The First Whodunnit Transcript

Caroline: The world of detective fiction has recently passed an important milestone. It’s a hundred years since the appearance of Agatha Christie’s first novel, The Mysterious Affair at Styles. First serialised in the London Times in 1920, it appeared in book form first in the US at the end of that year and then in the UK in January 1921. A whole century since Hercule Poirot first stepped onto the printed page, and he’s barely aged a day.

The various commemorations and reappraisals that this milestone provoked made me think, though. Although Styles was undoubtedly an influence on the way in which detective fiction would develop over the next two decades of its golden age, Agatha Christie didn’t really break any new ground with it. She didn’t invent the idea of the amiable narrator and his brilliant detective friend, nor was she the first to set a mystery story in a country house. Her work with poisons in this plot is unusual, but she didn’t originate the concept so much as turn it to her advantage. As much as I respect and admire the Queen of Crime’s work, she wasn’t really one for firsts. Her genius was for taking elements of what already already existed in the genre and synthesising it with clever character work to produce compelling, readable plots. Her huge and enduring success bears this out: it doesn’t matter if you’re the first if everyone likes you the best.

But the renewed focus on The Mysterious Affair at Styles during its anniversary made me wonder where it all began. I’ve come across so many different origin stories for detective fiction over the years that I’ve been reading it and this centenary finally made me want to investigate further. What was the first whodunnit? Let’s find out.


Welcome to Shedunnit. I’m Caroline Crampton.


You probably already think you know what the first whodunnit is. I certainly thought I did. Keep that title in the back of your mind, but don’t stop listening, because this is a much more complicated, and interesting, question than it first seems.

When you start any investigation, it’s always a good idea to know what it is you’re looking for and why you’re looking for it. So before we go delving through the shelves, let’s take a moment to think about our quarry and our motivation for chasing it.

I used the word “whodunnit” deliberately, because I didn’t want to specify that we are looking for either a full length novel or a short story or anything in between. I think “whodunnit” can refer to a work of fiction of any length that concerns the solving of a mystery, often but not exclusively a murder. And it’s the solving that really matters, because although plenty of other genres include works about deaths and crimes, centring the narrative around working out the “who” is really the core tenet of detective fiction. Everything else is immaterial compared to that: form, style, setting, period. All of it. The first whodunnit can be set in Ancient Egypt or in deep space, I don’t care. As long as there’s someone detecting who did a crime, it counts.

As for why I really want to know about the first whodunnit, I think there are two reasons for that. Firstly, it’s because of our culture’s general reverence for round numbers. Now that it’s been a century since the beginning of the golden age of detective fiction, we’re entering a period of anniversaries and reappraisals. A hundred years is a nice hook, so I think we can expect lots more coverage of the type that we’ve seen for the centenary of The Mysterious Affair at Styles. Against all of that, I’d quite like to know where it all really began. And then my second motivation for this is because of how self referential detective fiction is, especially in the early to mid twentieth century. I’ve talked before on the show about how mystery writers love to incorporate real life cases in their books, and they also really enjoy referencing each other. Agatha Christie once wrote a whole series of short stories where she had her sleuths Tommy and Tuppence impersonate detectives created by other writers. This general propensity for borrowing and collaboration on the page shows how much of a cohesive tradition detective fiction is. And to understand a tradition, I think you need to know where it started.


It’s 1886. A young British doctor has written a strange, bumpy novella about a deductive genius, a character partly based on one of his professors at medical school. He struggles to find a publisher for it, and eventually signs a terrible deal with a company that gives him £25 in exchange for all the rights, meaning that he can’t ever make any more money from it no matter how many copies it sells. But at least this company does find a home for the story, and in November of the following year A Study in Scarlet is serialised in Beeton’s Christmas Annual, a publication run by food writer Mrs Beeton’s husband. The magazine doesn’t have a huge readership but the work receives some nice reviews, enough that it is republished as a standalone novel in 1888. The doctor will go on to write dozens more stories about this same character, building on the foundation he laid in this initial tale.

Also in 1886, on the other side of the world, a young legal clerk working in Melbourne, Australia is feeling frustrated because no theatrical producers would even look at the plays he was writing, let alone produce them for the stage. Feeling desperate, he asks a bookseller what kind of book he sells the most of. He learns that the detective novels of the French writer Emile Gaboriau are what’s flying off the shelves at the moment, so he resolves to write something in the same vein. The resulting book is turned down by every publisher who sees it, but when he publishes it himself in October of 1886, 5000 copies sell in the first three weeks. By the end of the year, it’s estimated that every literate adult in Melbourne has read it.

