study ling(uistic)s

I don’t think modern linguist(ic)s should bother discussing “universal grammar” and the ideas connected to this concept any longer. 

“Universal grammar” is a notion based on speculative introspection – Martin Haspelmath (p.c.) used to call it “philosophy of syntax” – and dogma rather than empirical evidence. Worse, proponents of “UG” usually keep staring at human language though an anglo-centric lens and plain out refuse to acknowledge facts that do not fit their dogma. This is unscientific methodology and in itself already reason enough to discard the theory.

Apart from dogmatism, there a handful of other problems concerning this theory.

It is based on circular reasoning. First, English is taken as the model for human language, widely ignoring data from typologically extremely different languages. Any “results” of UG-based ruminations on language structure or language acquisition are then generalized and taken to be evidence for the uniformity of human languages and proof for the universality of features or parameters posited in the beginning. 

It is mechanistic. There is no wholesale one-to-one correspondence between units of any given two languages. The rule are one-to many correspondences in translation. This is true not only for lexical units and idiomatic expressions, but also for grammatical morphemes and categorial distinctions. Not all of these differences can be made compatible by subsuming them under some abstract categories. 

It is insensitive to context. As any translator or interpreter can tell you, it is impossible to translate more than the most basic expressions without taking into account extralinguistic factors which make up the communicative setting, the so-called situational context of an utterance. Any model of automatic translation or of automated speech production must be capable of taking sociolinguistic, pragmatic and contextual information into account. Yet, this is not provided for in UG.

 It assumes unnecessary complexity. Some grammatical properties are taken to be universal. It has been demonstrated by several linguists working with natural languages in the field that languages exist and do function without some of these properties. Among them are things like recursion (Everett) or the existence of different lexical classes and parts of speech or of some kind of inflection (Gil 2005).  

It assumes innateness of linguistic structures and a genetically hardwired “language faculty”. Everyone who ever watched children acquire language knows what to think of that. So do neurologists, geneticists and psychologists. If there actually was an innate “universal grammar” that only needed parameter setting, feral children or other children growing up without relevant input should automatically produce some kind of self-made language which displays the “default setting” grammar that is claimed to be genetically preprogrammed.

Unfortunately, UG has nevertheless grown a huge number of loyal followers who themselves became influential figures in linguistics. Chomsky rose to stardom because he had a nice theory at the right time, when people at M.I.T. communications lab got an awful lot of money from the Pentagon and others to develop machine translation in the light of the experience in WWII and the Korean war. Military intelligence* needed to have huge amounts of intercepted communications in Japanese respectively Korean translated quickly yet efficiently. Admitting that one’s theory of “universal grammar” – which is an essential prerequisite for “easy” automated translation – is flawed or entirely useless would be killing the goose that lays golden eggs.

The longer one keeps up this (self-)deception, however, the harder it is to let go of it. That’s why these theories, though repeatedly proven use- and pointless, are being vehemently defended by those making a living off maintaining the illusion that “universal grammar” exists.

Literature:

How Noam Chomsky’s world works by David Hawkes

Most writings by Daniel L. Everett and/or on Piraha are directly opposing Chomsky’s views on UG.

*The* eulogy on UG:  The myth of language universals: Language diversity and its importance for cognitive science by Nicholas Evans and Stephen C. Levinson in BBS 32 (2009), 429-492.

Vyvyan Evans (2015): There is no language instinct.  http://aeon.co/magazine/culture/there-is-no-language-instinct/

David Gil (2005): Isolating-monocategorial-associational language. In Cohen, H. and C. Lefebvre eds. Handbook of categorization in cognitive science. 347-379. Oxford: Elsevier. 

Ewa Dąbrowska: What exactly is Universal Grammar, and has anyone seen it? In:  Front. Psychol., 23 June 2015, 

See also: The linguistics wars

*) Yes, that is an oxymoron.

01. masterpost: start studying

hey! this is my first masterpost, in which I will link you useful stuff to motivate you and make you comfortable to really get stuff done! (even though I’m not the best at it, too) 

motivation:

- quote 1 || 2 || 3 || 4 || 5 || 6

- pictures and tips: 1 || 2 || 3 || 4

- videos 1 || 2 || 3 || 4

things to get you in the right mood:

- cute graphic

- study playlists 1 || 2 || 3 || 4 || 5

- study spaces 1 || 2 || 3 || 4 || 5 || 6 || 7 || 8

stationary:

- here a really kawaii korean shop x

- and another one x

- kikki.k x

- japanese stationary x

tips:

- list of things to do when you want to start studying

- some quick study tips

- when you really don’t feel like studying

I’m in Canberra where there are a bunch of French schools, and even though I do hear a lot of French, Italian is definitely heard more often, there are a lot of Italian communities around. Although here in the capital there are a lot of all kinds of languages, what with all the embassies around. 

However, in the modern language, nouns referring to female humans are often found in class III rather than class II. Already in the earliest description of the language, Uslar (1890), it is recorded that nouns referring to young women behaved as class III nouns. A woman would only start being referred to and referring to herself using class II nouns at some time between marriage and the birth of her first child; if she did this too early, or kept doing it too late, it would be seen as a ridiculous affectation of an inappropriate age status, and if others did it too early or too late it was considered an insult. (The distinction between classes II and III is thus comparable to the distinction between the Mrs and Miss titles in English, although the line is drawn slightly differently.) The use of class II has become more and more restricted since then; Khaidakov (1963) found that it tended to be restricted to immediate family members of an older generation, and it would be impolite to address even an old woman using a class II term if she was not a relative. In one dialect, that of Arakul’, it has gone so far that class II has been eliminated entirely.

Why did class III—a class originally reserved for non-humans—expand so as to include most women? When we put it like that, it seems odd that women would be offended by not being included in class III rather than the other way around. The probable explanation is that it is a generalization of a single exception which already existed in the language involving the word duš ‘girl’. This word might have always been in class III, rather than class II. Cf. German Mädchen ‘girl’, which is famously neuter rather than feminine; in the case of this word it’s because it’s a dimunitive, and a morphological rule that dimunitives are neuter takes precedence over the semantic rule that female humans are feminine. I don’t know if there is a similar morphological explanation in the case of duš, but there’s also a semantic reason why words for ‘girl’ are not prototypical feminine gender words: children tend to be seen as having a neutral gender. And if the idea that ‘girl’ would be assigned feminine gender still seems implausible, consider the fact that the sense of ‘girl’ can be a specialization of an older sense of ‘child’, in general. In fact, the English word girl was originally gender-neutral. And we know that the whole development has taken place in the Indo-Aryan language Konkani. Konkani has the traditional three Indo-European genders: masculine, feminine and neuter. The word čeḍũ is neuter, and originally meant ‘child’, but came to mean specifically ‘girl’ in recent times, and the neuter gender has been extended to other words such as bayl ‘woman’ when they refer to young women, as opposed to old women. The feminine gender has not been eliminated yet in Konkani, although it’s conceivable that it might be, eventually. So the change which has been happening in Lak is not just a one-off oddity, but something which seems to be encouraged by a combination of morphological, semantic and sociolinguistic circumstances which tend to co-occur across different cultures and languages.

11.23.15.