2011-11-04

Things they didn't tell me about English, part 3/n.

Phonotaxis/phototactics... For the most part those are comparable against any and all languages. But then there are always subtle differences. Between English and Finnish the most pronounced differences have to do with the long versus short consonants and vowels, which Finnish has and English mostly does not.

English does retain certain minimum pairs from Proto-Germanic, such as "uh" vs. "oomph". Short versus long vowels. Or in consonants some even rarer, colloquial cases like "thick" versus (Irish/Scots Gaelic) "Theean". But as you can see, those cases are pretty rare, and it's pretty hard for me to come up with such examples evenif I try.

In Finnish, that sort of thing is par of the course. In consonantal contrast, we might have something like "kato"/"look (colloquial imperative)" versus "katto"/"roof". In vowels, maybe "kaato"/"felling something over", and funnily enough even the homonym "kato"/"loss of something".Not as a verb form now, but as a noun in base form.

Be as it may, both the length of consonants and those of the vowels carry meaning in Finnish. At the same time in English, they usually don't. "The Thing" and "Thhee Thing" might come across as dialectal oddities, but they do not change the meaning of "the" as a preposition.

These are then just examples for both Finnish and English speakers. What's at issue here is the contrast between long/short vowels/consonants, and not any particular word. You Finnish speakers try to learn English, you again have the short end of the staff here: it's pretty difficult to unlearn differences you already know and hold dear. Like, "little" just has one "t", "litté" has 2.5 of it, and "lite" then has a mere 0.5-0.8 of the t-ness.

In there, the first "e" is very short, and almost elided altogether.The second, accented one is something like 2.0x as long and ends up sounding like "i". The third "e", in its context, is about 0.5x long, pure "e", with an abrupt stop at the end.

In the Finnish context, the last "e" would be the one in the word "ketterä"/"nimble". Very short.The middly long could happen e.g. in the Eastern vernacular word "keppana"/"a pint". For the long "e" we'd nee something like "Eerikinkatu"/"Eric's street.". The final lenght being: "Eeronkatu"/"Eero's street."h

2011-11-03

Things they didn't tell me about English, part 2/n

The phonemic inventory. Oh yes, that's a good one. Do you really think the English "th" has something to do with the Finnish "t" or "h"? If so, you're sorely mistaken.

Typically what is meant in written English by "th" is a consonant that simply doesn't exist in Finnish. That goes for many other vowels and consonants in the English language besides that one. As it happens, our languages have divergent "phonological inventories"; we just happen to use different sounds to communicate with each other. Even if we share most of the Latin alphabet in which we write both of our languages, the sounds we assign to them broadly vary..

The total, Finnish inventory is narrower than the English one. To compensate for that Finnish for example retains full contrast between long and short forms of both consonants and vowels, and is phonologically speaking a bit more malleable.

In written Finnish we have the vowels "aeiouyäö". In the spoken one, in IPA, "ɑeiouyæøə". In particular, there is no schwa in there, but simply a fully articulated "ø"/"ö". You'll notice the similarity between IPA and the Finnish alphabet, already. It is no mistake: as a language with a young, Latin derived alphabet at the time of the ratification of the first IPA, the Finnish, much phonetic at that time writing system exerted its due influence.

The consonants are in written, alphabetic order "bcdfghjklmnpqrstvwxz".  When we speak those they translate pretty much straight onto spoken sounds via IPA, after you elide a couple of them: "bdfghjklmnprstv".

To bring it further down into the native Finnish soundset, that translates something as in: "aeiouyäö" and "hjklmnprstv", because certain foreign consonants have to be dropped out as homophones. Also, the one consistent thing not captured by the alphabet is what we call "äng-äänne", as in the velar nasal, denoted variously by "nk" as a short one preceded by "k" and as "ng" as the long, standalone one. In both IPA and saami, it's denoted as "ŋ". Besides that, all kinds of sandhi, diphtongization, palatalization, and whatnot are of course going on at the same time, but are purposely left out of the transcription/orthography.

Be as it may, you will have already noticed how few consonantal phonemes we have in proportion to the wovels. That's part of why Finnish sounds kind of like thin singing to a foreign ear at first sight, and even at the second one, something like "Japanese spoken with an Italian accent". It's because of the unreasonable relative amount of vowels, half of which are then even lenghtened/held; and the relatively soft pronunciation of the narrow inventory of consonants that is left.

English, its phonological inventory is much more difficult, because it can't even be written down using anything approximating a usual Latin alphabet. The Finnish one can, because it was designed to be so, much as Korean Hangul was designed to be a phonetic writing system from the start. English, it's a total mess: you really have to put down two different notations if you want to talk about it. One written, and the other one spoken. So here it goes/hurts...

English vowels. Thanks to a long history of borrowing and endolinguistic mutation/mutilation, you pretty fuckers came up with a ton of them. Half of them I can even pronounce properly, in context. The other half is there, but in pronunciation, I remain insecure:

a (harken), e (zest), ɛ (bed), ɪ (bit), i (seesaw), ɑ (awe), ɒ (oh), o (prom), ʊ (hook), u (prune), æ (plan), ɝ (the r in rotation; it's actually a kind of vowel/consonant hybrid), ʌ (plus in Scottish/Cockney), ə (ehm...)

