They are the solutions that you can’t forget, no matter how much you try. You desperately want to re-read and experience that delicious jolt you got the first time, but you can’t. There’s something so original or special or boundary-breaking about these books that often they end up at the top of many “best of” lists for the ending alone, whether the entire work is actually deserving of that accolade or not.

Scott Ratner, a fellow GAD enthusiast, who knows at least as much about Agatha Christie as I do (having written and performed for years the tour de force Kill a Better Mousetrap) and who likes to ask the bigger questions and wallow in the minutiae as much as I do, recently posted on the Agatha Christie Appreciation page on Facebook:

“Certain Christie works (The Murder of Roger Ackroyd, Peril at End House, Hercule Poirot’s Christmas,Crooked House, A Murder Is Announced, The Mousetrap, The ABC Murders) are dependent upon the reader never even momentarily entertaining the possibility of the true culprit’s guilt for their full impact, while others (Death on the Nile, Five Little Pigs, And Then There Were None, Cards on the Table) are not. For that reason, I consider the titles in the first named group more “fragile” and susceptible to transparency. Am I alone in this?”

Most of the responses skirted Scott’s point entirely; instead, people talked about being fooled by Christie whether the ending was tricky or not. Meanwhile, the post inspired such a rush of feelings and ideas in me and prompted me to prepare a response too lengthy to put in a comments section on social media. So here we go: a little biography, a little analysis. But first, a warning, inspired by the late, great Noah Stewart:

Please be warned that this essay concerns works of detective fiction; part of its potential enjoyment is based on surprising the reader. If you read any further, I’ll be revealing crucial elements of several books, sometimes including the identity of the murderer and other relevant plot details. If you come upon a title here that you haven’t read, it will have lost its power to surprise you to greater or lesser extent, and that would be a shame. So please go and read these books before you spoil your own enjoyment. If you proceed past this point, you’re on your own.




I started reading GAD fiction very early: I was nearly twelve when I picked up And Then There Were None in a drugstore. I didn’t seek out unusual plotlines or surprise endings. Since my little mind was not yet built for crime-solving, everysolution came as something of a surprise.

I had learned the basic structure of a classic mystery even earlier – not from books but from movies. The Saturday matinees on TV were full of old dark house mysteries and Charlie Chan films: a closed circle would be strictly drawn around a finite set of suspects, murder would be committed (sometimes a series of them!) and the hero/detective investigated, pondered over clues and testimony and alibis and such, and wandered around dark places  until the killer was exposed in the final reel.

Quite often, the killer was of the genus “Least Likely Suspectus” in that they were quite likable, or they had an alibi, or they seemed to be on the side of law and order. In his excellent 1972 retrospective of classic mystery cinema, The Detective in Film, William K Everson discusses some of the oft-repeated tricks found in these movies, such as the romantic triangle, where the girl is torn between a kind suitor and a rude suitor. Clearly, the nice guy is more suitable, but for some reason the girl can’t commit to him. That’s because the girl is using an unconscious form of Mur-DAR – she senses instinctively that the “nice” guy is a dangerous lunatic. One should also beware the kindly guardian, the silly old woman, and the suspect in a wheelchair. In fact, the only character who could be 100% crossed off the list during the 1930’s would be the one played by the famous guest star villain (Lugosi, Atwill, Karloff, et al) who was there to foam at the mouth and be lurking at the scene of every crime and who always had a good, non-guilty reason for acting this way! Meanwhile, the actual killer had been mistakenly eliminated by the viewer on the basis of their personality or position in the narrative, setting up a lovely surprise at the end.


By coincidence rather than design, I cut my teeth on writers who specialized in the surprise ending. Nobody could do this better than Agatha Christie, but if you study her, you come to see that she actually worked variations on the same few themes. But if you read her casually or as a naïve child, or if you set out to not try and match wits with her, you can be fooled over and over. The titles that Scott references above, along with several others, are predicated on the belief that readers tend to cross certain characters off their list of suspects, including those investigating the case, the perceived intended victim, and the narrator of the story. A favorite trick of Christie’s is to pile suspicion on a certain suspect, only to pull the rug out of the case built against them at the last minute, leaving us to flounder for a replacement murderer . . . and then doubling back at the end to show us that the obvious suspect was indeed the killer!

