s that see it, in all the souls that conceive
it." Here he gets nearer the idea of criticism as portraiture, and
practically every critic of importance has been a portrait-painter. In
this respect Saint-Beuve is at one with Macaulay, Pater with Matthew
Arnold, Anatole France (occasionally) with Henry James. They may portray
authors rather than books, artists rather than their work, but this only
means that criticism at its highest is a study of the mind of the artist
as reflected in his art.

Clearly, if the reviewer can paint the portrait of an author, he is
achieving something better even than the portrait of a book. But what, at
all costs, he must avoid doing is to substitute for a portrait of one kind
or another the rag-bag of his own moral, political or religious opinions.
It is one of the most difficult things in the world for anyone who happens
to hold strong opinions not to make the mind of Shakespeare himself a
pulpit from which to roar them at the world. Reviewers with theories about
morality and religion can seldom be induced to come to the point of
portraiture until they have enjoyed a preliminary half-column of
self-explanation. In their eyes a review is a moral essay rather than an
imaginative interpretation. In dissenting from this view, one is not
pleading for a race of reviewers without moral or religious ideas, or even
prepossessions. One is merely urging that in a review, as in a novel or a
play, the moral should be seated at the heart instead of sprawling all
over the surface. In the well-worn phrase it should be implicit, not
explicit. Undoubtedly a rare critic of genius can make an interesting
review-article out of a statement of his own moral and political ideas.
But that only justifies the article as an essay, not as a review. To many
reviewers--especially in the bright days of youth--it seems an immensely
more important thing to write a good essay than a good review. And so it
is, but not when a review is wanted. It is a far, far better thing to
write a good essay about America than a good review of a book on America.
But the one should not be substituted for the other. If one takes up a
review of a book on America by Mr. Wells or Mr. Bennett, it is in
ninety-nine cases out of a hundred in order to find out what the author
thinks, not what the reviewer thinks. If the reviewer begins with a
paragraph of general remarks about America--or, worse still, about some
abstract thing like liberty--he is a