epelled by each theory and style of art, the simply natural and
elaborately scientific, as it came before him; and in his impatience of
each, to have been betrayed into a tissue of inconsistencies somewhat
difficult to unravel.

(6) He had been before speaking of Boucher, Director of the French
Academy, who told him that 'when he was young, studying his art, he
found it necessary to use models, but that he had left them off for many
years.'

ESSAY XIV. THE SAME SUBJECT CONTINUED

The first inquiry which runs through Sir Joshua Reynolds's Discourses
is whether the student ought to look at nature with his own eyes or
with the eyes of others, and on the whole, he apparently inclines to
the latter. The second question is what is to be understood by
nature; whether it is a general and abstract idea, or an aggregate of
particulars; and he strenuously maintains the former of these positions.
Yet it is not easy always to determine how far or with what precise
limitations he does so.

The first germ of his speculations on this subject is to be found in two
papers in the _Idler._ In the last paragraph of the second of these, he
says:

'If it has been proved that the painter, by attending to the invariable
and general ideas of nature, produces beauty, he must, by regarding
minute particularities and accidental discrimination, deviate from the
universal rule, and pollute his canvas with deformity.'

In answer to this, I would say that deformity is not the being varied in
the particulars, in which all things differ (for on this principle all
nature, which is made up of individuals, would be a heap of deformity),
but in violating general rules, in which they all or almost all agree.
Thus there are no two noses in the world exactly alike, or without a
great variety of subordinate parts, which may still be handsome, but a
face without any nose at all, or a nose (like that of a mask) without
any particularity in the details, would be a great deformity in art
or nature. Sir Joshua seems to have been led into his notions on this
subject either by an ambiguity of terms, or by taking only one view of
nature. He supposes grandeur, or the general effect of the whole, to
consist in leaving out the particular details, because these details
are sometimes found without any grandeur of effect, and he therefore
conceives the two things to be irreconcilable and the alternatives of
each other. This is very imperfect reasoning. If the m