When I decided to branch out to a second author, nobody told me that Ellery Queen could also provide a top-notch surprise ending. What stroke of fate made me peruse the bookstore shelf, with all its titles, and decide to begin with The Greek Coffin Mystery?  I remember so clearly the moment I read the reveal, dropped the book and let out a cry! I was filled with a desperate need to share this experience, but I dwelled in a houseful of non-believers. I did not re-read the book for forty years, and when I did, I was . . . underwhelmed. The characters were wooden, all except for Ellery, who was obnoxious. The plot creaked, and while I could appreciate the detection, I didn’t particularly enjoy it. It was a profound lesson in the truth that sometimes a jawdropper is not enough.

When I started reading Carr, Marsh and Brand, I learned that surprise endings were perhaps the exception to the rule. Ngaio Marsh nevertraded in surprise, and this didn’t hurt her writing one jot. (Other things did, it turns out.) And while both Brand and Carr provided some wonderful shocker finales (Tour de Force, The Rose in Darkness, The Burning Court, She Died a Lady, Dark of the Moon), they were also going after other things. Brand actually preferred casting suspicion around, making you fall in love with her characters so that the final reveal would not so much surprise as hurt.

greenfordanger            tour-de-force

Compare her two arguably strongest titles: Green for Danger and Tour de Force. Both of them follow a similar pattern: a half dozen or so well-drawn characters in a closed setting (wartime hospital operating room, island getaway) are suspected of murder. Inspector Cockrill turns up the heat, trying to melt the characters’ reserve or turn them against each other. A second death ensues, and the suspicion is passed around like crudités on a tray. The reveal, when it comes, is satisfying on both a logical and emotional level. But only in Tourdoes our jaw drop when the truth is laid bare. (Not for JJ, alas.) This, I contend, makes Tour de Force a more fragile book, for it contains a secret you do not want to learn in advance.

Carr, with his Rube Goldberg contraptions and miles and miles of footprints, offered puzzles that stymied without necessarily shocking. Granted, you’re hearing from a reader who is unlikely to gasp in wonder when the “how” is revealed, and greater fans of impossible crimes probably have a stronger emotional reaction to the expose of a good method. (That may be why the Carr titles I selected above shock us not with the “how” but with the “who.” Another blogger would no doubt come up with a different list.)

the emperor's snuffbox

The problem with reading as much classic crime fiction as I have is that one can’t help but begin to grasp certain techniques and patterns that narrow down the possible significance of events and clues. Impossible crime fans talk about this all the time: there are only so many ways a person can be stabbed in a locked room, die on a smooth, sandy beach, or make a house disappear. Sooner or later, the savvy reader starts to beat the author at his own game.


This is when detective fiction starts to become a fragile thing. My thoughts go back two years, to when I read Carr’s well-regarded non-series novel, The Emperor’s Snuff Box. Some call this the most “Christie-like” of Carr titles: it doesn’t rely on any locked rooms or missing footprints, but it trades in pure misdirection. I have to admit that I had saved up this title and was excited as hell to read it. Except, you see, I’ve read too much Christie. I solved this case, and I solved it immediately. It’s a good book, but it relies on a classic trick. Spotting that trick as quickly as I did ruined the book for me.


EXAMINING THE FRAGILE TITLES (Remember my warning about spoilers . . . )

Let’s examine the list of Christie titles Scott gave us above – and to this I will add Murder on the Orient Express. In four of these titles, Christie assigns the role of murderer to somebody that readers have been trained to exclude from suspicion: the policeman is a good guy, the child is an innocent, and as the storyteller, the narrator is essentially omniscient, nearly as much of an outsider as the detective, and has to be trusted if the story we’re reading is to be believed. Granted, in The Mousetrap, the killer is posing as somebody above suspicion, which cheapens the ultimate effect just a bit. Far more effective is the moment in another title when Poirot murmurs, “One forgets sometimes that policemen are human, too.”

There are two elements here that I feel lead to the sense of “fragility,” as Scott puts it. First is the fact that most readers, if you put them to it, will admit that they don’t want to solve the mystery. And in a case where the author spreads the wealth of suspicion with as much skill as she comes up with a logical solution that eliminates the possibilities to one, the reader is far less likely to come up with the right answer. And if he does come up with the right answer, his success is accompanied by a certain satisfaction that he has followed the clues, put aside the red herrings, and done an all-around good job. The lack of a twist actually supports this feeling.


