Back to starting page



A case study of the sign in fiction


Part 2 |Part 3

In Semiotica 44-1/2 (1983), pp. 75-93.
0037-1998/83/0044-0075 $2.00 Mouton Publishers, Amsterdam

The three endings in Pygmalion


The question how fiction (or belles lettres or literature or poetry) differs from other linguistic statements has been thoroughly examined through the ages. In this case study I want to shed some light upon a fraction of this problem, but an important one: I will try to show one decisive difference between the range of reference of signs in fiction and reference in real life. The main point is that even if the esthetic sign might be ambiguous and open to several possible interpretations-especially the supersign consisting of a piece of literature as a whole-in fiction we obviously impose interpretational restrictions upon the signs forming the possible world that is suggested to us in parts. 'In parts' means that we ourselves fill out enormous amounts of nongiven information as selfevident and/or irrelevant to the story told. For instance, there are very few novels-and presumably even fewer plays-that give information about the principal characters concerning their digestion and possible problems connected therewith (apart from crude jokes in certain genres). Nevertheless, we presume that fictitious characters do behave and function exactly as do living persons. As Grice has pointed out, in everyday conversation we speak only about relevant items. But contrary to everyday life, in fiction there cannot simply be said to exist items or persons that are not explicitly mentioned or implicitly suggested. In real life, people might have illnesses we never learn anything about but that may be determining factors for their behavior. But there is definitely no reason to suspect that old Hamlet suffered from cancer or that young Hamlet had a sister, hitherto unknown, who married Fortinbras after the play's end. In real life, though, men we know as slightly as we know Hamlet (for only a few hours) may very well have sisters we do not hear anything about.
Even if deus-ex-machina was an acceptable theatrical device at one time, we seem at present to be not quite happy with unprepared solutions that bring conflicts to a happy ending. This is a logical function of the interpretational restrictions in fictional signs, and ultimately this is the reason why the proposal of Bernard Shaw in the afterword to Pygmalion, that Eliza is going to marry Freddy, is contradictory to the play and the principles ruling our expectations of fictitious texts.
In order to demonstrate the thesis stated above, I will analyze Pygmalion from this point of view, and I will also introduce semiotic subcodes regarding our expectations of characters in plays and perhaps even fictitious characters as a whole.
It is a well-known observation that a rule or convention is never more apparent than when it is broken. The convention-or, in semiotic terminology, the code-I want to illustrate is that which governs our expectations about the relationships of the principal characters in a play. The study was inspired by the relationship of Eliza and Higgins in Shaw's Pygmalionl, and the question of how that relationship develops when the play is over. The audience has been led to expect a union between the two main characters, Eliza and Higgins. But in a postscript, Shaw insists that Eliza marries not Higgins but Freddy Eynsford Hill. I shall try to show that in proposing this sequel Shaw offends against a semiotic principle2: that the primary tensions or conflicts of a play are acted out by its main characters-in itself a fairly obvious proposition. In the first section of this essay I shall sketch the background of the different versions of Shaw's Pygmalion. In the second section I shall discuss what constitutes a 'main character' in a play, and what expectations we have of such a character. In a final section I shall discuss Pygmalion from the viewpoints that have emerged earlier. By showing the existence of a set of semiotic conventions, I hope to demonstrate that Shaw's hypothesis about Eliza's destiny is unreasonable.

