#661: The Mystery of the Yellow Room (1907) by Gaston Leroux [trans. ???? 1909]

Mystery of the Yellow Roomstar filledstar filledstar filledstarsstars
Can a book still be a masterpiece if it’s not brilliant? In the case of Gaston Leroux’s debut The Mystery of the Yellow Room (1907) — which plays up to and anticipates so many of the established and forthcoming trappings of detective fiction — I’d say yes.  The focus on propelling the plot at a time when even those who were focussed on plot weren’t exactly propulsive is both admirable and impressive, and the creativity Leroux brings to a subgenre that would utilise the secret passage for another 60+ years is staggering.  But it would be folly to claim that age has not caught up with it and that this was in the same class as the genre’s genuine masterpieces of the 1930s.

I’ve previously called this one of my fifteen favourite impossible crime novels, and in rereading it I think I’d now revise that opinion, but in a way that doesn’t alter my respect for it.  The murderous attack on Mathilde Stangerson in her yellow-wallpapered bedroom while her father and trusted factotum are outside, the windows being barred and the door locked and yet no sign of her assailant save a bloody handprint when the door is broken in, unfolds at a rapid pace and has intrigues aplenty at its filigree’d edges.  Why do the concierge and his wife insist they were woken by the shots when it’s established that no shots could be heard from their quarters?  Why does Rouletabille saying to the innkeeper Daddy Mathieu “We shall have to eat read meat — for now” have such a profound effect on that man?  In an era when multiple plot strands would infest mystery stories to the dilution of their supposed core, Leroux has a firm had on his characters and their role in the action — pretty much everything plays a direct part in explaining the crimes.

And the crimes themselves — the vanishing of the attacker from Mlle. Stangerson’s room, then again from a corridor when hemmed in on all sides, and finally the death by stabbing of a man undoubtedly shot and bearing not a single bullet wound on his corpse — are baffling, intelligently presented, and clearly labyrinthine enough to drive even seasoned investigator and pride of the Sûreté Frederic Larsan somewhat batty.  A sense of genuine hair-pulling bedevilment permeates every corner, in spite of Leroux’s avowed intention to provide you with every piece of information at Rouletabille’s fingertips: two maps of buildings are provided, reporter (and so outsider) Rouletabille is given unfettered access to crime scenes and congresses of the legal minds involved (“In our profession and for the general welfare, we have to put up with such mortifications and bury selfish feelings,” our narrator Jean Sainclair wryly informs us, setting the pattern for every amateur sleuth yet to come), interviews with the police are relayed with word-perfect recall.  Leroux wants to play fair before ‘playing fair’ was really a thing, and deserves credit for that.

And yet for every moment that contributes to laying a pattern that a great many others would follow — the warring sleuths and their contrary interpretations, the obscure declarations later shown to have great meaning, the servants being nameless until about the 60% mark, the subtle misdirections plied by the criminal and the equally subtle ways they give themself away — there’s much here that innovates because of how it infuriates.  Freeman Wills Crofts doubtless showed us a drawing of the footprint that proves so crucial in the first section of his own debut The Cask (1920) because of the infuriatingly late new interpretation put on the oft-discussed footprints herein, the “second half” of the guilty party comes out of nowhere, and goddamn isn’t Rouletabille the luckiest man alive to just happen to be in place to overhear the two key obscure phrases that enable him to unlock the whole thing (sure, Agatha Christie did this with Appointment with Death (1938), but that was one phrase, and at least she had the insight to make it the opening line of the book).

For every delight — lines which could be interpreted as direct influences on works by G.K. Chesterton and Edogawa Rampo, the Holmes-esque casual references to earlier cases in Rouletabille and Larsan’s pasts — there’s an infuriating influx of vagueries: it’s never addressed how a man can be shot at three times from close range and yet not have a single bullet-wound on him, the attack on Mlle. Stangerson is made ‘possible’ by an assumption that requires an oversight we’re assured can’t have happened, and that corridor-based vanishing is decidedly less well written on the page than in my memory: I can believe early 20th century readers falling for it, but not the men on the scene at the time.  Knowing the guilty party going in this time around — and, like The Curse of the Bronze Lamp (1945) by Carter Dickson, there’s a weird insistence on calling them a “murderer” despite no actual, y’know, murder having taken place — I really appreciated how gosh-darned lucky they got at times, too: not that luck in impossible crimes bothers me unduly, but good heavens there’s someone who better have an altar to Serendipity in their family room.

