Golden Age of Detective Fiction Forum

July 9, 2008

Memorable Clues

Filed under: General GAD — Jon @ 11:24 am

Since I last chimed in on this subject, I have been checking up on the excellent “memorable clue” suggestions that several of you have kindly submitted at my request. I still have quite a few left to check out (I’m neither a speedy reader nor researcher), but so far they have seemed to support my original theory (well, maybe the less ambitious term “notion” is more accurate), which I will now divulge. Please forgive how poorly I explain this; I assure you that my ideas are stronger than my ability to express them.

Though people read Golden Age detective fiction for a variety of reasons, certainly one of the key appeals of the genre is a sensation which one might be call “sudden retrospective illumination.” This sensation is referred to by many other names, including “paradigm shift,” “epiphany,” “the Homer Simpson effect (D’Oh!)” or, in Aristotelian terms, the convergence of “anagnorisis” (recognition) and “peripeteiea” (reversal). By whatever name, it entails the seemingly paradoxical simultaneous experience of surprise and inevitability (or, at least, deterministic causality).

Well, as I feared, I haven’t explained it at all well, but I suspect that most of you know what I’m talking about ( I also have another theory [by Anne Elk!] that the appeal of this sensation is tied to a subconscious validation of our very existence… but I’ll bore you with that one another time). At any rate, I believe that for many of us, this sensation is largely what defines a great whodunit denouement, and Dorothy L. Sayers described the joy of it thus:

“The aim of the writer of this type of detective story is to make the reader say at the end, neither: ‘Oh well, I knew it must be that all along,’ nor yet: ‘Dash it all! I couldn’t be expected to guess that’; but: ‘Oh, of course! What a fool I was not to see it! Right under my nose all the time!’ Precious tribute! How often striven for! How rarely earned!”

Now, so far as I’ve been able to discern, the clues suggested all provide this sensation… well, perhaps I should more accurately say, the relationship between the clues and the truths they are ultimately shown to indicate provide it. As such, they allow for solutions which surprise us, and yet are entirely consistent with all data we’ve been given earlier.

Almost without exception, however, these memorable clues also have one other important common denominator: while they are consistent with the ultimately revealed solution and, more importantly, serve to bolster the sense of inevitability of that solution (“What a fool I was not to see it! Right under my nose all the time!”), almost none of them deductively prove anything. While they may indicate possible discrepancies in the earlier, apparent scenario (i.e. what seems to be the case prior to the denouement), they don’t logically demonstrate that scenario to be impossible. Rather, they work to bring cumulative strength to the probability of the true scenario, serving, as Pooh Bah from Gilbert & Sullivan’s THE MIKADO would say, as “corroborative detail, intended to give artistic verisimilitude to an otherwise bald and unconvincing narrative.

Not that truly deductive clues do not abound in detective fiction, but you’ll find very few of them among our “most memorable” collection. For, surprising as this may be, the consensus of “memorable clues” clearly suggests that proving a scenario true by discounting all other possibilities (no matter how unassailable the logic) does not have nearly the power on the memory as does reinforcing its truth via a multitude of “circumstantial” elements.

I know that this theory holds for me. I have read several works in which it is proven that only T could have been the killer, because U, V, W, X, Y and Z couldn’t fulfill the (usually opportunity-based) requirements to be the culprit. However, as there are few details which indicate that T was the killer (beyond the elimination of other possibilities), I am not entirely satisfied by the denouement. Though I admire it in many respects, I believe that Queen’s THE GREEK COFFIN MYSTERY holds that weakness (though not nearly as damagingly as many other works).

Conversely, there are several powerful whodunits in which nothing (or nearly nothing) is proven, and yet the denouement has a powerful, memorable effect, both surprising and “inevitable.” FIVE LITTLE PIGS and HE WHO WHISPERS are two of my favorites which come to mind, though many of the other most notable works of the genre also qualify. Everything “clicks” in the denouements of these works — it all seems ultimately inevitable — yet none of the clues provided are truly univocal; they all could be accounted for with other explanations. It is only their cumulative effect which seems overwhelmingly convincing. Moreover, a large percentage of these powerful clues are behavioral discrepancies, about which nothing can be proven (a sudden change in the behavior of a character certainly indicates something, but though it can strongly indicate what that something is, it can never be proven).

I believe that the explanation for this perhaps surprising conclusion about deductive vs. corroborate clues is that, despite the importance of rational thought to the experience of detective fiction reading, the effect of “sudden retrospective illumination” is ultimately a primarily visceral one — it hits us at a gut, rather than intellectual, level.

