apt to begin than to end with marriage, and death has come
to be regarded as a rather cheap and conventional expedient for cutting
the knots of life.

From the fact that "the difficulty becomes greater the nearer we
approach to reality," it further follows that the higher the form of
drama, the more probable is it that the demands of truth and the
requirements of dramatic effect may be found to clash. In melodrama, the
curtain falls of its own accord, so to speak, when the handcuffs are
transferred from the hero's wrists to the villain's. In an
adventure-play, whether farcical or romantic, when the adventure is over
the play is done. The author's task is merely to keep the interest of
the adventure afoot until he is ready to drop his curtain. This is a
point of craftsmanship in which playwrights often fail; but it is a
point of craftsmanship only. In plays of a higher order, on the other
hand, the difficulty is often inherent in the theme, and not to be
overcome by any feat of craftsmanship. If the dramatist were to eschew
all crises that could not be made to resolve themselves with
specifically dramatic crispness and decisiveness, he would very
seriously limit the domain of his art. Many excellent themes would be
distorted and ruined by having an emphatic ending forced upon them. It
is surely much better that they should be brought to their natural
unemphatic ending, than that they should be either falsified or ignored.

I suggest, then, that the modern tendency to take lightly Aristotle's
demand that the drama should have a "beginning, a middle, _and an end_,"
arises from the nature of things, and implies, not necessarily, nor even
probably, a decline in craftsmanship, but a new intimacy of relation to
life, and a new sincerity of artistic conscience. I suggest that the
"weak last act," of which critics so often complain, is a natural
development from which authors ought not, on occasion, to shrink, and of
which critics ought, on occasion, to recognize the necessity. To elevate
it into a system is absurd. There is certainly no more reason for
deliberately avoiding an emphatic ending than for mechanically forcing
one. But authors and critics alike should learn to distinguish the
themes which do, from the themes which do not, call for a definite,
trenchant solution, and should handle them, and judge them, in
accordance with their inherent quality.

Let us, however, define our terms, and be sure that we know what we are
talking about. By an "unemphatic ending" I am far from meaning a
makeshift ending, an ending carelessly and conventionally huddled up.
Nor do I mean an indecisive ending, where the curtain falls, as the
saying goes, on a note of interrogation. An unemphatic ending, as I
understand it, is a deliberate anticlimax, an idyllic, or elegiac, or
philosophic last act, following upon a penultimate act of very much
higher tension. The disposition to condemn such an ending off-hand is
what I am here pleading against. It is sometimes assumed that the
playwright ought always to make his action conclude within five minutes
of its culmination; but for such a hard-and-fast rule I can find no
sufficient reason. The consequences of a great emotional or spiritual
crisis cannot always be worked out, or even foreshadowed, within so
brief a space of time. If, after such a crisis, we are unwilling to keep
our seats for another half-hour, in order to learn "what came of it
all," the author has evidently failed to awaken in us any real interest
in his characters.

A good instance of the unemphatic ending is the last act of Sir Arthur
Pinero's _Letty_. This "epilogue"--so the author calls it--has been
denounced as a concession to popular sentimentality, and an unpardonable
anticlimax. An anticlimax it is, beyond all doubt; but it does not
follow that it is an artistic blemish. Nothing would have been easier
than not to write it--to make the play end with Letty's awakening from
her dream, and her flight from Letchmere's rooms. But the author has set
forth, not merely to interest us in an adventure, but to draw a
character; and it was essential to our full appreciation of Letty's
character that we should know what, after all, she made of her life.
When Iris, most hapless of women, went out into the dark, there was
nothing more that we needed to know of her. We could guess the sequel
only too easily. But the case of Letty was wholly different. Her exit
was an act of will, triumphing over a form of temptation peculiarly
alluring to her temperament. There was in her character precisely that
grit which Iris lacked; and we wanted to know what it would do for her.
This was not a case for an indecisive ending, a note of interrogation.
The author felt no doubt as to Letty's destiny, and he wanted to leave
his audience in no doubt. From Iris's fate we were only too willing to
avert our eyes; but it would have been a sensible discomfort to us to be
left in the dark about Letty's.

This, then, I regard as a typical instance of justified anticlimax.
Another is the idyllic last act of _The Princess and the Butterfly_, in
which, moreover, despite its comparatively subdued tone, the tension is
maintained to the end. A very different matter is the third act of _The
Benefit of the Doubt_, already alluded to. This is a pronounced case of
the makeshift ending, inspired (to all appearance) simply by the fact
that the play must end somehow, and that no better idea happens to
present itself. Admirable as are the other acts, one is almost inclined
to agree with Dumas that an author ought not to embark upon a theme
unless he foresees a better way out of it than this. It should be noted,
too, that _The Benefit of the Doubt_ is a three-act play, and that, in a
play laid out on this scale, a whole act of anticlimax is necessarily
disproportionate. It is one thing to relax the tension in the last act
out of four or five; quite another thing in the last act out of three.
In other words, the culminating point of a four-or five-act play may be
placed in the penultimate act; in a three-act play, it should come, at
earliest, in the penultimate scene.[1]

In the works of Mr. Henry Arthur Jones we find several instances of the
unemphatic last act--some clearly justified, others much less so. Among
the former I unhesitatingly reckon the fourth act of _Mrs. Dane's
Defence_. It would not have been difficult, but surely most inartistic,
to huddle up the action in five minutes after Mrs. Dane's tragic
collapse under Sir Daniel Carteret's cross-examination. She might have
taken poison and died in picturesque contortions on the sofa; or Lionel
might have defied all counsels of prudence and gone off with her in
spite of her past; or she might have placed Lionel's hand in Janet's,
saying: "The game is up. Bless you, my children. I am going into the
nearest nunnery." As a matter of fact, Mr. Jones brought his action to
its natural close in a quiet, sufficiently adroit, last act; and I do
not see that criticism has any just complaint to make.