When, however, the ending is a piece of trickery, everything works out well in the end for the reader – UNLESS he guesses correctly. Then it’s all ruined. Granted, we are all so deeply brainwashed into believing in the innocence of cops and children and narrators that it would take an unhappy accident to shake us of those beliefs. But if we pay too close attention to the dirt on the chair, or the laws of genetic inheritance, or the way a man throws back his head and laughs . . . well, then, we are left sitting there with a hundred pages to go, fuming at our own cleverness.

And as I stated above, the more you read, the more fragile the situation gets. The first time you follow the detective as he tries to figure out who is trying to kill X, you probably give little thought to why the attempted killings keep failing. You emit a gasp when you learn that the potential victim is actually the killer. If the author wants to try the same trick again – and Christie used it a half dozen times – she has two options: try and shake up the pattern so well that loyal readers don’t recognize it, or admit at the outset that the intended victim might be faking for some purpose. Of course, once that supposition is said aloud, most of us cross it right off our list. We count on Christie to fool us. But then, how often has she used the “Backtrack to the Most Likely Suspect” gambit?


A Murder Is Announced is one of my favorite Christies, and I’m pleased to say it fooled me entirely. Not only that, but I can re-read this title again and again because I’m so intrigued by how Christie clues the book. Peril at End House came nearly twenty years before AMIA and basically uses the same trick, right down to a few of the same clues! And yet, is nota favorite of mine, although many fans adore it, and I do not return to it often. I think it does what it sets out to do quite brilliantly, but if by chance you glom onto the truth early on, – and, really, the personality of the murderer makes this far more possible than AMIA– the novel is far less fun.

The other problem is that, all too often, the author places too much reliance on the trick ending to stamp the imprimatur of excellence on a title. The ending of Dark of the Moon is grade A, but it marks the finale of a grade C- Carr. The Murder of Roger Ackroyd is justly famous, even if it was far from the first example of its kind. But the book itself is a pretty standard village mystery, right up till that ending.

When I first read Orient Express, I had no idea a mystery could end this way. I had no inkling of Van Dine’s Commandments (Christie breaks rules eleven and twelve here), and I dropped the book and cried out much as I did with the Queen. I still like the book, but if you re-read it, as everyone gets interviewed – twice! – you might find yourself asking why a jury couldn’t consist of six or seven people rather than twelve!


The ABC Murders plays a different sort of game, one that stands a better chance of not being seen through. Christie convinces readers that she is writing a different sort of book from her “typical” whodunnit. ABC describes a cat and mouse game between Poirot and a serial killer, whom the author “reveals” at the start. She underscores this “truth” by switching narrative point of view from Hastings to the killer. Christie relies – quite successfully here – on the reader making assumptions through this dual structure. She neversays that Mr. Cust is ABC; she doesn’t have to because we do that job for her!



Scott offers a list of “less flashy” titles in comparison to the shockers above. I think I disagree with his inclusion of Death on the Nile: maybe it’s because of the novel’s high place in my list of preference, or the movies or just the idea behind it. It’s a trick Christie had used before, going all the way back to the beginning of her career, and she would use it again soon after. I think what sets Nile apart from the other examples – and places it in this list of Scott’s – is that almost more than the success of the trick is the emotional power of the central triangle. This is something Christie experienced firsthand in her marriage to Archie Christie, and the effect it had on her writing – the realism in the 1930’s and 40’s with which she invoked the passions between a charming rotter and two women, one ruled by passion, the other by a ruthless pursuit of what she wants –  elevates these titles above the norm.

christie, agatha cards on the table later dell                      02fdbefcc9bedc018709c4319506a589

Titles like Cards on the Table and Five Little Pigs– those novels where Christie presents a more relatively “straightforward” investigation – are few and far between. Seriously, I looked for the novels where the killer isn’t eliminated from suspicion early on through trickery, and there aren’t many: I came up with Dumb Witness, Appointment with Death, Murder Is Easy, The Moving Finger, Death Comes As the End, Sparkling Cyanide, 4:50 from Paddington, and Ordeal by Innocence. (Yes, there are others at the end of the canon, but I choose not to list them.)

