Appreciations and Criticisms by G.K. Chesterton

Edwin Drood

CHAPTER XXII EDWIN DROOD 

Pickwick was a work partly designed by others, but ultimately filled up by Dickens. Edwin Drood, the last book, was a book designed by Dickens, but ultimately filled up by others. The Pickwick Papers showed how much Dickens could make out of other people's suggestions ; The Mystery of Edwin Drood shows how very little other people can make out of Dickens's suggestions. Dickens was meant by Heaven to be the great melodramatist; so that even his literary end was melodramatic. Something more seems hinted at in the cutting short of Edwin Drood by Dickens than the mere cutting short of a good novel by a great man. It seems rather like the last taunt of some elf, leaving the world, that it should be this story which is not ended, this story which is only a story. The only one of Dickens's novels which he did not finish was the only one that really needed finishing. He never had but one thoroughly good plot to tell; and that he has only told in heaven. This is what separates the case in question from any parallel cases of novelists cut off in the act of creation. That great novelist, for instance, with whom Dickens is constantly compared, died also in the middle of Denis Duval. But any one can see in Denis Duval the qualities of the later work of Thackeray; the increasing discursiveness, the increasing retrospective poetry, which had been in part the charm and in part the failure of Philip and The Virginians. But to Dickens it was permitted to die at a dramatic moment and to leave a dramatic mystery. Any Thackerayan could have completed the plot of Denis Duval; except indeed that a really sympathetic Thackerayan might have had some doubt as to whether there was any plot to complete. But Dickens, having had far too little plot in his stories previously, had far too much plot in the story he never told. Dickens dies in the act of telling, not his tenth novel, but his first news of murder. He drops down dead as he is in the act of denouncing the assassin. It is permitted to Dickens, in short, to come to a literary end as strange as his literary beginning. He began by completing the old romance of travel. He ended by inventing the new detective story. It is as a detective story first and last that we have to consider The Mystery of Edwin Drood. This does not mean, of course, that the details are not often admirable in their swift and penetrating humour; to say that of the book would be to say that Dickens did not write it. Nothing could be truer, for instance, than the manner in which the dazed and drunken dignity of Durdles illustrates a certain bitterness at the bottom of the bewilderment of the poor. Nothing could be better than the way in which the haughty and allusive conversation between Miss Twinkleton and the landlady illustrates the maddening preference of some females for skating upon thin social ice. There is an even better example than these of the original humorous insight of Dickens; and one not very often remarked, because of its brevity and its unimportance in the narrative. But Dickens never did anything better than the short account of Mr. Grewgious's dinner being brought from the tavern by two waiters: "a stationary waiter," and "a flying waiter." The "flying waiter" brought the food and the "stationary waiter" quarrelled with him; the "flying waiter" brought glasses and the "stationary waiter" looked through them. Finally, it will be remembered the "stationary waiter" left the room, casting a glance which indicated "let it be understood that all emoluments are mine, and that Nil is the reward of this slave." Still, Dickens wrote the book as a detective story; he wrote it as The Mystery of Edwin Drood. And alone, perhaps, among detective-story writers, he never lived to destroy his mystery. Here alone then among the Dickens novels it is necessary to speak of the plot and of the plot alone. And when we speak of the plot it becomes immediately necessary to speak of the two or three standing explanations which celebrated critics have given of the plot. The story, so far as it was written by Dickens, can be read here. It describes, as will be seen, the disappearance of the young architect Edwin Drood after a night of festivity which was supposed to celebrate his reconciliation with a temporary enemy, Neville Landless, and was held at the house of his uncle John Jasper. Dickens continued the tale long enough to explain or explode the first and most obvious of his riddles. Long before the existing part terminates it has become evident that Drood has been put away, not by his obvious opponent, Landless, but by his uncle who professes for him an almost painful affection. The fact that we all know this, however, ought not in fairness to blind us to the fact that, considered as the first fraud in a detective story, it has been, with great skill, at once suggested and concealed. Nothing, for instance, could be cleverer as a piece of artistic mystery than the fact that Jasper, the uncle, always kept his eyes fixed on Drood's face with a dark and watchful tenderness; the thing is so told that at first we really take it as only indicating something morbid in the affection; it is only afterwards that the frightful fancy breaks upon us that it is not morbid affection but morbid antagonism. This first mystery (which is no longer a mystery) of Jasper's guilt, is only worth remarking because it shows that Dickens meant and felt himself able to mask all his batteries with real artistic strategy and artistic caution. The manner of the unmasking of Jasper marks the manner and tone in which the whole tale was to be told. Here we have not got to do with Dickens simply giving himself away, as he gave himself away in Pickwick or The Christmas Carol. Not that one complains