Plus Voice

A Linguist's Log

[solcarrus-deactivated20230616]

The organization of contexts through which to make sense of writing.

Fuller, Michael A. (2004) An Introduction to Literary Chinese (Revised Edition)

[dedalvs]

A textbook called Grammatical Theory by Stefan Müller just came out, and man, he's dropping some truth bombs!

Furthermore, the claim that first language acquisition is effortless and rapid when compared to second language acquisition is a myth as has been shown by estimations by Klein (1986: 9): if we assume that children hear linguistic utterances for five hours a day (as a conservative estimate), then in the first five years of their lives, they have 9100 hours of linguistic training. But at the age of five, they have still not acquired all complex constructions. In comparison, second-language learners, assuming the necessary motivation, can learn the grammar of a language rather well in a six-week crash course with twelve hours a day (500 hours in total).

And then...

In Chapter 13, it was shown that all the evidence that has previously been brought forward in favor of innate linguistic knowledge is in fact controversial. In some cases, the facts are irrelevant to the discussion and in other cases, they could be explained in other ways. Sometimes, the chains of argumentation are not logically sound or the premises are not supported. In other cases, the argumentation is circular. As a result, the question of whether is innate linguistic knowledge remains unanswered. All theories that presuppose the existence of this kind of knowledge are making very strong assumptions.

In short: Human linguistic abilities aren't magic, and children aren't elves.

In meme form:

[plusvoice]

Get in [linguists], we're doing WORDLE

[singing "Somewhere" from West Side Story, replacing the word "place" with "wordle"]

Get your gramle on here.

[plusvoice]

Dependency and Constituency

Some quick notes on dependency and constituency relationships

[plusvoice]

Reminded of this beautiful syntactic treebank resource today by Wikipedia of all places! If you're into dependency parsing, templatic morphology, or Quranic Arabic, be sure to check it out here.

[gatheringbones]

“To learn a second language, you have to be willing to give your self up, the self encoded in your first one. You are no longer a person who speaks with facility and authority. You are less than what you were as a child: You cannot transact a phone call without help, discuss matters more complex than the color of fruits and vegetables. You cannot signal who you are. Most of us, by the time we’re adults, speak in so many words. We convey information through tone: I am sad, or I am displeased, or Is it not clear? I am important. Our speech acquires layers so that directness, when employed, has power through force and rarity: “I don’t like what you did.” But at the beginning in learning a language, you can only be direct. You can say “Tea is required here,” not “Can I get a cup?”—a vast difference in terms of your popularity. In half language, you’re half what you were, half an overgrown child. You speak like a child, are received as a child. In this other state, you lose abilities. “I was amazed at how quickly my English …” “Fell apart?” a cognitive neuroscientist named Arturo Hernandez, who’d also done time abroad, said, and laughed. This was a year or two after my return, and we were comparing notes. “There’s this very weird thing that happens where your language starts to bust apart. It’s because there’s language in your head and there’s language in the environment.” The one absorbs the other, he explained, the external one filters into your thoughts, becomes, to some extent, your inner one. “We think of language as ours,” he said, “but it’s not. It’s on the news, and we speak it with people. We use other people’s language all the time. It all makes you question, What is knowledge? What about that—is knowledge in our heads or in our environment? And if it’s in our heads, how fast can it break down?” He mentioned that overseas, when you’re aiming for fluency, you try to suppress your first language. “You don’t want to use it,” he said, then paused. “It’s interesting. Language is a lot more fragile than we think it is.””

— katherine russell rich, dreaming in hindi

[plusvoice]

“First, I take the position that all language universals are implicational universals. Only under that assumption can one avoid arbitrary decisions as to what will count as a language. Most work on language universals has systematically ignored child language, written language, sign language, non-native language (Roman Jakobson’s English, etc.), and the language of aphasics and schizophrenics. For example, linguists happily accept the claim that all languages have nasal consonants even though they are perfectly aware of languages that do not, for example, American Sign Language. It is of course reasonable for linguists to ignore ASL here: the universal relates to the vocal medium of spoken languages. However, the irrelevance of ASL to some language universals does not make it irrelevant to all discussion of language universals – it clearly is relevant to discussions of universals of constituent order or of semantic distinctions in the lexicon. If ‘language’ is understood broadly and universals are taken to be implicational, exclusion of any types of language from the domains of universals must be for cause: the antecedent of the implicational universal must give the grounds on which the particular varieties of language are to be excluded (for example, ‘If the medium of expression of a language is vocal, it will have nasal consonants’, excludes ASL by virtue of its medium of expression being nonvocal), and the linguist must attempt to state universals in their greatest generality, that is by specifying as narrowly as possible the conditions that would remove a variety of language from the applicability of the putative universal.