In recent French drama, _La Douloureuse_, already cited, affords an
excellent instance of a quiet last act. After the violent and
heartrending rupture between the lovers in the third act, we feel that,
though this paroxysm of pain is justified by the circumstances, it will
not last for ever, and Philippe and Hйlиne will come together again.
This is also M. Donnay's view; and he devotes his whole last act, quite
simply, to a duologue of reconciliation. It seems to me a fault of
proportion, however, that he should shift his locality from Paris to the
Riviera, and should place the brief duologue in a romantic woodland
scene. An act of anticlimax should be treated, so to speak, as
unpretentiously as possible. To invent an elaborate apparatus for it is
to emphasize the anticlimax by throwing it into unnecessary relief.

This may be a convenient place for a few words on the modern fashion of
eschewing emphasis, not only in last acts, but at every point where the
old French dramaturgy demanded it, and especially in act-endings.
_Punch_ has a pleasant allusion to this tendency in two suggested
examination-papers for an "Academy of Dramatists":

A--FOR THE CLASSICAL SIDE ONLY.
1. What is a "curtain"; and how should it be led up to?

B--FOR THE MODERN SIDE ONLY.
1. What is a "curtain"; and how can it be avoided?

Some modern playwrights have fled in a sort of panic from the old
"picture-poster situation" to the other extreme of always dropping their
curtain when the audience least expects it. This is not a practice to be
commended. One has often seen an audience quite unnecessarily chilled by
a disconcerting "curtain." There should be moderation even in the
shrinking from theatricality.

This shrinking is particularly marked, though I do not say it is carried
too far, in the plays of Mr. Galsworthy. Even the most innocent tricks
of emphasis are to him snares of the Evil One. He would sooner die than
drop his curtain on a particularly effective line. It is his chief
ambition that you should never discern any arrangement, any intention,
in his work. As a rule, the only reason you can see for his doing thus
or thus is his desire that you should see no reason for it. He does not
carry this tendency, as some do, to the point of eccentricity; but he
certainly goes as far as any one should be advised to follow. A little
further, and you incur the danger of becoming affectedly unaffected,
artificially inartificial.

I am far from pleading for the conventional tableau at the end of each
act, with all the characters petrified, as it were, in
penny-plain-twopence-coloured attitudes. But it is certainly desirable
that the fall of the curtain should not take an audience entirely by
surprise, and even that the spectator should feel the moment to be
rightly chosen, though he might be unable to give any reason for his
feeling. Moreover--this may seem a super-subtlety, but one has seen it
neglected with notably bad effect--a playwright should never let his
audience expect the fall of a curtain at a given point, and then balk
their expectancy, unless he is sure that he holds in reserve a more than
adequate compensation. There is nothing so dangerous as to let a play,
or an act, drag on when the audience feels in its heart that it is
really over, and that "the rest is silence"--or ought to be. The end of
Mr. Granville Barker's fine play, _The Voysey Inheritance_, was injured
by the fact that, several minutes before the curtain actually fell, he
had given what seemed an obvious "cue for curtain." I do not say that
what followed was superfluous; what I do say is that the author ought to
have been careful not to let us imagine that the colloquy between Edward
and Alice was over when in fact it had still some minutes to run. An
even more remarkable play, _The Madras House_, was ruined, on its first
night, by a long final anticlimax. Here, however, the fault did not lie
in awakening a premature expectation of the close, but in the fact that
we somehow were more interested in the other characters of the play than
in the pair who held the stage throughout the long concluding scene.

Once more I turn to _La Douloureuse_ for an instance of an admirable
act-ending of the quiet modern type. The third act--the terrible
peripety in the love of Philippe and Hйlиne--has run its agonizing
course, and worked itself out. The old dramaturgy would certainly have
ended the scene with a bang, so to speak--a swoon or a scream, a tableau
of desolation, or, at the very least, a piece of tearful rhetoric. M.
Donnay does nothing of the sort. He lets his lovers unpack their hearts
with words until they are exhausted, broken, dazed with misery, and have
nothing more to say. Then Hйlиne asks: "What o'clock is it?" Philippe
looks at his watch: "Nearly seven." "I must be going"--and she dries her
eyes, smoothes her hair, pulls herself together, in a word, to face the
world again. The mechanical round of life re-asserts its hold upon them.
"Help me with my cloak," she says; and he holds her mantle for her, and
tucks in the puffed sleeves of her blouse. Then he takes up the lamp and
lights her out--and the curtain falls. A model "curtain"!

* * * * *

[Footnote 1: The fact that a great poet can ignore such precepts with
impunity is proved by the exquisite anticlimax of the third act of
D'Annunzio's _La Gioconda_.]




_CHAPTER XIX_

CONVERSION


The reader may have noticed, possibly with surprise, that some of the
stock terms of dramatic criticism occur but rarely in these pages, or
not at all. One of them is _dйnouement_. According to orthodox theory, I
ought to have made the _dйnouement_ the subject of a whole chapter, if
not of a whole book. Why have I not done so?

For two