of his giving himself away; there was no better gift. What was the mystery of Edwin Drood from Dickens's point of view we shall never know, except perhaps from Dickens in heaven, and then he will very likely have forgotten. But the mystery of Edwin Drood from our point of view, from that of his critics, and those who have with some courage (after his death) attempted to be his collaborators, is simply this. There is no doubt that Jasper either murdered Drood or supposed that he had murdered him. This certainty we have from the fact that it is the whole point of a scene between Jasper and Drood's lawyer Grewgious in which Jasper is struck down with remorse when he realises that Drood has been killed (from his point of view) needlessly and without profit. The only question is whether Jasper's remorse was as needless as his murder. In other words the only question is whether, while he certainly thought he had murdered Drood, he had really done it. It need hardly be said that such a doubt would not have been raised for nothing; gentlemen like Jasper do not as a rule waste good remorse except upon successful crime. The origin of the doubt about the real death of Drood is this. Towards the latter end of the existing chapters there appears very abruptly, and with a quite ostentatious air of mystery, a character called Datchery. He appears for the purpose of spying upon Jasper and getting up some case against him; at any rate, if he has not this purpose in the story he has no other earthly purpose in it. He is an old gentleman of juvenile energy, with a habit of carrying his hat in his hand even in the open air; which some have interpreted as meaning that he feels the unaccustomed weight of a wig. Now there are one or two people in the story who this person might possibly be. Notably there is one person in the story who seems as if he were meant to be something, but who hitherto has certainly been nothing; I mean Bazzard, Mr. Grewgious's clerk, a sulky fellow interested in theatricals, of whom an unnecessary fuss is made. There is also Mr. Grewgious himself, and there is also another suggestion, so much more startling that I shall have to deal with it later. For the moment, however, the point is this: That ingenious writer, Mr. Proctor, started the highly plausible theory that this Datchery was Drood himself, who had not really been killed. He adduced a most complex and complete scheme covering nearly all the details; but the strongest argument he had was rather one of general artistic effect. This argument has been quite perfectly summed up by Mr. Andrew Lang in one sentence: "If Edwin Drood is dead, there is not much mystery about him." This is quite true; Dickens, when writing in so deliberate, nay, dark and conspiratorial a manner, would surely have kept the death of Drood and the guilt of Jasper hidden a little longer if the only real mystery had been the guilt of Jasper and the death of Drood. It certainly seems artistically more likely that there was a further mystery of Edwin Drood; not the mystery that he was murdered, but the mystery that he was not murdered. It is true indeed that Mr. Cumming Walters has a theory of Datchery (to which I have already darkly alluded) a theory which is wild enough to be the centre not only of any novel but of any harlequinade. But the point is that even Mr. Cumming Walters's theory, though it makes the mystery more extraordinary, does not make it any more of a mystery of Edwin Drood. It should not have been called The Mystery of Drood, but The Mystery of Datchery. This is the strongest case for Proctor; if the story tells of Drood coming back as Datchery, the story does at any rate fulfil the title upon its title-page. The principal objection to Proctor's theory is that there seems no adequate reason why Jasper should not have murdered his nephew if he wanted to. And there seems even less reason why Drood, if unsuccessfully murdered, should not have raised the alarm. Happy young architects, when nearly strangled by elderly organists, do not generally stroll away and come back some time afterwards in a wig and with a false name. Superficially it would seem almost as odd to find the murderer investigating the origin of the murder, as to find the corpse investigating it. To this problem two of the ablest literary critics of our time, Mr. Andrew Lang and Mr. William Archer (both of them persuaded generally of the Proctor theory) have especially addressed themselves. Both have come to the same substantial conclusion; and I suspect that they are right. They hold that Jasper (whose mania for opium is much insisted on in the tale) had some sort of fit, or trance, or other physical seizure as he was committing the crime so that he left it unfinished; and they also hold that he had drugged Drood, so that Drood, when he recovered from the attack, was doubtful about who had been his assailant. This might really explain, if a little fancifully, his coming back to the town in the character of a detective. He might think it due to his uncle (whom he last remembered in a kind of murderous vision) to make an independent investigation as to whether he was really guilty or not. He might say, as Hamlet said of a vision equally terrifying, "I'll have grounds more relative than this." In fairness it must be said that there is something vaguely shaky about this theory; chiefly, I think, in this respect; that there is a sort of farcical cheerfulness about Datchery which does not seem altogether appropriate to a lad who ought to be in an agony of doubt as to whether his best friend was or was not his assassin. Still there are many such incongruities in Dickens; and the explanation of Mr. Archer and Mr. Lang is an explanation. I do not believe that any explanation as good can be given to account for the tale being called The