Studies in the Literature of Sherlock Holmes

From The Arthur Conan Doyle Encyclopedia

Studies in the Literature of Sherlock Holmes is an article written by Ronald A. Knox first read in 1911 and published in 1912 and considered as the cornerstone of the Sherlockian literature.



Editions

  • First read by Knox at Merton College (Oxford) for membres of the Bodley Club (10 march 1911)
  • Read by Knox at the Gryphon Club of Trinity College (13 march 1911)
  • Published in The Blue Book: Conducted by Oxford Graduates (july 1912)
  • Published in Essays in Satire (1928)
  • Published in Sherlock Holmes Compendium by Peter Haining (1980)


Studies in the Literature of Sherlock Holmes

If there is anything pleasant in life, it is doing what we aren't meant to do. If there is anything pleasant in criticism, it is finding out what we aren't meant to find out. It is the method by which we treat as significant what the author did not mean to be significant, by which we single out as essential what the author regarded as incidental. Thus, if one brings out a book on turnips, the modern scholar tries to discover from it whether the author was on good terms with his wife; if a poet writes on buttercups, every word he says may be used as evidence against him at an inquest of his views on a future existence. On this fascinating principle, we delight to extort economic evidence from Aristophanes, because Aristophanes knew nothing of economics: we try to extract cryptograms from Shakespeare, because we are inwardly certain that Shakespeare never put them there: we sift and winnow the Gospel of St. Luke, in order to produce a Synoptic problem, because St. Luke, poor man, never knew the Synoptic problem to exist.

There is, however, a special fascination in applying this method to Sherlock Holmes, because it is, in a sense, Holmes's own method. 'It has long been an axiom of mine,' he says, 'that the little things are infinitely the most important.' It might be the motto of his life's work. And it is, is it not, as we clergymen say, by the little things, the apparently unimportant things, that we judge of a man's character.

If anyone objects, that the study of Holmes literature is unworthy of scholarly attention, I might content myself with replying that to the scholarly mind anything is worthy of study, if that study be thorough and systematic. But I will go further, and say that if at the present time we need a far closer familiarity with Sherlock's methods. The evil that he did lived after him, the good is interred with him in the Reichenbach. It is a known fact, that is, that several people contracted the dirty and deleterious habit of taking cocaine as a result of reading the books. It is equally obvious that Scotland Yard has benefited not a whit either by his satire or by his example. When Holmes, in the 'Mystery of the Red-Headed League,' discovered that certain criminals were burrowing their way into the cellars of a bank, he sat with a dark lantern in the cellar, and nabbed them quietly as they cam through. But when the Houndsditch gang were found to be meditating an exactly similar design, what did the police authorities do? They sent a small detachment of constables, who battered on the door of the scene of operations at the bank, shouting, 'We think there is a burglary going on in here.' They were of course shot down, and the Home Office had to call out a whole regiment with guns and a fire brigade, in order to hunt down the survivors.

Any studies in Sherlock Holmes must be, first and foremost, studies in Dr. Watson. Let us treat at once of the literary and bibliographical aspects of the question. First, as to authenticity. There are several grave inconsistencies in the Holmes cycle. For example the Study in Scarlet and the Reminiscences are from the hand of John H. Watson, M.D., but in the story of 'The Man with the Twisted Lip,' Mrs. Watson addresses her husband as James. The present writer, together with three brothers, wrote to ask Sir Arthur Conan Doyle for an explanation, appending their names in the proper style with crosses after them, and an indication that this was the sign of the Four. The answer was that it was an error, an error, in fact of editing. 'Nihil aliud hic latet', says the great Sauwosch, 'nisi redactor ignoratissimus.' Yet this error gave the original impetus to Backnecke theory of the Deutero-Watson, to whom he assigns the Study in Scarlet, the 'Gloria Scott', and the 'Return of Sherlock Holmes'. He leaves to the proto-Watson the rest of the Memoirs, the Adventures, the Sign of Four and the Hound of the Baskervilles. He disputed the Study in Scarlet on other grounds, the statement in it, for example, that Holmes's knowledge of literature and philosophy was nil, whereas it is clear and the true Holmes was a man of wide reading and deep thought. We shall deal with this in its proper place.

