The question is: is constructing a 'transparent white body' like constructing a 'regular biangle'?
I can look at a body and perhaps see a matt white surface, i.e. get the impression of such a surface, or the impression of transparency (whether it actually exists or not). This impression may be produced by the distribution of the colours, and white and the other colours are not involved in it in the same way.
(I took a green painted lead cupola to be translucent greenish glass without knowing at the time about the special distribution of colours that produced this appearance.)
And white may indeed occur in the visual impression of a transparent body, for example as a reflection, as a high-light. I.e. if the impression is perceived as transparent, the white which we see will simply not be interpreted as the body's being white.
Nor can we say that white is essentially the property of a--visual--surface. For it is conceivable that white should occur as a high-light or as the colour of a flame.
A body that is actually transparent can, of course, seem white to us; but it cannot seem white and transparent.
But we should not express this by saying: white is not a transparent colour.
'Transparent' could be compared with 'reflecting'.
An element of visual space may be white or red, but can't be either transparent or opaque.
Transparency and reflection only exist in the dimension of depth of a visual image.
- Remarks on Colour, III, 138-140, 145-150