Gaston Dorren's Blog, page 9

Both inspiring and disgusting

I love Lexicon Valley, the podcast on linguistics. In a show from June, titled The tragedy of English spelling, John McWhorter (upper picture) interviewed Anatoly Liberman (lower picture). Let that line-up sink in for a second: both of them are not only highly regarded professionals, they’re also great popularisers of the science of language. As an admirer of both men, I was listening breathlessly.

And then, along comes this adorable passage.

Here are – I repeat – two renowned linguists. They are discussing language change. And then all of a sudden, they owe up – McWhorter half-heartedly, Liberman without holding back – to a sentiment that I share and that I’ve expressed elsewhere, but always with a pang of guilt.

Here’s the passage:

Liberman: ‘I’m quite conservative when it comes to language change. It’s very interesting and very inspiring to study the history of language. It’s rather disgusting to be part of it. Haven’t you felt that? We all have that feeling.’

McWhorter: ‘I don’t like to admit that I feel it, but I know what you mean.’

What a relief. It happens to the best of us.

 

Vietnamese (2): very early discoveries

Over 30 years ago, I studied some Danish from a book. As a result, I understand a lot of the written language, but my idea of what it sounds like is sketchy at best. That was a mistake I didn’t want to repeat with Vietnamese. So the first thing I’ve been concentrating on these past few days is pronunciation and how it relates to spelling.

The good news is that Vietnamese is much, much more consistent in this respect than is Danish (or English, or French). The not-so-great news is that quite a few of the spelling conventions are counterintuitive. And the really bad news is that many of the Vietnamese phonemes are hard to distinguish for my European ears. Plus there’s tones, of course, but I knew that, so it’s no news, just bad.

What’s odd about Vietnamese spelling? Well, for one thing, since it was designed by Europeans speaking Romance languages (such as Alexandre de Rhodes, pictured), it has inherited some Italian and Portuguese spelling quirks. For instance, the g, which is always pronounced hard (as in go), is spelled gh when followed by an e or i. That makes sense in Italian for reasons that you can look up elsewhere, but serves no purpose whatsoever in Vietnamese. A /k/ sound is spelled k before e and i, but c before o, a or the full vowel u, and qu when the u is pronounced as a mere glide, as in quả. Also, the /f/ sound is spelled as ph, as in Latin words of Greek origin. A few centuries ago, ph was apparently pronounced as an aspirated p, /ph/, so that’s why. It’s a reminder that even a good spelling system goes bad if it’s not occasionally updated to reflect evolving speech. Think Great Vowel Change.

In some other choices, the alphabet designers would have been well-advised to follow a European model, but didn’t. To give just one example, the s in Vietnamese is pronounced /sh/, whereas the x is pronounced /s/. This is the very opposite of what you would expect on the basis of Portuguese, and s for /sh/ is weird by any standard (except Hungarian).

More troublesome than this are the subtle differences between vowels, and especially between several diphthongs and triphthongs. The ư sounds a lot like the u (though I can tell them apart when I hear them side by side), and the ơ and the â are both schwas (you know, as the e in over), except that the ơ is longer. And as for the differences between iêu and ươu or ươi and uôi – please ask me again in a couple of lifetimes.

Add to this the visual clutter of tonal marks, which often make me overlook that there’s an ư or ô on the page rather than a mere u or o; and add to that the phonetic interference of the tones with all these vowels, and you’ll be getting an idea of what I’m up against.

To hack myself a path through this jungle of diacritics and vocal nuances, I have downloaded dozens of sound files from forvo.com, with native speakers pronouncing as many words. I’ve included several minimal pairs (and triplets, et cetera), in the hope that at some point chất, chát, chắt, chật and chặt will all begin to sound wildly dissimilar to me. (None of these words means ‘cat’ by the way. For that, Vietnamese has the endearing mèo.) Actually, I should make this list longer, because chậc, chắc, chạc and chạp are at this stage pretty indistinguishable too, given that these final plosives (t, c, ch and p) are pronounced very weakly.

Oh well. No-one said it would be easy. And at least my wife has had a couple of good laughs over my vocal antics.

****

You can find all my posts about Vietnamese 

Vietnamese (1): why, and how to begin

Attending the Polyglot Conference in Thessaloniki, late last month, has inspired me to do a bold and daring thing: start learning Vietnamese. While the language may appear small in comparison to its northern neighbour, Chinese, it actually has no fewer than 80 million speakers. Moreover, one of these, a friendly and wonderfully efficient woman named Tuyết, happens to clean our house every fortnight, so I may have an opportunity for practice there – or at the very least, I will be able to surprise her one of these days. (Days? Make that ‘weeks’. Or ‘months’, more likely.)