The 'Gloria Scott' is condemned by Backnecke partly on the ground of the statement that Holmes was only up for two years at College, while he speaks in the 'Musgrave Ritual' of 'my last years' at the University; which Backnecke supposes to prove that the two stories do not come from the same hand. The 'Gloria Scott' further represents Percy Trevor's bull-dog as having bitten Holmes on his way down to Chapel, which is clearly untrue, since dogs are not allowed within the gates at either university. 'The bull-dog is more at home' he adds 'on the Chapel steps, that this fraudulent imitation among the divine products of the Watson-genius.' A further objection to the 'Gloria Scott' is that it exhibits only four divisions out of the eleven-fold division (to be mentioned later) of the complete Holmes-episode, a lower percentage than is found in any other genuine story. For myself, however, I am content to believe that this irregularity is due merely to the exception character of the investigation, while the two inaccuracies are too slight (me judice) to form the basis for so elaborate a theory. I would include both the 'Gloria Scott' and the Study in Scarlet as genuine incidents of the Holmes-biography.

When we come to the 'Final Problem', the alleged death of Holmes, and his subsequent return in an unimpaired and even vigorous condition, the problem grows darker. Some critics, accepting the Return stories as genuine, regard the 'Final Problem' as an incident faked by Watson for his own purposes; thus M. Piff-Pouff represents it as an old dodge of the thaumaturgist, and quotes the example of Salmoxis or Gebeleizis among the Getae, who hid underground for two years, and then returned to preach the doctrine of immortality. In fact, M. Piff-Pouff's verdict is thus expressed: 'Sherlock Holmes has not at all fallen from the Reichenbach, it is Vatson who has fallen from the pinnacle of his mendacity.' In a similar vein, Bilgemann asserts that the episode is a weak imitation of Empodocles on Etna, the alpenstock being left behind to represent the famous slipper which was revomited by the volcano. 'The episode of the "Final Problem"', in his own immortal language, 'has the Watsons-applecart completely overturned.'

Others, Backnecke of course among them, regard the 'Final Problem' as genuine, and the Return stories as a fabrication. The evidence against these stories may be divided into (a) those suggested by changes in the character and methods of Holmes, (b) those resting on impossibilities in the narrative itself, (c) inconsistencies found by comparison with previous narrative.

(a) The true Holmes is never discourteous to a client: the Holmes of the 'The Adventure of the Three Students' 'shrugged his shoulders in ungracious acquiescence while our visitor... poured fourth his story.' On the other hand, the true Holmes has no morbid craving for serious crime' but when John Hector Macfarlane talks of the probability of being arrested, the detective is represented as saying 'Arrest you! This is most grati — most interesting.' Twice in the Return he gibes at his prisoner, a habit from which the true Holmes, whether from professional etiquette of for other reasons, invariably abstains. Again, the false Holmes actually calls a client by her Christian name, an impossible thing to an author whose views had not been distorted by the erroneous presentation of him in the play. He deliberately abstains from food while at work: the real Holmes only does so through absent-mindedness, as in the 'Case of the Five Orange Pips'. He quotes Shakespeare in these stories alone, and that three times, without acknowledgement. He gives way to ludicrously bad logic in the 'Dancing men'. He sends Watson as his emissary in the 'Solitary cyclist,' and this is elsewhere unparalleled, for in the Hound of the Baskervilles he himself goes down to Dartmoor as well, to watch the case incognito. The true Holmes never splits and infinitive; the Holmes of the Return-stories splits at least three.

(b) It is likely that a University scholarship paper – nay, an Oxford scholarship paper, for the Quadrangle is mentioned in connexion with it – should be printed only one day before the examination? That it should consist of only half a chapter of Thucydides? That this half-chapter should take the examiner an hour and a half to correct for the press? That the proofs of the half-chapter should be in three consecutives slips? Moreover, if a pencil was marked with the name JOHANN FABER, how could the two letters NN, and these two only, be left on the stump? Prof. J. A. Smith has further pointed out that it would be impossible to find out from the superimposition of the tracks of front and back bicycle tyres, whether the cyclist was going or coming.

