Books are wonderful – so what could be better than a book? A book about books, of course!
After I saw John Dickson Carr: A Critical Study book mentioned both in Douglas Greene’s biography of Carr and on Tangled Yarns, I felt I had to check it out.
Douglas Greene described it as “filled with insights about Carr’s writing and attitudes”, and I have to say, it met my expectations! There are some great insights here, and it’s extremely quotable. Only on one occasion did I have to stop and look up some terminology – it’s perfectly accessible for a non-literary dolt like me. I imagine, however, that it’s only really intended to be read by the Carr fanatic or those who want to study his work.

Part I is an overview of Carr’s works, separated by detective, subject, and category. Sections centred around a particular detective examine their methods and principles, if any are detectable. Thankfully, they are not biographies of the detectives. Joshi says, “… I have probably not recorded absolutely everything known about them: this sort of fanatical fact-gathering is very popular in certain schools of detective criticism, but unles these little tidbits can be shown to have some wider significance, it is all so much useless and irrelevant information”. As you can see, Joshi’s analysis comes with some critical stings every now and then! Though I must admit that, due to his aim at covering Carr’s writing from every angle, some of his observations do fall into this. For example, did you know that the number of deaths per book decreased over the course of Carr’s career? Well, now you too can be mildly interested by this fact.

Now some years ago Jim, Dan, and Ben (of The Invisible Event, The Reader is Warned, and The Green Capsule) did a fun podcast all about how you can learn more about Carr’s development as a writer by studying his work chronologically rather than by detective. I agree with them and would love to see more investigation in that direction – but Joshi makes some excellent points here that make a lot of sense. It’s very believable that Carr wrote his pseudonymous novels with a different goal in mind to those he wrote under his own name.

Of the detective “series” the one which is the most different from the others is the Bencolin series. As far as I can tell, many current Carr aficionados view these as impressive curiosities which represent Carr before he really gets into his stride. For Joshi, though, these early works rank among the best of the Carrs. He points out that they have greater interest in horror and, interestingly, in emotion, than the books that came after them. Though I don’t much like the Sharon Grey bits (Joshi is more of a fan), that relationship is given greater importance, and feels more emotionally real (despite the overwrought emotions) than most of Carr’s later “romances”. Unfortunately I still have two Bencolins left to read and had to skip the discussion of Joshi’s favourite, The Corpse in the Waxworks.

Joshi’s clear favourite among the detectives is Dr. Gideon Fell, whom Joshi analyzes alongside Superintendent Hadley. He posits that it’s the Fell/Hadley combo which helps Carr to achieve his greatest heights of detective fiction; I do wonder if that’s a bit post-hoc, since Hadley’s departure co-incides with Carr’s decline in writing ability. What do you think? I never particularly thought of Fell and Hadley as an iconic pair, but maybe Hadley is more important than I give him credit for?
Speaking of Hadley, his finest performance has to be in The Arabian Nights Murder. Joshi has strong feelings on this book:
“…The Arabian Nights Murder is not merely Carr’s best pure detective story, but very likely the greatest pure detective story even written.”
(He clarifies that he doesn’t think it’s Carr’s best novel).
I hear some people prefer The Problem of the Green Capsule – but the gauntlet has been thrown!
Since I’ve come to love the “weird and ridiculous crime” Fells more as I read more of them – I’m a big fan of The Arabian Nights Murder myself – I’ll quote part of Joshi’s conclusion to this section as it encapsulates why they are so brilliant:

“In sum, it is the atmosphere of half-controlled lunacy, both in events and narration, that makes the novels unique… It is as if everything he Fell comes into contact with is bent slightly askew; and it is only he… who can set it right again.”

The chapter on Merrivale may be more controversial for some readers. Joshi is not such a fan:
“…not one… is to be compared in scope or brilliance with the four or five best Fell novels, or even with the best of the Bencolin series…”.
I don’t agree; She Died a Lady was the book that finally got me into Carr, and The Judas Window is perhaps my favourite Carr of all. He does acknowledge that the Merrivales are more experimental in tone, setting, and storytelling style than the Fells, and even calls The Judas Window “a narrative tour de force“. I can’t square his descriptions of the highlights of Merrivale’s career (he finds The Reader is Warned to be the best of these) with his dismissal of the worthiness of the Carter Dickson books.
However, there are some interesting insights in this chapter. Joshi points out that many Carr books are set a few years, or even a decade, in the past, and this begins to happen more frequently around the Second World War. This was not something I’d noticed – looking back from the modern era it can be hard to notice these small differences. Joshi also believes that Merrivale’s “buffoonery” was always present from the start – with the description of him as a “fighting Socialist” at the same time as being a Baronet and figure of the establishment always intended to make him a ridiculous figure.

I skipped much of the chapter on the “Other Detectives” as I haven’t read them; Joshi appreciates the experimentation shown in the novels featuring them, but pours scorn on Patrick Butler.