Mystery of Edwin Drood, if the tale practically starts with his corpse. If Drood is really dead one cannot help feeling the story ought to end where it does end, not by accident but by design. The murder is explained. Jasper is ready to be hanged, and every one else in a decent novel ought to be ready to be married. If there was to be much more of anything, it must have been of anticlimax. Nevertheless there are degrees of anticlimax. Some of the more obvious explanations of Datchery are quite reasonable, but they are distinctly tame. For instance, Datchery may be Bazzard; but it is not very exciting if he is; for we know nothing about Bazzard and care less. Again, he might be Grewgious; but there is something pointless about one grotesque character dressing up as another grotesque character actually less amusing than himself. Now, Mr. Cumming Walters has at least had the distinction of inventing a theory which makes the story at least an interesting story, even if it is not exactly the story that is promised on the cover of the book. The obvious enemy of Drood, on whom suspicion first falls, the swarthy and sulky Landless, has a sister even swarthier and, except for her queenly dignity, even sulkier than he. This barbaric princess is evidently meant to be (in a sombre way) in love with Crisparkle, the clergyman and muscular Christian who represents the breezy element in the emotions of the tale. Mr. Cumming Walters seriously maintains that it is this barbaric princess who puts on a wig and dresses up as Mr. Datchery. He urges his case with much ingenuity of detail. Helena Landless certainly had a motive; to save her brother, who was accused falsely, by accusing Jasper justly. She certainly had some of the faculties; it is elaborately stated in the earlier part of her story that she was accustomed as a child to dress up in male costume and run into the wildest adventures. There may be something in Mr. Cumming Walters's argument that the very flippancy of Datchery is the self-conscious flippancy of a strong woman in such an odd situation; certainly there is the same flippancy in Portia and in Rosalind. Nevertheless, I think, there is one final objection to the theory; and that is simply this, that it is comic. It is generally wrong to represent a great master of the grotesque as being grotesque exactly where he does not intend to be. And I am persuaded that if Dickens had really meant Helena to turn into Datchery, he would have made her from the first in some way more light, eccentric, and laughable; he would have made her at least as light and laughable as Rosa. As it is, there is something strangely stiff and incredible about the idea of a lady so dark and dignified dressing up as a swaggering old gentleman in a blue coat and grey trousers. We might almost as easily imagine Edith Dombey dressing up as Major Bagstock. We might almost as easily imagine Rebecca in Ivanhoe dressing up as Isaac of York. Of course such a question can never really be settled precisely, because it is the question not merely of a mystery but of a puzzle. For here the detective novel differs from every other kind of novel. The ordinary novelist desires to keep his readers to the point; the detective novelist actually desires to keep his readers off the point. In the first case, every touch must help to tell the reader what he means; in the second case, most of the touches must conceal or even contradict what he means. You are supposed to see and appreciate the smallest gestures of a good actor; but you do not see all the gestures of a conjuror, if he is a good conjuror. Hence, into the critical estimate of such works as this, there is introduced a problem, an extra perplexity, which does not exist in other cases. I mean the problem of the things commonly called blinds. Some of the points which we pick out as suggestive may have been put in as deceptive. Thus the whole conflict between a critic with one theory, like Mr. Lang, and a critic with another theory, like Mr. Cumming Walters, becomes eternal and a trifle farcical. Mr. Walters says that all Mr. Lang's clues were blinds; Mr. Lang says that all Mr. Walters's clues were blinds. Mr. Walters can say that some passages seemed to show that Helena was Datchery; Mr. Lang can reply that those passages were only meant to deceive simple people like Mr. Walters into supposing that she was Datchery. Similarly Mr. Lang can say that the return of Drood is foreshadowed; and Mr. Walters can reply that it was foreshadowed because it was never meant to come off. There seems no end to this insane process; anything that Dickens wrote may or may not mean the opposite of what it says. Upon this principle I should be very ready for one to declare that all the suggested Datcherys were really blinds; merely because they can naturally be suggested. I would undertake to maintain that Mr. Datchery is really Miss Twinkleton, who has a mercenary interest in keeping Rosa Budd at her school. This suggestion does not seem to me to be really much more humorous than Mr. Cumming Walters's theory. Yet either may certainly be true. Dickens is dead, and a number of splendid scenes and startling adventures have died with him. Even if we get the right solution we shall not know that it is right. The tale might have been, and yet it has not been. And I think there is no thought so much calculated to make one doubt death itself, to feel that sublime doubt which has created all religion -- the doubt that found death incredible. Edwin Drood may or may not have really died; but surely Dickens did not really die. Surely our real detective liveth and shall appear in the latter days of the earth. For a finished tale may give a man immortality in the light and literary sense; but an unfinished tale suggests another immortality, more essential and more strange.

