What does an author owe their reader? And what does the reader owe to the person who wrote the novel that they pick up in the bookstore, or the supermarket, or even from a cardboard box that says, ‘Please Help Yourself’? 

Do we owe each other anything? 

I was studying for my final exams in my BA in English Literature, when I stole some time for myself and curled up with a murder mystery novel. This was precious time, time I should have spent doing something ‘more important’ instead. I didn’t realise it at the time, but that time was also to become the germinating idea of my doctoral thesis. I had, unknowing and randomly, selected a novel which was one of the most controversial murder mystery novels ever written – not for its content, but for breaking its contract with the reader.* 

I was so incensed at the ending that I threw the book against a wall. 

I am a creative writer, that is to say I write novels, but I am also a reader, too. As a writer I want the freedom to write anything I choose, but as a reader I have certain expectations about what I am reading. As a writer I want to break rules, but as a reader I need rules to guide me through the novel. 

Ever since Aristotle wrote ‘The Poetics’ 2000 years ago, people have argued about what the audience, or the reader, has a right to expect from the architect of the story. Aristotle declared that a story must have a beginning, a middle and an end. The film maker Jean-Luc Goddard agreed but added the caveat ‘not necessarily in that order’. 

This illustrates the problem exactly. Both writers and readers agree that there is an implied contract between them but neither of them knows what the terms are. Readers tend to want lots of clauses, writers want none. 

My PhD by Practice explores the terms of this somewhat illusive reader/writer contract through the form of a mystery novel, and through deconstructing a range of literary and commercial fiction, so that an explicit contract can be devised that gives writers the freedom to create new and exciting fiction, and leave readers feeling satisfied, excited, amazed or moved, according to the book they have selected. 

I am currently finishing my second year on a part time programme, which I have found extremely satisfying, despite the tribulations of the last year. I am lucky that being a creative writer has meant that I am used to spending long periods of time alone at my desk, although perhaps not quite this long or this alone. I do miss attending the PGR workshops run by PAHC and hope that they run again soon. While attending sessions over zoom is fine, what I found most inspiring about the in-person workshops was not the tutors, though they were great, or the topics, though they were useful, what was inspiring was listening to other PGRs talking about their projects, and feeling their passion for things I know nothing about. I am looking forward to more of that. 

However, what I most love about being on the PGR programme is being able to read novels all day and call it something ‘more important’, instead of having to steal that time away, and read in secret. 

*If you want to know the book that I was reading, it was Agatha Christie’s ‘The Murder of Roger Ackroyd’. Some readers have called it brilliant. Those readers are mostly writers. 

And they are wrong.  

**Spoiler Alert: the narrator dunnit. 

Hazell was funded through the Manchester Metropolitan Graduate School’s Research Support Award to attend the 2021 International Conference on Narratology. You can find out more about the award and upcoming deadlines by visiting the PGR Development Moodle area.  

Related Posts

2 comments on “Reader Expectations and Writer Obligations

  • 3rd August 2021 at 3:39 pm
    Dr Howard Mason

    “The Murder of Roger Ackroyd” is probably my favourite Christie. I wouldn’t say it was “brilliant”, but I would say it was satisfyingly surprising. Does it break the rules? I’m not sure it does. Christie might deflect our attention from the truth, but there’s nothing in the narrative that contradicts the ending. Certainly it confounds the reader’s expectations, but that’s what good mystery fiction is supposed to do, isn’t it?

    • 9th August 2021 at 8:51 am

      Hi Howard,

      Thank you for your comment on my PGR blog post about The Murder of Roger Ackroyd.

      I’m so glad you commented, because I could argue about this book all day (and will, given even the slightest encouragement).

      My principal objections are:

      • The narrator is presented as a stand-in for Captain Hastings, and is specifically compared to Hastings at least 5 times, which reassures the reader that the narrator is completely reliable in their narration, though probably unreliable in their deductions.
      • The narration is presented as a story told to the reader by an interested by-stander for the first 195 pages of a 221-page novel.
      • Then it becomes the official record of the investigation, of the kind written about Sherlock Holmes by Dr Watson, who the narrator, is also compared to.
      • Hercule Poirot then takes the official record and makes deductions from it, outside the four corners of the novel, which the reader is not privy to.
      • Then the narrative becomes a kind of confession, and we have the slightly ludicrous spectacle of the murderer narrating the moment when the Great Detective names the murderer (only he doesn’t) and having to do the traditional Hastings part of ‘Poirot you astound me with your brilliance’ at the same time as doing the traditional snarling part of the cornered murderer. Which feels kinda awkward
      • And then the narrator revisits the manuscript, pointing out all the times he lied to the reader. So, when he says, ‘I hesitated with my hand on the door handle wondering if there was anything I had left undone’, the bit he had left undone was the bit where he had killed Ackroyd and failed to tell us about it.
      I could go on (and on) but you get the point.

