The Floating Admiral is one of the best-known works of detective fiction.  As it states on the cover––and the spine––of the first edition it was written by ‘Certain members of the Detection Club’, those certain members being G. K. Chesterton; Dorothy L. Sayers; Canon Victor L. Whitechurch; G. D. H. & M. Cole; Henry Wade; Agatha Christie; John Rhode; Milward Kennedy; Ronald A. Knox; Freeman Wills Crofts; Edgar Jepson; Clemence Dane; and Anthony Berkeley.  The book was published by Hodder and Stoughton in 1931.

The premise behind The Floating Admiral is an interesting one.  Each one of the notable authors was to write a chapter in turn, before handing on the baton of the story to the next writer.  It was hoped that in this way, and with no one author knowing what their fellow writers envisaged regarding the direction of the plot or the ultimate conclusion, the story would more closely resemble the stages of a real police investigation.

And did it succeed in its aims?

Personally, I think Anthony Berkeley, who wrote the concluding chapter of the book might have experienced misgivings about the experiment not dissimilar to my own as a reader, when he called his chapter: “Clearing up the mess.”  For, in truth, The Floating Admiral is a bit of a mess.  The tone from one chapter to the next is understandably a bit uneven, and this must have surely been even more apparent at the time of publication, when it couldn’t have been read under the historically unifying umbrella of ‘Golden Age’.  Some authors’ commitment to the project seem more enthusiastic than others: Dorothy L. Sayers contributed a 6-page introduction, a solid 40-page chapter and an earnest 22-page elucidation of her conclusions; whereas Agatha Christie wrote a rather light 10-page chapter and just two short pages by way of conclusion in the book’s appendix.  The book is revealing, though, as a testament of each author’s ultimate longevity: the rather turgid chapter by Milward Kennedy; the worthy, but dull chapter by Ronald A. Knox; and the rather flighty piece of whimsy by Clemence Dane perhaps revealing why they no longer remain such household names as the likes of Agatha Christie, Anthony Berkeley, and Dorothy L. Sayers.

Ultimately, I think what I find most disappointing about The Floating Admiral is that it contains, what for me is, the greatest crime a writer of detective fiction can make: it provides a non-unique exposition.  Perhaps I should explain what I mean by that.  I want to get to the end of a detective novel and be presented with a conclusion that is logical, unique and incontrovertible based on the evidence that has gone before it not, as is the case with The Floating Admiral, a conclusion that plausibly explains the facts of the case, but where any number of other explanations could equally do the same, such that it feels as though the author has simply pulled a random rabbit out of a hat at the last minute––although, to give him his due, it is amazing that Anthony Berkeley managed to come up with any kind of exposition given the mess of evidence that had come before.

My own views are rather echoed by one of the authors.  In her own conclusion, Clemence Dane is honest enough to admit about one aspect of the plot that: “It’s quite inexplicable to me.” (p. 346) and that: “I am, frankly, in a complete muddle as to what has happened.” (p. 347).

For me, The Floating Admiral is an interesting experiment, but one that does not work.  I remain similarly unsure of a lot of so-called ‘writing by committee’.  There has been a lot of speculation both on this blog site and in the wider media about the effect of AI and specifically ChatGPT on creative writing.  ChatGPT is the ultimate aggregator of words; the largest ‘writer by committee’.  It will be interesting to see if it can produce a convincing murder mystery to rival the best works of the genre.

© Fergus Longfellow

Fergus Longfellow is not a team player.