness. Whether or not we say that my
perception includes a sensation of redness is terminological. I think I
would prefer not to say it, partly because it suggests that there is some
incorrigible component in the perception, and partly because 'sensation'
seems a better word for pains, feelings of discomfort, etc.
Ordinarily (i.e., excluding cases of self-deception and unconscious action),
if I am trying to do something, I know that I am trying to do it; and I know
it automatically. I do not know it from sensory experiences, though they
usually accompany my trying. I just have a way of knowing what I am willing,
similar to my way of knowing what I am thinking. In each case, it is
non-sensory and immediate, thus a priori. Although I have been convinced of
this for more than a quarter of a century, I was heartened when I found a
similar view expressed recently by a contemporary philosopher (see Lucy
O'Brien's 'On Knowing One's Own Actions' in 'Agency and Self-Awareness' ed.
Roessler and Eilan).
A trying is an action if it eventuates in a bodily movement (see Hornsby,
'Actions'). Thus a trying is an event, and events are entities. And
ordinarily I know when one of my tryings happens. And I know it immediately
because I do it. The content of the trying reveals the aspects of the action
under which it is intentional; but the action will have innumerable other
aspects, many of which may surprise me.
An obvious consequence of what I am saying is that a priori knowledge need
not be universal or necessary. Kant was mistaken there (as with so much
else). But it should be noticed that while 'I think' only ever expresses
subjective knowledge, the proposition that each subject can know a priori
the proposition expressed by 'I think,' is a candidate for objective
knowledge and is a universal proposition. It will be admitted to objective
knowledge if, through argument, examples, clarifications and psychological
experiments (if relevant), we get to a position where we can reach agreement
on it. This indeed is how philosophy proceeds. It is how maths and logic
proceed too. But it is not an infallible procedure: something accepted today
may be rejected tomorrow, just as in science (Newton in, then Newton out),
and in maths (axiom of parallels in, then axiom of parallels out), and in
logic ('every property defines a set' in, then 'every property defines a
I have never read Schopenhauer. I know Brian O'Shaughnessy was very
influenced by him and has written about our immediate knowledge of our will.
But O'Shaughnessy regards our awareness of our willing or trying as sensory,
which seems plain false to me (see his paper in the 'Agency and
Self-Awareness' book cited above).
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the hist-analytic