      Roland Barthes said that the problem was that ‘the reader looked for [the murderer] behind every ‘he’ in the plot: he was all the time hidden under the ‘I.’ Agatha Christie knew perfectly well that, in the novel, the ‘I’ is usually a spectator and that it is the ‘he’ who is the actor.’

      I was just writing about the Ackroyd book when I got your message, and I’d written that ‘The form of the novel is like space. Its dimensions are infinite. It can accommodate any story that the author wishes to tell. But, just as in space, there are some basic laws of physics that the novelist must either adhere to, or subvert, in such a way as they acknowledge the existence of these laws.’

      So, if she wants to have a story narrated by the murderer, Christie can do that, and in fact she did do that at least twice more in other novels. But those other novels didn’t try to shove that round peg of a plot into the square hole that is Hercule Poirot!

      Anyway, thanks for your comment. I appreciate it. Just as soon as I work out how to authorise it, I will.



      Hi Hazell
      Thank you for your email. At least, I think I want to say thank you, but it led to, let us say, a heated late-night discussion between my partner, a devout and redoubtable Christie fan, and me.

      On the other hand, it was Judith who managed to find our copy of Ackroyd, which turns out to be a Penguin, first published in 1948, reprinted in 1955, with a cover price of 2s 6d. She bought it in Lancaster Market between 1971 and 1974 for the princely sum of 8p. I confess I hadn’t read it for quite some time. Anyway, I’ve quickly re-read it and feel reasonably well-equipped to engage with your objections.

      Your first, I have to say, hardly looks like an objection at all. Yes, Christie does try to insinuate to the reader that Sheppard is a surrogate Hastings and therefore we should trust him, but more fool us if we fall for it! Surely this is just the legitimate sleight-of-hand of the mystery writer. As the doyen of crime fiction criticism, Julian Symons, observed, writing about Ackroyd in Bloody Murder, ‘every successful detective story in this period involved a deceit practised upon the reader’.

      I question whether the narrative becomes an ‘official record of the investigation’. Yes, Sheppard claims that he wants to be seen as Watson to Poirot’s Holmes, but that’s as far as it goes. It is a manuscript that records Sheppard’s (flawed) account of events, nothing more. Nor am I convinced that Poirot ‘makes deductions from it, outside the four corners of the novel, which the reader is not privy to’. The reader is never privy to Poirot’s deductions until he is prepared to reveal them, usually not till the end. There is nothing to suggest there is anything in the manuscript that has not been communicated to the reader.
      Your major objection, it seems to me, is that Sheppard ‘lied to the reader’. This, I dispute. If Sheppard lies at all, it is to other characters in the story, but we have no reason to doubt that he has faithfully recorded the conversations he was privy to. After all. Poirot says the manuscript is ‘a very meticulous and accurate account’ (p. 225). Later, he ‘murmurs’ to Sheppard, ‘It was strictly truthful as far as it went – but it did not go very far, eh my friend?’ (p. 236).
      Of course, that is the crux of the matter. Sheppard is an unreliable narrator who withholds information. But to describe this as ‘lying’ is a stretch to say the least.