No doubt authors read this and learned a huge amount from it (though I wish a few more of them had followed Sainclair’s lead when he writes that “I shall not here report the long examination to which Monsieur Darzac was subjected”, as it would have saved a great many readers a great many tedious inquests) and it contains — in observations about the press and its desire to print the most salacious news stories, and in Larsan’s wry reflection about the police inventing accomplices when the real criminal cannot be found — points of view that feel several decades ahead of their time.  But the purpose of a stepping-stone is for people to rest on it a while and then move on, and the novel of detection, and the impossible crime, most certainly did both.  Its focus and innovation, its layering of strands and insistence on the rational, its wild playfulness and brilliant creativity in exploring multiple false solutions commend it and make this novel worthy of the utmost respect.  But also, don’t be afraid to read it and find it wanting.  The genre probably wouldn’t have emerged and improved as it did without this, but it did emerge and it did improve.


See also

Fiction Fan @ Fiction Fan’s Book Reviews: The plotting is great, enhanced by a couple of detailed floor plans allowing the reader to try to get to the solution before Rouletabille. (I failed miserably!) The initial mystery of the locked room is only one of the “impossible crime” features – there is another halfway through which is not only baffling but quite spooky, and there are other sections where Leroux creates a beautifully tense atmosphere. But overall the book leans more towards entertainment with lots of humour, especially in the rivalry between Rouletabille and Larsan.

Wyatt James @ GADetection wiki: Reread this a week or so ago. I have no idea why it is ranked so highly as a detective story (by Carr and the like). Even the ‘impossible crime’ solutions are ludicrous… Worse are the contradictions and improbabilities – the ultimate villain had an iron-clad alibi (he was in London at the time), so how could that have been faked? How come footprints are found on hallway rugs in the house even when no mud was tracked in and whoever laid them did not have cleats on his shoes? How can you take seriously Roulettabile’s superimposing a piece of paper over a footprint in the mud and cutting out an outline of it with scissors, then identifying another footprint as being by the same person by superimposing his cutout? The melodramatic plot itself is both improbable and doesn’t really apply to any of the characters who did what they supposedly did. The butler did it (not in this case) because he was secretly in love with Mlle. and therefore had to kill her! Leroux was a very sloppy writer. This book is absolutely absurd.

Xavier @ At the Villa Rose: French scholars’ lack of interest in the traditional detective novel may explain why so few of them ever cared about the book’s role in the shaping and development of the genre. Most regard it as a trophy to be paraded every time the need arises to remind the world that French crime writers can do it just as well as their Anglo-Saxon colleagues when they consent to.


Mystery of the Yellow Room WWHaving this time read the 2018 HarperCollins Detective Club reprint shown above, The Mystery of the Yellow Room becomes the first novel I’ve read in two different translations.  My previous encounter was via the Wordsworth edition shown left, which is copyrighted 2010 but does not name its translator (the reprint doesn’t either, frustratingly, but since it’s the 1909 first English translation I can well believe that such details may well have not seemed important to History).  The differences aren’t huge, but a few stand out — such as the use of ‘Daddy Jacques’ and ‘Daddy Mathieu’ in the 1909 text where the Wordsworth called them ‘Old Jacques’ and ‘Mathieu’ respectively, or the anglicising of Bête du bon Dieu in the later version.  A fuller comparison of the two texts might be interesting if done by someone with a talent of that sort of thing, but I wanted to pull out a few instances where the phrasing — and in some cases even the contents of the novel — vary to an interesting degree.

