[hist-analytic] The "Analytic A Posteriori"

Danny Frederick danny.frederick at tiscali.co.uk
Fri Sep 4 11:52:54 EDT 2009

Hi Bruce,


Thanks for the response on colour. I am afraid I must make a further
response (I can resist the urge no longer).


Your position seems to me to be inconsistent. You maintain:


(a)            if two patches are of different determinate colours, then we
can distinguish their colours under optimal conditions;


(b)            if two patches are of identical determinate colour, then we
cannot distinguish their colours under optimal conditions;


(c)             we can make mistakes such that


(i)               sometimes under optimal conditions we cannot distinguish
the colours of two patches which have different determinate colours, and


(ii)            sometimes under optimal conditions we can distinguish the
colours of two patches which have the same determinate colour.


But (c)(i) flatly contradicts (a); and (c)(ii) flatly contradicts (b).


It seems to me that the only way, consistent with your aims, of extricating
yourself from these contradictions is to distinguish different senses of
'optimal conditions.' Thus in (c), 'optimal conditions' will be taken to
mean 'conditions that we think are optimal.' The trouble with this is that
it renders your (a) and (b) either wholly empty, or both empirically empty
and wholly arbitrary. Let's take the second disjunct first.


Your (a) and (b) affirm the existence of optimal conditions under which
identity is equivalent to indiscernibility. But this sort of purely
existential statement is unfalisifable and thus empirically empty: no matter
how long and actively we have searched for such optimal conditions without
success, it is always possible that there are such conditions. Further, it
is an arbitrary stipulation: why should the world be so organised that there
will be perceptual conditions (if only we could find them) under which we
cannot be mistaken about colour identity/difference?


You can avoid this commitment to empty and arbitrary empirical claims by
saying that (a) and (b) define what optimal conditions are but leave it an
open question as to whether there are any such optimal conditions. But this
reduces (a) and (b) to pointless verbal play.


Your position is therefore either self-contradictory, or empirical but
arbitrary and untestable, or a merely stipulative definition that does no
work. Although this is not a reductio ad absurdum of it, one must wonder why
anyone would hold it.


Best wishes,



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rbjones.com/pipermail/hist-analytic_rbjones.com/attachments/20090904/6c109adb/attachment-0002.html>

More information about the hist-analytic mailing list