Occasionally I wonder how valuable computerized analysis of works of literature will ever be. To reassure me that I’m not wasting my time, Jan Christoph Meister tells me1 what Johann Wolfgang von Goethe thought:
Goethe discusses the pros and cons of morphology as a science, and eventually concludes: “Its arrangement of phenomena calls upon activities of the mind so in harmony with human nature, and so pleasant, that even failures may prove both useful and charming.”
The citation is to a translation of Goethe’s collected works. 2 If we want the old man to be saying something about the study of literature, we’re making an analogy, because Goethe is talking about morphology as an essential predecessor to what we now call “biology” (because he was right). Literally, he’s talking about the study of form without regard to function, which is all my programs can do with poetry. I like even better the sentences just before Meister’s quotation.
The advantages of morphology are that it is made up of widely recognized elements, it does not conflict with any theory, it does not need to displace something else to make room for itself, and it deals with extremely significant phenomena.
That sounds like a feasible goal for a small project like mine. This connection across 220 years shows that once again it’s useful to be, as the philosopher Adam Ant phrased it, “an eighteenth-century brain, in a twenty-first-century head.”
Daniel Stride, kiwi clarissimus, points us to a YouTuber who does a not-bad job of speculating on what JRR Tolkien would have thought about generative AI. A useful application of AI would be to turn that 20-minute talking-head video into a blog post I could read in five. A really useful application of AI would be to take a video of a conference presentation and turn it into a written document of what I wished I’d said. But I digress. What was I talking about? Oh, yes.
“Girl Next Gondor” thinks that JRRT would have seen Large Language Models as a vindication of his ideas about language in “On Fairy-stories”: that we can abstract the word “green” from grass and “sun” from the sky and conceive of a “green sun”, which is the fundamental act of fantasy. That’s pretty close to what LLMs do. When the temperature is low, they only connect green with green things, but if you turn the temperature higher the model will connect adjectives with a much more diverse set of nouns.
Daniel kind of agrees, but notes that of the various kinds of magic, LLMs do not engage in the good kind. They don’t produce enchantment, because there is no enchanter. The person thinking of the fantastical situation is writing a prompt of a few dozen words, not, say, a novel. The resulting block of text is not really a work of art. LLMs produce a kind of mindless, inescapable magic in which we blunder around. We are not enchanted; maybe the word “emprompted” could be coined and pressed into service.
I’m less optimistic. Tolkien saw this coming — at his “Hobbit Dinner” in Amsterdam in 1958 he said, “the Age of Paper is ending; the Age of the Gadget begins.” Looking around the world, he said “… I see that Saruman has many descendants.” LLMs are definitely the work of one of those. As is generally the case, an LLM “cannot make, not real new things of its own.” [LR 6.01.109] It just twists existing language to whatever purpose it’s given. But we know what a maker of twisted language is in the Legendarium: that’s the essence of dragons. Like Glaurung, an LLM assembles words to achieve its goal without regard for truth. The goal for an LLM is maximizing likelihood conditional on the prompt, where the goal of the dragon was ruining Hurin’s life, but the effect can be the same.
Why would something as innocuous as maximizing a function turn out as evil as a dragon? In Letter 153 to Peter Hastings, Tolkien was talking about military contractors developing weapons, but the words he used sound painfully applicable to the LLM-mania of our current crop of tech oligarchs. They may not be intrinsically evil, but “things being as they are, and the nature and motives of the economic masters who provide all the means for their work being as they are, are pretty certain to serve evil ends.”
I just re-read That Hideous Strength by C. S. Lewis for the first time since I started studying the Inklings. It’s a completely different experience, now that I know the subtexts of what he’s talking about. Everyone gets the “Numinor” references, but for someone who’s read Owen Barfield and Charles Williams, the book turns out to be stuffed with easter eggs. 3 This line jumped out at me particularly: “… they had all, by various routes, come too far either to consider [Merlin’s] art mere legend and imposture, or to equate it exactly with what the Renaissance called Magic.” (p.200) It reminded me of Galadriel’s line, “this is what your folk would call magic. I believe; though I do not understand clearly what they mean; and they seem also to use the same word of the deceits of the Enemy.” (LR2.07.080) That change in the meaning of the word “magic” is intriguing. Galadriel comes from an earlier Age than the hobbits. She’s from the Elder Days, whereas the Shire is post-medieval.4 Hobbits think and talk like we do. Therefore, they use the word “magic” in a way she doesn’t get. Galadriel’s line emphasizes her distance from them, as if it needed more emphasis.