(c) As to actual inconsistencies. In the mystery of the 'Solitary Cyclist' a marriage is performed with no one present except the happy couple and the officiating clergyman. In the 'Scandal in Bohemia' Holmes, disguised as a loafer, is deliberately called in to give away an unknown bride on the ground that the marriage will not be valid without a witness. In the 'Final Problem', the police secure 'the whole gang with the exception of Moriarty.' In the 'Story of the Empty House' we hear that they failed to incriminate Colonel Moran. Professor Moriarty, in the Return is called Professor James Moriarty whereas know from the 'Final Problem' that James was really the name of his military brother, who survived him. And, worst of all, the dummy in the Baker Street window is draped in 'the old mouse-coloured dressing-gown'! As if we had forgotten that it was a blue dressing-gown that Holmes smoked an ounce of shag tobacco at a sitting, while he unraveled the dark complication of 'The Man with the Twisted Lip'! 'The detective,' says M. Papier Mache, 'has become a chameleon.' 'This is not the first time', says the more ponderous Sauwosch, 'that a coat of many colours has been as a deception used! But in truth Sherlock, our modern Joseph, has altogether disappeared, and the evil beast Watson has him devoured.'

To this criticism I assent: I cannot assent, however, to the theory of the deutero-Watson. I believed that all the stories were written by Watson, but whereas the genuine cycle actually happened, the spurious adventures are the lucubrations of his own unaided invention. Surely we may reconstruct the facts thus. Watson has been a bit of a gad-about. He is a spendthrift: so much we know from the beginning of the Study in Scarlet. His brother, so Holmes finds out by examining the scratches on the keyhole of his watch, was a confirmed drunkard. He himself, as a bachelor, haunts the Criterion Bar: in the Sign of Four he admits having had too much Beaune for lunch, behaves strangely at lunch, spekes of firing off a double-barreled tiger-cub at a musket, and cautions his future wife against taking more than two drops of castor-oil, while recommending strychnine in large doses as a sedative. What happens? His Eligah is taken away from him: his wife, as we know dies: he slips back into the grip of his old enemy; his practice, already diminished by continued neglect, vanishes away; he is forced to earn a livelihood by patching together clumsy travesties of the wonderful incidents of which he was once the faithful recorder.

Sauwosch has even worked out an elaborate table of his debts to other authors, and to the earlier stories. Holmes's stay in Thibet with the Grand Lama is due to Dr. Nikola; the cipher of the 'Dancing Men' is read in the same manner as that in the 'Gold Bug', by Edgar Allen Poe; the 'Adventure of Charles Augustus Milverton' shows the influence of Raffles; the 'Norwood Builder' owes much to the 'Scandal in Bohemia'; the 'Solitary Cyclist has the plot of the 'Greek Interpret'; the 'six Napoleons' of the 'Blue Carbuncle'; the 'Adventure of the Second Stain' is a doublet of the 'Naval Treaty, and so on.

We now pass on to the dating of the various pieces, so far as it can be determined by internal evidence, implicit or explicit. The results may be tabulated thus:

(1) The 'Gloria Scott' – Holmes's first case. (2) The 'Musgrave Ritual' – his second. (3) The Study in Scarlet -- Watson first appears, i.e. the first of the We-Stories. date 1879 (4) 1883, the 'Speckled Band'. (5) 1887, April, the 'Reigate Squires'. (6) Same year, the 'Five Orange Pips'. (7) 1888, the Sign of Four – Watson becomes engaged. (8) The 'Noble Bachelor'. Then comes Watson's marriage, followed closely by (9) The 'Crooked Man'. (10) The 'Scandal in Bohemia', and (11) The 'Naval Treaty', apparently in that order.