Another good thing about Vietnamese is that it uses the Latin alphabet, so I won’t have to learn thousands of characters, only the meaning of nine diacritics: the tilde, the acute, the dot below, and so on. Moreover, Vietnam is an attractive tourist destination, so I may one day go on holiday there. (Or perhaps not. I’ve been flying a lot recently, and I should get back to practicing what I preach, climate-wise.) The main downside is that, like Mandarin, Vietnamese has tones, and even more: six instead of four. But hey, I have a good ear for tone in music and I grew up with a pitch language, Limburgish, so that should give me something of a head start.

Still, I haven’t yet mentioned the main reason to embark on this adventure, which is that I’d like to report on it in Babel, the book I’m working at. Babel will be about the world’s twenty largest languages, and Vietnamese is one of them. I’m sure the language has all sorts of surprises up its sleeve, and I’m looking forward to being amazed and amused and frustrated and rewarded – which I hope will make for an interesting chapter in Babel.

Right now, I’m trying to figure out what might be a good way to start cracking this rather exotic nut. There’s an Assimil course on its way to my address (the German version; it also exists in French, but somehow not in English). I already owned a language guide from the German series, which tends to be good. And there is of course a lot of free stuff online: the big names, such as Duolingo, Memrise, Forvo, but also dozens of videos featuring a Saigon-based native speaker and teacher named Annie.

Time to take the plunge. I’ll keep you posted.

 

Dad’s polyglots – a recipe

A father recently sought my advice about the linguistic education of his two young children. Among the many friendly and interesting emails I get from readers of Lingo, this one really stood out, because his was a question I’d never given much thought to before. Our brief correspondence is reproduced below, anonymised, very lightly edited and, of course, with the father’s permission.

Gaston,

I just finished reading Lingo and wanted to extend my compliments. I’m recommending it to all my friends interested in languages. (I’m American, so unfortunately I can count all these folks on one hand).

I’d like to put a question to you. I have two children under the age of 5, and am interested in giving them the gift of a portfolio of languages while they’re still young enough to learn them very easily. But I want to choose wisely.

They’re already getting English and Chinese, but I’m wondering what other languages might be attractive (and if there is a preferred order in learning them). For example, is there a Romance language that commands a middle ground such that once learned, the other Romance languages can be easily acquired? The same goes for a middle-ground Slavic language, etc. German appears to be an interesting choice because it’s relatively widely used, and once mastered, the Scandinavian languages and Dutch seem to fall in line without too much of a fight.

I’m also interested in getting my kids exposed to a good diversity of grammar and sounds before my kids’ abilities to effortlessly internalize these things wither and die.

Thanks in advance for any perspective you might be able to provide and again, congratulations on an excellent book!

J

———-

Hi J,

Thanks for your email, so full of praise and interest.

I will try to answer your question, but please realise that asking other people like me the same question might well result in as many different answers. Also, I feel – without being a father myself – that children develop best if they can play to their hearts’ delight and choose from options offered to them, rather than by strong stimulation of some skills over others. Having said that, I do wish my parents had made me study the piano…

Chinese is certainly a wonderful choice, for a whole range of reasons: a tonal language (very useful) with a massive number of speakers, a highly un-English and typical East Asian grammar and a script that requires a lot of exercise and which gives a degree of access to Japanese. German is a very good choice, as, to my mind, would be Spanish. This last one may sound rather bland from a US perspective, but it is both the largest and a fairly typical Romance language, much more so than French.  I’m not sure about the Slavic languages. Russian is obviously the largest by far, but I don’t know how typical it is. Still, probably the best choice. Farther afield, I’m even less sure. Hindi? Very many speakers, but English will see one through in India. Arabic? Nearly as hard as Chinese, an exceptional grammar and phonology, and I’m not convinced the culture is all that vibrant. For access to the Muslim word, I’d recommend Persian or Indonesian: two nice and easy languages with 100 million speakers or more each.

Anyway, even ‘only’ Chinese, German, Spanish and Russian may simply be too many. German would be the first to drop: it shares a fair amount of basic vocabulary with English and it has a case-based grammar like Russian. I do agree that it is a fine introduction to Dutch and Nordic, but then, the Dutch and Nordics have good English (as do many Germans).