Next, Joshi gives a great deal of consideration to Carr’s historical fiction. He singles out The Murder of Edmund Godfrey as particularly admirable. Apparently, since it was considered as a more serious historical work, it received press outside of the usual crime fiction circles, including a historian describing the book as “not to be despised” (put that quote on the cover).
Considering the more adventure-based historicals, Joshi calls The Devil in Velvet “one of Carr’s masterworks” for the way it places a 20th century character into the 17th century.
Although the historicals lack a “higher purpose”, Joshi finds they are very successful as entertainment.

To close out Part I, Joshi takes a look at Carr’s short stories and radio plays. He finds Carr to be weak as a short story writer – for him, Carr requires the canvas space of a novel to work his magic with atmosphere and plot convolutions. He even dismisses The House in Goblin Wood with, “we have heard this sort of thing too many times before.”
With his work on the Sherlock Holmes stories, Joshi says “Carr …lapses into originality almost in spite of himself”, making him also ultimately imperfect as a pasticher. He also says that “The trouble with serious pastiche… is that any sort of creativeness or originality is by definition ruled out.” Pretty harsh words (that I disagree with).
The radio plays Carr worked on – including adaptations of other works – Joshi considers to be “some of the most vivid and thrilling works in his entire oeuvre.”

Part II analyzes various aspects of Carr’s work, beginning with evidence in the texts of his personal beliefs. Joshi finds evidence of Carr’s conservatism, and of his inability to portray “views opposed to his own without making them grotesquely ridiculous.” For the most part, Carr avoids “dealing with significant social, political or philosophical issues” and so portays characters with similar beliefs to his own. I was nodding (or sighing) along, but nothing surprising here.

Time for the really juicy section, “The Theory and Practise of Detective Writing”. There is a wealth of material to study, both in the novels and in essays, when it comes to Carr’s philosophy about detective stories. Joshi’s first point is that Carr claims detective fiction “does not belong to the domain of realism but to that of fantasy.” Though Joshi considers this “revolutionary”, it reminds me of G.K. Chesterton’s quote that “…in these stories the hero or investigator crosses London with something of the loneliness and liberty of a prince in a tale of elfland…”
But perhaps what was not so revolutionary back then has become more so over time. In essence the impossible crime in Carr’s work takes the same role as the presence of magic in a fantasy.
Next Joshi examines Carr’s approach to the key questions of crime fiction: “who”, “how”, and “why”. Carr focuses on the “how” more than most crime writers, to the extent that the “why” questions in his work are usually a subset of “how”: “Why was the crime committed in this way?” rather than “Why was the crime committed at all?” Engagement with the latter is restricted to making sure the crime is not committed “without due cause”.
Joshi says that, “Since the method of Carr’s murders is so intricate and recherche, it can ordinarily be fastened upon a single individual… once this method is ascertained, the motive is a mere afterthought”.
In Joshi’s view, Carr’s treatment of detective fiction as a type of fantasy cannot cover for the “[violation of] not fictional probability but human probability.”
However, he does praise Carr for considering the “why” of the impossible crime, and analyzes the four reasons for creating an impossibility which are discussed in the Merrivale books:

1. The murder is designed to look like suicide;
2. The murder is designed to look like a supernatural manifestation;
3. The “impossible” situation is the result of an accident.
4. If the method of murder cannot be ascertained, there can be no conviction in a trial.

Of these, “accident” is the most plausible and indeed apparently the most commonly used by Carr.
Next, Joshi flags up an under-appreciated (and occasionally frustrating…) technique that Carr was a master of – one inherited from the Sherlock Holmes and Father Brown stories: The Enigmatic Clue. The classic example is Holmes’ “curious incident of the dog in the night-time”. It’s a clue which is presented openly to the reader – and to the non-detective characters – but which seems completely incomprehensible. On a re-read, it will all make sense. Carr uses this technique throughout his work to add to the already formidable bafflement caused by his impossible and incredible crimes.
Joshi concludes that Carr’s self-restrictions in focusing on playing the “Grandest Game” and on impossible crimes left him a “major writer in a minor field.”

Next is a detour into examining Carr’s use of the supernatural; an element which gives Carr’s work a distinctive tone and atmosphere not matched by other detective fiction authors. Joshi believes that the standard interpretation of the supernatural in detective fiction – that dispelling it represents the restoration of order and rationality to the universe – is not the full story with Carr. While Carr deliberately punctures the supernatural atmosphere with “the obviously human” – for example, the requirement to purchase a knife – this only proves the existence of the real horror, which is “the pervasiveness of human evil”. Joshi picks out Castle Skull, Poison in Jest, and He Who Whispers, to which I would add The Man Who Could Not Shudder.
The chapter concludes with an analysis of The Burning Court, which Joshi considers Carr’s best novel. I haven’t read the book (but I have been spoiled), so I had to skip this.