To some period in the year '88 we must assign 12, 13, and 14, that is, the “Stockbroker's Clerk', the 'Case of Identity', and the 'Red-Headed League'. In the June of '89 we have (15) the 'Man with the Twisted Lip', (16) the 'Engineer's Thumb' (summer), and (17) the 'Blue Carbuncle' (somewhere in the octave of Christmas). The 'Final Problem' is dated '91. Of the remainder, 'Silver blaze', the 'Yellow Face', the 'Resident Patient', the 'Greek Interpreter', the 'Beryl Coronet', and the 'Copper Beeches' are apparently before Watson's Marriage, the 'Boscombe Valley Mystery' after it: otherwise they are undated.

There remains only the Hound of the Baskervilles. This is explicitly dated 1889, that is, it does not pretend to be after the Return. Sauwosch, who believes it to be spurious, points out that the Times would never have had a leader on free Trade till after 1903. But this argument from internal evidence defeats itself: we can show by a method somewhat akin to that of Blunt's Undesigned Coincidences in Holy Scriptures that it was meant to be before 1903. The old crank who wants to have a law-suit against the police says it will be known as the case of Frankland versus REGINA – King Edward, as we all know, succeeded in 1901.

I must not waste time over other evidences (very unsatisfactory) which have been adduced to show the spuriousness of the Hound of the Baskervilles. Holmes's cat-like love of personal cleanliness' is not really inconsistent with the statement in the Study in Scarlet that he had pinpricks all over his hand covered with plaster – though this is also used by Backnecke to tell against the genuineness of the earlier production. A more serious question is that of Watson's breakfast-hour. Both in the Study in Scarlet and in the Adventures we hear that Watson breakfasted after Holmes: in the Hound we are told that Holmes breakfasted late. But, then, the true inference from this is that Watson breakfasted very late indeed.

Taking, then, as the basis of our study, the three long stories, The Sign of Four, A Study in Scarlet, and The Hound of the Baskervilles, together with the twenty-three short stories, twelve in the Adventures, and eleven in the Memoirs, we may proceed to examine the construction and the literary antecedents of this form of art. The actual scheme of each should consist, according to the German scholar, Ratzegger, followed by most of his successors, of eleven distinct parts; the order of them may in some cases be changed about, and more or less of them may appear as the story is closer to or further from the ideal type. Only A Study in Scarlet exhibits all of the eleven; The Sign of Four and 'Silver Blaze' have ten, the 'Boscombe Valley Mystery' and the 'Beryl Coronet' nine, the Hound of the Baskervilles, the 'Speckled Band', the 'Reigate Squires', and the 'Naval Treaty' Eight, and so on till we reach 'the Five Orange pips', the 'Crooked Man', and the 'Final Problem' with five, and the 'Gloria Scott' with only four.

The first part is the Proömion, a homely Baker Street scene, with invaluable personal touches, and sometimes a demonstration by the detective. Then follows the first explanation, or Exegesis kata ton diokonta, that is, the client's statement of the case, followed by the Ichneusis, or personal investigation, often including the famous floor-walk on hands and knees. No. 1 is invariable, Nos. 2 and 3 almost always present. Nos. 4, 5 and 6 are less necessary: they include the Anaskeue, or refutation on its own merits of the official theory of Scotland Yard, the first Promenusis (exoterike) which gives a few stray hints to the police, which they never adopt, and the second Promenusis (esoterike), which adumbrates the true course of the investigation to Watson alone. This is sometimes wrong, as in the 'Yellow Face'. No. 7 is the Exetasis, or further following up of the trial, including the cross-questioning of relatives, dependents, etc., of the corpse (if there is one), visits to the Record Office, and various investigations in an assumed character. No. 8 is the Anagnorisis, in which the criminal is caught or exposed. No. 9 the second Exegesis (kata ton pheugonta), that is to say the criminal's confession, No. 10 the Metamenusis, in which Holmes describes what his clues were and how he followed them, and No. 11 the epilogos, sometimes comprised in a single sentence. This conclusion is, like the Proömion, invariable, and often contains a gnome or quotation from some standard author.

