[hist-analytic] Response to Danny Frederick

Danny Frederick danny.frederick at tiscali.co.uk
Sat Aug 29 15:08:09 EDT 2009

Hi Bruce,

I do not think you have disposed of the objections I raised to what you say
about determinate colours and discernibility. You now say that two
determinate colours are the same just when no colour difference will be
detected when they are both observed under conditions where their true
colour is visually apparent to good observers.

That seems empty: of course no colour difference will be detected if the
TRUE colour is visually apparent and the colours are the same.

But I doubt that we can make much sense of 'the true colour.' Are we ever in
conditions in which we can identify the true colour? Consider my example of
same colour/discernible difference that concerned the effects of light and
shade: the bit of the lawn in bright sunlight looks yellow, the bit in the
shade of the tree looks grey, even though they are (by hypothesis) the same
shade of green. You object that the lighting conditions should be the same.
But which lighting conditions give us the 'true colour'? Can there be such a
thing? We can make an arbitrary selection of lighting conditions and call
that 'true;' but that is just arbitrary.

There is another problem too. What shade of colour we perceive seems likely
to depend upon our mood. We know from studies of pathology that mood
influences perception. I do not know what evidence there is that colour
perception is influenced by mood, but I feel sure there will be such
evidence if relevant experiments have been done. Why? I went through a
period of depression some years ago during which the whole world looked grey
to me. I could still distinguish colours, but they all looked grey. Yes, the
cup looked red; but it looked grey too. It may be that nothing can be red
all over and grey all over; but it can LOOK red all over and grey all over.
The shade of red I saw on the cup, i.e., that which was 'visually apparent,'
must have been different to the shade I was used to seeing; otherwise I
could not have thought it looked grey. You object that the 'true colour' is
the one 'visually apparent' to 'good observers.' Depressed people clearly
will not count as good observers. Okay; but which observers do we select as
the good ones? Depression will not be the only thing that makes a
difference. Again, it seems, the selection will ultimately be arbitrary,
which impugns the notion of 'the true colour.'

Let's drop the notion of the true colour of a thing (using 'colour' in its
ordinary non-technical sense). What you intend to say, I think, is something
like this: for any person in state A, under environmental conditions B, he
can distinguish two shades of colour if and only if the two shades are of
different determinate colours. Some work will have to go into spelling out A
and B, but let's be generous and assume that can be done.

Even so, my objection from the spreading effect still stands because the
effect occurs under conditions of good lighting for all observers. What
shade we see a particular colour depends on the shades surrounding it. By
changing those contextual shades, we change the visual appearance of the
colour even though the determinate colour itself does not change. The
observer distinguishes two shades even though he is observing the very same
determinate colour. For a discussion of this, and examples that you can
view, see E Gombrich, 'Art and Illusion,' fifth edition, p.260 and plate IV,
which is just before p.51. (Incidentally, I give bibliographical references
so that anyone interested in pursuing the matter further can do so. It is
intended to be helpful, not an insult.)

I think this may throw some light on my depressive period. Nothing relevant
had changed in the external world. But the personal context (my mental
environment) in which my colour perceptions took place had changed: all the
colours of the world were surrounded by the gloom inside me and that had the
effect of changing the colours that were 'visually apparent' to me. And I
noticed the difference: the cup that looked a bright red in the past now
looked a duller red. Do 'happy-go-lucky' people see the world as more
brightly coloured? It seems plausible to me, though probably untestable.

To summarise so far: two instances of the same determinate colour will be
distinguishable if the surrounding context has changed in relevant ways.
Therefore, discernibility does not imply difference. This depends purely on
the spreading effect.

The third example I gave, of poor lighting, was intended to show that colour
difference does not imply discernibility. Your response to this is to invoke
standard lighting conditions. My reply is to wonder, again, whether this is
not arbitrary. When the lighting is poor, I might not be able to distinguish
green and blue. Improve the lighting and I can. Suppose that under these
conditions I cannot distinguish two blue surfaces (with regard to colour).
But if we add more light, or if we make the light purer in some way, perhaps
I will. It seems it will always be possible that, if we improved the light
in some way (perhaps by using a technique not yet invented), the two patches
of apparently identical colour would become distinguishable. If so, then
whatever stage we are at, it may be the case that different colours are

There is a fourth example that I omitted, probably because it is so
familiar. This is the failure of transitivity of indiscernibility. I cannot
distinguish A from B, and I cannot distinguish B from C, but I can
distinguish A from C. If we combine this with 'distinguishable if and only
if distinct,' we get a contradiction.

At the moment I am in the process of buying a house and preparing to move.
On top of that we have finally got some good weather, so I will be spending
as much time as I can outdoors. So it may be a week or two before I respond
on the matter of analyticity.



More information about the hist-analytic mailing list