jbdp@jenny.UUCP (Julian Pardoe) (10/28/85)
In his good (if a little confusing) article dick@tjalk.UUCP (Dick Grune) writes: > They have awfully long words, by our standards. The standard example being the Turkish avrupalilastirilamiyanlardan- siniz (no dots on the i's, cedilla under the first s): you are one of those who are incapable of being Europeanized -- sounds like an artificial creation, but of course in Ataturk's day Europeanization was a hot topic, and words like this do come naturally to speakers of these languages. The word is parsed avrupa-li-las-tir-il-ami-yan-lar-dan-siniz or Europe-<coming from...>-become-cause--<passive>--unable--being--<plural> -from-`you are'. (Lots of `Lar's and `Ler's are usually a sure sign of Turkish, by the way, as are `den's and `dan's.) > Note: Finnish is by far the most difficult language I've seen, no > insult intended. Yes! My Teach Yourself Finnish gives twenty rules for the formation of the plural stem and soime fifteen or more for the formation of the genitive plural! However many of the forms such as `miesten' as the genitive plural of the stem `miehe-' are predictable once you get familiar with the language. Something that wasn't mentioned in the `softening' of consonants that takes place in Finnish and Estonian. In Finnish there are three series, basically tt -> t -> d; pp -> p -> v; kk -> k -> 0/v/j. Basically a consonant is softened when the syllable which it starts becomes closed (i.e. changes from CV to CVC). Thus the genitives of pojka (boy) and kirkko (church) are pojan and kirkon. Often assimilation takes place: ranta (shore) -> rannalla (on the shore). Ranta shows what the Finns do with consonant clusters: this word comes from the Swedish strand. Skruntch would probably end up in Finnish as ransi. The Finns have only a small number of consonants -- no `b', no `d' except as a way of spelling the softest form of `t', no `f', no `g' except to spell a long `ng' sound (Helsinki -> Helsingin just as ranta -> rannan), no sh (except in a few borrowed words), no z (I'm not even considering letters like `q'), so we have words like kirahvi (giraffe), tohtori (doctor), Ranska (France, or Prantsu in Estonian; they've no `f' either!) Estonian uses the letter ~o (pronounced as a Russian hard i: `bI'). It's most unusual feature is that whereas F. and H. make do with two grades of length Estonian has a third: over-long, though this isn't represented in spelling except for the `t', `k' and `p' sounds, where the series are (in the order short-long-overlong): d-t-tt, g-k-kk, b-p-pp. (Like F. E. doesn't distinguish between voiced and voiceless consonants, so the letters `d', `g' and `b' are spare.) Softening of consonants seems (to me anyway) much less regular than in F: more variations are possible, and the cause (i.e. closing of a syllable) isn't so obvious. This maybe because the word endings have been worn down. As Dick Grune points out the Estonian word is often the Finnish word with the last vowel cut off. Put this vowel back and you get the genitive, which is the base for other cases in the singular (Tallinn -> Tallinna (gen) -> Tallinnas (in T.), Finnish Tallinna -> Tallinnan / Tallinnassa; E. laev (ship) -> laeva, Finnish laiva) -- which suggests to me that maybe the genitive has an ending like the Finnish `-n' but this got worn away: this would explain the softening which often occurs in the genitive of a noun (sport -> spordi). One of the distinguishing features of these languages is vowel harmony (though modern Estonian has lost it). F. and H. basically divide the vowels into two classes: front and back, and suffixes endings occur in front and back forms. Thus in H. (to choose rather artificial examples) we have f"ustetlen (smokeless) and h'azatlan (houseless). In Finnish `a', `o' and `u' are front vowels, with `"a', `"o' and `y' as their back equivalents. `E' and `i' are neutral (though tending towards front vowels). Thus Oulussa (in Oulu), Helsingiss"a (in Helsinki), vuodelta (from the year), vedelt"a (from the water). Turkish has a more complex scheme. Basically there are two harmony families: a / e (``two-fold harmony'') and \i / i / u / "u (``fourfold harmony'', where \i is the dotless i); and the rules are after 2 4 from (= dan^2) our (= [i]miz^4)... a, \i a \i father: baba-m\iz-dan e, i e i house: ev-imiz-den u, o a u car: oto-muz-dan "u, "o e "u bridge: k"opr"u-m"uz-den I could go on... Julian Pardoe ------------- University of Cambridge Tel: +44 223 352435 ext. 265 Computer Laboratory Arpa: <@ucl-cs: jbdp@cl.cam.ac.uk> Corn Exchange Street Janet: jbdp@UK.AC.Cam.CL CAMBRIDGE, CB2 3QG UUCP: mcvax!ukc!cl-jenny!jbdp Great Britain