Not here! by Olena Panasovska
Galadriel has a Ring that she uses “to preserve all things unstained”. That’s Elrond’s phrasing (LR2.02.240), but “on the land of Lorien there was no stain,” (LR2.06.160) so Galadriel and Elrond are of one mind on this. What kind of stain are they talking about? Lewis’s line gave me a new way to read that: the stain is Renaissance rationalism, the kind that gave us the twin children Science and Magic. Galadriel is preserving a pre-modern concept of Nature that doesn’t have a place for Early Modern concepts like those. She is using her Ring to make sure that there will be no Renaissance while she has the power to prevent it. She’s not alone: Elrond is called a “lore-master” by Denethor (who should know). A lore-master is a worthwhile person to have around, but he doesn’t create new knowledge like a scientist does. Rivendell has libraries, but I’d be surprised if it had a laboratory. And Gandalf, who held the third Ring, delivered a memorable warning against foolish scientism.5
I have learned, whenever I get an idea for one of these bloviations, to check what Tom Shippey has to say on the topic. (He’s always there before me.) He extracts from the Inklings’ writings an opposition of magia, goeteia, scientism, and religion.6That Hideous Strength is about goeteia and scientism ganging up on religion, which defends itself successfully when it’s reinforced by Merlin’s magia. Shippey identifies a common theme among the Inklings, lamenting the loss of authority by religion as the new knowledge. That may be the stain Galadriel is talking about.
But why should she, in a world without religion as we think of it, care? Religion to her is completely irrelevant — she’s actually met most of the people a religion would be aimed at worshiping, though until Frodo’s visit, they’re not taking her calls. Why is she afraid of a Renaissance? Well, that one has already been figured out. Tolkien thought that the Renaissance (in its guise as the Age of Discovery) had put an end to Faërie by sending explorers around the globe and finding just more land and more people. He suggested that shrinking to Shakespearean proportions was their defense. Perhaps Galadriel was willing to diminish to an extent, but not all the way to flitting around an English garden on moth wings. And then, of course, will come the Enlightenment, and fairies are doomed.
Poetry assistant and archaeologist Martin Rundqvist has published a follow-up to his last November’s post about how the standard software for carbon-dating gives its answers in an idiosyncratic way. In a nutshell, the answer to “how old is this sample” comes out with confidence intervals of 68.2% and 95.4%.7 These intervals aren’t necessarily things that make sense — they could be totally disjunct periods. In MR’s case the interval is the union of 776-782 CE, 879-994 CE, and 1007-1012 CE. What use is that to anyone?
What happened in 776?
His latest post announces a solution to the biggest problem with that. In Bayesian terms, the person requesting the test has important prior information that shouldn’t be neglected when the results are computed. This problem has been solved by a software tweak so the scientist can put in a historical period and get out the probability that the sample came from that period. That’s how statistics ought to be done. Bravo, y’all!
One other thing about that figure got me interested, though. That wiggly gray line up top is the correction the software has to make for the background atmospheric concentration of carbon -14. Different years have different amounts of C-14 to start with, so the decay over time begins at a different place. There’s a big spike in the late eighth century. (The spike points downward because this is a correction factor.) It was strong enough to make the test conclude (against prior knowledge) that there was a chance the sample could have come from then. What’s that all about? It’s from outer space! In 2012, Fusa Miyake and her team measured the C-14 content of each individual ring in some old trees and calculated the year-by-year changes. People are still arguing over what the exact cause was: Solar flare? A passing comet? I’d put my bet on a coronal mass ejection, for what that’s worth. A feature film about radioactive Vikings is already in the script-development stage, I have no doubt.
Once again, I find myself thinking back to the days of my graduate study, and my self-pitying wails over the amount of work I had to do. Ha! All I did was sit at a desk. I didn’t have to peel ancient tree trunks, ring by ring, and if my razor blade slipped a millimeter the sample was ruined. Theoretical physicists have it easy.
I just re-read Protector by Larry Niven in the Ballantine paperback edition.8 It’s a great example of classic science fiction. The reader is barraged with one cool idea after another. Just the thing for a nerdy teen-aged boy.