The end of chapter 6 was the first instance that struck me: Rouletabille outside the door of the Yellow Room for the first time declaring in the 1909 translation “There is the door behind which some terrible scene took place” is a suitably dramatic note on which to end a chapter, and stood out because the first time I’d read that sentence in the Wordsworth edition it was the far more prosaic “Here’s the door behind which something happened!”, which is a startlingly bland sentiment at such a dramatic moment and shows no improved appreciation for the text after a century of reading.

Elsewhere, the Wordsworth edition includes entire lines of prose and dialogue that the HarperCollins doesn’t.  For instance, the scene in which Rouletabille, M. Stangerson, Larsan, and Old Jacques aim to capture the villain before their miraculous vanishing in the corridor reads as follows just as Rouletabille is about to clamber through the window and force the mysterious personage out into the hallway and the waiting arms of his confrères:

Ah!  Now for rapid action!  I had to be the wind, the hurricane, lightning itself!  But, alas! there were some necessary movements to be made and, while I was making them — my knees on the windowsill, my feet on the floor — the man, who had seen me at the window had bounded to his feet…

Where the HarperCollins version has the following:

Now for quick action!  it was indeed time for that, for as I was about to place my legs through the window, the man had seen me, had bounded to his feet…

…with the 1909 editors perhaps feeling that Rouletabille going on about how quick he intended to be had the ironic effect of slowing down the narrative.  Who knows?  And then later in that same chapter, as the villain runs into the corridor and vanishes, Wordsworth would have it play out like this:

“He could not have escaped!” I cried in a fit of temper, for my anger was greater than my terror.

“I actually touched him!” Frederic Larsan exclaimed.

“He was there!  I felt his breath on my face!” cried Old Jacques.

“We were touching him!” M. Stangerson and I repeated.

And once more we all said, like maniacs, “Where is he?  Where is he?  Where is he?”

Where the 1909 version pares this back a little:

“It is impossible he can have escaped!” I cried, my terror mastered by my anger.

“I touched him!” exclaimed Frederic Larsan.

“I felt his breath on my face!” cried Daddy Jacques.

“Where is he — where is he?” we all cried.

I could go on, but with neither editor nor translator available to account for these differences it will simply have to remain a point of largely academic interest.  I’m aware that translation between languages is a matter of some interpretation – the difference between ‘running’ and ‘fleeing’, say — and that some editorial changes made be made necessary by cultural elements that are too labyrinthine to wish to get into (interestingly, the Wordsworth edition goes to the trouble of providing a footnote to explain Rouletabille’s nickname which the 1909 edition doesn’t…), but the excision (or addition?) of whole lines and phrases makes an interesting point of reflection.  Maybe I liked this more first time because the Wordsworth edition provides enough wider touches to make the book feel richer — I suppose only a page-by-page comparison would make that evident, and not even I would inflict that upon you, dear reader.

Anyhoo, I found this difference in texts interesting; any insight would be greatly appreciated.

28 thoughts on “#661: The Mystery of the Yellow Room (1907) by Gaston Leroux [trans. ???? 1909]

    • I started the sequel, intended to review it on here, fairly soon after starting this blog, and I could not get on with it. I wonder what proportion of the learned nerds who devour detective fiction have read them both…


      • “Both?” For some reason the only Rouletabille review I have on my blog is that of the third novel, which nobody ever mentions… My vague memories of the second was that it was really melodramatic, and it isn’t as if the first one wasn’t!


        • I say “both” because I assume most detection fans would start with the first, go on to the second, and then possibly no make it to the third 🙂 I understand the series became quite adventure-centric after TPotLiB, so I imagine the proportion of Yellow Room readers who read The Secret of the Night is pretty darn small.

          Maybe I should dig them up and see just how much they veer from the course; have always been a fan of actually experiencing stuff there’s a lot of received wisdom about, but …Lady in Black was hard yards, IIRC, so I might put this project off for a little bit…


  1. I read the book in Spanish, and remember a prosaic line equivalent to “Here’s the door behind which something happened!”, so I presume that one is in the original French. If the Wordsworth translation consistently has more lines, then probably it is more accurate than the 1909 one. English translations from French from those times are notorious for often cutting the text at will.