These would be some of my thoughts, for what they’re worth. I hope your children will enjoy the learning experience – and if not, be spared the suffering.

Thanks again for your note,

all best,

Gaston

——–

Gaston,

I was very pleased to see your reply in my inbox this morning.  Thanks very much for your perspective, I value it greatly.

And please rest assured that I promise to avoid being an overbearing tiger dad with my kids.  My goal is to lead them to some very large and inviting pools of water, and hope they drink.  If they don’t, we’ll move on until we find other academic, athletic and/or artistic things they truly enjoy.

I have to admit though that I’m very encouraged with their language instruction so far.  It’s mostly singing, dancing and games (all in Mandarin), and while I walk away with absolutely nothing at the end of a session, they’re at ages where seemingly everything sticks immediately or with a minimum of repetition — and they’re incredibly eager to learn new words. It’s fascinating to watch. And for what it’s worth, they call their Chinese class “dance class”, which I find encouraging.

Russian and German may be challenging where I live (Hawaii) but Spanish should be easy.  Indonesian might be possible too.

Thanks again for taking the time to help out a hobbyist.  I’m looking forward to your next book.

J

——–

Chinese, Indonesian, Persian, Russian, English, German and Spanish, along with some others, will all feature in my next book, Babel, planned for 2018.

Science update: the articiple

In a surprise turn of events, the Large Hadron Collider near Geneva has generated a new part of speech that does not fit into the Standard Model of Grammar. After smashing participles and articles into each other, experimental linguists found not only particles, as expected, but also words of a hitherto unknown category, provisionally labelled as articiples.

A collision between the article the and the participle gone produced the particle to (which in nature only ever exists in infinitives such as to be) accompanied by the novel word ghene. Now that the experimenters know what to look for, they expect to find other articiples. Until then, it’s hard to ascertain the meaning, function or even pronunciation of ghene.

The discovery has already led to frenzied speculation among theoretical linguists. ‘I believe that smashing articles into any part of speech may yield an “art of speech”, given the right conditions’, said Oene Daasma, a theoretical linguist at the University of Harderwijk, the Netherlands. ‘And in the privacy of our coffee corner, I’ve heard my colleague Fetze Alsvanouds think out loud that verbs might be turned into adverbs by adding no matter what, preferably nothing or even less. These are exciting times for theoretical linguistics.’

⩗⩗⩗⩗⩗⩗⩗⩗⩗⩗⩗⩗⩗⩗⎞

The scribbles on the right are not just doodles, a badly drawn rough sea or an attempt by a 5-year-old to emulate grown-ups’ fascinating handwriting. A real adult has written a real word here: minimum.

Even if you had figured that out for yourself, you’ll agree it’s not easily legible. That’s due to a shortcoming in our alphabet: the similarity between hand-written i, n, u and m. In many words confusion is never far away, which is why monks, clerks and other writers have come up with all sorts of clever tricks.

First and most obviously: dotting the i’s. That seems self-evident now, proverbial even, like crossing the t’s, but at one point it must have been a brilliant new idea, given that the old Romans didn’t do it (just like the person who wrote minimum, but that was negligence). With capitals, we still don’t, wrongly assuming that there’s no risk of confusion there. I for one think that Io, one of Jupiter’s moons, had better be spelled İo.
In non-classical Latin, the second of two i’s was often given a tail, first in hand-writing, then also in print. This title page of a 18th-century book by one Blasius Altimarus, is a case in point: see the words Blasii, regii and consiliarii. In Dutch, where double i was frequent, ij then proceeded to develop into a separate letter, which explains why the name of the country’s largest lake is spelled IJsselmeer, not Ijsselmeer.
In English, the letter combination um was considered troublesome in short words. To fix the problem, writers did what English writers do best: dream up a few more  irregular spellings. Two of these have survived to our day: come and some. Which makes you wonder, why not gome for gum? The answer is that gome was common too, but printers managed to suppress this rather less frequent word in the days when they were trying (with limited success) to clean up English spelling. Should we give it another try now? Better not, perhaps. Turning some into sum would be confusing, as it has acquired a different meaning. Changing the spelling come into cum would be as great an idea as changing pawn shop into porn shop.
In Norwegian, m is the only consonant that is not doubled at the end of a word when the preceding vowel is short. I’m not sure if the rule was introduced to prevent confusion, but I’m sure it does.
In German, the u is often written as ŭ, to emphasise the difference with n. In the pre-war Kurrent type of writing, this stroke above the vowel was compulsory, because there was even less of a difference between u and n than in modern handwriting.
When in 1928 Turkish got its own Latin alphabet, a dotless i (ı) was introduced, with a sound value different from the kitchen or garden i. The Alphabet Commission ‘never stopped to ask themselves what the dot was doing there in the first place,’ as the Turkologist Geoffrey Lewis wryly observed. The result of adding another n’ish and u’ish letter was, of course, more confusion. To solve their self-inflicted problem, the commission suggested that the ı be written as ĭ. Turkey’s president Atatürk, the driving force behind the new alphabet, did exactly this, as well as writing ŭ for u, German-style. Nowadays, the habit is obsolete and the Turks just try to live with ı.