There’s almost a sense of relief in a book like this for both author and reader. Any of nearly a dozen people could have poisoned Mrs. Boynton, or clubbed the Symmington’s maid, wiped out Imhotep’s clan, or slaughtered Rachel Argyle. Christie’s aim here is to misdirect, but to do so in a less flashy way than in Ackroyd. A murder that must be familial turns out to be the work of an outsider. Conversely, the death by an intruder turns out to be an inside job. The exciting aspect about these titles is how they focus on evidence both physical and psychological: the character of the victim provides the detective with important clues in Appointment with Death, The Moving Finger, and – if we consider Jacko Argyle a victim – Ordeal by Innocence. The endings to these books may not be fancy, but they are satisfying.


What I have come to appreciate about Cards on the Table and Five Little Pigs that I didn’t embrace when I first read them, due to my youth, was how brilliantly Christie works with a small cast of characters upon whom she spreads equal suspicion. Cards has a beautiful hook: the victim is a collector of oddities, and the four people in the room with him when he dies are his most bizarre set – four accomplished murderers. As the past lives of this quartet are slowly revealed, it becomes our job to try and match the correct killer to thismurder. As Poirot insists, we have to seek the psychology of the case. Sure, there are twists along the way, but they merely illuminate that each suspect is worthy of the suspicion cast upon him or her. No character is wasted.


What Five Little Pigs adds to this is depth of feeling that elevates this title to the top of the canon. To some extent, every character contributes to Caroline Crale’s fate; everyone is guilty of something, whether it’s sheer prejudice, or the hiding of a small fact, or the desire one person feels for another that causes them to mis-remember. The ending may not provide a twist in the manner of Crooked House, but what it lacks in shock value makes it no less effective, both as a logical culmination of the facts and as a powerfully emotional ending, one of the strongest in the canon.

I’ll let you in on a secret that proves And Then There Were None is brilliant: I went into my first Christie experience with full knowledge of who the killer was, and it did absolutely nothing to ruin the joy of this book. Sure, uncovering the identity of the monster slaughtering these nine fellow monsters is part of the joy of the novel. Honestly, though, it’s the characters themselves, along with the unbearable tension Christie creates over these chilling three days, that is the true accomplishment of the novel. That, and her ability to get us to feel for characters like Vera and Lombard, two of the most repellent murderers to spring from the author’s pen. There is nothing fragile here about the set-up or execution. No wonder this has become the mystery to enter the realm of cultural literacy, (Look up Hirsch!), the one book in the genre that we allmust read. Everything about it is dazzling. And while I would recommend newcomers going into it with as little prior information as possible, its fame precedes it, a fact that, to my mind, does nothing to mar the joy of experiencing it. When I produced the play, I all but ignored Christie’s script at first, concentrating on the novel. (It shocked me how Christie sacrificed the book’s all-consuming power to create a “lighter” entertainment.) I incorporated dialogue and action from the book to create what was, for me, a more satisfying experience. The audience gasped not just for the killer’s reveal but for the fate of the guests.


So there you have some thoughts. I would be fascinated to hear what other think. Do authors play with fire when they come up with surprise endings? Are there inherent risks at play, like the possibility that all is ruined if a reader susses out the ending? (Witness JJ and Tour de Force.) What if all that precedes the shocker is banal by comparison? Or must it be banal for the surprise to really work?

Let’s discuss . . .


  1. I love the choices you made here, Brad. Your examples really show some of the best of the jaw-dropper endings. And you make such an interesting point about Christie and her endings. If you look closely, and really think about it, she does use only a few variations. But they work, and she was able to use a lot of solid plot to make it all seem different in each case.


    • At what point in a writer’s career, Margot – after how many books? – does she start to wonder how to recycle previously used ideas? The wonder of Christie is that some people don’t recognize the repetition and others (like you and me) actually admire her for the way she varies similar tropes.

      Liked by 1 person

  2. Excellent analysis, Brad, and yes, you have come much closer to the heart of what I was talking about than anyone else has. However, in one respect I think I still might be somewhat misunderstood in my meaning, and this is likely due to my usual failure to make myself clear (and my inclusion of Crooked House in my “fragile” group probably didn’t help).