    I was fortunate to read this book at a young age, when all my mystery experience was 3 Investigators and Hardy Boys, some Sherlock Holmes, and a few Father Brown stories. I fell in love with it, and was certain it was the most perfect mystery novel of all time. Reading your review all the flaws you mention are indeed apparent(*), but I still think the novel is a masterpiece. I think one of the most extraordinary things in it is the identity of the murderer. Long before Christie taught readers to suspect ~everyone~, indeed long before she used a similar twist in a Poirot novel, Leroux hides the murderer’s identity in what in 1907 must have been seemed to all readers just a stock character required by the genre and utterly above suspicion, even though the clues pointing to their identity are fair and obvious in retrospect.

    (*) I am not sure though what you mean by “the attack on Mlle. Stangerson is made ‘possible’ by an assumption that requires an oversight we’re assured can’t have happened”–could you elaborate, in rot13 if needed?


    • Oh, I completely agree about the brilliance and the intelligence Leroux shows — the various different solutions he suggests, the way our guilty party has the temerity to come up with a blatant piece of misdirection and still get away with it — because, of course, why should they come under suspicion? — is genius, and laid a blueprint in many ways for so many others to follow.

      Thanks, too, for your insight on the translations. That sounds about right — the fact that the translator from 1909 doesn’t even get a mention (hell, it might even have been translated by more than one person) gives some indication of how little fealty was owed to the original script and the simple fact of it being written in another language. So, yes, I’d say you’ve hit the nail on the head.

      As to my over-obscure reference to events (I really should have cleaned that up, eh?) I refer you to the following:

      Jr’er nffherq gung gur ebbz unf orra fhowrpgrq gb gur uvturfg fpehgval, ohg ab-bar abgvprq gur unvef ba gur zneoyr orqfgnaq gung jbhyq nppbhag sbe gur jbhaq ba ure urnq.

      Not quite a ‘Chesterton’s Invisible Man’ oversight, but still frustrating…


  2. Thanks for the link! I’ve only read it once, in the Detective Club reprint, and I found the high drama style of that translation added to the fun, once I got used to it – it felt nicely foreign whereas sometimes translations are so smooth they make books feel as if they could have been set anywhere. I do agree that the genre developed and that later books are more tightly plotted, but I found this one had retained its entertainment value despite the occasional weakness.


    • There’s a real trick, I agree, to translations feeling enough like a translation to tease out the origins of a text — at (thankfully rare) times they can feel too blandly homogenised, and it’s always extremely pleasing when, as here, a French text retains its Frenchness, in whatever way that is achieved. That’s yet another skill in the translator’s toolbox, I suppose: having the cultural awareness of where the text originates and knowing how to carry that over into the ‘new’ language.


  3. Shockingly, looking at my Goodreads pre-blog rating I gave this book 3/5 – same as you! If I didn’t have lots of other new books to read and old books to re-read I would be tempted to see if the rating stayed the same, (not that I can see it going up). My main impression is of the density of the writing, it definitely felt like a story I had to plough or trudge through.


    • Interestingly, I wouldn’t say it was a trudge — more that, for me, the plot just doesn’t hold together as well as I remembered. I found it pretty light on its feet, all told, especially when you put it up against similar genre-pieces of the era (two of the Sherlock Holmes novels have divergences into weird Frontier-style Westerns for one thing…).

      However, I’m delighted to find us seeing the same thing, Kate, even if we are viewing it from slightly different perspectives 🙂


    • Ahead of rereading this, I was convinced that I’d love it just as much as I had the first time, and then I caught myself not quite buying what it was selling, or seeing the various parts not quite line up…and, well, I wrangled for a while with how much this was due to the book and how much of it was down to my general discomfort in life at the moment. I half felt that I couldn’t post a three-star review because it would finish off whatever slim vestiges of my reputation remain.