That young doctor, of course, was Arthur Conan Doyle and his deductive genius is Sherlock Holmes, a character who was to become so popular that all subsequent fictional detectives have arguably had to define themselves against him as the archetypal sleuth. But that first story, A Study in Scarlet, isn’t as convenient an origin point for the whodunnit as we might want it to be: it wasn’t very successful upon publication, and the entire second half of it is about Mormonism in Utah in the 1840s. If you haven’t slogged your way through the book itself you might not be aware of that – most subsequent adaptations rather conveniently leave that part out.

The second writer I mentioned there is more intriguing, though. That thwarted playwright was Fergus Hume, and his book was called The Mystery of a Hansom Cab. It has a fairly recognisable plot and structure, without any Mormonism tangents. A corpse is found in a cab in the early hours of the morning and a police investigator called Detective Gorby is given the task of identifying the victim and tracking down the murderer. Gorby has something of Holmes’s quick habits of deduction, although this facet of his character isn’t quite so codified and emphasised. The story is also interesting for the glimpse it provides of poverty in late nineteenth century Melbourne, with Hume noting in a later edition that much of this side of the book came from his own observations.

Above all, though, what makes me interested in The Mystery of a Hansom Cab in relation to our search for the first whodunnit  is its runaway popularity. After the book sold tens of thousands of copies in Australia, a consortium of English investors bought the copyright from Hume so it could be published in the UK. He made the colossal error of accepting only £50 for the full rights, meaning that when the book then sold 200,000 copies in its first six months of availability in London, he didn’t make any extra money.

There is an argument, I think, for saying that The Mystery of a Hansom Cab is the first whodunnit to become a global bestseller, and that is significant. When something is so ubiquitous that dozens of parodies and rip offs are published and it’s selling 3000 new copies a day, it’s clear that it has had an impact. Arthur Conan Doyle himself was aware of the book, but didn’t like it at all — in a letter to his mother just before his own first story was published he described it as “one of the weakest tales I have ever read”. While it’s certainly not the worst nineteenth century detective novel I’ve read, I would agree that The Mystery of a Hansom Cab isn’t a literary triumph.  But so many people read it that I think it should forever have a place in our understanding of how the whodunnit developed. Hume certainly never equalled its impact. He moved from Australia to Britain and wrote another 130 books, but none of them took off in quite the same way. His isn’t the first whodunnit, by any means, but it is a waypoint on our journey to it. Sometimes, to make your mark, you don’t need to be first. You just need to be really, really popular.


Stepping further back in time, there are some other good suspects for our investigation to be found in the 1860s and 1870s. The big hitter of these decades is The Moonstone by Wilkie Collins, which you will find mentioned as “the first detective story” all over the place. The poet TS Eliot felt strongly that this was the foundation stone — he called it “the first, the longest, and the best of modern English detective novels” and said that Collins had invented the genre.

First published in 1868, The Moonstone is a book that contains multitudes. It concerns the looting of the famous diamond of the title from an Indian shrine and then its subsequent theft from the person of the young woman who has inherited it from her corrupt uncle. There are protofeminist themes, elements of the Gothic and a discussion of class within the plot, as well as its detective elements. Collins based Sergeant Cuff, the competent and efficient Scotland Yard man who takes the case, on the real life Inspector Whicher who worked on the infamous Constance Kent murder case in 1860. There’s an amateur sleuth on the case too, and the two approaches are constantly compared and contrasted. It has a lot of recognisable aspects of a whodunnit, from the physical reconstruction of the crime to the red herrings Collins strews throughout the plot.

The Moonstone is definitely a detective story, but was it first? I don’t think so — I think Collins was following on from several other significant whodunnits from the 1860s and 50s. L’Affaire Lerouge by Emile Gaboriau is one such. It was first published in French in 1865, and was a big success — this was one of the books that Fergus Hume was trying to emulate with The Mystery of a Hansom Cab. It, too, follows a police investigation into a murder and introduces an amateur sleuth alongside the official investigation. Gaboriau’s focus on building the case using evidence through the book feels significant, too, as a move away from the sudden, unsupported revelations prevalent in the sensation or supernatural fiction of the time. Similarly, we might look to Lady Audley’s Secret by Mary Elizabeth Braddon, first published in 1862, which also features elements of the Constance Kent case as well as an amateur sleuth. And then there’s Charles Dickens’s Bleak House from 1852, which I think can also be read as a detective story, with Inspector Bucket gradually picking his way through a complicated mass of chancery and secrecy to solve a murder.