    My reason for assigning Death on the Nile to the latter group is not a matter of the scope or power of its surprise, but because I don’t think it fits in with the characteristic of the first (fragile) group that the power of their surprises is dependent upon the guilt of the true culprit never— even momentarily— being considered as a possibility by the reader. This consideration of possibility can range from momentary (did Simon or Jackie do it? No, they both have alibis… move on) to extensive (it looks like Doctor Roberts did it… no it looks like Rhode did… no it looks like the Doctor… no, it looks like…). But either way, it entails exoneration in the reader’s mind at some point in the process. Yes, Simon and Jackie are cleared quickly, but they ARE considered.

    But the culprits in the “fragile” group, as I see it, are never even considered as possibilities. And more importantly, I believe that if they are EVEN MOMENTARILY considered— if the reader even for a second wonders if they could be the killer— the power of these works to deceive and surprise is significantly reduced. That’s the fragile aspect, as I see it.

    One might say that these culprits are exonerated prior to the reading of the book, by a categorical exoneration inherent in the conventions of the genre and the expected functions of the character roles: The detective is not the murderer (the function of the detective is to solve the case). The narrator is not the murderer (the function of the narrator is to be our trusted conduit to the story) . The victim is not the murderer, nor is the intended victim (the function of the victim is to supply the case, and as target of the murderer is therefore not the murderer). And yes, everybody is not the murderer (the murderer is ONE of the suspects, the function of the other suspects being to provide necessary obfuscation as to make the solution not instantly apparent). As I mentioned, Crooked House doesn’t properly belong in the category as, while we don’t generally suspect children, the role of child does not serve a specified function in the genre.

    I’m pretty sure I would’ve been fooled by A Murder is Announced if I had not read Peril at End House. But by the time I read AMIA I had not only read Peril (and several others employing that device), I had also read Ackroyd, Hercule Poirot’s Christmas, Orient Express… by this time, there wasn’t a name mentioned in the story I didn’t consider as a possibility (I even read books considering the possibility of the READER being the killer!). And so, though I wasn’t sure Letitia Blacklock was the killer, the fact of her being injured actually turned my suspicions toward her rather than away from her. And the fact that I did suspect her (I actually hoped it wouldn’t turn out to be her), resulted to her unmasking offering me no pleasing sense of Major surprise, even though I did not foresee several of the other key aspects of the solution.

    Compare this to Cards on the Table— I really don’t think reading of all other Christie novels in any order would make the solution transparent, and this is at least in part because the book’s surprises are not dependent upon my never looking in the direction of the culprit. Indeed, I believe my consideration of Doctor Roberts at multiple times in the story actually enhanced my surprise at his eventual unmasking. Same with And Then There Were None— suspecting the judge is an important part of the process of exonerating him.

    The solutions of her “non-fragile” works still often have revolutionary surprises that turn the reader’s world upside down… Most of us never entertain the possibility that Simon was still in league with Jackie. Most of us never entertain the possibility that Amyas was not planning to leave Caroline. Most of us never consider the possibility that the Seven Dials is a benign organization. But these are not surprises dependent upon us never entertaining a culprit possibility. The genre— by its very name “whodunit”— has us on the lookout for WHO could possibly be the killer. But we can’t possible consider all the possible scenarios and character relationships. That’s why I consider works hinging on these other non-culprit-identity surprises as being generally sturdier.

    The current advertising for The Mousetrap in London is “Suspect Everyone,” which I deem a pretty foolhardy tagline, especially considering its solution. And even worse when one realizes there isn’t much else TO that solution! (Incidentally, I really do appreciate the plug for my show, but I feel the need to point out that its title is KILL a Better Mousetrap. I gotta be vigilant about such things!) I can’t see how basically taunting an audience with “you’ll never guess who the murderer is!” helps Christie’s reputation, especially as it makes it all the more likely that they’ll guess correctly.

    Liked by 3 people

    • Incidentally, I would never recommend one of the “fragile” category to a first time Christie reader. Yes, the rewards of being hit with an Ackroyd level surprise is great, but the possibility of its transparency is too great, IMO, and I feel the level of disappointment might be sufficient to preclude my ever getting them to read a second. I’ll usually suggest Five Little Pigs, Death on the Nile, After the Funeral, or And Then There Were None (though I’m sometimes concerned about that signature at the bottom of the last page, and the tendency of people to flip back…) After one or two of these, I’ll hit them with Ackroyd or Orient Express.