In one scene, set on the artificial moon Kobold,9 Brennan has constructed a Moebius strip that’s as wide as a sidewalk. Brennan has technology that can control gravity, which he has used to make it possible to walk on the strip and have gravity always pulling you towards the surface. A character named Alice gives it a try, and walks a lap around the strip. After that, on the bottom of Page 155, she’s referred to as “Sally”. In the next scene, she and Roy climb the stairs in a 3-D copy of M.C. Escher’s “Relativity”, after which she is called Alice again.
This put me on high alert. “Alice” and “Sally” are the same sounds, but the consonants are flipped from places 2 and 4 to places 3 and 1. Hypothesis: Niven is playing a word-game here, where a trip around an non-orientable surface scrambled her name, then another mind-bending walk straightened it out. Lewis Carroll loved playing pranks like that on his Alice.
Sadly, no. She’s “Sally” again once on page 166. And now that I’ve noticed that, the constellation “Saggitarius” is mentioned several times, and M.C. Escher is spelled “Esher”. It’s just a bad editing job. Oh, well. I guess close reading isn’t a universally-useful technique.
Renowned Inklings scholar and Causer of Things to Happen Sørina Higgins is organizing a conference on Jan 31 & Feb 1 of next year, called “Fahrenheit 2451”, dedicated to the idea that the human race has thought of some ideas that we ought to save, no matter what disasters may befall. Which ones would you pick?
It’s in San Francisco at the Internet Archive. There are still two weeks left to get your ideas in.
In Appendix A of The Lord of the Rings we learn that in his youth Aragorn journeyed through Rohan and Gondor and won fame under the name of Thorongil. (LR A.1.iv.63-67.) He made quite an impression on the people of those countries at the time, but a few decades later things are different. We don’t hear of any old men in Rohan who recognize him, even though he rode with Thengel, father of Theoden. Denethor apparently recognizes Aragorn through the image in his palantir, so his appearance can’t have changed too much. But then Prince Imrahil, who is of similar age, says “Shall we not now send for the Lord Aragorn?” when Aragorn is standing a few feet away. (LR 5.08.041) Imrahil doesn’t notice him until he speaks.
But here’s the thing – Aragorn is “at least 6 ft. 6.”. 10 Even great warriors like Boromir aren’t that tall. 11 How is a bean-pole like Aragorn not immediately recognized by everyone over the age of 50? His reception in Edoras should have been less laden with suspicion, and more like, “Eala, Thorongil! Long time no see! How’s the weather up there?”
Thought 2:
The odd shrinkage of Elves. Medieval fairies could be small, or very large, or human-sized. “Taking the broadest known parameters, we find that size can range from fourteen feet high to a being small enough to sit on a cowslip.” 12 Victorian fairies are uniformly tiny, though. Tolkien wondered about this in “On Fairy-stories”. 13
As for diminutive size: I do not deny that the notion is a leading one in modern use. I have often thought that it would be interesting to try to find out how that has come to be so; but my knowledge is not sufficient for a certain answer.
The Professor goes on to speculate that elves started to shrink during the Age of Discovery, when Europeans sailed all around the world and found no Faërie anywhere. In order for Elves to continue to be a thing people could believe in, they needed to be able to hide from view. It was advantageous for them to shrink. The process was exactly parallel to evolutionary pressure: just as squirrels are now the right size to fit through the holes in a chain-link fence, Elves met the challenge by shrinking until they could hide practically anywhere, even behind garden flowers.
Synthesis:
Evolution by natural selection is tempting, but it is fraught with difficulties when we’re dealing with immortal creatures. Fortunately, there’s an easier answer, provided by Tolkien himself. The way the cloaks of Lothlorien hide you from view must be to make you look short. Otherwise, there’s no way Aragorn could have maintained his anonymity while standing half a head above everyone around. Apparently it’s not just Galadriel and her maidens who can make those; it’s a common skill among Elves. Maybe it’s the only way they know to make clothes. It also explains the cloaks’ preternatural effectiveness in getting Sam and Frodo to Mount Doom — knock a foot off the perceived height of hobbits and they’re just barely macroscopic.
I’m not just making this up. Let’s jump over to “Smith of Wootton Major”.14 Alf’s dramatic revelation at the end:
’Would you spare a few moments for the King of Faery?’ the other answered. To Nokes’s dismay he grew taller as he spoke. He threw back his cloak.”
In 1939, Prof. Tolkien’s knowledge was not sufficient, but by 1967 he had figured it out.
