So long as the romantic is sure of his truth and certain of his
power to convince the reader, he need not support his truth by an
accumulation of evidence imitated from the actual life he has
observed. But on the other hand, there is nothing to prevent his
doing so; and unless he be very headstrong--so headstrong as to be
almost unreliable--he will be extremely chary of his freedom. He
will not subvert the actual unless there is no other equally
effective means of conveying the truth he has to tell. Many times
a close adherence to actuality is as advisable for the deductive
author as it is for the inductive; many times the romantic writer
gains as much as the realist by confining his fiction to his own
environment of time and place. Scott, after all, was less successful
with his medieval kings and knights than with his homely and simple
Scottish characters. Hawthorne, in "The Marble Faun," lost a
certain completeness of effect by stepping off his own New England
shadow. "Dr. Jekyll and Mr. Hyde," with its subversion of the actual,
is the sort of story that might be set out of space, out of time;
but Stevenson enhanced the effect of its imaginative plausibility
by setting it in contemporary London. More and more, in recent
years, the romantics have followed the lead of the realists in
embodying their truth in scenes and characters imitated from
actuality. The early stories of the thoroughly romantic Mr. Kipling
were set in his own country, India, and in his own time; and it was
not until his actual experience had broadened to other lands, that,
to any great extent, his subjects broadened geographically. In his
stories of his own people, Mr. Kipling just as faithfully portrays the
every-day existence he has actually observed as any realist. His
method is romantic always: he deduces his details from his theme,
instead of inducing his theme from his details. He is entirely
romantic in the direction of his thought; but it is very suggestive
of the tenor of contemporary romance, to notice that he has taken the
advice of the realists and seldom gone beyond his own experience.

The range of romance is therefore far wider than the range of
realism; for all that may be treated realistically may be treated
romantically also, and much else that may be treated romantically is
hardly susceptible of realistic treatment. Granted that a romantic
have truths enough in his head, there is scarcely any limit to the
stories he may deduce fr