      The fact is that Christie plants enough clues along the way to enlighten any reader with the wit to grasp their significance (not me, unfortunately):
      1. Sheppard feels ‘a momentary throb of anxiety’ at the mental image of Ralph and Mrs Ferrars in intimate conversation (42). Why? With hindsight, because he was afraid she had revealed to Ralph that he, Sheppard, was the blackmailer.
      2. Sheppard insists Ackroyd read Mrs Ferrars letter, for a similar reason. We are then told there is a ten-minute gap between the letter arriving and Sheppard leaving the room (45). What had he been doing in the interim? And surely we could have made something of the fact he looked back to see ‘if there was anything I had left undone’.
      3. Sheppard makes an ‘instant decision’ to relate the events of the evening to Inspector Davis. Why would an innocent man have to ‘make a decision’ to tell the police what he knew? (62).
      4. Why did Sheppard hope that his visit to the Three Boars ‘would remain unnoticed?’ (71)
      5. He gives Poirot ‘a careful narrative’ setting out what we already knew (74).
      6. We are told Sheppard had made some bad investments (107).
      7. Sheppard tries to persuade Poirot that the pulled-out chair, a key piece of evidence, is in fact of no account (140).
      8. Sheppard realises he had been ‘unbelievably stupid ‘about the boots’ (151).
      9. Further evidence of Sheppard’s financial concerns: he is desperate not to lose a large sum at Mah Jong.
      10. Poirot notices Sheppard is not startled by Flora’s admission that she hadn’t, after all, seen Ackroyd alive at 9:45; Sheppard knew he was already dead.

      You quote with approval Barthes’s remark that Christie had deceived the reader into assuming the murderer would be hidden behind a ‘he’ when in fact he was hidden behind the ‘I’. Well, yes, exactly! Well done her. I refer you again to Symons.

      Does Christie set out to mislead and deceive the reader? Of course. But you argue more than this: that she breaks the rules. But what rules? At the time Christie wrote, S. S. Van Dine’s ‘Twenty Rules for Writing Detective Stories’ and Ronald Knox’s ‘Detective Story Decalogue’ were widely known and generally adhered to by detective fiction authors, in spirit, if not always strictly in practice. Ackroyd naturally attracted much controversy, but on the whole was regarded within such circles with approval, not least by Dorothy L. Sayers and other members of the Detection Club.

      I hope that this at least gives you some food for thought!


      PS It seems to me that there is at least some overlap between our PhD studies. My project is entitled, ‘A Regular Guy: A “Realistic” Cold War Spy Novel’. It is practice-based, centring on the spy novel A Regular Guy, but the more I work on the Analytical Commentary the more I realise that defining what counts as “realistic” spy fiction depends less on extra-textual facts and more on a collaboration between reader and author.

      Hi Howard,

      Thanks for your email, which I enjoyed enormously, and also for taking the trouble to reread Christie’s book, in order to rebut me. I appreciate the effort, and I hope that my rebuttal of your rebuttal will be useful prep for my upcoming Annual Review!

      Agatha Christie, like all mystery writers, of course, is trying to confound the reader and slip things past them. If the evidence is there, and the reader doesn’t spot it, or doesn’t spot the significance of it, well more fool them. And, of course, the reader is half hoping to be tricked. It is the reason that they continue to read mystery stories in such vast numbers. But there are rules to the game. You mentioned Julian Symons book, Bloody Murder, which is an interesting history of the mystery genre.

      Speaking of the rules of detective fiction, Symons writes “The attitude from which they sprang was that the detective story was a kind of game played, as Knox put it, between ‘the author of the one part and the reader of the other part’. When one talked about rules, he went on, it was not ‘in the sense in which poetry has rules… but in the sense in which cricket has rules — a far more impressive consideration to the ordinary Englishman.’… To infringe the rules was, to say the least, extremely bad form.” (beginning, chapter 8).

      Since both Ronald Knox and SS Van Dine, both very successful mystery writers of the time, had written a version of the rules of detective fiction which explicitly forbad what Christie did, I think we can conclude that what Christie did was certainly considered ‘bad form’ at the time, and not cricket. Both sets of rules are reprinted in Howard Haycraft’s book The Art of the Mystery Story, which is also excellent.

      Van Dine’s Rule 1 stated “The reader must have equal opportunity with the detective for solving the mystery. All clues must be plainly stated and described.”

      If you are unwilling to agree that ‘hiding behind the I’ is a deceit, how about the lesser (in my mind) offence of the Dictaphone? Since you have just re-read it, you will probably remember that up until the denouement there is only one very brief mention of a Dictaphone, which covers a couple of lines only. A few days before the murder a salesman had visited the house. “Mr Ackroyd has some idea of purchasing a Dictaphone… It would have enabled us to get through a lot more work in a limited time. The firm in question sent down their representative, but nothing came of it. Mr Ackroyd did not make up his mind to purchase”. The Dictaphone which was not purchased is never mentioned again until the denouement (on page 205 in my book, and fifteen pages from the end), when it turns out that not only did Ackroyd purchase a Dictaphone, but it was crucial in establishing time of death. Poirot admits, pretty lamely, ‘Why he concealed this matter from you, I do not know’.