There are bound to be other spelling rules and writing habits that somehow originated in the visual similarity of the handwritten i, n, u and m, or perhaps other letters. If you are are aware of any, please let me know!

 

The importance of the German praying man

Traduttore traditore, usually translated as ‘the translator is a betrayer’, is probably the only Italian expression in my active vocabulary. And other than dictionaries and suchlike, Umberto Eco’s La ricerca della lingua perfetta must be the only Italian book on my shelves. So there is something peculiarly congruous about my discovering, earlier today, a disconcerting translation error in that book.

On page 98 and 99 of the English-language edition, In Search Of the Perfect Language, I came across a passage claiming that for the German reformer Martin Luther, ‘German was the language closest to God.’ Statements of that sort can easily be found about Hebrew, Arabic, Tamil, Korean and some other languages, possibly including German. Yet flowing from Luther’s quill, it somehow seemed out of character.

I decided to track down the original. It’s not an easy task to Google for a quote whose exact original wording you don’t know, but I’m fairly confident I’ve found Eco’s source. It runs, …weil sie den deutschen Beter näher zu seinem Gott führt als alle Fremdsprachen. In literal translation, it means ‘because it (= the German language) brings the German praying man closer to his God than all the foreign languages’. Incidentally, it wasn’t Luther himself who wrote this; it was the twentieth-century German historian Arno Borst who thus paraphrased the reformer’s view on the matter.

It’s hardly an outrageous or even remarkable observation, I’d say. If talking to other human beings is easiest in one’s first language, then the same is undoubtedly true when it comes to conversing with the Supreme Being.

So what is it that makes Borst’s original line sound so much more reasonable than Eco’s translation? The crucial bit is the object, ‘the German praying man’. In omitting this, Eco makes it seem as if for Luther, German has a wondrous capacity for bringing anybody, regardless of their mother tongue, closer to God – almost as if God himself spoke German. That statement would fit neatly into Eco’s argument that modern Europeans took the glorification of their languages to preposterous heights. They did, to be sure, and other, lesser-known Germans of the era may well have been among them. But not Luther.

Or was it the English translator who lapsed here, rather than the great author? No, James Fentress is innocent. In the Italian version, the crucial words are equally absent: il tedesco è la lingua che più di tutte avvicina a Dio. No praying man to be found.

I’ve just realised that I know one more Italian expression: se non è vero, è ben trovato – ‘even if it’s not true, it’s a good story.’ But this particular book of Eco’s is non-fiction, so ben trovato won’t cut it. And the line about Luther is more than just ‘not true’ in some innocent detail; it looks suspiciously as if it’s been doctored.

From plural to singular, three times over

Plurals and singulars are not hewn in stone. Plurals, especially those of foreign extraction, are regularly mistaken for singulars, and -sometimes – vice versa.

The word stamina, for instance, was really the Latin plural of the word stamen (a term you may remember from biology class, albeit in a very different meaning), but has in English long been a singular. The same has happened with agenda and, more recently, data. It is happening under our eyes with phenomena. People get worked up about it, but there’s nothing new under the sun. Even the respectable opera was once a plural, and only became a singular because Italians couldn’t be bothered with Latin grammar – and why should they?

Another example, closer to home, is skates. It entered English from Dutch as a singular, but the final -s and the fact that these objects usually come in pairs conspired to get skates interpreted as an English plural in no time at all.

A very different piece of metal that often comes in pairs is the long bar that, fastened to sleepers, allows trains and subways to run from A to B: the rail. Having borrowed this word in the plural, the Russian now call each of these bars a rels, with a regular plural relsy. (Or strictly speaking rel’s and rel’sy, where the apostrophe has nothing to do with possession or greengrocers.)

So far, I’ve only given examples of ‘number confusion’ with loanwords, but it happens with inherited vocabulary too.