Although the Study in Scarlet is in a certain sense the type and ideal of a Holmes story, it is also to some extent a primitive type, of which elements were later discarded. The Exegesis kata ton pheugonta is told for the most part, not in the words of the criminal, but as a separate story in the mouth of the narrator: it occupies a disproportionate amount of the total space. This shows directly the influence of Gaboriau: his Detective's Dilemma is one volume, containing an account of the tracing of the crime back to its author, who is of course a duke: the second volume, the Detective's Triumph, is almost entirely a retailing of the duke's family history, dating back to the Revolution, and we only rejoin Lecoq, the detective, in the last chapter. Of course, this method of telling the story was found long and cumbrous, but the French school has not yet seen through it, since the 'Mystery of the Yellow Room' leaves a whole unexplained problem to provide copy for 'The Perfume of the Lady in Black'.

But the literary affinities of Dr. Watson's masterly style are to be looked for further afield than Gaboriau, or Poe, or Wilkie Collins. M. Piff-Pouff especially, in his Psychologie de Vatson, has instituted some very remarkable parallels with the Dialogues of Plato, and with the Greek drama. He reminds us of the blustering manner of Thrasymachus when he first breaks into the argument of the Republic, and compares the entry of Athelney Jones: 'Oh, come, now, come! Never be ashamed to own up! But what's all this? Bad business, bad business! Stern facts here, no room for theories,' and so on. And when the detective comes back crestfallen after a few days, wiping his brow with a red handkerchief, we remember how Socrates describes the first time in his life when he ever saw Thrasymachus blushing. The rival theories of Gregson and Lestrade only serve to illustrate the multiformity of error.

But the most important point is the nature of the Scotland Yard criticism. Lecoq has his rival, but the rival is his own superior in the detective force, thwarts his schemes out of pique, and actually connives at the prisoner's receiving notes through the window of his cell. The jealousy of a Lestrade has none of this paltry spirit about it; it is a combination of intellectual pride and professional pique. It is the opposition of the regular force to the amateur. Socrates was hated by the sophists because they took money, and he did not. The cases in which Holmes takes money, explicitly at any rate, are few. In the 'Scandal in Bohemia' he is given £1, 000, but this would seem to be only for current expenses, and my well have been refunded. At the end, he refuses the gift of an emerald ring. He will not allow the City and Suburban Bank to do more than pay his expenses in connection with the 'Red-Headed League'. He says the same elsewhere: 'As for my reward, my profession is my reward.' On the other hand he takes £4, 000 from Mr. Holder when he has recovered the missing beryls for £3, 000. In A Study in Scarlet, when setting out in business, he says: 'I listen to their story, they listen to my comments, and then I pocket my fee.' In the 'Greek Interpreter' he affirms that detection is a means of livelihood with him. And in the 'Final Problem' we hear that he has been so well paid for his services in several instances to crowed heads that he is thinking of retiring from business and taking to chemistry. We must suppose, therefore, that he did sometimes take payment, but perhaps only where his clients could well afford it. None the less, as compared with the officials, he is a free lance: he has no axe to grind, no promotion to seek. And further, there is an antithesis of method. Holmes is determined not to be led away by side issues and apparent pressure of facts: this it is that raises him above the level of the sophists.

If the sophists have been borrowed from the Platonic dialogue, one element at least had been borrowed from the Greek drama. Gaboriau has no Watson. The confidant of Lecoq is an old soldier, preternaturally stupid, inconceivably inefficient. Watson provides what the Holmes drama needs – a Chorus. He represents the solid, orthodox, respectable view of the world in general; his drabness is accentuated by contrast with the limelight which beats upon the central figure. He remains stable amid the eddy and flux of circumstance.

Ille bonis faveatque, et consiletur amicis, Et regat iratos, et amet peccare timentes; Ille dapes laudet mensae brevis, ille salubrem Justitium, legasque, et apertis otia portis. Ille tegat commissa, deosque precetur et oret wut redeat miseris, abeat fortuna superbis.

It is professor Sabaglione that we owe the profoundest study of Watson in his choric character. He compares such passages at that in the 'Specked Band':

Holmes: 'The lady could not move her bed. It must always be in the same relative position to the ventilator and the rope – for such we may call it, since it was clearly never meant for a bell-pull.'