Second, I wish to dissociate myself from an assumption that is so popular among linguists that it is difficult to find anyone who disputes it, namely the assumption that people who talk the same have the same linguistic competence. I have recently been advocating (McCawley 1976a; 1977a) a conception of language acquisition in which many details of acquisition are random or are influenced by ephemeral details of linguistic experience. Such a scheme for language acquisition need not lead to gross inhomogeneity in a linguistic community, since there is ample opportunity for revision of learning that has made the speaker grossly divergent from his neighbors. Moreover, it is easy to think of alternative linguistic rules and underlying forms that yield exactly the same well-formedness data and exactly the same pairings of meaning and expression; indeed, linguists are perpetually arguing about such alternative analyses, for example, the analysis in which the English regular plural ending is /iz/ and a rule deletes its vowel under one set of circumstances, versus the analysis in which the ending is /z/ and a rule inserts a vowel under other circumstances. Maybe for some people plurals work the one way and for other people the other way. The assumption that all normal adult members of a linguistic community have the same internalized analysis in such cases is gratuitous. In the rare cases where linguists have looked for interpersonal variation in language, they have generally found it. For example, Haber (1975) reports that speakers who ostensibly form English plurals the same way give a broad range of responses on tasks requiring the formation of plurals of novel words, with each speaker having his own ways of dealing with novel plurals. There is also considerable individual variation in the morphemic relations that speakers of English perceive, as one can readily verify by asking one’s friends whether pulley is related to pull or tinsel to tin.

Third, and closely related to the last point, I claim conscientious objector status in the ongoing war against ‘excessive power’ of grammatical devices. We have all been taught to limit our descriptive devices as tightly as possible, preferably to those that were good enough for our scientific forefathers, since seemingly innocuous devices may turn out to harbor within them the dread Turing machine. It has accordingly become common for linguists to attempt to resolve disputes among competing analyses by drafting sweeping restrictions on grammars so as to give one of the competing analyses a legal monopoly. I will argue below that many proposed language universals have served only to allow linguists to construct cheap arguments for their favorite analyses and that those arguments have given the illusion of significance only because their alleged role in the war effort against ‘excessive power’ has obscured important respects in which they are extremely implausible."

McCawley 1982. "Language universals in linguistic argumentation." from Thirty Million Theories of Grammar.

[plusvoice]

Getting psyched for thirty million theories of grammar (by James McCawley)

[plusvoice]

“Experience, aptitude and individual differences in native language ultimate attainment” | Dąbrowska. 2018. Cognition.

This article demonstrates the remarkable heterogeneity of language attainment across individuals, even in core aspects of grammar. Rather than suggesting that up to 40% of individuals have “incomplete acquisition” of their native language as Dąbrowska asserts, these findings challenge the notion of (complete) acquisition altogether. When linguists and other cog/neuro researchers think about what "knowing a language” looks like in the brain, 40% of individuals cannot be swept aside as having incompletely acquired their native language. This natural variation in linguistic competence and performance across constructions crops up throughout the acquisition and sentence processing literatures, and although it is not obviously fully attributable to transformational syntax, nor fully attributable to measures of intelligence and working memory, it is clearly characteristic of language at the population level. The extent of this variation severely undermines nativist accounts of language, but without clear, predictive domain-general accounts, these data remain a taunting reminder of how much we have yet to learn about language. 

[plusvoice]

I have no beef with computational phonology, but this strikes me as particularly navel-gazing. It smells like the kind of computationalist hubris that brought us neo-phrenology. Besides, this paragraph would be far stronger if the first two sentences were cut.

[plusvoice]

Baldo et al. (2011). Role of the precentral gyrus of the insula in complex articulation. Cortex.

[plusvoice]

Douglas & Martin. (2004). Neuronal Circuits of the neocortex. Annual Reviews of Neuroscience.

[plusvoice]

IFG demonstrates highest degree of phase locking to speech envelope

Assaneo et al. 2019. Nat Neuro.

[plusvoice]

Absolutely bonkers if true.

Tanner & Van Hell. 2014. Neuropsychologica.

[plusvoice]

full citation: Boomershine, A., Hall, K.C., Hume, E. and Johnson, K., 2008. The impact of allophony versus contrast on speech perception. Contrast in phonology: Theory, perception, acquisition, 13, pp.145-172.

[plusvoice]

Something has gone horribly awry. Neither of those accents looks grave at all.

[plusvoice]

A hammer, and the world is your nail.