      BTW, I wouldn’t say I know as much about Christie as you do, Brad, although I may possibly wrestle with these ideas nearly as often!

      Liked by 2 people

      • Okay, I fixed the title of your play, just in case Harold Prince is lurking around, reading my blog. I do feel grateful that I read Christie so young and that she was essentially my first time. I got so few of them correct and enjoyed them all the more for that.

        SPOILERS: Yes, I get your point about how enjoyable it was to get to the end of Cards – or any title – and discover the killer is someone you considered and then threw away. I think I was luckier than you because I plumped for Mrs. Lorrimer. That means I had the exquisite sensation of being proved correct, only to have my cleverness pulled out from under me. But I love those moments – and you don’t find them in the “fragile” titles – where Christie waves the real killer in your face for a moment and then expertly steers you away. I love that Mrs. Oliver is always essentially correct: she plumps for Dr. Roberts immediately, and in Dead Man’s Folly, her choice of killer for the murder game matches the identity of the real killer. Of course, she blows it pretty badly in Mrs. McGinty’s Dead, which is my favorite Mrs. Oliver book for many reasons. But it’s an odd book to try and apply to your theory. On the one hand, we are meant to eliminate all men from the equation; and yet, somehow, we can’t quite do that. There’s a sense that the person who pushed Poirot in the train station was a man. I think the idea is explained by a male protecting a female from Poirot’s nosiness, but I probably have to read the book again.

        Anyway, thanks for putting that post up. It was fun to ponder these titles and all those wonderful surprises.


        • Come to think of it, there’s one other liability to the “fragile” titles. As “high concept” solutions, I think they lend to an under-valuation of Christie’s ingenuity. After all, you and I know that there’s really much more to the cleverness of The Murder of Roger Ackroyd than simply that the murderer was the narrator, or to Murder of on the Orient Express than that everyone is the killer. Not just these big concepts but also the details of the execution were brilliant. But because of the “high concept” aspect, people speak of them as if that is their entire merit— merely the “thinking outside the box” aspect. On the other hand, any reflection upon the solution to Death on the Nile involves recognizing the complexity of Christie’s puzzle plotting.

          None of this is to suggest that works such as Ackroyd or Orient Express are in any way minor or lesser Christies. Quite the contrary. But I do think the fragilility aspect must be recognized as a liability weighing against their assets.


      • ***This is as vague as possible but could still potentially spoil After The Funeral****

        Interesting that you say you would recommend ATF because I would consider After the Funeral as a “fragile” title according to your thesis. The trick has two parts-the mislead about which “murder” is important, and the fact that the murderer seemingly has no believable motive for either crime. So he/she is never really considered as a viable suspect. If you DO see through these two tricks (which granted, does take a VERY astute reader) I’m not sure what else the book has to offer-the other characters are not particularly memorable, the crime and detection parts are pretty standard.


        • Actually, I tend to agree with Rick on the categorization of After the Funeral— it’s a tricky one— diffusing the clarity of my categorical distinctions— but I think its deception really is ultimately based upon our never really considering Miss Gilchrist; here not because of any categorical exoneration based on genre function, but because of the primary deception of centering attention on the wrong “murder.” Indeed, despite being psychologically consistent with her character, the wedding cake poisoning is perhaps the most potentially damaging aspect to the deception, as it really raises our suspicions of Miss Gilchrist, rather than exonerating her.

          I think one probably should put this one in the fragile category— or at least the Crooked House “dependent upon our never considering” category— though I feel it is much more more deceptive than others in those categories because its primary trick (making us look at the wrong central death) was not nearly as commonly played.

          I love the book, and I consider it very deceptive— but technically I think it must go in the fragile realm.


      • Rick, the challenge to your inclusion of AtF is that the killer is vetted as a viable option right away (the very reason for their elaborate scheme) and is essentially dismissed because of the wedding cake. I love this book so much, and I love the characters, especially the older generation.


      • The interesting part about the poisoned cake is that it made the inspector suspicious of Miss Gilchrist. Therefore the possibility that Gilchrist poisoned the cake herself is directly mentioned in the book.