Watson: 'Holmes, I seem to see what you are hinting at. We are only jus in time to prevent some subtle and horrible crime.'

with the well-know passage in the Agamemnon:

Cassandra: 'Ah, ah, keep away the bull from the cow! She takes him, the black-horned one, in a net by her device, and smites him; he falls in a watery vessel – I speak to thee of the Mystery of the Treacherous Cauldron.'

Chorus: 'Far be it from me to boast of any particular skill in oracles, but I deduce from these words some impending evil.'

Watson, like the Chorus, is ever in touch with the main action, and seems to share the full privileges of the audience; yet, like the Chorus, he is always about three stages behind the audience in the unraveling of the plot.

And the seal, and symbol, and secret of Watson is, of course, his bowler. It is not like other bowlers; it is a priestly vestment, an insigne of office. Holmes may wear a squash hat, but Watson cleaves to his bowler, even at midnight in the silence of Dartmoor, or on the solitary slopes of the Reichenbach. He wears it constantly, even as the archimandrite or the rabbi wears his hat: to remove it would be akin to the shearing of Samson's locks by Delilah. 'Watson and his bowler' says M. Piff-Pouff, 'they are separable only in thought.' It is his apex of wool, his petasus of invisibility, his mitra pretiosa, his triple tiara, his halo. The bowler stand for all that is immutable and irrrefragable, for law and justice, for the established order of things, for the rights of humanity, for the triumph of the man over the brute. It towers colossal over sordidness and misery and crime: it shames and heals and hallows. The curve of its brim is the curve of perfect symmetry; the rotundity of its crown is the rotundity of the world. 'From the hats of Holmes's clients,' writes Professor Sabaglione, 'deduce themselves the trains, the habits, the idiosyncrasies: from the hat of Guatson deduces itself his character.' Watson is everything to Holmes – his medical adviser, his foil, his philosopher, his confidant, his sympathizer, his biographer, his domestic chaplain, but above all things else he stands exalted in history as the wearer of the unconquerable bowler hat.

And if the rival detectives are the sophists and Watson is the Chorus, what of the clients, and what of the criminals? It is most important to remember that these are only secondary figures. 'The murderers of the Holmes cycle,' M. Papier Mache assures us,' are of no more importance than the murderers are not in Macbeth.' Holmes himself often deprecates Watson's habit of making the stories too sensational, but he does him an injustice. The authors of crime are not, in Watson, of personal interest, like the Duke in Gaboriau; they have no relation to the detective other than that which subsists between the sleuth-hound and its quarry – the author of the 'Mystery of the Yellow Room' was a bungler when he made Jacques Rouletbille the criminal's natural son – they are not animated by lofty of religious motives like the high-flown villains in Mr. Chesterton's Innocence of Father Brown. All clients are model clients: they state their case in flawless journalese; all criminals are model criminals; they do the cleverest thing a criminal could possibly do in the given circumstances. By a sort of Socratic paradox, we might say that the best detective can only catch the best thief. A single blunder on the part of the guilty man would have thrown all Holmes's deductions out of joint. Love and money are their only incentives: brutality and cunning their indefeasible qualities.

And thus we arrive at the central figure himself, and must try to gather together a few threads in the complex and many—sided character. There is an irony in the process, for Holmes liked to look upon himself as a machine, an inhuman and undifferentiated sleuth-hound. L'Homme, c'est rien; l'oeuvre, c'est tout, was one of his favourite quotations.

