Analytic languages and their function. (4)
lise.menn at colorado.edu
Mon May 29 16:19:56 UTC 2006
ok, a few notes on child language here, though this is tangential to
the issue of the theoretical cycle of typologies, and problematic in
its relations to pidgins. [And on the topic of pidgins, remember
that - definitionally! - speakers of pidgins have some OTHER
language as their native language. They are not naive as to how
languages are constructed - e.g. that they have embedding - even
though the various speakers of the pidgin - again by definition -
have different native languages to draw on. Sociologically it seems
very weird to assume - as the old bioprogram claims apparently do -
that a substantial fraction of children growing up surrounded by a
pidgin (and thus on their way to becoming the first generation of
creole speakers) do not also hear parents and parents' friends
speaking a fully-developed human language, even if they themselves do
not become fully able to speak that language.]
First: Children learning English and other languages often have a mix
of longer unanalyzed phrases, words + 'fillers', and single words.
Ann Peters has been publishing on this since 1977; go to http://
A detailed diachronic study of one case is a dissertation (available
via University Microfilms) by my student Andrea Feldman, Colorado; a
portion of her data are in
Feldman, Andrea, & Lise Menn (2003). Up close and personal: the
development of filler syllables. Journal of Child Language, 30:4,
Secondly, having a modest amount of inflectional morphology (say a 2-
way marking of person or case distinction) is not tantamount to
having a full grammar - NOR is it the same as having a totally
monomorphemic language. Children who use accusative case for things
they are throwing or eating and nominative case for someone who is
doing something do have a case contrast - but it's very restricted -
and it may be almost completely predictable by the animacy of the
noun (nominatives animate, accusatives inanimate), as early pronoun
use may be in English. It does not make sense to be dichotomous
here, to say 'either there's a grammar or there's no grammar'.
Third: What does it mean to say that a child who says 'dog' is or is
not using a noun? If she is pointing to a dog, you can say that
she's referring to a concrete object. Whether that is sufficient for
it to be a noun is going to depend on your theory of grammar.
Suppose she also says 'byebye' (an extremely common first word) when
people are leaving her house. And 'no' when she pushes things away
that she doesn't want to eat. You've got a contrast now, but that's
still not sufficient; 'dog' doesn't become a noun just because 'bye-
bye' and 'no' aren't nouns, and adding a verb to the mix won't do it
either. Again, consider all the properties you expect a word to have
before you call it a noun or a verb. Whatever your list is, evidence
that children's early words have the full set of properties that you
want will not be forthcoming - but evidence for SOME of them will be
People working in the tradition of Martin Braine - now, most
visibly, Michael Tomasello & colleagues - argue for a gradual
development of language-wide 'parts of speech' from very local
privilege-of-occurrence classes. Try Veneziano's 'The Emergence of
Noun and Verb Categories in the Acquisition of French', pdf at http://
cogprints.org/3044/01/Veneziano.pdf for one well-articulated approach.
On May 28, 2006, at 4:02 PM, A. Katz wrote:
> Steve Long wrote:
>> I think this is where the disagreement is. There's good reason to
>> believe that those Turkish speakers starting out with inflectional
>> morphology are not doing anything different from analytic speakers
>> out with non-inflected words. The grammatical implications have
>> for the most
>> part not entered into either of their speeches. Many researchers
>> have tried
>> hard to find grammar invery early development, but I believe that a
>> disciplined objective review will show that it is just not there.
> Conversely, I'm not sure that beginning English speakers at the one
> stage are actually using nouns when they say things like "dog" or
> or "Mommy". I don't think the category noun can have any
> significance if
> there isn't another category to contrast it with.
>> I'm not sure what theory of mind means, but it is true that
>> grammar more
>> accurately maps identities, processes and relationships between
>> things in
>> time and space. But just because early languages did not have much
>> grammar did notmean they had not developed into distinctively human
> Theory of Mind refers to an individual's awareness of the minds of
> their context, desires, beliefs and motivations. Humans are known for
> their unusually well-developed theory of mind, but not all humans are
> equally endowed with this ability. Autistics are said to have
> considerable deficits in theory of mind, regardless of how high
> their IQ may be
> in other areas. Deficits in theory of mind make language
> acquisition very
> difficult, because mind-reading is a good way to jump start first
> learning. However, later in life, autistics often catch up in language
> ability, despite lingering deficits in theory of mind.
> Shifting between 1st and second person is something that come
> naturally to
> speakers with a normal theory of mind. If you don't have TOM, it's
> harder to guess when to use first and when to use second person.
>> Hmmm. My dog knows the difference between yes and no very well.
>> Generalized (abstracted) across a whole slew of situations. Which
>> reinforces the idea that the real functions of language may be more
>> centered in the listener than the speaker.
> Really? Does your dog answer yes/no questions? I have yet to meet a
> who can.
> When I ask my dog if she wants to go out, it's clear that she
> the question. But if she doesn't want to go out, she never gives any
> signal to say so. She just stands there glumly and won't follow me
> to the
> door. If she wants to go out, she wags her tail excitedly and goes
> to the
> door before I do.
> That is not answering "yes" and "no." That is involuntarily showing
> emotional states using body language. She has no volitional control
> that wagging tail.
> There is nothing to keep her from nodding her head "yes" or shaking it
> "no," as she sees me doing. But she never does.
> In what context has your dog used "yes" and "no"?
>> What was early human language like? Well, we can be pretty
>> certain that
>> a full-blown Latin grammar was not inherited from our "pre-human"
>> ancestors. Then how much grammar did the earliest human speech
>> have? I
>> see no reason to think it had anymore than the simplest, least
>> grammatical pidgin. Any other conclusion calls for a deus ex machina
>> that somehow implanted grammar full-blown inthe form of a
>> Chomskayan language
>> mechanism, which I believe violates the
>> naturalistic assumption of science.
> No, there are other conclusions that do not require us to adopt
> mechanisms or violate the naturalistic assumptions of science.
> The idea of individual words carrying meaning without a system of
> contrasts is not naturalistic. When you imagine people talking in
> verbs and prepositions, and you think of them building up their
> grammatical system gradually, one word at a time, that is not
> naturalistic. How long do you imagine our primitive ancestors spent
> at the
> one word -- or rather, one morpheme stage? How long a period from
> the use
> of the "first" word to the discovery of the second word? There is a
> critical mass required before a language can be useful. A one word
> language serves no function, and nobody will wait around to learn the
> second word.
> Our ancestors, before they were truly human, already had a
> system much like that of present-day primates. Do you think they would
> discard it for something whose communicative function was much
> weaker, in
> the hopes of one day working their way up to modern language?
> Dr. Aya Katz, Inverted-A, Inc, P.O. Box 267, Licking, MO
> 65542 USA
> (417) 457-6652 (573) 247-0055
Lise Menn Office: 303-492-1609
Linguistics Dept. Fax: 303-413-0017
295 UCB Hellems 293
University of Colorado
Boulder CO 80309-0295
Professor of Linguistics, University of Colorado, University of Hunan
Secretary, AAAS Section Z [Linguistics]
Lise Menn's home page
"Shirley Says: Living with Aphasia"
Japanese version of "Shirley Says"
Academy of Aphasia
More information about the Funknet