      Me neither. But apparently, also never mentioned till just this moment, Ackroyd had “a childish love of surprising people. Meant to keep it up his sleeve for a day or so. Probably was playing with it like a new toy.”

      On page 214 (five pages from the end) Poirot comes up the idea of a McGuffin or “something in the nature of a time lock, or even of a simple alarm clock” being added to this Dictaphone no one knows about, by a person “with the necessary mechanical knowledge”.

      That’s a big chunk of information we never had.

      On page 215 Poirot states that the murderer was a person who Ackroyd knew well enough to know that he had purchased a Dictaphone. But hadn’t we just found out 10 pages ago that Ackroyd had kept the purchase to himself because of his childish love of surprises?

      And the culprit has ‘timed the Dictaphone to go off at half-past nine.’ How did the device work? Dunno. How did he know what time to set it for? No idea. How did he know Ackroyd would be alone, in his study, and would not invite the other guests or members of his household or secretary or anyone to join him and Sheppard after dinner? Not one damned clue.

      We only find out on the last page that Ackroyd had given the Dictaphone to Sheppard who had ‘fashioned a device’ to make it replay at a certain time, and stop replaying too, presumably.
      According to Van Dine’s rules, that clue was not plainly stated and definitely was not described.

      Rule 2: “No wilful tricks may be played on the reader other than those legitimately played by the criminal on the detective himself.” You might argue that the whole manuscript was a trick played on the detective because it is the report of the investigation that he gives to Poirot, but, at the time that Dr Sheppard begins the story he doesn’t know who Poirot is, and spends several chapters thinking of him as a foreign hairdresser named Porrot. The culprit, and indeed all the suspects, can lie to the detective, and each other to their hearts content, but, I argue that the narrator can never lie to the reader, unless the author has somehow prepared the reader for this.

      Christie wrote another novel, Endless Night, which had the exact same plot device, but, crucially, she did not try to shoehorn it into a Poirot style novel, and it caused no controversy whatsoever.

      Christie breaks a whole load of Van Dine’s rules in this book, but I think that the most egregious is Rule 4, “The detective himself, or one of the official investigators, should never turn out to be the culprit. This is the bald trickery on a par with offering someone a bright penny for a five-dollar gold piece. It’s false pretences.’

      Ronald Knox has similar concerns, and his Rule IX explicitly states: “The stupid friend of the detective, the Watson, must not conceal any thoughts which pass through his mind.”

      Dr Sheppard is explicitly compared to both Captain Hastings and Dr Watson in this novel, and, as the narrator, filling the role of the honeymooning Hastings, the reader is perfectly entitled to presume that he is filling that role. Indeed, even if he had considered Sheppard as a suspect from the beginning, which you could legitimately argue that he should have done as a careful reader, there is still no way that they could have reached the right conclusion, which, again breaks Van Dine’s first rule, and incidentally the Detection Oath that Christie swore when she joined the Detection Club (Do you swear never to conceal a vital clue from the reader).

      You rightly point out that Dr Sheppard has told no outright lies, but St Augustine would have had no trouble with pronouncing him a liar, since such obvious twisting of words such as
      “The letter had been brought in at twenty minutes to nine. It was just on ten minutes to nine when I left him, the letter still unread. I hesitated with my hand on the door handle, looking back and wondering if there was anything I had left undone.”, cannot possibly lead a reader, even a particularly suspicious and perceptive one, to conclude that that the ‘anything’ referred to a murder, just committed. You mentioned the ten-minute time period, but has Dr Sheppard not just told us that that time period was taken up with checking and re-checking the window, discussing the letter, reading part of it, and Sheppard repeatedly trying to persuade him to read out the rest? Indeed, he has told us so much of what he did do, that you might wonder how he found time to also stab the old man in the neck, whip out the Dictaphone which he had brought for the purpose, move the chair, check the sightlines, and set the timer on the ‘time lock device’!

      Thanks again for all your comments. Your project sounds very interesting. Are you doing a creative PhD? I always thought John Le Carre’s books had the perfect realistic spies, and that George Smiley was as far from James Bond as it is possible to be. The Spy Who Came in From the Cold is still a classic, I think. Do tell me more, if you feel comfortable.

      Thanks again



Leave a Reply

Your email address will not be published. Required fields are marked *