The final section tackles “Style and Characterisation”. Joshi brings up the famous Dorothy L. Sayers review of The Mad Hatter Mystery:
“In short, he can write – not merely in the negative sense of observing the rules of syntax, but in the sense that every sentence gives a thrill of positive pleasure,”
before calling this “largely exaggeration. The fact is that Carr was not distinguished purely as a prose stylist.” Instead, Joshi claims, “Carr’s true virtues as a prose stylist are an unrelenting vigour that matches the bizarrerie of his plots… and an ability to change the atmosphere with startling suddenness”, qualities which build “cumulatively over the course of an entire work” (part of why he believes Carr’s short stories don’t stand out).
After considering Carr’s standard narrator figures – of whom Jeff Marle is apparently the most successful, for his contrast with Bencolin – Joshi moves on to the aspect of Carr’s work that appears to excite him the most: structure.
When Joshi calls The Arabian Nights Murder “perhaps the single most complicated work in detective fiction”, he means it as a compliment:

“It is not merely that he conceived incredibly elaborate plots, but that he unfolded them in so convoluted a fashion that we are occasionally distracted from the events of the tale to marvel instead at the author’s structural virtuosity.”

He suggests that the secret to how Carr’s books can be so readable and yet have such dazzlingly complex plots is in the portioning out of key information and explanations throughout the book, so that come the denouement, the only questions remaining are “who” and “how”. Also, the final questions are so compelling that the delivery of the solutions is not boring.
Joshi includes this key quote from an article about Carr. This description would have been from the period of the Bencolin novels, when Carr was living in New York:

“First, in the preparation of a new book, he draws up what he calls a ‘clue outline,’ marking points in the narrative at which he plans to plant the signposts to the guilt of whichever fiend he is building. Then, with elaborate pains, he makes working sketches of the characters, sometimes promoting minor players to star roles, and the reverse, as he goes, depending on how they respond to the call to creation. He jots down snatches of dialogue as he visualizes the characters and hears their speech in his mind. Plotting is easy for Carr – he habitually sees the entire network of human relations as a slough of intrigue – and the blocking out of the separate scenes is perhaps his favorite chore. He hates the actual writing, however, and does it in anguish, emerging from each long sessions hollow-eyed and spent.”

Two Authors in an Attic (New Yorker, 8th September 1951) – Robert Louis Taylor

This quote is such a tease! Enough info to make it seem like his process is laid bare… except not really!
Now for characterization, Joshi’s… less favourite aspect of Carr’s work. He starts out swinging with “[Carr] could not draw character at all” before admitting exceptions such as Fell and other key characters such as Justice Ireton from Seat of the Scornful. He then goes on to make the same statement about female characters, but here it’s harder to find exceptions; Joshi calls Carr’s typical females “overgrown children”, and compares their strikingly similar descriptions, eg “[full of vigour] without being in the least pretty”. The romances involving these characters are also “major drawbacks” for Joshi.
Joshi explains, but doesn’t excuse, the poor characterization by Carr’s focus on the plot and puzzles; maybe Carr could have done better if he’d wanted to.

Now for the conclusion, which Joshi has saved all his harsh criticism for.
He takes Carr to task for close-mindedness with regards to literature. Carr considered telling a story as “The first duty of any novelist”, and disdained any literature that aimed for something different. Joshi then goes on to consider Carr’s “rank” in literary terms, which seems a little pointless to me. What am I supposed to do, start at the top and read my way down? Carr is in the second rank, and below Margery Allingham and Rex Stout, in case you were wondering.
After castigating Carr’s work for lacking depth, he relents and celebrates Carr’s strengths. “No writer can keep one reading more than John Dickson Carr.”
When he’s firing on all cylinders, Carr’s style is “almost exhilerating in its forward energy”, producing “a consuming interest in the puzzle”. His plots are “baffling but never confusing“.
Carr’s tone is unique among detective writers in combining “intricate plot, supernatural chills, and boisterous humour.” The balance of these elements varies across the breadth of his work – the sardonic and grotesque in the Bencolin novels, the slapstick giving way to the grim in the Merrivales.
Though the puzzles are excellent, the true heart of a Carr work, says Joshi, is the “arresting tableau” (that is, a scene that creates a vivid image in your mind. Not a tableau where someone is arrested). Carr’s approach to detective fiction as a work of fantasy, combined with his mastery “of tone, mood, and atmosphere” all works together to make these striking scenes as effective as possible.
I’m glad the analysis ends on a note reminding us of just why Carr is so unique, of why we keep coming back, of what he does that only he could do.

The closing bibliography covers Carr’s publications, original and translated, as well as works about Carr and reviews of Carr’s work. The Douglas G. Greene biography doesn’t cover quite the same areas as this bibliography, so it could be quite useful to anyone else wanting to study Carr.