Welsh III

I have a 60+ day streak on Duolingo Welsh now, so I suppose it’s time for an update.

I keep thinking of my first entry about learning Welsh, in which I was at sea in an absolute lack of cognates. That changed pretty quickly, and now, as I’m getting into a bit of grammar, not only are a lot of words direct cognates of English, but even the shape of the sentences and their connecting words is beginning to feel comfortable.

To take a sentence Duolingo really loves, “Dwi’n gweithio yn yr garej” means “I work in the garage.” “Garej” is a direct cognate, and “yn yr” sounds so much like “in the,” or maybe I’m finagling it as “in (that) there.” It’s nice. There are certainly prepositions that sound nothing like English’s–“mewn” means “as a,” as in, “Dw’in gweithio mewn mecanic,” or “I work as a mechanic”. Those require some help, but the Welsh grammar is still similar enough that I feel like it’s holding my hand more often than not.

The biggest stumbling block for me now is the construction of noun phrases with adjectives coming after their nouns rather than before. I ought to be used to this, because Romance languages do it too, but I’m still stumbling over strings of modifiers like “dillad gaeaf newydd Megan,” which directly translates to “clothes winter new Megan’s,” or in English order, “Megan’s new winter clothes.”

I’m figuring out more and more about Welsh phonology and how it’s, written. A single F is almost always pronounced like English V; if you want the voiceless F sound, you write it FF. Similarly, LL is the unvoiced lateral, a delightful sound that English doesn’t use… but DD is the voiced version of English TH. We don’t separate the voiced and voiceless versions of this sound in English orthography, though we do use them both. The vowel U in Welsh is often pronounced like EE in English, and W and Y are flat-out vowels in Welsh. I think, because I’m not coming up with counterexamples, that K isn’t used in Welsh orthography. Instead G, via phonological transformations, takes its place. A point I need to pay special attention to is in which situations S is pronounced as English SH, and when it is just S.

I am getting to the point, when Duolingo wants me to listen to a sentence, that I know what it means without turning on my English filter. I still have to translate it to prove to Duolingo that I know… but I can grasp the meaning without translating in my head. This is a key transition in learning a language.

Proud of myself as I am, I recently had a good kick in the teeth re. my actual proficiency in Welsh (after two whole months, poor me). I have been watching The Crown, and recently saw the episode in which Charles is sent to study Welsh in Wales. Friends, I did not understand one single spoken word of Welsh in the entire episode. The only thing I knew was “Croeso y Cymru,” which was written on a sign. So no, Duolingo at two months out still has nothing to do with actual functioning in the actual language.

Alas.

Welsh language history

Hello. It’s a rainy Saturday here. I’m dilly-dallying on taking the Christmas decorations down; once they’re down I need to start a winter program of weeding the garden, and I’m just not ready for that. Gardening is so exhausting that one needs a wintertime break… unfortunately here in the PNW the winter isn’t a dormant season. If I can get after the creeping underground weed roots now, it will pay dividends for the rest of the year.

Anyway. When not engaging in graft, I am working on Humanist Majiscule lettering–now with dip pens, hooray–and taking the Duolingo Welsh course. As I said in my last entry, the course got out of hand and I had to refer to Wikipedia. Doing so has left me faint of heart (witness the Wikipedia entry on Welsh colloquial morphology and weep) but full of curiosity.

English is my native language, and it has a History. As popularly understood, this history consists of following other languages down dark alleys and mugging them for stray bits of grammar and vocabulary. This analysis feels a little self-congratulatory to me, but it has truth. English is a widely spoken language because it was the language of a lot of people who did a lot of colonizing.

Welsh, on the other hand.

After a minimal amount of poking around on Wikipedia, I have learned that prior to the Saxon settlement of Britain there was probably a single, mutually intelligible Brythonic Celtic language spoken over the whole island, south of the Firth of Forth. This language somehow survived the Roman occupation of Britain, and from previous investigation I think that’s because of Vulgar Latin. Latin and Celtic languages have very similar grammars, apparently, which made it easy for Brythonic people to kind-of sort-of speak Latin by swapping Latin words into their own native tongue. This was called Vulgar Latin. When the Romans left Britain, I imagine (and there is evidence later in this post) that people simply “relaxed” back into their own language altogether, though with a lot of Latin vocabulary sticking around.

The Saxons, however, spoke Saxon, a Germanic language, not a Romance one, and one with a distinct grammar. They settled the central part of the island, pushing the Celtic holdouts to its corners and effectively cutting off those peoples’ ability to travel by land. This led to the formerly unified Brythonic language differentiating into Welsh, Cornish, Manx, etc.