The three endings

In the course of its transformations into film and musical, Shaw's Pygmalion has been equipped with three different endings.
When the curtain comes down on the fifth act in the original version of Pygmalion (1912), Eliza states quite definitely that she has no intention of returning to Higgins (and Pickering) at Wimpole Street, while Higgins for his part seems tolerably convinced that she will nevertheless do exactly that.
MRS HIGGINS. The carriage is waiting, Eliza. Are you ready?
LIZA. Quite. Is the Professor coming?
MRS HIGGINS. Certainly not. He cant behave himself in church. He makes remarks out loud all the time on the clergyman's pronunciation.
LIZA. Then I shall not see you again, Professor. Goodbye. (She goes to the door).
MRS HIGGINS (coming to Higgins) Goodbye, dear.
HIGGINS. Goodbye, mother. (He is about to kiss her, when he recollects something). Oh, by the way, Eliza, order a ham and a Stilton cheese, will you? And buy me a pair of reindeer gloves, number eights, and a tie to match that new suit of mine, at Eale & Binman's. You can choose the color. (His cheerful, careless, vigorous voice shews that he is incorrigible).
LIZA (disdainfully) Buy them yourself. (She sweeps out).
MRS HIGGINS. I'm afraid youve spoiled that girl, Henry. But never mind, dear: I'll buy you the tie and gloves.
HIGGINS (sunnily) Oh, dont bother. She'll buy em all right enough. Goodbye.
They kiss, Mrs Higgins runs out. Higgins, left alone, rattles his cash in his pocket; chuckles; and disports himsey in a highly self-satisfied manner. (Shaw 1951: 751)
The reader or spectator (for simplicity's sake I shall in future refer to both as the audience) is left in uncertainty. This ambivalence or openness-to use a term that won a place in semiotic terminology with Umberto Eco's Opera Aperta (1962)-may certainly be experienced by many as a stimulus and a challenge; but it has never really been accepted either by theatre people or by their public. During the first rehearsals in London the cast were already dissatisfied with the ending (Henderson 1932: 566). When the play was filmed in 1938, the ending was changed. In the film Eliza returns to Wimpole Street, where Higgins receives her with the famous line, 'Eliza, where the devil are my slippers?' (This curtain line was retained in the musical and in the film based on it).(3)
The new ending seems less open than the original one, and naturally involves a first step toward a happy ending-in the sense of the relationship between the two main characters developing in a romantic or erotic.direction. It gives the audience even more cause to interpret Eliza's and Higgins's future along the lines that Shaw, in his postscript to the play (first printed in 1916) treats with such characteristic irony: 'The rest of the story need not be shewn in action, and indeed, would hardly need telling if our imaginations were not so enfeebled by their lazy dependence on the ready-mades and reach-me-downs of the ragshop in which Romance keeps its stock of "happy endings" to misfit all stories' (Shaw 1972: 782). In fact, however, the new film-ending gives no very clear picture of what sort of future-if any-Eliza and Higgins are going to share. Higgins has offered to have her back on a friendly basis: 'You and I and Pickering will be three old bachelors together instead of only two men and a silly girl' (1972: 781). Even if Shaw did alter the ending for the film version, he left unanswered the question of whether Eliza and Higgins ever get married. (Any other kind of sexual bond was clearly unthinkable to Shaw, who, rebel though he may have been, was still very much a Victorian.) It is reasonable to assume that it was this degree of openness that allowed him to make the concession of letting Eliza return to Wimpole Street.(4)
Shaw's own conception
Shaw's own conception of 'what happens afterwards' is set out in the postscript I have already quoted, where he insists-with some fairly elaborate arguments-that Eliza does not marry Higgins but Freddy. Shaw may have convinced certain dramatic theorists (Bentley 1947: 123; Styan 1968: 128) about this sequel, but, as I have already pointed out, he cannot be said to have convinced either actors or audience. Alan Jay Lerner, who wrote the book of the musical, observes in his introduction to My Fair Lady: 'I have omitted the sequel because in it Shaw explains how Eliza ends not with Higgins but with Freddy and-Shaw and Heaven forgive me-I am not certain he is right' (1958: vi).(5)
The brilliant ending of the 1938 film is retained in the musical.6 Although this ending-according to Shaw (see note 4!)-is no guarantee of a romantic 'happy ending', Shaw did not keep it in a later version, published by Penguin in 1941, that contains various other scenes from the film. But neither has he kept the original ending of 1912. The ending of the final version after Higgins has given Eliza shopping orders runs like this:
LIZA (disdainfully) Number eights are too small for you if you want them lined with lamb's wool. You have three new ties that you have forgotten in the drawer of your washstand. Colonel Pickering prefers double Gloucester to Stilton; and you dont notice the difference. I telephoned Mrs Pearce this morning not to forget the ham. What you are to do without me I cannot imagine. (She sweeps out).
MRS HIGGINS. I'm afraid youve spoilt that girl, Henry. I should be uneasy about you and her if she were less fond of Colonel Pickering.
HIGGINS. Pickering! Nonsense: she's going to marry Freddy. Ha ha! Freddy! Freddy! Ha ha ha ha ha!!!!! (He roars with laughter as the play ends).
(Shaw 1972: 781ff.)
Eliza's last words in the 1912 version, 'Buy them yourself', have made way for a longer exposition of the issues, which makes it clear that Higgins really would find it rather hard to manage without Eliza. Her telephone call to Mrs. Pearce about the ham is touchingly considerate. The speech is plainly put there to show thatj for all her anger, Eliza is still carrying out the duties that Higgins regards as his right.
The last lines of all are of even greater interest in this connection. This time Mrs. Higgins does not offer to buy the tie for her son. (In the first version, moreover, Higgins's refusal means that he has freed himself from her, and thus contradicts Shaw's own view in the postscript). Instead she steers the conversation over to the personal plane, to Colonel Pickering, and so provides a lead-in for Higgins's extraordinarily interesting curtainline. It is interesting partly because one can see in it 'Higgins's' answer to Shaw's postscript. It is clear that Higgins does not believe for a moment that Eliza is going to marry Freddy, and it is easy enough to see in 'Higgins's' line Shaw's own answer to his postscript. Higgins resembles Shaw closely in many respects, and here Shaw seems to be laughing at himself-and at those who believed in his arguments in the postscript.
If anyone else but Shaw in his role of author had written an essay along the same lines as the postscript, I hardly think anybody would have considered the possibility of Eliza's marrying Freddy, let alone have been persuaded that it was probable. Shaw's well-known love of questioning conventions, and his status as an aesthetic and social critic, suggest that the postscript might perhaps be interpreted rather as a general onslaught on our 'lazy dependence on ... happy endings'.
Now, of course, we have to ask ourselves in what respects it would be a less 'happy' ending for Eliza to marry Freddy rather than Higgins. To assume that she must necessarily marry somebody is surely 'lazy dependence' in itself! When Shaw rebukes 'people in all directions' for assuming that Eliza marries Higgins 'for no other reason than that she became the heroine of a romance' in which Higgins is 'the hero' (Postscript, 1972: 782), he is in fact reacting against a semiotic principle of drama that determines the audience's expectations of what will happen between the hero and heroine of a play. This rule makes it extremely unlikely, not to say unreasonable, that Eliza should marry Freddy.

Part 2 || Part 3

Starting page