      • Yes, that is the aspect of the cake poisoning that IS effectively deceptive. The inspector’s overt suspicion of it actually works to exonerate Miss Gilchrist in the mind of the reader, in the topsy turvy world of detective fiction, which is governed by what I call the deception expectation principle— a principle carried to extreme in a work such as Witness for the Prosecution. Unlike the way it would be in the real world, the more overt suspicion dumped upon Leonard Vole, the more ready is the reader to believe he is innocent. Once a suspicion is openly expressed by a character, it is far less compelling to the mind of the reader.


  3. Well, I know a novel where the narrator is the murderer but he is never considered as a possibility since his thoughts and actions clearly indicate that he is innocent. A clever trick is used here. The killer has an accident as a result of which he has temporary amnesia. After some time , his memories return to him but the events relating to the murder remain forgotten.


  4. I think a lot depends on whether or not the reader knows beforehand that the books in fragile category do ‘something special’.

    I read Ackroyd, Orient Express and Christmas when I was 12 without being aware of their reputation and all of them completely fooled me. Knowing in advance that these books are considered classics can make them transparent!

    Liked by 1 person

    • Very true. But that’s part of the fragile aspect— the fact that reputation (or knowledge of another use of the same device) can endanger the deception. One doesn’t have to worry about that nearly as much with the non-fragile group. Indeed, even a suspicion of a faked death doesn’t give away the identity of the killer in And Then There Were None. It’s her most famous novel, but not an instance of the Big Fragile Surprise.


    • How many obscure mysteries have been ruined before the fact by a reviewer who references “a certain mind-blowing Christie novel of the ‘20’s?” We have all contributed to certain novels’ reputations, and it does not bode well for the pure enjoyment by future readers. I liken it to my colleague at work who has decided to binge-watch Game of Thrones this summer and is careening from room to room with her hands over her ears, desperately trying to avoid spoilers. We crave discussing these twists, which makes shows like GoT suffer from the same fragility.


  5. This post include spoilers to the killers from Peril at End House and A Murder is announced:

    I think the difference between these two books is, that Peril at End House played the trick more straight forward (maybe because it was the first time Christie did it). It really only works if you are not suspecting Nick at all. We even know that she was in the house with Maggie, when Maggie was shot.

    Miss Blacklock in AMiA on the other hand, has sort of an alibi. We see that she got wounded herself, so if we suspect her, we must also solve, how she did it. Nick isn’t suspected, because Poirot believes that she’s the intended victim. Miss Blacklock isn’t suspected, because it really seems like she is a victim. This is a big difference, IMO.

    Also, it might worker better if you read the Christie books in order. I didn’t. And once you read some of the other “the intended victim did it” novels, you probably suspect Nick straight from the beginning (or at least consider the possibility). But since it was the first time Christie did it, Readers back then might be less prepared.

    That said, despite of it having a solution I guessed, I think Peril at End House is brilliantly done. I reread it regularly, which is more than I can say about other novels I consider obvious (Lord Edgware, they do With Mirrors).

    Liked by 2 people

    • You make the differences quite clear between the two; I just lumped them together for the obvious general reason. Reading them was, for me, a very different experience despite the similarity of the general ending. That’s the beauty of Christie. I would go so far as to say AMIA is more flawed than PaEH – too many disguised relations – but I love it more, for many reasons.


      • AMIA may be more flawed in terms of plotting, but IMO it packs a bigger emotional punch than End House. Though I did feel sad in the scene with the victim’s parents in End House.

        Liked by 1 person

      • I grant that A Murder is Announced is more diffused and subtle in its execution than Peril at End House, and therefore more potentially deceptive. But there is more than just Poirot’s belief to convince us of Nick’s peril— her own dismissal of her danger is another factor (the more such danger is pooh poohed, the more we are psychologically convinced).

        And I don’t think that the fact that Miss Blacklock has an apparent alibi really bolsters that deception all that greatly. I know from my own personal experience that once you’ve read a few GAD novels, a character’s alibi does little to stop you from suspecting them (even if you can’t figure out how they’ve done it), and once you’ve come across the apparent-intended-victim-as-culprit ploy, you begin suspecting any person who apparently narrowly escapes death, regardless of their apparent lack of opportunity. After all, opportunity deceptions were the most frequent type in the Golden Age— to the point where giving a character a cast-iron alibi was the surest way to cast suspicion upon him in the reader’s mind.