And what of Edgar Allen Poe, who I think is most often credited with writing the first whodunnit? I bet that’s the name that first came to mind for lots of you when you heard the theme of this episode. Well, I think there’s a very strong case to be made. His story “The Murders in the Rue Morgue” from 1841 features a lot of the tropes that some of the later nineteenth century works like The Moonstone, The Mystery of the Hansom Cab and A Study in Scarlet that I’ve mentioned would go on to develop more fully. His detective, C Auguste Dupin, espouses a method of deduction called ratiocination, broadly understood to mean reasoning based on logic and observation. The story is narrated by his friend, inhabiting the role we would now, post Conan Doyle, call that of the “Watson”. The reader is mislead, just enough, that Dupin’s solution is a surprise despite his friend’s seeming transparency. It’s all very familiar — indeed, it has been said that “The Murders in the Rue Morgue” constitutes in itself almost a complete manual of detective theory and practice”. It’s the obvious answer. Case closed.

But I’m not completely satisfied. Was there really no whodunnit before 1841? Did Edgar Allen Poe really invent a whole genre with a few short stories and then go back to the spooky stories and verse that made up the bulk of his output? I must admit, I was hoping for a more satisfying origin point.

After the break: It turns out, we are following in the footsteps of Dorothy L. Sayers

Ad music

Welcome to the intermission, the brief break in the episode where I tell you about a way that you can support the continued existence of the podcast. Sometimes that’s by listening to an advert and maybe clicking on the sponsor’s link (just clicking makes a big difference, by the way, even if you don’t want to buy anything, because it demonstrates that you heard and understood what I said). Today, I’d really love it if you’d pause this episode right after I finish this sentence, and spend ten seconds doing two quick things. Firstly, check if you’re subscribed in your app, and if you aren’t, hit that button. By doing that, you’re telling your platform that you like the show and want to be notified every time I publish a new episode. Then, pull up a text or a new social media post and recommend the show to a friend who you think will like it. Old fashioned word of mouth is still the absolute best way to introduce new people to podcasts, and without the backing of a big publisher or company I rely solely on you, listeners, to help me get the word out about Shedunnit. So, off you go. And now that you’ve done that, let’s get back to our search for the first whodunnit.

Ad music

It was when I was investigating the claim of Edgar Allen Poe to be the author of the first whodunnit that I found that Dorothy L. Sayers had already walked down this path ahead of me. In her introduction to a 1929 anthology of mystery stories called The Omnibus of Crime, she had also tried to establish what the earliest example of this genre was.

In doing so, she separates out three different branches within the tradition that make it easier to assess what belongs and what doesn’t. These are: detection, mystery, and horror. The first is self explanatory — the story must include some element of detection. Books like The Moonstone and Bleak House easily pass this test because of the presence of professional police detectives, but you might say that Lady Audley’s Secret doesn’t because we have a character trying to find things out who has no prior history as a sleuth. Debatable, that one, but it’s useful to think about. The detective is, Sayers says, “the latest of the popular heroes, the true successor of Roland and Lancelot”. I really like this allusion to the heroes of medieval romances and Arthurian legends, with the detective the modern equivalent of the knight on a quest. It places the searching, inquiring requirement of the mystery genre within a larger and much longer literary tradition.

The second of Sayers’ three branches is mystery. This also seems obvious once you say it out loud: of course the first whodunnit must be a mystery. There must be a problem to solve, a knowledge gap that must be closed. It can’t be immediately obvious what the outcome is going to be. Well, yes.

The third is more interesting: horror. There’s a lot of cross over in the early nineteenth century between Gothic fiction, sensation fiction and proto detective fiction, with horrifying events being woven into a narrative of investigation. The work of Edgar Allen Poe is a notable example of this. The vast majority of his output was firmly within the Gothic, exemplified by stories like “The Masque of the Red Death” and “The Fall of the House of Usher”. But in addition to the Dupin stories I’ve already mentioned, there are others that seem to infuse their horror with elements of mystery, such as “The Imp of the Perverse”, in which a narrator boasts of having got away with an undetectable crime.

Standing alongside Poe in this regard is the Irish writer Sheridan Le Fanu, the master of sensation fiction. Sayers was very keen on him herself, and gave her obsession to a character in her own whodunnits, Harriet Vane. In Gaudy Night, Vane spends a great deal of time on researching a possible biography of him, and in the Sayers continuations written by Jill Paton Walsh, she has become quite a noted expert on his work.