That's bad enough already. But the consonants are even worse, because of the sheer number and variety of them. They might not require exotic IPA symbols to describe most of them. But the elision-by-homophones which I did above with the Finnish consonants can't really be done here, because all of the consonants actually have minimal pairs. Thus, the canon of them looks like a dog's puke. Which I'm required to regurgitate every day, by the way, as my firm's official language:

b (bad), d (dinghy), g (gun), h (height), j (the y in you; again not quite consonant, not quite vowel), ʤ (jump), k (kill), f (florist), l (linen), m (mine), n (nice), ŋ (trailing ng in mining), p (porous), r (Scottish trilled r as in curd or are), s (sift), ʃ (shot), t (tan), ʧ (chip), θ (thaw), ʒ (zion), ð (the th in father), w (what, another approximant between consonant and vowel), v (via), z (zed, zardoz). Complicating the picture, quite a number of English consonants are aspirated, which carries onto their subjugate vowels even at the root of a word. As in /pin/ versus /pʰin/, which are fully allophonetic in English, with the latter, aspirated version sounding more authentic in most dialects and the first pronunciation often used by naïve Finnish learners of the language.

If you look closely and squint a bit, you'll notice that most of the sounds overlap with the Finnish ones, and with an English dialect or two. We don't, as humans, usually mess around with the sounds we let out too much. Yet a Finnish learner of English, phonogy-wise gets the bitter end of the stick: even in approximation, English simply has a broader inventory of both vowels and consonants.

So let's then do a symmetric difference between the two sets to see what might require extra practice from each side of the language barrier, and at the same time what is already known.

In vowels, Finnish has "ɑeiouyæøə".
English OTOH has "aeɛɪiɑɒoʊuæɝʌə"

This means that English learners of Finnish already know everything but: "yø". That's easy enough.

Vice versa, Finnish learners of English miss: "a
ɛɪɒʊɝʌ". So it's no wonder many eminent Finnish people fuck up their English vowels, and as a result sound like a hinterland peasant.

So what about the consonants? Let's see...

Basic Finnish has: "bdfghjklmnŋprstv", though we do employ one or others for loan words. What was left out can and are being fully substituted with these ones.
English, it has: "bdghjʤkflmnŋprsʃtʧθʒðwvz". Quite a lot in comparison, as I already said.

So for a Finnish Learner of English, the set of nasty, foreign sounds would now be: "ʤʃʧθʒðwz". Quite a handfull.
For an English learner of Finnish, the corresponding set would be: none whatsoever. This repertoire is already fully covered.

Thus, based on phonology alone, Finnish is actually rather a simple language for an English speaker to learn. At the same time, we do have to learn totally foreign phonemes while being trained in our third-to-second, most important language.

But there is still at least one phoneme which is exceedingly difficult for an English speaker to get, evenas it's an sich present in the English language. That's the trilled "r". You can do it, alright, for example in a word like "rustic", Scottish style. But in Finnish (and Spanish and Portuguese), that trill is much, much deeper, and could be sustained for pretty much as long as your breath carries it.

As in the favourite Finnish curse, "perkele". Obviously you should be able to hold the trill for arbitrary lengths of time when ordering your beer.

Binary autopsy, digital eulogy and the post-mortem clouded shrine

Nowadays people still write eulogies. They try to sum up in a few paragraphs another person's lifetime of contributions and effect upon society. More and more the person writing up a notable life is also an uninterested, salaried employee of this or that news outlet, devoid of any emotional connection to either the person being wrapped up or his true meaning to the field he leaves behind in mourning.

What prompted me to write this piece was the near-simultaneous demise of both Steve Jobs and Dennis Richie, and the waves they caused in both blogosphere and computer science minded academia. One was jubilated, the other one nigh forgotten, despite a vast difference in their contribution to lasting human knowledge.

So it's that knowledge aspect which then bothers me, even looking into my own eventual passing. I was no Steve Jobs or Dennis Richie, neither in charisma nor in technological insight/prowess. I never left my mark that an archeologist might find; not as a billion once shiny iPhones, nor in the form of code that underlies the hundred billion further, fundamentally digital artifacts that they're going to be digging up in ten kiloyears or so. By those standards, I'm absolutely nothing; I haven't left a lasting mark; I could easily be gone and soon forgotten, with the surrounding society hardly noticing anything ever changed.

Except that I don't feel I'm a total idiot, with nothing to give, never having thought of anything novel. In my musings, I believe there are occasional nuggets of gold, every here and there. Though minor by the standards of Jobs or especially Richie's, they're still there, and add to the sum body of human knowledge. And they are entirely digital, again more like Richie's and less like Jobs's.

Which finally brings me to the point: increasingly a person's life and achievement is encapsulated in his correspondence, personally accumulated library—read: hard drive—and the rest of his electronic goings-about. That's no longer just a collection of letters or works, the like of which they've built for notable figures since written-history-day-one. No, our generation's legacy is increasingly a day-to-day, acute description of us as a whole, as networked beings, and much easier to put together afterwards than the "letters and correspondences" of some singular historical figure of the past.

So why is it that even now all of what is heard about a dead person, even a notable one, is a somber epitaph? Why doesn't it come ripe with a link to a well-secured package of everything digital the person ever did? By modern standards, most of the facts of his life? The standing record of what, how, when and who you were?

Why is it that we still don't perform a binary autopsy, write a truly digital eulogy, and entomb a person's complete life work in a well-preserved, forward error corrected shrine of a torrent, floating in the Cloud?

Because that's what I'd like done once I'm gone. It shouldn't be too difficult anymore even for the Joe Nothings like me. Yet most especially, it'd even make the Richies of our day grin from their grave onto the Jobs's, as it should be.