I’ve been away for a while!
The end of undergraduate senior year does that to a person, really.
Regardless, I’m back, more or less. I’ll be spending next year being a hermit in New Hampshire and applying to Ph.D. programs in linguistics, both of which should be different kinds of exciting. In the mean time, I hope to be posting here a bit more. I hope to get up a few left-over posts about the end of the conlang seminar, along with more about 8’0i (my consign) soon; I hope to resume (by which I mean begin) the Monthly Conlang project hopefully starting in July; and I’ve got some new music that I have to get up and post about.
Most immediately, though: I’m about to leave on a two-week-long drive across the country, from San Francisco to Dublin, NH. I’ve included my itinerary, below: As you can see, I’m taking a rather southern route, so that I can swing through Alabama for the National Sacred Harp Convention, which I’m very much looking forward to.
I’m sure that I’ll be tweeting, and possibly posting, from along the way.
Dear Ghana,
Your food is tasty! Really, I’ve come around on this — I was dubious at first, but I’ve gotten to really like it! But seriously — folks? This is the most impractical method of food consumption I’ve seen yet.
Soup is great! There are many wonderful ways to eat soup — with a spoon, drinking straight from the bowl, dipping with bread, pouring over rice… Many, many perfectly reasonable ways to consume soup. But with your fingers, using only a wad of totally non-absorbent paste? This is not one of them.
Along similar lines: I like goat meat, I really do! It’s very tasty, especially when sopping in a good spicy groundnut soup. But prying off pieces of scalding hot meat with ones fingers, one-handed, without dumping the whole bowl of food in your lap — this would not be an easy task, even if the meat in question weren’t naturally rather tough, full of slippery blubber, and covered in (ferociously hot) slimy okra stew.
It’s good food! I like banku a lot, really, I do! But there’s got to be a better way to get it from bowl to stomach.
Sincerely,
Leland
I keep worrying that I’ll go look in a library at some point and discover that everything that I’ve learned about Twi in the past three weeks is sitting on a shelf somewhere, nicely summarized. In fact, this very nearly happened the other day — on the library at the University, while I was interviewing C. There, on the shelf: A book that claimed to summarize the Akan tense system.
Except that I don’t believe it.
Don’t get me wrong: It pointed out a few things that I’ve been missing — for instance (painfully obvious in retrospect) that two affixes that I’ve been treating as homophones actually differ in tone. Oops.
But on the other hand, it claims evidence for distinctions in meaning that I haven’t seen, and proposes affixes that I can’t find. I’m not sure whether this is primarily dialectal difference, or what, but I definitely don’t believe that (for instance) the -i suffix is really a perfect aspect marker — in my research, it definitely turns up only as a Fante dialect form of the -ye suffix that my professor has elegantly analyzed as a bizarre form of do-insertion, and thus basically past tense.
So: I’ve dodged a bullet this time. I still have hope that the adverb placement tests I’ve been doing with SVCs will help me to formulate a nice syntactic account that hopefully hasn’t been done before. We’ll see.
How nice it would be, to be writing a first grammar of a language, and not have to worry about this…
I’ve finally started in on my secondary, personal purpose for this trip: Studying some local music.
Given that I play in the gyil (West African xylophone) ensemble back at Swat, I knew I’d be studying that some here; I had my first lesson yesterday. I’ll be studying 6 hours a week with the inimitable Bernard Woma (and/or his associate Jerome). It’s a bit of a drag to haul out to his house in Gao-gao (about 1.5 hours by tro-tro each way), but so far well worth it: I’ve started a few basic pieces. Hopefully I’ll have good calluses before too long, though — I got a nasty blister within the first half hour of playing, yesterday.
But even more exciting for me is that I’ve just begun to study the seprewa, the Ghanaian equivalent of the Malinke kora and the instrument of the Ghanaian “troubadours” going back to the 1600s. It’s a stringed instrument, a “lute-harp”, and considerably less well-known than the kora, which it differs from in number of strings (10-15 for the seprewa, 21 for the kora) and the make of the resonating body (wood for the seprewa, calabash gourd for the kora). Otherwise, it seems to me to be very similar: Played almost identically, always solo as an accompaniment to the voice, and used mostly in praise and storytelling.
Unlike the gyil (and other Ghanaian music), it is a fully diatonic instrument; the gyil uses a version of the pentatonic scale. (It’s not clear to me exactly how the seprewa scale was tuned, traditionally: It’s very possible that it was tuned significantly differently from the Western diatonic, but these days electronic tuners abound, and force it into more-or-less equal temperament.) The seprewa has a very bright sound, comparable in volume to the Western harp, I’d say. My teacher is one Kyere Matin, at the University of Ghana (Legon) music department, and so far he and I are getting along famously.