What struck me about this, before I’d taken a minute to think about it, was how recently it all happened. Wikipedia suggests that the Battle of Dyrham in 577 was the no-going-back point of the Saxon settlement, the point at which the pan-Brythonic tongue was doomed. Welsh as a language isn’t considered to have emerged until the 9th century, though.

I mean. Not much more than a thousand years ago. The Modern Welsh period begins in the 14th century, a time when we have lots of historical documents and know a fair amount about what life was like and what people thought about it. And that’s when Modern Welsh really got going! Only just then!

This really gobsmacked me, though it was because I wasn’t thinking clearly at first. Old English, for example, isn’t really a thing until the 9th or 10th century, after all. Early Modern English is Shakespeare, in the late 1500s. That’s the 16th century. But neither language sprung up overnight, nor did they “not exist” before they are first attested. Language is a continuum. It changes constantly, and often over startlingly short periods of time, and just because English and Welsh didn’t “emerge” until the 9th century doesn’t mean people weren’t speaking a mutually intelligible, practically identical tongue ten years before the turn of the 9th century. And ditto for that tongue and ten years before that. And ditto and ditto and ditto, back to the unknown and frankly unknowable time before the noises we made were “language.”

I have a PhD in linguistics, but I haven’t thought much about these things because my degree is an American one, and diachronic linguistics–the kind that studies how languages change over time–is anathema to American linguistics. We’re all about synchronic linguistics, the study of how languages are at one particular moment. It’s a very different discipline and frankly a lot drier, but it has given us leaps and bounds in knowledge with which to do synchronic linguistics.

J.R.R. Tolkien was a linguist long before Noam Chomsky fathered the modern field of synchronic linguistics. Tolkien, the lucky bastard, did diachronic linguistics, and mostly phonology at that. You can see in his drafting of Elven languages that it’s what he was primarily occupied with: phonological change leading to dialects.

Here’s the question I have, though. Languages do change within a single human lifetime, but there must be a limit on how much. From the Welsh example and a couple of others I kind-of remember (I don’t want to cite them now because I don’t want to do the research to confirm them now, or I’ll never get this post published), 300 years seems to be about the amount of time required for a human language to differentiate itself enough to be recognized. (Differentiating languages is a losing proposition–a language is a dialect with an army and a navy, yada yada yada–but just go with it for now.)

Tolkien’s elves, though, are effectively immortal. They can live for thousands of years. I don’t know what it’s like to live for thousands of years, though I am old enough to have forgotten things I used to believe I’d always remember. I wonder, however, if it’s really likely that an intelligent organism designed to live for thousands of years could ever look at its own old diaries and discover that they were written in a completely different language from the one it now uses. What do you think?

Anyway: my new observation about Welsh is that when I find a cognate, it’s almost always with a Latin root, almost never with a Germanic one, and I guess the history I laid out at the beginning of this post explains why. It’s from Vulgar Latin.

The Welsh for “milk” is llaeth, awfully similar to French lait. And the Welsh days of the week are named after Roman gods, not the Norse ones English uses.

Dydd Llun in Welsh, Lunedi in French, Monday in English. All meaning the day of the Moon, but from different linguistic roots.

Dydd Mawrth or Fawrth (phonology!) in Welsh, Mardi in French, but Tuesday in English. Mars’ day… or Tyr’s day.

Dydd Mercher in Welsh, Mercredi in French, Wednesday in English. Mercury’s day or Oden/Woden’s day.

Dydd Iau in Welsh, Jeudi in French, Thursday in English. Jupiter’s or Thor’s day.

Enough of that. You see my point.

Welsh II

Hello hello, I’m finally back. I had the flu and we had a serious snow & ice spell, and then it was Christmas and then it was New Year’s and now I’m trying to get my head back in the game.

I am still working on Welsh in Duolingo. I’m at a twenty-some day streak now, and today the old familiar “dw i’n” and “dych chi’n” was no longer good enough–suddenly it’s throwing all kinds of alternate formations at me, which translate into the same “I am/Am I” and “You are/Are you” in English.

Time to take it to Wikipedia.

The opening of Wikipedia’s article on Colloquial Welsh Morphology (today, January 5, 2023) states that “the morphology of the Welsh language has many characteristics likely to be unfamiliar to speakers of English or continental European languages like French or German.” To which I say, oh crap.

The good news is it goes on to say that “there is no case inflection in Modern Welsh.” To which I say, hooray.

It goes on to explain that in addition to having several dialects, which is amazing to me, indicative of the extraordinarily long and intense history of the language and its speakers but probably not relevant to my Duolingo studies, Welsh can be spoken at several levels of formality. Aha. Perhaps this is what I’m seeing. But also, “verbs inflect for person, number, tense, and mood, with affirmative, interrogative, and negative conjugations of some verbs.”