Corey Olsen gave a talk at Mythmoot XI on the subject of Gimli’s song about the glory days of Moria. (LotR, II iv.) He began with the fact that the song’s rhythm is iambic tetrameter, steady as a metronome, and its rhyme pattern is a plain-vanilla aabb. This is kind of boring, so why is the song so interesting? Olsen’s answer is that there’s a counter-melody interwoven with the rhymed verse. When you look for stressed-syllable alliteration, a different verse-form emerges. Here are the last four lines. Look at the D’s and W’s:
But still the sunken stars appear
In dark and windless Mirrormere;
There lies his crown in water deep.
Till Durin wakes again from sleep.
Hypothesis: the te-tum-te-tum of the iambics is just the rhythm section, and there’s a melody going on in the alliteration that’s much more varied and interesting. Olsen thought of a hip-hop singer over top of a mechanical rhythm track; I’d have said a baroque concerto over a continuo. The alliterative patterns aren’t rigid or repetitive enough that it’s easy to say for certain that Tolkien created the counter-structure intentionally.
We can confirm or refute the hypothesis if we look at the full distribution of sounds in the poem. First thing I did was tweak the program I wrote for finding Old English alliterative long lines. Out with the Sievers types; now I just want to flag the initial sounds of stressed syllables. Start at the beginning of a text. Count how many stressed syllables there are before I reach the next repetition of its sound. Then do the same for the second stressed syllable. Repeat until we run out of text.
To see what this distribution ought to look like for different kinds of texts, let’s take some easy examples. First, the least-poetic thing in the world: Choose 50 random sentences from the Simple English Wikipedia. (Drop sentences with numbers in them.) I got this from Goldhahn, et al., 15 The histogram of intervals looks like this:
Fig. 1. Unpoetic intervals
This is a geometric distribution, more or less. If all stressed sounds were equally common in English, and people didn’t like the sound of alliteration, the distribution would be exactly geometric. But some letters are more common than others, so there are random peaks and valleys, and people like alliteration even when they’re not writing poems, so there’s a bit of a spike at zero. So far, so good. Let’s cut the graphs to hide the extreme outliers from now on — extremes tell us mostly about the length of the text, not so much about sounds.
Now let’s pin down the other end of the spectrum of possibility with the “Song of the Mounds of Mundberg”. Tolkien said in Letter 187 that it was “written in the strictest form of Anglo-Saxon alliterative verse.”16 Old English long lines have alliteration patterns of either axay or aaax, so we’d expect the histogram of distance through the text to be heavily weighted at 0 and 1. And so it is:
Fig. 2. Intervals in strict Old English form.
More than half of the time, an alliterating sound comes with the next or next-but-one syllable. This is a good benchmark for the low end of alliteration intervals.
Last, we’d like a statistic to compress the information. The median over all intervals in the poem is not sensitive to the extreme tails of the distribution, but catches how scrunched-up to the left everything is.
Comparing Poems and Prose
Let’s look at a variety of texts that span a range of unintentional and intentional alliteration. I am indebted to Paul Deane’s overview of alliterative verse for good examples of verse-forms I’m not so familiar with.
Wikipedia, as above. Tagged “wiki”.
“The Mounds of Mundburg”. Strict Old English, as above. Tagged “mundburg”.
Gimli’s Khazad-dum poem, where we started. Tagged “khazad”.
The section of prose right before Gimli’s poem, for comparison. Tagged “prose”.
“Far over the Misty Mountains cold” from The Hobbit. Thorin & Co. aren’t trying to alliterate, but you can’t avoid it when you’re talking about the Misty Mountains. This is what happens when Tolkien’s natural penchant for alliteration gets polished into verse. Tagged “mistymts”.
Legolas’s song about Nimrodel. Not trying to alliterate at all.
“Rosemary” by Andrew Frisardi. A modern adaptation of strict Old English verse form.
“Sea Change” by John Beaton. This is in a Middle English style of alliteration.
“Old English, New World“, by me. An attempt to reconcile modern Southeastern-US word order with Old English form. Tagged “shenandoah”.
“Do you hear me, Mere-Watcher?” by Laura Varnam. This is purely modern free verse, using alliteration to allude to Beowulf. 17 Tagged “merewatcher”.
The histograms in Fig. 3 give us a feel for what’s happening. “wiki” and “prose” seem to be the most spread out, as we’d expect.