I especially like that, for Kyere Matin at least, the process of teaching seprewa inevitable starts with building one. So, yes, I’ve just purchased an instrument (inevitable, really); he’ll start work on it tomorrow, and hopefully finish it over the weekend. I’ve asked to be included in the process as far as possible, but at the moment it seems like that will mostly just be observing. In the mean time, I’m starting to learn on his instrument, and enjoying it greatly.
A small detail: There’s a neighborhood around here that somehow gets named after it’s traffic circle (despite the fact that such circles are everywhere). In local English, “Circle” comes out as [s3:ko] (using 3 for epsilon). I had a devil of a time understanding this the first time around.
Related, though not actually about English proper: Tro-tro money-men (not sure if they have an official title — the folks who collect the fare and lean out the window shouting the name of the destination) tend to apply some creative reduplication to the name of their route. So “Circle” comes out, more or less, “Sekseksekseksekseko! Ku’sekseksek!” (The “ku” is from “Kumasi”, for the full name of the circle itself.) Those headed to Tema Station, Accra tend to shout “T’accraccracra! T’accraccraccracra!” My real favorite is the tro-tros going to the new station in Achimota, which say something like “Achimotamotamotamota Stayshstayshstaysh!” Because these calls are the only way of knowing where a tro-tro is going (so far as I can tell, anyway), I’ve learned to interpret them relatively well.
The best, though, is when folks try to apply Twi style reduplication to English. In Twi, adjectives used predicatively get reduplicated, so that we have nsu nyunu “cold water”, but nsu nyunu-nyunu! “The water is cold!” One place that this comes up for me commonly is the phrase M3 ti Twi kakrang-kakrang, “I hear (understand) Twi a little bit.” The best part, though, is that often when I use Twi with taxi drivers or what-not, they observe, in English, that “You learn Twi small-small!” Small is so-far the only adjective I’ve seen this happen with, I think, but it’s come up rather often.
Not actually unintelligible, and mildly frightening:
- 3NAILS + 1CROSS = 4GIVEN
Easy mistake to make:
- EXCEPT THE LORD
Names of shops that I found amusing:
- JESUS JESUS JESUS (They sold mobile minutes.)
- Macedonia Repairs and Christ in Law Auto Body (Not all in-laws are to be dreaded? Or perhaps this refers to our Elected Savior Obama? Speaking of which…)
- Hotel Obama (Spotted off the main road through Legon.)
On that last one: Obama-mania is definitely alive and well, around here. The bookstore at the mall has an entire shelf of Obama related books; pictures of him are everywhere; I even found a series of Nigerian-produced comic books telling his life story (starting with his father’s journey to America). I saw a Ghanaian flag yesterday with a football slogan on it: “Black Stars [Ghana's team]: YES YOU CAN”. I have to assume that this is Obaman in origin.
The tro-tro is the Accra public transportation system, though it’s often referred to as a “group taxi” because all the buses (read: minivans) are privately owned and operated. I like it rather a lot, whenever I’m not trying to go anywhere at rush hour…
The backs of tro-tros typically have decals with typically religious sayings. Most of these are Christian and in English (“PRAISE GOD”), but a few are in Twi and a few are clearly Muslim. A surprising number, however, are either rather misguided or just totally inexplicable. Here are a few of my favorites:
- NOT ALL FRIENDS ARE GOOD (Ominous, much?)
- YOUR TIME WILL COME (Ditto!)
- OH GOD (Ditto ditto?)
- 1+1=3WHO (WTF?)
I’m sure there are more that I’ve forgotten; I’ll post more as I find them.
I’ve found that the most common problem I run into with my speakers when doing grammaticality tests is that they just don’t hear ungrammatical sentences. A typical exchange goes like this:
Me: “Would it be correct to say, ‘Kofi ntemtem saa-ye.’”? (“Kofi quickly danced.”)
Informant: “Yes.”
Me: “Could you say that sentence for me?”
Informant: “Kofi ntemtem osaa-ye.” (“Kofi was quick. He danced.”)
(NB: the e’s in ntemtem should be epsilons; the o on osaa should be open.)
Notice the extra o- in that last statement. The informant automatically interpreted what I’d said in way that did make sense in Twi! As the translations indicate, this amounts to basically splitting the sentence in two and throwing in an extra pronoun. Sometimes I’ll repeat my question four or five times in a row, and even try to be explicit (“No, without the o-”), and still not get them to actually hear what I’m saying.
The only real defense against this is, as above, to rigorously ask them to repeat back what I’ve said. Sometimes in repeating it back they’ll give me an entirely different form (“Ntemtem Kofi saa-ye” is common, for the above example), and I’ll have to correct them, but if after a while I can’t actually get them to say what I’m asking for, I have to conclude that the sentence I’m trying for isn’t grammatical after all.
Unfortunately this method is far from foolproof. For a while, I constantly had conversations like this:
Me: “How about, ‘Kofi twaa ene dua no too fom.’?” (“Today Kofi cut the tree down.” A more literal translation would be “Kofi cut today the tree fell down.”)