      What I really hoped to get across in this review is how knock-you-over impressive the whole thing is, even though it doesn’t quite hold up: the innovation, the multiplicity of explanations and threads — it’s an incredible work, and so spry and easy to read given its age. But, yes. it’s a classic in the way that, er, something old is a classic: it improved what precede it, informed what followed, and had a huge influence in changing the former into the latter. And it would be a bit of a slap to the Golden Age to suggest that nothing produced therein — not even a meaningful amount of stuff — could have bettered this. ‘Cos that wouldn’t be quite so Golden, would it 🙂

      I’ll never get over the first time I found out who the killer was, though. That’s one of those things that stays with you.

      Liked by 1 person

  4. Pingback: Gaston Leroux (1868 – 1927) – A Crime is Afoot

  5. My US first edition copy …Yellow Room has no translator credit. I went through all the rest of my US Leroux books and no luck. But in my Maurice LeBlanc books which were published about at the same time as Leroux I found the translator was Alexander Teixeira de Mattos, an extraordinarily prolific translator from the early 20th century of French and German works into English. He did all of LeBlanc’s books and he also is credited with the first English translation of Leroux’ The Phantom of the Opera (1911). I cna’t say if he did …Yellow Room. Considering the terrible errors — the most glaring being the use of murderer — I would surmise that De Mattos had nothing to do with the translation. Hannaford Bennett was the English translator of novels by Gaston Leroux published after 1912. It’s odd that the practice of crediting a translator was so inconsistent and haphazard. I’m guessing it had a lot to do with clout and reputation. De Mattos and Alfred Allinson (who did the Fantomas novel English translations) were respected scholars and probably had it written into their contracts to get credit on the title page, not just the copyright page where translators were often relegated.

    As further elucidation on the condensation of the 1908 translation I can offer these bits of publishing trivia. In the late 1920s and into the 1930s the novels of Maurice Renard were very poorly translated into English and abridged by Florence Crewe-Jones. Renard’s science fiction/horror novels, especially Doctor Lerne (re-titled New Bodies for Old in the US edition), often included sexual references that were completely eliminated. In Doctor Lerne, as one example, the protagonist conducts horrific interspecies grafting experiments and some passages dealt with altered sexual drive as a side effect in the blending of animal, pant and human life. Crewe-Jones has been aggressively criticized by Brian Stableford (who did new English translations of Renard’s books for Black Coat Press) for her decision to cut out passages she personally objected to. Editors at the time rarely compared the two versions prior to publication. The first US and UK editions of The Hands of Orlac, also by Renard and translated by Crewe-Jones, had huge chunks of the story missing from her translation. Condensation or abridgement and even expurgation might have been part of a translator’s job depending on who was doing the work and who was publishing the book.


    • The face of translating really has undergone a change, hasn’t it? The idea of a text being bowdlerised at the whim of one of the only people to know what it says is insane in the current connected world, but that’s how times change, I suppose. Heaven knows what famous horror stories there must be of books ending up in (or out) of English with very different contents — it happened to Christie going from the UK to the US, right? So it could happen to anyone…


        • I think I remember the Icelandic translation of one Christie — maybe Lord Edgware — taking the translator ages because of the difficulty in getting a particular clue into the language. I suppose less assiduous souls would just leave it out, but it seems odd to me to go to the effort of translating something and then do a knowingly poor job…but, hell, I guess it happens in other areas of work, too, so it’s not like they’re the only ones cutting corners 😄


  6. The Wordsworth translation reads like a true French sensation novel. Just reading the passage that ends with “Where is he? Where is he? Where is he?” reminds me of the kind of storytelling you find in the work of Maurice Renard as well as Marcel Allain & Pierre Souvestre’s fantastically fun and genuinely thrilling Fantomas novels. The repetition of certain ideas by several characters, the rhythm of that repetition, the use of the words terror, dread and horror in abundance — this is the hallmark of early 20th century sensation fiction in France. Detective, horror and thriller writers all employed these techniques. I prefer translations that capture this undeniably French flavor. Thanks for alerting us to the differences. I may have to replace all my vintage US Leroux hardcover editions (admittedly all thick tomes, so what could be missing?) with newer Wordsworth translations!


    • Yeah, I’d agree that there’s that element of sensation fiction about it — perhaps the earlier translations focussed on making it palatable for the English market, and the art of translating took some time in catching up with the spirit of the original.

