In Turning Darkness into Light, Marie Brennan’s latest novel set in the Lady Trent universe, Audrey Camherst, Lady Trent’s granddaughter, is tasked with translating a newly found cache of ancient Draconean tablets with the utmost secrecy. She travels to Lord Gleinleigh’s estate, where she meets his niece Cora, who is assigned to be her assistant—and spy on Audrey for her uncle. Gleinleigh wants the tablets to be translated before the Falchester Congress, a summit between humans and Draconeans scheduled to happen in ten months. Audrey requests that her Draconean friend Kudshayn be allowed to travel to Gleinleigh’s estate to help her, and, to her surprise, Gleinleigh says yes. So Audrey, Kudshayn, and Cora work to copy and translate the tablets that seem to tell the story of the creation of Draconeans and their early history.
In the interest of disclosure, Marie emailed me and asked if I wanted to read her book for my column, and I was sold based on “archaeologists translating dead languages.” This is 100% my jam. I haven’t read any of the other Lady Trent books, but that didn’t matter, because there’s enough description and backstory to go on—so if this sounds like your jam, but you’re not familiar with Lady Trent and her adventures, don’t let that stop you. I must also admit that about halfway through, I stopped taking notes—partly because of heavy-duty spoilers (seriously, there are a couple things I really want to talk about, but they give away the ending), but also because I was having too much fun just reading the book. If you’ve read it, don’t spoil the ending in the comments!
Historical linguistics has two main branches, both of which deal with how language has changed over time. The first branch is the one most people think of when they think of historical linguistics: reconstruction, also known as comparative linguistics. There are reconstructionists who work in all the language families. The Indo-Europeanists are also called Indo-Germanists, because a lot of the early philologists were German, and it was at the height of 19th-century European nationalism. In German, the Indo-European languages are still called indogermanische Sprachen. What reconstructionists do is take existing language data and work backwards from the oldest written records to hypothesize what the ur-language looked like. It is purely hypothetical, because there are no written records of Proto-Indo-European. It’s based on the assumption that sounds shift uniformly, although each subfamily has its own shifts. The Germanic languages shifted differently from the Celtic languages, which shifted differently from the Greek languages, which shifted differently from the Iranian languages, etc.
The other main branch involves working from the oldest written records forward to the modern day (or vice-versa). For Germanists, this is a much shorter time frame (the oldest written record is the Gothic Bible, 188 pages of a 4th-century translation of a Greek text) than it is for Romance philologists, whose written records go back to ancient Rome. (The field of Germanic linguistics is small enough that you can meet all the big names at the annual conference. Romance linguistics, however, is huge.) This is the branch I have worked in, as well as the very new sub-branch, historical sociolinguistics.
As part of my education, I have translated texts from dead languages (Gothic, Old Norse, Old English, Old Saxon, some Middle High German) and am intimately familiar with the type of mistakes one can make as a new translator and learner of said language. So imagine my glee when Brennan provided a translation of the first tablet by Cora, who is just learning Ancient Draconean, and it read basically identically to a lot of my own first efforts at translation in those dead languages I just listed. The note I wrote reads, “Cora’s first translation is so wonderfully stilted in some places & awkward in others—weird/idiomatic structures translated literally.”
Listen with your wings in the ditches and the rocks in all corners.
Through me I say how clay was made, dirt and water and ceiling and wind and grains and animals of the ground and flounders and sky, the three heart reeds and the four that were three later. Stone my words for the coming year, because mind records are the one real forever. When this clutch is recorded, we live with them, and the goodness of their treasure will keep the going generations doing things.
Pretty awkward, right? But a reader can get the general gist of the text, even if parts of it are nonsense. Compare it to Audrey’s translation:
Hark, spread your wings to hear, from the canyons to the heights of stone, in every corner of the world.
Through me this clay will speak of how everything was made, the earth and the waters, the heavens and the wind, the plants and the beasts of the land and the rivers and the sky, the three peoples and the four who afterward were three. Preserve my words for the ages to come, for memory is the only true immortality. So long as these four are remembered, they will live in us, and the blessings of their deeds will remain.
Dictionaries of dead languages are compiled by people who have read a lot of texts, and sometimes you find a word with dozens of potential meanings, frequently in the same semantic field, and without the context of having read many texts yourself—in those instances, you just have to guess which one makes sense, or write down a few plausible candidates. You can see this with Cora’s “ditches” and Audrey’s “canyons.” These are both long, narrow holes in the ground, but on a different scale. See also “ceiling” and “heavens.” Modern German Himmel, Old Norse himinn, and Gothic himins all mean both “sky/heavens” and “Heaven”—so this is absolutely plausible. Also, I just love the phrase “mind records are the one real forever.”
There are a lot of things in this book that are very true to the experience of people working with what amounts to archaeological records.