In short, oh crap.

I’ll be spending some time in the next day or two perusing this article and whatever other internet resources I need in order to make myself a table of what I should be looking for in the coming lessons. Duolingo’s approach is effective and certainly leads to instinctual use of the language, but I used to be a linguist so I want a grammar, dangit.

I have also noticed that, as in English, the way a Welsh word is spelled doesn’t correlate one-to-one with the way it’s pronounced. “U” at the end of words, in particular, can have an -ee sound, as in prynu. It’s interesting to note which languages are like this (in my experience, English and Welsh) and which are pronounced absolutely to-the-letter (Swahili, Italian) and which aren’t pronounced to-the-letter but can still be predicted from their spelling (French.) There are languages, like German, which govern their orthography, and periodically make official spelling reforms, and then there are languages like English and, apparently, Welsh. Og help us.

That might be unfair of me, though. Wikipedia’s article on Welsh phonology notes that “the phonology of Welsh is characterised by a number of sounds that do not occur in English and are rare in European languages,” which I was prepared for going into this course. In particular, I knew that it has a voiceless lateral, a fact that delighted me when I took phonology classes. It has a glottal phoneme too, which also didn’t surprise me because Tolkien’s Elvish languages have them and he based those partially on Welsh. Welsh is written with the Roman alphabet with relatively few diacritics, though, so there you have it. The Roman alphabet just isn’t equipped to handle it, and that isn’t Welsh’s fault.

I have noticed one phonological process so far. “Good morning” is bore da, “good afternoon” is prynhawn da, “good night” is nos da, but “good evening” is noswaith dda. That dd stands in for either the voiced or unvoiced dental fricative (I don’t know if it’s both or only one), and it appears to be a carryover from the -th at the end of the preceding word. Very nice. I approve, both of phonological processes and of dental fricatives.

And that’s about all I have to say about my progress in Welsh. I am having fun.

Welsh

Yesterday I loaded up Duolingo hoping to find a course on either Icelandic or Old Norse, and didn’t find one. This forced me to choose something they did have. I’ve done Norwegian with them in the past (apparently Norwegian was their first course, and it’s the best-developed!) so I skipped past other Nordic/Scandinavian languages because seriously, Icelandic or bust, and ended up settling on Welsh.

Twenty-five years ago, in ye olden days of the internet, a Welshman on a Compuserve forum assured me that it would be very inconvenient indeed to live in Wales without speaking Welsh. Nothing I’ve seen gives me the impression it would be inconvenient to live in either Scotland or Ireland without speaking their respective Gaelics, though (and Duolingo didn’t specify which flavor theirs is), so Welsh it was.

And oh. OH. Oh my.

In school I studied French, which has so many cognates with English that one feels one has a foothold. Then Italian, which has fewer English cognates but a lot of commonalities with French. And then Norwegian–which was glorious. A few cognates, but the grammar is so easygoing and familiar that I can honestly say it was the easiest language I’ve ever studied.

Welsh, on the other hand.

I am completely at sea with Welsh. I cannot find cognates. The lessons I did yesterday included greetings and the words for “man,” “boy,” “woman,” and “girl.” All I can find to hang onto with those four words is that the word for man begins with D, like dad, and the word for boy with B, and that both words for mother and girl begin with M like mother but I’ll just have to memorize which is which to tell them apart.

History. Of course, this is history’s fault. The Roman invaders were welcomed with open arms, and because Latin and the Celtic languages (apparently, I’ve heard) have similar grammars, suddenly everyone was happy to speak vulgar Latin while the Celtic languages, and cultures and peoples, were pushed to the side, and here we are. Romance languages and Celtic languages should be equally separate from English, a Germanic language, and yet. French so easy. Welsh so overwhelmingly different.

Duolingo doesn’t lift a finger to explain grammar to you, you have to figure it out yourself, so because I haven’t looked it up elsewhere on the internet (yet) I don’t know precisely what to expect… but a long time ago I did get graduate degrees in linguistics. For your entertainment, here are my vague impressions of what to expect. I don’t vouch for their accuracy and in fact expect to be wrong:

1. I think it’s a head-last language. Like Latin. Yesterday Duolingo had me translating “I am Owen” as “Owen I am,” but with Welsh words obviously. Head-last.
2. Because I’ve seen so many very long Welsh words, I think perhaps it is agglutinating? Maybe only for nouns? I like that. I like the way English lets you create compound nouns with abandon. It’s a nice feature.
3. I think it has a case system. Again, because Latin. I haven’t done a language with a real case system before. Help.