That is not to suggest that deductive clueing is unimportant to the genre. Indeed, they are extremely useful, often giving the mystery’s solution intellectual credibility. However, their most important function is often as a precursor to the more memorable, non-deductive clues. For example, a process of deductive elimination may prove that Mr. Jennings, and only Mr. Jennings, had the opportunity to drink the full contents of the whisky glass. But it is the clue that Mr. Jennings, a well-known teetotaler, drank the whisky (and the ultimate explanation for this bizarre behavioral discrepancy) that will be most remembered.

One exception to my notion — and it is indeed an important one — is the famous “curious incident of the dog in the nighttime” clue from Conan Doyle’s SILVER BLAZE. It can be summed up as a simple logical syllogism:

1. The dog would bark if the visitor to the stables was a stranger.

2. The dog did not bark.

3. Therefore the visitor to the stables was not a stranger.

Though one could argue that it too doesn’t positively provide absolute logical proof (the first premise is not entirely solid; there are other possibilities which could account for the dog’s silence: the dog could be drugged, it might have be switched for another dog, etc…), I will grant that it fairly well proves its point.

Then, why is the deductive “dog in the night-time” clue memorable?

I can find two possible explanations:

In the first place, the logical syllogism of this clue is tied intimately with a behavioral discrepancy (unlike the Mr. Jennings clue above, in which the deductive process only leads us to the behavioral discrepancy). Thus I’d suggest that it is the why? aspect of the behavioral discrepancy and its explanation, rather than the deductive proof of that explanation, which is most viscerally powerful.

In the second place, this clue fits into that relatively rare category of clues which are clearly presented, long before their final explanation, as clues; we know that it is of importance, it is only the nature of its importance that is unknown to us until later. John Dickson Carr referred to this type of clue (of which the title phrase of his THE CROOKED HINGE is another non-deductive example) as the “enigmatic” clue, as it openly presents an enigma to the reader. The majority of clues in mysteries, on the other hand, consist of plot details which have their status as clues only made apparent at the time they are explained. The “enimatic” clue is undoubtedly among the most difficult type of clue to create, for, to put an indicator openly in front of the reader—in essence, to say to him “this is important; I challenge you to guess what it means” — and then to provide him with an answer that is both surprising and satisfying is quite a feat. Having been baffled by something so clearly and openly put in front of his face, the reader can only be greatly impressed. The “curious incident of the dog in the nighttime” achieves this, thus explaining its power.

Again, I apologize for my inability to articulate this all clearly—it’s a tricky subject. Hopefully, some of you got an idea of what I was trying to say. If so, please give me your thoughts on the matter.

– Scott


  1. I was both glad and sad to read your analysis of clues in Golden Age detective fiction. Glad to see that someone takes such fiction seriously enough to make a reasoned and excellent analysis.

    Sad because I am afraid your analysis might negatively affect my enjoyment of detective fiction.

    Let me explain: I realize that Poe thought of such fiction as a problem in logic. When I read detective fiction, however, I like to follow the detective as he uncovers and interprets each clue. It is the chase not the result that I enjoy. I read for pure enjoyment. If I read a story as a problem in logic, I tend to over-think it.

    I like hard-boiled detective novels as well as the Golden Age ones. In the hard-boiled novels, clues often seem out place. That is, they don’t seem to function logically in the story.

    Comment by coffined — July 9, 2008 @ 3:28 pm | Reply

  2. One of my favorite mysteries, I think, epitomizes the type of “under my nose all the time” clues you mention. John Dickson Carr’s “The Blind Barber” is a pure armchair detective story – the action takes place on board a transatlantic liner. A murder takes place – but the body disappears and nobody seems to be missing. When the ship arrives in England, a witness goes to Dr. Gideon Fell’s home and tells him the story of what happened – and Fell, of course, comes up with the correct solution.

    Dr. Fell adds to the problems faced by the reader by providing sixteen enigmatic clues (as you define them) – enigmatic in name: “The Clue of Invisibility,” “The Clue of Seven Razors,” and so forth. And, of course, when the solution is revealed at the end – and in the edition in my library, the explanation is footnoted with the page numbers on which the information was given to the reader – they all leave us with that feeling of, “How could I have missed them?” I can think of few other novels where we are given so many memorable clues – and, generally, can do so little with them.

    Comment by Les Blatt — July 26, 2008 @ 11:44 pm | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

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

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

Blog at

%d bloggers like this: