new ideas

[Brian MacWhinney]

Dear Aubrey,

Right, there is certainly a big emphasis this time of year on  
marketing.  That hadn't occurred to me.

I much agree with your line of thinking that foregrounds principles  
such as spatial symmetry, embedding, and recursion as precursors to  
similar functions in language.  We all perceive the salience of  
symmetry in houses and drawings.  But, like Lise, I would argue that   
symmetry cannot apply directly to language on the sentential level,  
since judgments of symmetry require the copresence of all pieces and  
language evolves through a rapidly fading temporal medium.  Lise's  
example of saying your phone number backwards was lovely.  There may  
be some symmetry effects at the level of the word and syllable,   
since those units are perceptually copresent.  There could also be  
some symmetry effects on the discourse or rhetorical levels, but  
those would be probably backed up by long-term training in rhetorical  
form.

In regards to the speed of evolutionary change, my remarks about  
glacial change applied not to historical change or creolization, as  
in the case of the emergence of Nicaraguan sign language, but to the  
emergence of traits in the species.  When I think of language  
evolution, I am thinking about that, not about linguistic diachrony.

Having said this, I remain curious what evidence there is for  "truly  
new" evolutionary ideas.  Are you suggesting that binding and  
structural dependency are evolutionary new ideas?  Let us focus first  
on binding, which is a clear phenomenon and which you mention in your  
message.  I have become convinced that binding constraints emerge not  
from c-command but from pragmatic factors derived from perspective- 
switching (see my chapter in Pecher and Zwaan, 2005, "Ground  
Cognition).  In this account, the child learns the binding principles  
by figuring out the cues to perspective switching.  If we believe  
that the binding phenomena arise directly from c-command than that  
starts indeed to look like very much like a new evolutionary idea.   
But if binding is expressing perspective switching and flow, then it  
seems to be more closely pinned to cognitive developments that have  
emerged over millions of years in the context of shared social  
action.  In that case, this is not a truly new idea, but an old  
function expressed by new forms, as Liz, Werner, Mayr, and Darwin  
would say.

Structural dependency is another matter.  Children are clearly able  
to process structures with linked grammatical relations built up  
through recursion. But is recursion itself a new idea or an old idea  
with new clothing?  We see the core of recursion in the child's early  
learning and processing of item-based patterns.   One can see  
precursors of item-based processing in the motor world, but the shift  
of item-based slot filling to language does seem rather remarkable.   
Is it a truly new idea in evolutionary terms?  As you know, that is  
at the core of much recent argumentation and I would say that the  
jury is still very much out on that one.  Perhaps some other readers  
have ideas about that.

--Brian MacWhinney

On Dec 19, 2005, at 9:08 PM, aubrey at pigeonpostbox.co.uk wrote:
Dear Brian

You're right in guessing that I wasn't invoking divinity. My  
metaphors were festive season ones. But in order to extend the idea  
of new parts from old to phenomena like binding, endocentricity,  
structure dependency, I think that the old has to be both  
identifiable and such as to help explain why things are the way they  
are. I just don't see how this can be done so as to account for the  
main outlines of these theories and various others. Take  
endocentricity and its most outward appearance with edgemost maximal  
projections. In human perception generally there seems to be a  
default expectation of symmetry. In single-family dwellings built in  
Britain for the last 150 years, a double-fronted layout is rare. But  
in a child's picture of a house, I have yet to see the normal  
terraced layout with the front door on one side. The same preference  
for symmetry seems to be attested in metaphor - with the centre  
favoured over the edges.

In both syntax and phonology, a symmetrical layout would be easy to  
define, with words and sentences built strictly from the middle. But  
whatever the number of cases where this might be appropriate, it  
seems to me that they are so few in number, and that the  
preponderance of asymmetry and directionality in headedness, Wh  
movement, syllable structure, and more, should be treated as highly  
significant.

I make no guesses as to the likely triggering or rate of spread of  
linguistic change. These seem to me to be some of the most  
fascinating questions in linguistics. But from the extreme case of  
the school for the deaf in Nicaragua, where an entirely new language  
is said to have emerged in a single childhood, it seems to me that  
linguistic theory must be at least capable of accounting for change  
at the wildfire end of the scale rather than the glacial.

I certainly wasn't assuming that the speed of change might provide a  
way of telling whether a given phenomenon was a case of new from old  
or speculating as to how this might be done. None of the cases are  
simple.  Obviously the argument needs to be in detail. I was allowing  
that there might be a number of cases of new from old, and listed  
some cases where this seems to me most plausible. I was suggesting  
only that the new from old model may not be the only one, and that  
some changes may have been just by the odd roll of the genetic dice -  
to get back to the festive season,

Aubrey

On 17 Dec 2005, at 21:18, Brian MacWhinney wrote:

Dear Aubrey,

    Can you provide an example of an evolutionary "new idea" that  
does not arise from old parts?  Without concrete examples of this,
I have no idea about how to distinguish canny marketing from crude  
cannibalism.   Of course, intervention from a Divine Marketing  
Department will work, but I don't assume that you have that in mind.   
Maybe what you have in mind is something like a "powerful idea" that  
arises in the usual way in one evolutionary configuration, but then  
spreads like wildfire because of the adaptive advantage it provides.
Of course evolutionary wildfires are usually something more like  
glaciers that advance at the pace of a millimeter a millenium, right?

--Brian MacWhinney

On Dec 16, 2005, at 8:51 PM, aubrey at pigeonpostbox.co.uk wrote:

Speaking as one who believes in Darwinism, and (more or less) in the  
restrictive Hauser, Chomsky, Fitch hypothesis, and thus not in  
Uriagareka's exaptation, it seems to me that indeed language  
improvises with cannibalised parts, association, projection, range,  
etc,, but roughly once in every million years the evolutionary  
Marketing Department comes up with a new idea. This gives us the  
eight or so language universals, structure dependency, endo- 
centricity, binding principles, etc., since the point of human  
divergence. The interesting questions, it seems to me, are: in what  
order of things did the canny marketing prevail over the crude  
cannibalism? And: Why?

Aubrey Nunes
PhD, FRSA, MRCSLT
Director Pigeon Post Box, Ltd.,
52, Bonham Road,
London SW2 5HG
0207 652 1347










-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/info-childes/attachments/20051220/6cef3c55/attachment.htm>