      The difference a great translation makes — see Lucia Graves’ work in bringing Carlos Ruiz Zafon’s books over from Spanish — is incredible, and makes me really appreciate the work done by the likes of Ho-Ling, Louise Heal Kawai, John Pugmire, Geoffrey Sainsbury and others in giving a real sense of what those puzzle novels must have been like first-hand.

      Interesting, too, to think that later, cheaper editions might be “better” on account of fealty to the original text. How very upside down!


  7. Just looking at that final exchange you’ve quoted, the Wordsworth is more faithful to the original text, which reads (and for the most part literally translates) as follows:

    “«Il est impossible quil se soit enfui! mécriai-je dans une colère plus grande que mon épouvante!

    Je le touchais, s’exclama Frédéric Larsan.

    Il était là, jai senti son souffle dans la figure! faisait le père Jacques.

    Nous le touchions!» répétâmes-nous, M. Stangerson et moi”

    Où est-il? Où est-il? Où est-il? …”

    Source: http://www.gutenberg.org/cache/epub/13765/pg13765.html


    • Edit: Apparently brackets are escape characters that caused my translations to be removed.

      “It’s impossible that he could have escaped!” I cried in an anger greater than my terror!
      “I touched him,” exclaimed Frédéric Larsan.
      “He was there, I felt his breath on my face!” said (literally ‘made’) Daddy Jacques.
      “We touched him!” repeated M. Stangerson and I.
      “Where is he?” x3


  8. I remember having the same sort of mixed reaction when I read this. Sometimes remembering that other people found it an inspiration and then did it better because they avoid the mistakes or excesses of the original, means one is willing to forgive the bits that do not work now or are a little OTT.

    There can also be cultural differences which can make books written by authors in different countries more of a challenge to identify with. However I remember this working better for me than the first Fantomas book although a selection of the Lupin stories in a Penguin edition worked quite well. On the other hand I prefer the John Buchan Hannay books to Sapper’s Bulldog Drummond stories so maybe it is more complicated than that.

    Translations also make a lot of difference – certainly when watching Chekhov and Ibsen plays I do recall feeling that different productions of the play were almost unrecognisable primarily because of the translation rather than the strength of the performers. On the comparisons you have provided it does seem the Wordsworth translation was preferable in giving the flavour of the original.


    • The first Lupin I read, a translation of The Eight Strokes of the Clock was so! excitable! Every! Single! Sentence! Seemed keen to! Be exciting! It! Was! Exhausting!

      I see that Pushkin Vertigo have republished Emma Orczy’s Old Man in the Corner stories, and it’d be great to have similar editions of well-translated Leblanc, because some of the stories are delightful and, to the best of my knowledge, all that’s available these days is a very few of them in anything like a professional translation.


  9. I’ve a copy of the second Dedalus edition (2003) and the copyright page says it’s a “substantially revised and edited” version of the anonymous 1908 translation done by Margaret Jull Costa, but no idea how widely this revised and edited version is used. The Dedalus translation is copyrighted 1977.

    You might be surprised to learn that I’m one of those learned nerds who has read both, but if you liked The Mystery of the Yellow Room, you’re best advised to avoid The Perfume of the Lady in Black. It actually managed to damage Yellow Room with its soap opera-like plot and revelations. There’s a good reason why its rarely mentioned as a direct sequel to Yellow Room.


    • TC, I wouldn’t be at all surprised if you’d read all the Rouletabille books — your ability to dig out (and actually get through!) ever single corner of obscurity in detective fiction is a marvel.

      My memory of TPofLiB didn’t exactly paint it in glowing colours (after all, I gave up on it for a reason…), and I was kept reading purely by the promise of another impossible crime, which had not appeared by the time I quit. I could be wrong, but I have a feeling that, having taken so long to get nowhere in that book, that was the week I put up my review of The Hollow Man instead. Far more worthy!

      Also, gleeps, how many English translations of Yellow Room are there? I know it’s 113 years old, but surely there can’t be a fourth or fifth one also doing the rounds…can there?


Leave a Reply to JJ Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.