Sherlock Holmes was descended from a long line of country squires: his grandmother was the sister of a French artist: his elder brother Mycroft was, as we all know, more gifted than himself, but found an occupation, if the Reminscences are to be trusted, in a confidential audit of Government accounts. Of Sherlock's school career we know nothing; Watson was at school, and one of his schoolmates was the nephew of a peer, but this seems to have been exceptional there, since it was considered good fun to 'chevy him about the playground and hit him over the shins with a wicket.' This seems to dispose of the idea that Watson was an Etonian. On the other hand, we have no evidence as to his University career, except the testimony (always doubtful) of one of the Return stories that he was unacquainted with the scenery of Cambridgeshire. Of Holmes's student days our knowledge is much fuller; he was reserved by nature, and his recreations – boxing and fencing – did not make him many acquaintances. One of his friends was Percy Trevor, son of an ex-convict, who had made his money in the Australian goldfields; another Reginald Musgrave, whose ancestors went back to the Conquest – quite the last word in aristocracy. He lived in a College, but what College? And at which University? The argument that his scientific bent would have naturally taken him to Cambridge defeats itself, for why should he have been only up two years if he wanted a proper scientific training? More and more as I consider the wealth of his two friends, the exclusive aristocracy of the one, and the doggy tendencies of the other, together with the isolation which put even so brilliant a light as Holmes's under a bushel – more and more I incline to the opinion that he was up at the House. But we have no sure evidence.

If he was an Oxford man, he was not a Greats' man. Yet when Watson describes his first impressions of the man at the beginning of the Study in Scarlet – the locus classicus for Holmes's characteristics – he wrongs him in saying that his knowledge of philosophy is nil, and his knowledge of literature is nil. The fact is, clearly, that Holmes did not let his talents appear till he had been living with Watson for some time, and had come to recognize his sterling qualities. In fact, he compares Hafiz with Horace, quotes Tacitus, Jean Paul, Flaubert, Goeth, and Thoreau, and reads Petrach in a G.W.R. carriage. He has no definite interest in philosophy as such, yet he holds certain definite views on scientific method. A philosopher could not have said, 'when you have eliminated the impossible, whatever remains, however improbable, must be the truth.' He could not have confused observation with inference, as Holmes does when he says, 'Observations shows me you have been to the Post Office' judging by the mud on Watson's boots. There must be inference here, though it may be called implicit inference, however rabid the transition of thought. Yet Holmes was no Sensationalist. What sublime confession of faith could a realist make that the remark in the Study in Scarlet: 'I ought to know by this time that when a fact appears to be opposed to a long train of deductions, it invariably proves to be capable of bearing some other interpretation.'

And here I must say a word on the so-called 'method of deduction'. M. Papier Mache has boldly asserted that it was stolen from Gaboriau. M. Piff-Pouff in his well-known article, 'Qu'est-ce que c'est la deduction?' declares roundly that Holmes's methods were inductive. The two fallacies rest on a common ground. Lecoq has observations: he notices footsteps on the snow. He has powers of inference for he can infer from such footsteps the behaviour of those who have left them. He has not the method of deduction – he never sits down and reasons out what is probable the man would have done next. Lecoq has his lens and his forceps: he has not the dressing-gown and the pipe. That is why he has to depend on mere chance, again and again, for picking up lost threads. Holmes no more depended on a chance than he prayed for a miracle. That is why Lecoq, baffled after a long investigation, has to have recourse to a sort of arm-chair detective, who, without leaving the arm-chair, tells him exactly what must have happened. It is wrong to call this latter character, as M. Papier Mache does, the original of Mycroft: he is the original, if you will, of Sherlock. Lecoq is but the Stanley Hopkins, almost the Lestrade, of his period. Holmes himself has explained for us the difference between observation (or inference) and deduction. It is by observation a posteriori that he recognizes Watson's visit to the Post Office from the mud on his trousers; it is by deduction a priori that he knows he has been sending a telegram, since he has seen plenty of stamps and postcards in Watson's desk.