The word truce started life as trewes, the plural of trewe, which is related to true. Since there is nothing particularly multiple about it, it began to be used as a singular, and the truces has been a bona fide plural in its own right for centuries now.

The Late Latin words poma (for ‘apple’, earlier ‘fruit’) and folia (‘leaf’) also used to be plurals. But Latin words ending in -a could either be neuter plurals (as in paraphernalia, et cetera et cetera) or a feminine singular (as in vagina and formula). Poma, folia and a few other plurals were ‘reanalyzed’, as the jargon has it, and became feminine singulars, which in turn developed their own plurals.

Let’s focus on folia, because its further development was somewhat adventurous. First of all, it got mangled pretty badly in the process that transmogrified Latin into Spanish: the final result was hoja, plural hojas, which is pronounced as /ohas/. But what interests us here is that Chamorro, a language spoken in the Pacific island of Guam, borrowed this Spanish plural and once more turned it into a singular. Ohas is now a proper Chamorro word, not for ‘leaves’, but for ‘leaf’.

Unfortunately, I’ve not been able to establish beyond reasonable doubt what he Chamorro plural looks like; my best shot is ohas siha, but correct me if ‘m wrong. Anyway, I would suggest that we borrow the Chamorro word. We could drop the space in the middle and then use it for, say, ‘palm leaf’ – this is the Pacific, after all. And of course, we’d add an s for the plural: every palm has several ohassihas.

That’s the life for a linguist: drinking your cocktail on the beach under Pacific, Mediterranean or Caribbean ohassihas. It might not be something entirely new under the sun, but the rarity would be well worth the trip: the plural of an original plural, made singular three times over.

Do you be, that’s another question

Grammatical irregularities in a foreign language can drive you mad, but grammatical regularities can be even worse – when you expected them to be irregular.

English is a second language to me, and I dutifully learnt 38 years ago that most verbs to be turned into questions require the auxiliary to do: ‘Where do you live?’ rather than the Shakespearean-sounding ‘Where live you?’ But this is not not true for to be: ‘Where are you?’ is fine.

So far, so good. Until the other day, when on the Wait But Why blog (much recommended) I came across this question: ‘How do you be a good person?’

Excuse me?!

Having recovered from my astonishment, I tried to analyse the sentence. The ‘logical’ alternative would of course be, ‘How are you a good person?’ But I can see why the writer shouldn’t want to put it that way. Unless I’m much mistaken, it sounds pretty offensive, suggesting as it does that the other person is everything but a good person. That was not what this writer meant. She just wanted to ask, ‘How do you go about being a good person?’, ‘What are you to do so as to be a good person?’

So I think that in her ‘how do you be’, be was sort of shorthand for ‘get to be’, ‘become’, ‘act so as to be’, all of which do require the auxiliary verb to do in questions. Usually to be and its other forms (am, were, etc.) are just an inconspicuous glue, connecting the subject to some adjective, adjective or both: the turtle is slow, a turtle is an animal, turtles are slow animals. In this case, however, it had a much stronger, ‘pregnant’, meaning. That’s why it behaved like one of those normal verbs which have strong meaning, like to act, to wallow, to dehydrate and thousands more. Or so I think.

Which made me wonder if I could come up with an example of my own. Here then is another sentence where a form of to be has a different, but similarly pregnant meaning: ‘A wise person doesn’t do much, a wise person just is.’ I believe, or perhaps I just hope, that if I  turned the latter phrase into a question, it would be okay to say, ‘Does a wise person just be?’

But as ever, that’s for native speakers to decide. Feel free to let me know!

 

 

My World in Words

Earlier this year, I was interviewed in the comfort of my home by Patrick Cox (see photo), a British-American radio journalist who specialises in language. I’d enjoyed dozens of his World in Words shows as podcasts, partly because they are so interesting, partly because I like Patrick’s friendly and intelligent style and his pleasant voice (and trust me, I’m not saying this about all interviewers). World in Words is probably the language podcast that I like best, with Lexicon Valley an excellent second.

Yesterday, he sent me a note saying that the episode featuring me has been put online. Listening to myself talk is among my least favourite things (here’s why), but I think Patrick has managed to make me sound fairly coherent – there’s skill for you. Here‘s the link, and if you scroll down a bit, you’ll see the contents of the podcast listed: nearly 15 minutes of Dorren talking about multilingualism and me (well, he asked) and even singing a song. The other 15 minutes are about Klingon.

Enjoy, and do let me know what you think (but please, break it gently).