Renga in Blue   Interactive fiction and the All the Adventures project

It is unknown who actually was the first to put words crossing in a grid, but the earliest example we have is from a piece of 2nd century wall found at a Roman house in Cirencester.

The same words are spelled across as down.

ROTAS
OPERA
TENET
AREPO
SATOR

Word squares became a popular pastime in 19th century England, and eventually, clues started to be made in order to generate them. For example, these clues from Guess Me (1872) make a 5 by 5 square:

1. To watch over
2. Below there
3. A fair lady's name
4. A memorial of the feast
5. A severe lawgiver

Other shapes, like a diamond or a triangle started to be used, but it wasn’t until Arthur Wynn’s Word-Cross for 1913 did “across” and “down” get turned into separate clues, with arbitrary shapes. This eventually kicked off a crossword craze that lasted all the way through the 1920s on both sides of the Atlantic.

Our protagonist for today’s story, Edward Powys Mathers, was on the British side. He originally distinguished himself as a translator and poet, but also was a prolific writer of puzzles under the pen name Torquemada (he of the Spanish Inquisition). There was, for example, the 1925 volume Cross-Words in Rhyme for Those of Riper Years.

We’ll try a sample from here in a moment.
As the title indicates, all clues are in rhyme, and already are much more elaborate than the “American” style crossword.

In 1926 started publishing crosswords with The Observer, and importantly, continued developing what is essentially his invention, that which is sometimes in the United States called the “cryptic crossword”. Unlike regular crosswords, the clues in a cryptic involve a double-indicator, one which is a “straightforward definition”, one which is generally wordplay, and the two are merged together in such a way it can be difficult to discern which is which.

I say “essentially” because it’s important to note that the puzzle type was being developed before “fairness rules” were established; Torquemada was followed by Ximenes (aka Derrick Somerset Macnutt) who quite clearly articulated that there should be a definition, a secondary wordplay, and absolutely no extra words, but that was not the case with all of Torquemada’s puzzles.

For example, try this clue, 3 letters:

The sluggard’s pet abomination
Got all mixed up when dealing with this one

Don’t try too hard; I honestly don’t understand the answer even after reading it. Still, let’s give a period postcard for some spoiler space.

From a postcard set in a 1930 catalog.

The “got all mixed up” seems to be an anagram signal, but as opposed to the letters being there, you have to figure out what the pet is first:

NTA, Anagram of ant, of whom the sluggard must be heartily sick.

What does NTA stand for? I’ll assume that’s something that made sense in the 1920s, but given how mysterious some of the other clues are, maybe not?

All this is important for the work in question today, that of the Torquemada Puzzle Book published in 1934.

Source.

By this point, Torquemada was a well-established puzzle writer, and embedded in the volume of EXCLUSIVELY UNPUBLISHED MATERIAL there was a 100-page mystery novel by the name of Cain’s Jawbone. It is a story over the span of six months where there are six murderers and six victims. There is the slight catch that a.) the identities of the victims are unknown b.) the identities of the murderers are unknown and c.) all 100 pages are out of order. The goal of the puzzle is to reconstruct all three.

A 2019 republication of Cain’s Jawbone as a standalone novel — with an accompanying contest which is still active until the end of 2022 — recently went viral; well, recent-ish, by Internet time. (I can say I was cool enough to know about the book before it went famous. I did not predict it would be the source of Youtube videos with nearly half a million views.)

My skill at cryptic crosswords is only middling-to-fair, so I’ve up to now resisted an urge to take a crack at the puzzle, but I took a gander through the text recently and it’s pretty good even just read as experimental literature. Edward Powys Mathers had genuine literature chops.

The sound of the bell, as of a boding gnat, just came to me. The finger causing it was, I knew, the index of a most skilful hand, one I had commanded, one that would pluck me from embarrassment, and yet one I vaguely distrusted. Really, if the lower orders don’t set us a good example, what on earth is the use of them? They seem, as a class, to have absolutely no sense of responsibility…. One had to be in the key for such things.

The impression is a tactile mix of a well-characterized narrator, social commentary, and what likely are some gnarly puzzles mixed in. A good cryptic crossword clue often has surface meaning indistinguishable from normal text. This clearly isn’t … normal text … but yet somehow fits directly into what I’d expect from experimental modernists in the 1930s.

No doubt — especially given the book cover indicates such — there is hidden wordplay that is part of the puzzle. But there’s also still a story here, and a narrative logic, and if I’m going to take a crack at solving Cain’s Jawbone the first thing I want to try to do is sequence by raw intuition.

I’ll be tackling more specifics next time, but I do want to linger a bit more on the narrator character. The narrator himself (herself?) may be one of the murderers.

In one way, of course, I was glad they were married. I had always been rather a stickler for purity in family life. That scandalous rumour of a Maltese landing on our island and seducing an ancestress of mine—or was it an ancestor?—from the path of duty, I never had and never would believe. If I had had a real education, instead of just listening to him, I could have told—it was bad that day—how I detested being called Hal. It was she that did it. But he was pleased in a way, and said to her, out of a book, the original ground of the transaction appears to have been sentimental : “He was my friend,” says the murderous doctor ; “he was dear to me.” Some Tom, not the one I killed in the matter of Jasmine, had done that, I gathered. He was enthusiastic and provided a chop for me, and said it was good he had visited England to-day for the first time.

Is “the one I killed in the matter of Jasmine” referring to “the murderous doctor” or is it referring to the narrator?

From what I’ve sampled so far, I also get the sense that chronology does not flow exactly (the setup material for book even indicates that the narrator’s mind will sometimes flit about “in the modern way”) and it’s certainly possible not everything said is reliable, not even accounting for some text probably being there for literal wordplay rather than indicating real events in the world universe.

I don’t expect to get a solution — nobody seems to be, except for one person — but I would like to get some insight on the puzzle design in general, and especially just what the frame is. Is there a specific way each name gets encoded? Is there a strong indicator across every page that allows checking the page sequence? I was going to commit to writing four posts, and like my writeup on Alkemstone, I’m hoping to at least get a fair notion of the central puzzle, even if I don’t know enough 1930s British references to make much headway. If nothing else, John Finnemore (the solver I just linked to) said “it’s a well-designed puzzle so when you start to make some headway, you know it” meaning despite the obtuseness, it’s probably at the very least a tighter design than Alkemstone.