L'auteur des Malheurs de Sophie et des Mémoires d'un âne n'a pas vieilli. Son univers délicieusement désuet continue d'enchanter les têtes blondes, et ses livres, constamment réédités, commencent à intéresser la critique littéraire et universitaire.

Par Michel Tournier, de l'Académie Goncourt

[10 novembre 2005]

PEUT-ON imaginer une plus somptueuse entrée dans la vie ? Sophie naît le 1er août 1799 - la même année que Balzac - le jour de la sainte Sophie, à Saint-Pétersbourg. On la baptise le 9 août dans la chapelle royale du Palais d'hiver avec pour parrain le tsar Paul Ier, un tyran déséquilibré et imprévisible, dont son père, Fiodor Rostopchine, était le premier ministre. En février 1801, Fiodor tombe en disgrâce. Il quitte Saint-Pétersbourg avec sa famille et s'installe dans sa propriété de Voronovo, à une cinquantaine de kilomètres de Moscou. Le 20 mars, un message impérial l'atteint : qu'il revienne d'urgence ! Il se précipite, mais, à Moscou, il apprend que le tsar vient d'être assassiné dans des circonstances effroyables par des conjurés. Il retourne à Voronovo et se consacre à l'exploitation de son domaine. Pour en hâter la modernisation, il fait venir toute une famille d'Ecosse - les Paterson - rompue à l'agriculture actuelle. La comtesse de Ségur utilisera ses souvenirs « écossais » dans son roman Un bon petit diable, dominé par la terrible Madame Mac Miche.

Le retour aux affaires de Rostopchine est favorisé par la grande duchesse Catherine, soeur du tsar Alexandre Ier. En mars 1812, il est nommé gouverneur de Moscou. Six mois plus tard, Napoléon marche sur la ville. Sophie - qui a treize ans - est réfugiée à une soixante de kilomètres. Elle est fascinée par les rougeoiements de l'horizon : Moscou brûle. On ne peut oublier que cette même nuit, Stendhal parcourait les rues embrasées de la capitale russe. Plus tard, les historiens s'accorderont pour tenir Rostopchine responsable de l'incendie de Moscou. Il publiera en 1823 un libelle pour se disculper.Pour Napoléon, c'est ensuite la retraite et la défaite. En octobre 1817, toute la famille Rostopchine s'installe à Paris, 21, rue du Mont-Blanc - devenue depuis rue de la Chaussée -d'Antin. Sophie a dix-huit ans, elle ne retournera plus jamais en Russie. Rostopchine ne passe pas inaperçu. Ogre bonasse et soupe au lait, il servira de modèle à sa fille pour le général Dourakine. Il illustre parfaitement le féminisme agressif de Sophie, décelé par Marc Soriano : tous les hommes sont des imbéciles - quand ils ne sont pas franchement scélérats -, l'intelligence et la sagesse étant des qualités exclusivement féminines. Chateaubriand écrit dans ses mémoires : « On a vu à Paris le comte Rostopchine, homme instruit et spirituel. Dans ses écrits, la pensée se cache sous une certaine bouffonnerie. Espèce de barbare policé, de poète ironique, dépravé même, capable de généreuses dispositions, tout en méprisant les peuples et les rois. Les églises gothiques admettent dans leur grandeur des décorations grotesques. »L'accueil qui est fait aux Rostopchine par l'aristocratie parisienne est mitigé : curiosité, admiration, mais secret mépris pour ces jeunes étrangères qui ne savent ni s'habiller ni se coiffer et parlent en roulant les r. Le jugement que Sophie porte sur les Parisiens est tout aussi nuancé. Le 15 août 1818, un fait divers scandaleux secoue la bonne société. On repêche dans la Seine le corps d'Octave de Ségur. Or, une lettre de lui circule : il s'est suicidé, désespéré par les infidélités de sa femme Félicité. Il laisse trois fils, Eugène, Adolphe et Raymond, dont l'aristocratie s'écarte désormais. Sans doute les émigrés russes sont-ils moins farouches, car Eugène est accepté comme mari de Sophie qu'il épouse le 14 juillet 1819 dans la chapelle privée du cardinal de la Luzerne, place du Louvre.

