Re: Iversen's second multiconfused log thread
Posted: Mon Dec 04, 2017 10:07 pm
vonPeterhof wrote:На самом деле английский перевод был ближе. Может быть в словарях есть форма несовершенного вида засекать? В любом случае, тут предполагается значение "to pinpoint" или "to spot", так что "to sight" в принципе подходит.
RU: Благодаря встроенному динамику von Peterburg я теперь также предпочитаю английский перевод!
EN: As I wrote in the nerd's thread a few moments ago: "when your goodnight reading the last three nights has been "Quechua für Peru-Reisende" in German from Kauderwelsch, "L'Albanais de poche" in French from Assimil and "Θεσσαλονίκη - Σύντομος Τουριστικός Οδιγός" in Greek from Μαλλιάρης Παιδεία. I have been on a family visit since Friday, and there is one think more to report, namely that my mothers hornbeam tree has been proven innocent in the case about disturbances to satellite TV reception. We found out that the guilty part was her golden rain (laburnum), which without permission has evolved from a bush into a sizeable tree ... and it stands precisely on the line from her parabola to the Astra satellite complex. So after some judicial pruning we can now watch German TV from somewhere between 40 and 50 TV stations. But it seems that almost all the stations in other languages like Spanish, Dutch, French and Chinese have been closed down. I have looked the fate of one of them (BVN in Dutch/Flemish) up on the internet and..
DU: "Belangrijk bericht inzake de doorgifte van BVN TV. BVN is verhuisd naar een andere transponder op de Astra 1L satelliet. Het signaal van BVN wordt sinds 1 augustus 2014 doorgegeven op de nieuwe transponder 48. Na 21 september 2014 is BVN niet meer te ontvangen via de huidige transponder 105."
So to get those channels back I'll have to make a new channel search and reestablish the favorite lists etc etc. (and weed out the 117 variants of WDR or NDR). Life was easier in the old days where we just had ARD and ZDF and NDR.
As mentioned about I have been reading a book about Quechua. It seems to be a mostly agglutinative language, and the author is adamant that
GE: "Wie in vielen Indianersprachen, gibt es auch in Quechua nicht die Deklination (Beugung der Hauptwörter). (...) Den Beugungen liegt in Quechua ein gänzlich anderen Grammatikverständnis zugrunde: Grammatische Bezüge von Wörtern innerhalb eines Satzes werden mit bestimmte Suffixen ausgedrückt (...). Dabei handelt es sich meist um Besitzanzeigen und Richtungsangaben, die mit einer Deklination, wie sie etwa der Lateiner erwartet, nichts, gar nichts zu tun haben!"
Well, I understand the importance of keeping German learners from blindly applying the rules and concepts of German on a very different language like Quecha, but is it really true that the affixes of agglutinative languages and flexives of inflected languages absoluty nothing have in common? The flexives of a language like German have two characteristics: they differ according to things like gender, and there is only a limited number of them. Quechua has a lot of different affixes, and they mostly remain unchanged (although there is at least one whose form changes according to whether the preceding sound is a vowel or a consonant ). But if you consider Finno-Ugrian languages they have they have lots of socalled cases with relatively constant endings (if you take vowel harmony into account), so they lie somewhere between German and Quechua.
One thing more: many if not most of the affixes of Quechua have different roles, so even if you see them as items with a meaning that meaning will be somewhat fluid and heterogenous. The same could be said about cases, although the picture is blurred because of things like prepositions which a connected with specific cases. The Dative in German has something to do with giving something to somebody in a very wide sense, and once upon a time there must have been a semantically motivated reason for using it in each and every of its applications - but right now it is difficult to see why "mit" should be followed by something in the dative case. Another language may use other cases in situations which we would see as quite parallel to those occurring in a German sentence, and some languages actually have cases which German hasn't got. But when I read about the many roles of for instance Quechua "-ta" the picture is just as confusing. It is characterized as a "Richtungsangabe" (indication of direction), but the examples show that this includes situations where the point is some action having a goal or something having a purpose - like in this example:
warmi mikuy-ta yanu-n
Frau Essen kochen-sie
Die Frau kocht das Essen
So if you say that -ta has a meaning, then you also have to admit that this meaning is fairly hard to pin down. And of course learners can't use their experiences from German to help them with this task(-ta), but it still ressembles the problems a learner has to deal with when getting a sense of what the dative or accusative covers in German. The main difference is that the actual endings in German are more complex, but only have to be learnt for a small number of gramamtical cases. The affixes of Quechua are more numerous, but don't vary much.
This reminds me of a self-inflicted problem in the generative linguistics of Chomsky and his followers. In "Syntactic Structures" from 1957 he compared three types of grammar: a finite state grammar, a constituent structure grammar and his own transformational generative grammar. Most of the good things he said came with the realization that some ambiguous phrases could be derived from two different phrases by using some appropriate transformations - so one surface formulation could hide two different deep structures, and then you could of course use the same idea on cases that weren't ambiguous. Unfortunately he retained the constituent structure grammar of the mid 50s as part of his theory and just added the transformations. In this way he brought some unnecessary assumptions into his systems, like a preference for binary divisions including the notorious NP + VP and the idea that semantics should be introduced as late as possible. And he and people inspired by him have been battling with the consequences of these choices forever since.
There are competing theories. My own preference is to say that things like words and flexives are the bread and butter of language, and grammar is just the description of observable regularities in their behaviour. Or in other words: where Chomsky's grammatical rules almost have the status of Platonian ideas I just see them as practical rules of thumb. And this implies that the meaning is there from the beginning (although I agree with Chomsky that we should base the formulations of grammatical rules on hardcore facts that actually can be observed and not on loose semantic talk). It also implies that there may be different kinds of descriptions at play in the analysis. S = NP VP is not a viable startingpoint for the analyses of most phrases in Indoeuropean languages. In most such phrases the finite verb organizes whatever roles the other elements can have, and there is nothing that dictates that those roles should be organized in a binary way. So what about NP VP? Well, if you have a phrase without a finite verb it may still have a structure where two elements are connected in some way (in the Danish grammatical tradition we have traditionally call it a "nexus"). There is for instance such a structure that involves "this" and "nonsense" in both the formulations "this is nonsense" and "I call this nonsense", and you could say that it is the same relation that exists between a subject and the verb - but there is no reason to do so unless the verb is a copula verb (or the void left by its absence in Russian). And for ergative languages the NP VP formula just serves to make the analysis more complicated.
So, to revert to Quechua and German: both the consequences of using a certain affix in Quechua and those caused by using a certain case ending in German should be described in terms that refer to observable behaviours, but in at the bottom of the matter it is a question of using words and affixes/endings with more or less vague semantics that tend to be used in some typical patterns. There is no grammar without a halo of semantical 'dark matter'.