Gleinleigh sees the ownership of these ancient artifacts as a type of prestige, but he doesn’t care about them otherwise. Audrey remarks that he gives no thought of how to display the works in a sensible fashion, that he has artifacts from one society haphazardly displayed next to those from a completely different society. There is, in fact, a running theme throughout the book questioning whether it is ethical for people outside a culture to own such artifacts, with discussion of the black market, looting, and hoarding for prestige, as well as repatriation.
Audrey wishes that so many records hadn’t been destroyed when humans revolted against their Draconean overlords, because (a) that history wouldn’t have been lost and (b) there would be more extant texts to do scholarship on.
Many (most) of the tablets in the cache are records: lists of queens, royal decrees, and tax records. This is also true in the real world. The extant corpus of Old Saxon is a poetic retelling of the Jesus story in Germanic alliterative verse, a few fragments of the book of Genesis, some prayers, and a handful of property records. Very famously, a Babylonian customer complained to a merchant of copper ingots that they were of poor quality, and this tablet remains to this day (and has become, somehow, a meme).
Cora, as she learns Ancient Draconean grammar and orthography, is deeply offended at how nonsensical it is. Audrey comments, “I have never seen someone so outraged by orthography,” although “Draconean writing is really quite irrational, when you get down to it. But it was the first time anyone had invented writing.” Brennan has Audrey describe said orthography: this tablet is old enough that there is no space between words, which makes separation of phonemes tricky; the reader has to figure out whether a consonant was geminated, because they didn’t write doubled consonants; they use triconsonantal root signs, which “might stand in for any one of a dozen nouns or verbs built from that root.” (This made me think of the Semitic languages, which build on triconsonantal roots and create the word with vowels.)
Brennan also hits on something I find frustrating about reconstructionists, especially the Old School ones: they “corrected” what they termed “scribal error” to match the form that was “supposed” to occur, based on their scholarship and deductions. Audrey comments, “Mistakes do happen, but they’re less common than we want to believe, and if we go around correcting ‘errors’ all over the place, we’re likely to make a mess of the whole thing.” Reader, I cheered.
Audrey notices a change of style in writing between some of the tablets, which she suggests is due to a recopying from a later scribe. This is definitely a thing that happens; the Peterborough Chronicle, for example, was written and recopied by monks over centuries, and parts of it were copied from other monasteries after theirs was burned in a 12th century fire. There are lots of opportunities for error, as well as the introduction of idiosyncrasies of a particular scribe or a newer form of an inflection, or even a newer borrowing.
A few other nice touches from the text: there is a comment about things that were obvious to the contemporaneous reader, like funerary offerings, that are lost to the modern audience. There are some things from Ancient Rome (I believe; I am not a classicist) that do not appear in any text, because all the Romans knew what they were so they didn’t write it down.
Draconean body language is drastically different from humans’. Draconeans have wings and different facial structure, so their body language developed around that.
Audrey and Kudshayn reach a section referring to “the Endless Maw” and speculate that a bit of trouble a colleague had with translating a phrase involving the “mouth determiner” is due to a later taboo against naming an evil entity in full. Taboo avoidance (and its cousin avoidance speech) is a really cool subject. We see it reflected today in words like “frak” or references like “He Who Must Not Be Named.” Language Log has a wonderful discussion of several instances of taboo avoidance in publications, which includes the quote “Well, [I summarily reject] that.” Perhaps in Battlestar Galactica, they would have said, “Well, frak that!” One of my friends’ major research interest is swear words and taboo avoidance, and I really want to sit down with her (remotely) and pick her brain on this for a future column on taboo avoidance in SFF. Because there is SO MUCH.
So, to sum up before I hit my word limit: So much of the experience of translation in this book is accurate, either to my personal experience or to the general reality of historical linguistics (and archaeology, I assume; I am not an archaeologist, but I love reading about it, and there is a lot of overlap between the two fields), that it’s obvious that Brennan either did a ton of research into it or has a lot of experience with this herself. (I’m going with B. There’s just too much insider-view for it to be A.) There’s also a bunch of digs at academia, which exude great verisimilitude.
What other books do similar things with translation work? What are your favorites? Let’s discuss in the comments!
- William Jones, the first person to suggest that Greek, Latin, and Sanskrit could be related through regular sound change. (He was also a colonizer in India.)
- A History of German (2nd edition 2018) This is a fairly accessible textbook on the history of German, written with the goal that it would be readable by non-linguists (it’s mostly successful). Joe’s dry sense of humor comes through in the text and keeps it from being deadly dull. (He is my MA advisor’s dissertation advisor, and I once got to explain shitposting to him.) The links under “additional reading” are quite varied and often for a general audience.
CD Covington has masters degrees in German and Linguistics, likes science fiction and roller derby, and misses having a cat. She is a graduate of Viable Paradise 17 and has published short stories in anthologies, most recently the story “Debridement” in Survivor, edited by Mary Anne Mohanraj and J.J. Pionke.