Cette œuvre immense charme par l'atmosphère brutale et rustique où elle plonge. Les fessées du Bon Petit Diable n'ont rien à envier à celles qui faisaient jouir Jean-Jacques Rousseau. On a souvent l'impression que la vieille Russie, avec ses neiges et ses traîneaux, n'est pas loin. On se pose des questions assez savoureuses sur la part de naïveté et la part de rouerie de la comtesse. C'est le cas, notamment, pour l'histoire assez noire qu'elle nous raconte dans son roman La Fortune de Gaspard, que Gérard Blain a magnifiquement adapté à la télévision.

Tout commence comme Le Rouge et le Noir, de Stendhal. Un brave homme de paysan maltraite son fils Gaspard qu'il surprend sans cesse plongé dans des livres. C'est qu'il a des ambitions et veut échapper à son humble milieu. Son modèle est un certain Féréor qui a créé, dans le village, une usine d'effilage et de profilage des métaux. Il est à la fois la providence du village qu'il fait vivre et un objet de haine, en raison de la dureté avec laquelle il exploite ses ouvriers. Tout le roman nous décrit la ruse et la bassesse dont fait preuve Gaspard pour séduire cet homme, lui-même dépourvu de tout sentiment humain. Finalement, il réussira à se faire adopter par lui, et leur «mariage» sera célébré par une messe à laquelle assistera tout le village abasourdi. Les sentiments de la comtesse à l'égard de ses héros sont indéchiffrables. On songe à la règle du jeu formulée par Flaubert à propos de son Dictionnaire des idées reçues: faire en sorte qu'à chaque ligne le lecteur se demande si on parle sérieusement ou si on se moque de lui, sans qu'il puisse jamais répondre à cette question.

Oui, le personnage et l'écrivain Sophie sont incomparables. Au milieu des ruines fumantes de la défaite de 1871, elle écrit à son petit-fils: «Dans peu d'années, nous serons comme le Phénix qui renaît de ses cendres plus glorieux que jamais.»