Let us now take two pictures of Sherlock Holmes, the one at leisure, the other at work. Leisure was, of course, abhorrent to him – more so than to Watson. Watson says he was reckoned fleet of foot, but we have only his own word for it, and Holmes always beat him; beyond this alleged prowess we have no evidence of Watson's athleticism, except that he could throw a rocket through a first-floor window. But Holmes had been a boxer and a fencer; during periods of enforced inactivity he fired a revolver at the opposite wall till he had 'marked it with the patriotic device V.R.' Violin playing occupied leisure moments when Watson first knew him, but later it seems to be nothing more than a relaxation after hard work. And – this is very important – in this music was the exact antithesis of cocaine. We never hear of the drug being used in order to stimulate the mental faculties for hard work. All the stimulus needed he derived from tobacco. We all know, of course, that he smoked shag: few people could say off-hand what his pipe was made of. As a matter of fact, his tastes were various. The long vigil in Neville St. Clair's house was solaced by a briar – this is when he is hard at work; when he sees his way through a problem bin inspection, as in the 'Case of Identity', he takes down 'the old and oily clay pipe, which was to him as a counselor.' In the 'Copper Beeches' he takes down 'The long cherrywood pipe with which he was wont to replace his clay when he was in a disputatious rather than a meditative mood.' On one occasion he offers Watson snuff. Watson, by the way, smoked Ship's tobacco when he went into lodgings with Holmes, but must have replaced it soon after with a sterner stuff, thinly veiled under the nom de plume of Arcadia Mixture. This expensive product he did not abandon even under the exigencies of married life; though his circumstances were not those of affluence, since he had linoleum laid down in the front hall. But the pipe is not to Watson what it is to Holmes: to Holmes belongs the immortal phrase: 'This will be a three-pipe problem.' He is one of the world's great smokers.

Now let us see Holmes at work. We all know how brisk he becomes at the appearance of a client; how, according to the inimitable phrase in the Reminiscences: 'Holmes sat up in his chair and took his pipe out of his mouth like a hound that has heard the View Halloo.' We have seen him in the mind's eye prowling round the room with his nose an inch from the ground, on the look-out for cigarette-ends, orange-peel, false teeth, domes of silence, and what not, that my have been left behind by the criminal. 'It is not a man,' says M. Minsk, the great Polish critic, 'it is either a beast or a god.'

It is this charge of inhumanity brought against Holmes that I wish specially to rebut. True, he is reported to have been found beating the dead subjects in the laboratory, to see whether or no bruises could be produced after death. True, he was a scientist. True, we get passages like that in the Sign of Four.

'Miss Morstan: From that day to this no world has been heard of my unfortunate father. He came home with his heart full of hope, to find some peace, some comfort, and instead ---

She put her had to her throat, and a choking sob cut short her utterance. 'The date?' asked Holmes, opening his notebook.'

But is it true to say that Holmes's anxiety to catch the criminal was not, like Watson's, due to a passion for justice, but a purely scientific interest in deduction? Such truths are never more than half-0truths: it would be hard to say that the footballer plays only for the goal, or that he plays only for the sake of exercise. Humanity and science in Holmes are strangely blended. At one moment we find him saying 'Women are never to be trusted, not even the best of them' (the coward!) or asserting that he cannot agree with those who rank modesty among the virtues, since the logician must see all things exactly as they are. Even his little sermon on the rose in the Naval Treaty is delivered in order to cover the fact that he is examining the window-frame for scratches. At another moment he is purchasing 'something a little choice in white wines,' and discoursing on miracle plays, on Stradivarius violins, on the Buddhism of Ceylon, and on the warships of the future.

But there are two specially human characteristics which come out at the very moment of action. One is a taste for the theatrical arrangement, as when he sends back five orange pips to the murderers of John Openshaw, or takes a sponge into prison with which to unmask the man with the Twisted Lip, or serves up the Naval Treaty under a cover as a breakfast dish. The other is a taste for epigram. When he gets a letter from a duke, he says: 'It looks like one of those social summonses which call upon a man either to be bored or to lie.' There is a special kind of epigram, known as the Sherlockismus, of which the indefatigable Ratzegger has collected no less than one hundred and seventy-three instances. The following may serve as examples:

'Let me call your attention to the curious incident of the dog in the night-time.' 'The dog did nothing at all in the night-time.' That was the curious incident,' said Sherlock Holmes.

And again:

'I was following you, of course.' 'Following me? I saw nobody.' 'That is what you must expect to see when I am following you,' said Sherlock Holmes.

To write fully on this subject would need two terms' lectures at least. Some time, when leisure and enterprise allow, I hope to deliver them. Meanwhile, I have thrown out these hints, drawn these outlines of a possible mode of treatment. You know my methods, Watson: apply them.