        Thus the transparency of AMIA to me, despite its impressively solid level of deceptive carpentry. It was admittedly only my extensive experience with the apparent-intended-victim ploy that made me fear (as I wish to be surprised) that Miss Blacklock was the killer. And I’ll also admit I didn’t fully figure out how she pulled it off. But as well done as it was, the ultimate revelation of her guilt brought me disappointment rather than surprise— and really only because my previous reading experience had led me to consider her as a possibility.

        And if you think about it, no one ever cites the clever alibi as the reason why people are surprised by the identity of the Ackroyd murderer. It is a clever alibi, but alone it would not warrant such surprise. No, first time readers of the Ackroyd ploy, like first time readers of the Peril at End House ploy, are flabbergasted because they had never even considered the possibility of the culprit’s guilt. So, the order in which we are are introduced to these “fragile” ploys has indeed an extraordinary level of impact. Which makes the invulnerability of such works as Cards on the Table, Five Little Pigs, and And Then There Were None all the more impressive.


      • I agree, AMIA suffers from being a bit flawed due its too many coincidences, specifically, “too many disguised relations” as you mentioned Brad, and perhaps the way Miss Marple rats out the murderer towards the end, but overall, despite it’s flaws, one of the many things that makes AMIA a bit better than PaEH (though I love the book and is up there at the list of great Christie mysteries, in my opinion) are the characters, the rich post-WW2 atmosphere, and the more interesting and unique setup that Christie pens, and who can forget the cake dubbed “Delicious Death”?


  6. There’s a tricky aspect at play in all of this in that we know what one reader will spot and gripe about another will be floored by; as a general rule I now spot killers about 90% of the time, and I’m getting to the stage of my GAD reading where this fragility could ruin a lot of books if there was nothing else to enjoy along the way. This is part of why I adore Carr’s The Problem of the Green Capsule so much — when those questions come out, whether or not you figure out early who the killer is, it’s a joyous section of uprooted expectations that I think is all the more powerful for how you’re not expecting it. Sometimes, looking ahead is so inevitable that you get to enjoy the things much closer to you on page 73 rather than page 273.

    The books that increasingly stick with me are the ones that aren’t about that Big Shock at the end, whether I saw it coming or not. I think this might be why I find myself edging slowly in the direction of slightly later crime writing (and certainly GAD-era inverted novels), because you’re never quite sure what sort of a surprise you’re going to get along the way (and this is why I think the middle reveal of The Judas Window is far more interesting than the end one).

    For the record, I’m not a fan of the structure of Five Little Pigs, as I’ve maundered on about elsewhere, but it’s one of the most brilliant reversals I think Christie ever wrote. I was led effortlessly into the assumption you;re supposed to make, and completely overlooked what is now a probably hoary and obvious implication. But them’s the breaks, and I’m delighted that I get to be surprised by the same trick at least a few times before it starts to become repetitious and frustrating. The joy of GAD is that authors knew this would happen, and so you;re never entirely sure where the next devastating surprise is going to come from.

    Liked by 1 person

    • For the record, I don’t think any author does a better job of making you jump back and forth in your suspicions than Carr. He rarely traffics in “surprise” endings, but he leads you up and down so many (locked) garden paths that I usually come out satisfied with the ending. His rare shockers are more dangerous, which is why I was so disappointed with Snuff Box (but you weren’t) and so thrilled (unlike you, if memory serves me correctly) by the ending of The Burning Court. So the first point you make here is absolutely correct!


      • No, I found Snuff-Box a little disappointing — much like you, I saw the trick very quickly (though the precise workings of how it’s deduced to be a trick in the narrative are very clever), and the further I get from Burning Court the more I like it. I’m never going to love it, and I wouldn’t say it qualifies for this notion of fragility in the way you’ve outlined above, but it’s very good, and as a piece of plotting it might even be peerless.


      • Well, Carr fans complain about the first, I’ll agree. But there’s a difference between “I saw through this trick” and “This book isn’t any good” — come come, let us talk like enlightened men.


  7. The release of The Gold Watch by Paul Halter has been somewhat delayed due to the hospitalisation of John Pugmire. He informed me about 4 days back that the book will be released as soon as he gets out of hospital.


Leave a Reply to Brad Cancel reply

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

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

Facebook photo

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

Connecting to %s