Ce message fait suite au précédent. Ne pas lire ce qui suit, si vous n'avez pas vu ou lu les oeuvres, dont il est question : Witness for the prosecution (Témoin à charge) de Billy Wilder, Beyond a reasonnable doubt (L'invraisemblable vérité) de Fritz Lang, The stage fright (Le grand alibi) d'Hitchcock, The strange affair of uncle Harry de Robert Siodmak, Le meurtre de Roger Ackroyd d'Agatha Christie. Certains auteurs de polars usent de ce crédit que nous accordons à ce qui nous paraît vrai, à ce dont nous ne pensons pas douter. Ce procédé est magistralement mis en œuvre dans certains films d’Hitchcock (Le grand alibi -Stage fright, par exemple où l’innocent apparemment accusé à tort se révèle bel et bien coupable à la fin, procédé repris dans L’invraisemblable vérité – Beyond a reasonnable doubt de Fritz Lang) ou dans le film de Billy Wilder intitulé Témoin à charge (Witness for prosecution). Dans les deux cas, « nous nous faisons avoir » (et nous sommes à la fois furieux et ravis du procédé, puisqu’il nous surprend). En effet, si nous pensons à Témoin à charge, un homme, encore jeune, est accusé du meurtre d’une vieille femme avec qui il se serait conduit en séducteur. Il hérite de sa fortune. Or, selon toutes apparences, il n’est pas coupable. Son épouse détient un alibi ; mais voici qu’au lieu de le défendre, elle l’enfonce, donne toutes les preuves de sa culpabilité ; parallèlement, on découvre le passé sordide de cette femme (qui est déjà mariée à un autre et qui a un amant). L’avocat de l’accusé (interprété par le savoureux Charles Laugthon) parvient à innocenter l’homme en enquêtant sur le passé de sa femme, sur la foi de qui reposent l’accusation. L’homme est acquitté et, très vite, nous comprenons qu’il est réellement coupable, et que sa femme et lui ont imaginé ce plan afin de le faire innocenter. En effet, quelle meilleure preuve de son innocence que de se faire accuser à tort, apparemment, par une femme vile et adultère ? Idée vicieuse mais efficace. L’avocat de la défense s’exclame : « J’avais pensé à tout, sauf à ça ! » La logique n’a pas été épuisée… Or, il se produit alors un autre retournement de situation : l’homme a une maîtresse (qui vient narguer l’épouse au tribunal, après le verdict) et s’est joué de sa femme ! Cette dernière, qui va être jugée pour parjure et est abandonnée, l’assassine aussitôt avec un coupe-papier qui traîne opportunément dans le tribunal ; justice est faite, finalement ! La troublante Marlene Dietrich joue également un rôle de premier plan dans le film d’Hitchcock susnommé. Le vrai-faux meurtrier, là encore, paie à la fin, par la chute providentielle d’un rideau de fer du théâtre. Le film est remarquable par la « tricherie » du cinéaste : l’histoire nous est racontée par le meurtrier sous la forme d’un flash-back mensonger et nous sommes incités à croire ce que nous voyons (les images du film justifient le propos de l’homme). C’est le deuxième élément, après l’épuisement de la logique, que nous souhaitions mettre en évidence : l’envie de croire. Le film de Siodmak repose malgré la valonté du cinéaste sur un procédé comparable (les studios lui ont imposé cette fin décevante), et le héros interprété par George Sanders (encore lui ! Mais je l'adore !) se réveille à la fin de cette aventure meurtrière. Nous nous sommes faits piéger dans et par son rêve. Le meurtre de Roger Ackroyd constitue sûrement la plus éclatante réussite du genre. Agatha Christie fut accusée de trahison, tout comme Hitchcock, ce dont elle se défendit dans son autobiographie[1]. La lecture de l’essai de Pierre Bayard[2], bien que très contestable dans le fond et la forme, se révèle fort plaisante et agréablement stimulante du point de vue intellectuel. [1] « Là [dans Le Meurtre de Roger Ackroyd ] j'avais trouvé une bonne formule, et je la dois en partie à mon beau-frère James qui avait dit d'un air maussade, quelques années auparavant : - Maintenant, tout le monde peut se révéler coupable, dans un roman policier, même le détective. Moi, ce que j'aimerais, c'est un Watson coupable. C'était une idée fort originale, et j'y repensai souvent. Il se trouva qu'une suggestion très similaire me fut faite par celui qui était alors Lord Louis Mountbatten. Il m'écrivit pour me demander si je ne pourrais pas envisager une histoire racontée à la première personne par quelqu'un qui se montrerait ensuite être l'assassin. (…) L'idée me paraissait ingénieuse, et j'y réfléchis longuement. Elle présentait d'énormes difficultés, bien sûr. Mon esprit rechignait à imaginer que Hastings pût tuer qui que ce soit, et de toute façon, il ne serait pas facile de monter une telle histoire sans tricher. Certes, beaucoup prétendent que Le Meurtre de Roger Ackroyd est une tricherie. Mais qu'ils le lisent avec attention, et ils verront qu'ils se trompent. Une phrase ambiguë permet de dissimuler les inévitables sauts dans le temps. Quand au Docteur Sheppard, il éprouve un malin plaisir à n'écrire que la vérité : pas toute la vérité, mais la vérité tout de même. » Agatha Christie, Une autobiographie, Ed. Le Masque, Paris, 2002, p. 413-414, je souligne. [2] Qui a tué Roger Ackroyd ?, Ed. de Minuit, « Paradoxe », Paris, 2001

Remords secrets