Le Fanu very much straddled the boundary between horror and mystery. Many of his stories are full of demons, vampires, ghosts and hellfire, but he also wrote an early locked room mystery called Uncle Silas, which shares many features with the work of Wilkie Collins. More than twenty five years before The Moonstone, Le Fanu published a short story called “A Passage in the Secret History of an Irish Countess“,  in which a body is found in a locked room with no sign of how the murderer can have escaped. But there’s no detective figure and no scheme of deduction at all, so as much as I enjoy Le Fanu’s work, we will have to look elsewhere for the first whodunnit.


There is a common theory that the reason that detective fiction didn’t flourish until the late nineteenth century is because society needed an established system of real life detectives for the public to appreciate them in fiction. It is certainly the case that official state police detectives didn’t come into being in places like Britain and France until the mid nineteenth century, but I’m less and less convinced that this is a good argument for Edgar Allen Poe or Wilkie Collins as the author of the first whodunnit. Investigators existed prior to the official police ones, and anyway the genre abounds with examples of canny amateurs who take up the mantle of sleuth when circumstances demand it.

If you follow this line of reasoning, then suddenly a much longer lineage exists for the whodunnit. And in this vein, Sayers included in The Omnibus of Crime four stories which she terms “the primitives”. There’s the tale of Hercules and Cacus, which is repeated in one version in Virgil’s Aeneid. It’s part of one of the twelve labours of the mythical hero Hercules — later retold by Agatha Christie as cases for her sleuth Hercule Poirot. A fire breathing giant called Cacus is terrorising the area around the river Tiber and steals some of the cattle that Hercules has just won. In order to disguise the crime, Cacus pulls the cows backwards by their tails to his cave lair so that the hero won’t be able to follow their footprints and recover them. Keen Sherlockians might notice that Conan Doyle borrowed this plot for one of his short stories. Anyway, this legend displays a very early recorded example of an author including the fabrication of false clues that the detective hero must then overcome to find the true solution to the mystery. Also in the anthology are two biblical stories that similarly demonstrate unmistakeable aspects of the whodunnit: the history of Bel and the history of Susanna. The former sees the biblical hero Daniel analysing ash as evidence to determine who committed a crime, and the latter interrogating the testimony of witnesses to arrive at the real truth. The earliest story in this volume is from the Greek author Herodotus, and it is thought to date from around 600 BCE. Not only is the tale of Rhampsinitus and the thief an undoubted mystery, it’s also a locked room mystery. Sayers declared it to be an example of “the psychological method of detection: plot and counter plot”.

Another favourite suggestion of mine is one of Aesop’s fables, “The Fox and the Sick Lion”, in which an elderly predator pretends to be too sick to hunt in order to lure sympathetic animals into his cave so he can eat them. The fox, though, is too smart to fall for this ruse because he notices that the footprints of others who have come to the lion’s aid lead only into the cave, not out, and thus deduces that the lion is not as harmless as he seems. “Sherlock Holmes could not have reasoned it more lucidly,” Sayers says.

But we can go still earlier than Aesop, Herodotus, or the Bible. The Greek tragedy Oedipus Rex by Sophocles is thought to have been first performed in the fifth century BCE, and sees the titular king trying to solve the murder of his predecessor in order to appease the gods and stop a plague. Step by step, this king/detective follows the clues and analyses the evidence until he can eventual reveal whodunnit in an explosive conclusion worthy of Agatha Christie herself. Aristotle was a fan: he wrote in the Poetics that “of all recognitions, the best is that which arises from the incidents themselves, where the startling discovery is made by natural means. Such is that in the Oedipus of Sophocles”. I read that as an argument for fair play in the detective story: like the writers of the golden age in the early twentieth century, Aristotle wants the clues to be laid out for all to see so that the audience has a chance of reaching the conclusion for themselves. This play also belongs to the minority of whodunnits that not only portray the solution to the mystery but also the justice dispensed afterwards — in a way, Oedipus belongs to the same tradition as G.K. Chesterton’s Father Brown, a sleuth who sees not only clues but sin and its consequence.


I think every fan of detective fiction, or indeed of genre fiction more broadly, has been told at least once that the books they love are not “proper” literature. There’s long been this persistent assumption that if something is enjoyable to read and popular with a lot of people, it must somehow be lesser. But by tracing the whodunnit’s lineage all the way back to Sophocles, we see just how ridiculous this is. Detection is a fundamental part of the way stories are told. It’s a basic building block of narrative.

To enjoy the first whodunnit the way its writer intended then, we’ll just need to brush up on our Ancient Greek.


This episode of Shedunnit was written and narrated by me, Caroline Crampton, and edited by Euan MacAleece. You can more information about this episode and links to all the books mentioned at I provide transcripts for every episode of the podcast too: find them at

Thanks for listening. I’ll be back soon with another episode.