Informant: “Yes, correct.”
Me: “Could you repeat that back to me?”
Informant: “Kofi twaa dua no too fom ene, yes.”
(NB: The e’s in ene are epsilons; the o’s in too are open.)
Whoops. I can’t tell you how many times I had exactly that conversation, or one just like it, until suddenly, today, one of my informants reproduced the sentence I’d said correctly, and explained that it was highly marked but possible. This, needless to say, was rather surprising. (I was told that for such a sentence to work, the meaning had to be, basically, “There was this specific tree that fell down yesterday because Kofi chopped it.” Like I said, highly marked, but possible.)
This is probably the most common issue in elicitation that I’ve had to deal with, though the fact that C. refuses to give me just one version of any sentence is also a bit of a problem… (I noticed yesterday that what he was doing that made it so impossible to listen to him was self-correcting between dialects mid-sentence, so that I’d never get a full sentence in either Fante or Twi. Ai ya…)
me rebu efu (the first “e” should be a lower-case epsilon) means “I am growing angry/annoyed” in Twi; it’s one of my favorite idioms in that, if you translate it literally, it means “my chest grows”.
A minor annoyance, really: I’ve been on a search for “subject-oriented” adverbs, and am now fairly certain that Twi has none.
A word of explanation: The placement of adverbs in a sentence is an extremely important syntactic test — it helps us know where the boundaries of various phrases fall. Adverbs are not a single class, really: There are several different types. The most common are adverbs of manner, such as “quickly”, which simply say something about the way in which the action proceeded. Slightly different, however, are “subject-oriented” adverbs such as “deliberately”, which say something about the way the agent of the action carried out that action. These can be used for slightly different types of tests.
But, it seems, Twi doesn’t have any! “Angrily”, “deliberately”, “accidentally”… I’ve tried various forms of all of these, and the result is always the same: The meaning is expressed using verbs. Sometimes (with, for instance, “angrily”) it becomes a serial verb construction meaning, roughly, “something happened, and that made my chest grow”. Other times (as with “happily”), it gets expressed using a different type of SVC and a noun, such that it means “something happened with happiness”. And then (for instance, with “deliberately”), we get even a third type of construction: Deliberately is expressed using another idiom, “to look well”, and so we get a type of SVC that says, more or less, “I looked well, and then did something.”
I should be happy, I suppose: These have opened up new SVC possibilities. But the tests using subject-oriented adverbs will be slightly hard to replace. I suspect I can do something similar with adverbs of time (“tomorrow”, “yesterday”, “soon”), but not quite. Oh well.
I’m now up to four informants — I’ve added E., a one-time Twi instructor at the University who now works selling odds-and-ends in the market. Unlike the others, he speaks Twi (rather than Fante), and, having at one time been an instructor, is rather wont to take control of sessions if I’m not careful. My first session with him was less than productive, but my second (just this morning) shed new light on one of the biggest puzzles I’m working on: Tense.
Monolingual English speakers are fairly unlikely to have thought about tense in any great detail, I think, but anyone who took (for instance) French in high school at least knows that it can be a very complicated problem. In particular, it’s necessary to separate between tense and aspect. Tense is usually past, present, or future — in other words, it locates the action in time with respect to the time the sentence is uttered. Aspect, on the other hand, expresses something about the temporal structure of the event itself, most simply whether it’s finished or still going. This is the difference in English between “I ran” (past tense), “I had run” (past tense, perfect aspect), and “I was running” (past tense, imperfect aspect).
Twi tenses and aspects are perhaps not so multivariate as in English and French, and yet they interact in very complex ways. Furthermore, negation screws with the picture as well: The tense marker usually analyzed as future tense in the affirmative is, for instance, totally unavailable in a negative sentence, while the continuous aspect marker (similar to English “-ing”) from the affirmative appears to become a future tense marker in the negative! And neither of these may overlap with the past tense marker. even though one would expect to be able to form a past continuous sentence (“I was running”, in English)!
Why am I focused on tense, when my primary goal is an understanding of SVCs? For two reasons: For one, tense interacts with SVCs in interesting ways. When I was listing questions I wanted to answer (in this post), I included tense interaction because I already knew that a few interesting things would come up — which verb gets marked for tense in an SVC, and how, is interesting and complicated. In fact, I suspect that SVCs will actually be essential to compiling any coherent picture of tense in Akan — just today, for instance, evidence from SVCs started to change the way I thought about a particular tense marker in a way that I think might explain some of the puzzles I talked about above.
But just as SVCs will be necessary for understanding tense, I think tense will be necessary for understanding SVCs. How verbs get tense has been, in the study of syntax, extremely important for the understanding of how entire sentences are put together into coherent wholes; it seems clear to me that how the multiple verbs of an SVC get tense will be essential for understanding how all those verbs are integrated into one sentence. I’m rather a long ways from being able to do that, however.