Fig. 3. Interval histograms
Ranking the texts by median interval of sound similarity in Fig. 4, the prose texts are at the high end and the Old English are at the low end, as the metric was constructed to do.
This metric left the Old Norse patterns in the middle of the pack. “Valhalla” does have a strong presence of 0 and 1-syllable intervals, but it has a heavy presence of longer ones, too. Wrapping the alliteration around the end of a line18 moves it up the graph.
The most interesting thing about Fig. 4 is that it doesn’t have an obvious breakpoint between alliterative and non-alliterative verse. I was expecting a big gap, but the possible uses of alliteration form a continuum, visible even in this small sample.
Fig 4. Ranked text samples.
The histogram for a poem with a free-form alliterative “melody” should have a secondary peak in its histogram at some larger spacing than the 0-1 that come from strict alliteration. We see something like that in Varnam’s poem, which definitely has such a melody. It’s a short poem, only half the length of the other texts, so its histogram can have large fluctuations around the ideal form. “Sea Change” has two linked stanzas, the second of which occasionally harks back to the first with K and S sounds, which puts a small peak about 15.
Alas for Prof. Olsen’s idea, Gimli’s song has barely closer alliteration than prose texts do. Figure 3 almost has a secondary peak at 5, but it’s not significantly different from random fluctuation. Now, I’m not one to claim poetry should have statistical significance, but the prose section just before Gimli’s song has more of those peaks than the song does. If anything, this method of measuring suggests Tolkien removed alliteration to make the poem, as John Ruskin totally didn’t advise.
Technical Details
There are 28 sounds on which we can alliterate here, since we count all vowels as one class, C is either K or S, etc. I’ve added the sound “KH” to the set since we’re dealing with Dwarves. Using the CMU notation, the sounds are “B” “CH” “D” “DH” “E” “F” “G” “HH” “JH” “K” “KH” “L” “M” “N” “NG” “P” “R” “S” “SH” “SK” “ST” “T” “TH” “V” “W” “Y” “Z” “ZH”.
The CMU pronouncing dictionary thinks “Shenandoah” is a four-syllable word with the accent on the “do”. John Denver probably deserves the blame for this geographical lapse. I overrode the dictionary entry to restore the name to its true dactylic nature. (The Boss said so.)
There are quite a few 2’s in the “Mundberg” graph that shouldn’t really be there. That’s because the intervening word is one that sometimes has a stress and sometimes doesn’t, and my program isn’t smart enough to tell which is which. This happens in all the texts. The extreme case is in “Mere-Watcher”: The last line of this poem is “So fucking what?” Under the usual rules, “so” would be unstressed. It’s stressed here, so I jumped into the program’s output and fixed it by hand to give it the proper weight. I imagine Seamus Heaney had the same problem, once.
Tolkien’s Collected Poems contains several we’ve never seen before. One of them is a bit surprising. Number 36 is a half-finished piece called “The Empty Chapel”, written in 1915 while JRRT was in basic training before shipping out to the front. One stanza in particular jumped out at me from Page 226:
Lo, war is in your nostrils and your heart
And burning with just anger as of old
Though stunted in dark places far from God
Though cheated and deluded and oppressed
Arise you, O ye blind and dumb to war
Come open your eyes and glorify your God
Come sing a hymn of honour to your Queen.
When I read that, I thought that Tolkien was right there with Wilfred Owen among the World War 1 poets. I heard echoes of “Dulce et Decorum Est“, and could hear irony dripping from them. I don’t think I’ve ever heard that kind of voice from Tolkien before. Yes. About that. It’s important to read poems more than once, and this is why. On third reading, I see that Tolkien was completely sincere. The “Queen” is the virgin Mary, about whom Tolkien would never have been less than reverent. This wouldn’t have worked out well for him.
The survivors of WW1 would become Gertrude Stein’s “Lost Generation”.19 Earnest poems about how the destruction wrought by the war somehow brings glory to god would fall flat on any reader’s ear. So how to oppose the ironic disillusionment of the Zeitgeist? Blank verse stating the opposite of everyone else won’t do. As well be shot for a sheep as for a lamb, and write poems about Faërie.
Occasionally I wonder how valuable computerized analysis of works of literature will ever be. To reassure me that I’m not wasting my time, Jan Christoph Meister tells me1 what Johann Wolfgang von Goethe thought:
