So far as I can see, the strongest reason against keeping a secret is that, try as you may, you cannot do it.  This point has already been discussed in Chapter IX, where we saw that from only one audience can a secret be really hidden, a considerable percentage of any subsequent audience being certain to know all about it in advance.  The more striking and successful is the first-night effect of surprise, the more certainly and rapidly will the report of it circulate through all strata of the theatrical public.  But for this fact, one could quite well conceive a fascinating melodrama constructed, like a detective story, with a view to keeping the audience in the dark as long as possible.  A pistol shot might ring out just before the rise of the curtain:  a man (or woman) might be discovered in an otherwise empty room, weltering in his (or her) gore:  and the remainder of the play might consist in the tracking down of the murderer, who would, of course, prove to be the very last person to be suspected.  Such a play might make a great first-night success; but the more the author relied upon the mystery for his effect, the more fatally would that effect be discounted at each successive repetition.

One author of distinction, M. Hervieu, has actually made the experiment of presenting an enigma—­he calls the play L’Enigme—­and reserving the solution to the very end.  We know from the outset that one of two sisters-in-law is unfaithful to her husband, and the question is—­which?  The whole ingenuity of the author is centred on keeping the secret, and the spectator who does not know it in advance is all the time in the attitude of a detective questing for clues.  He is challenged to guess which of the ladies is the frail one; and he is far too intent on this game to think or care about the emotional process of the play.  I myself (I remember) guessed right, mainly because the name Giselle seemed to me more suggestive of flightiness than the staid and sober Leonore, wherefore I suspected that M. Hervieu, in order to throw dust in our eyes, had given it to the virtuous lady.  But whether we guess right or wrong, this clue-hunting is an intellectual sport, not an artistic enjoyment.  If there is any aesthetic quality in the play, it can only come home to us when we know the secret.  And the same dilemma will present itself to any playwright who seeks to imitate M. Hervieu.

The actual keeping of a secret, then—­the appeal to the primary curiosity of actual ignorance—­may be ruled out as practically impossible, and, when possible, unworthy of serious art.  But there is also, as we have seen, the secondary curiosity of the audience which, though more or less cognizant of the essential facts, instinctively assumes ignorance, and judges the development of a play from that point of view.  We all realize that a dramatist has no right to trust to our previous knowledge, acquired from outside sources.  We know that a play, like every other work of art, ought to be self-sufficient,