Wednesday, April 2, 2008

Perspective, Language Evolution & Genes

As I wrote in my last post, according to Edmund Blair Bolles, evolang08 can be seen as an all-out attack on the Chomskian paradigm. Whereas Bolles’ thinks that the downfall of the Chomskian paradigm is near, I am more skeptical. Of course, many of the talks at evolang08 seem to have dealt severe blows to general conceptions of Chomskyan linguistics, but given that the field of language evolution studies was opposed to many of Chomsky’s ideas from the start, this seems a natural development.

So will the Chomskyan Paradigm simply go extinct in the sense of Kuhn’s (1962) “Structure of Scientific Revolutions”? Clearly in this field people like Juan Uriagereka are in the minority, and this speaks for the diminishing influence of some Chomskian ideas, especially ‘syntactocentrism’, i.e. the idea that "that it is 'as if syntax carved the path interpretation must blindly follow.'" (Chomsky 2007: 24) in general, but on the whole I think that the Chomskyan research paradigm is alive and well, speaking in terms of ‘believers’, as Derek Bickerton (2008) puts it, in the generative enterprise.

EBB himself writes that: “It is only the odd ones among them who are interested enough in the evolution of language to even think of attending such a conference, but the absence of all the leaders of the effort to understand the evolution of generative grammar—Steven Pinker, Marc Hauser, Tecumseh Fitch, Ray Jackendoff, and Paul Bloom — cannot be entirely a coincidence.”

An example of biologically informed research in linguistics – Biolinguistics – in the Chomskyan tradition is the open access journal with the same title which was inaugurated in 2007 and whose second issue appeared this week. The second issue features two very cool papers.

The first, by Boban Arsenijević (2008), of whom I must admit I’d never heard before, is about how we got “From Spatial Cognition to Language.” Arsenijević argues that cognitive mechanisms for spatial computation switched from domain-specific to a domain-general non-spatial use, enabling the production of generative conceptual structures, i.e. complex kinds of displaced concepts which could be located in some kind of ‘cognitive map’ which weren’t inextricably bound to a spatial situation anymore, but instead became a ‘decoupled representation’ (Sterelny 2003).

This cool proposal is also consistent with Brian MacWhinney’s “Perspective Hypothesis”, which argues that: Perspective taking is an essential property of human communication and higher-level cognition, and that

"1. Perspective taking operates online using images created in five systems: direct experience, space/time deixis, plans, social roles, and mental acts.

2. Language uses perspective taking to bind together these five imagery subsystems.

3. Grammar emerges from conversation as a method for supporting accurate tracking and switching of perspective.

4. By tracing perspective shifts in language, children are able to learn the cognitive pathways and mental models sanctioned by their culture.“ (Macwhinney 2005: 198)

I would argue that we can visualize this new representational perspectival format with Bühler’s (1934) ‘coordinate system of subjective orientation’, i.e. as a systemic space (Köller 2004) which through communication is made available to both interlocutors, becomes a shared, ‘mutually manifest’ (Eilan 2005) representation.

Similarly, in a paper for evolang 08, Benôit Dubreuil argues that the key shift to modern human behavior, such as language and other shared symbolic artifacts, was a “domain-general cognitive change”, namely “the capacity to fully represent the point of view of other, what psychologists call ‘level-2 perspective-taking’” (see, e.g., Flavell 1988).


There is another really cool paper in this issue of Biolinguistics called “Languages and Genes: Reflections on Biolinguistics and the Nature–Nurture Question” by D. Robert Ladd, Dan Dediu & Anna R. Kinsella of Edinburgh University. Ladd & Dediu published a really cool paper in the Proceedings of the Academy of Sciences in 2007 (Dediu & Ladd 2007), in which they described their two genes related to brain growth and development, ASPM and Microcephalin, which show signs of relatively recent natural selection and exist in a more recent (ASPM -D; about 5,800 years old, and MCPH-D, about 37,000 years old) and an older form (ASPM, MCPH).
It turns out that the geographical distribution of these genes is highly correlated with whether people speak tonal languages (like Chinese) or non-tonal languages (like English). In general, populations speaking tonal languages possess the older forms, and populations speaking non-tonal languages have the D (for derived)-versions of the genes, though there of course are exceptions and cases of contamination. In their present paper Ladd et al. (2008) discuss the importance of genetic findings for the linguistic theory. They clarify that they

“are not proposing any sort of deterministic relation between genes and language, only a very indirect and probabilistic one; we certainly are not suggesting that there are 'genes for Chinese'. But we believe that the broad outlines of an explanation based on the interaction of bias and cultural transmission are very plausible indeed“ (Ladd et al. 2008: 118).

They also argue against the notion that the linguistic effects of ASPM-D and MCPH-D were the driving forces of the genes’ natural selection, but rather see them as selectively neutral side-effects, as there are no obvious differences between tonal and non-tonal languages in regards to biological fitness, as both are equally suited to support and scaffold all sorts of varieties of complex human behaviors. (Ladd et al. 2008: 120)
To take an example, in the time of early skyscraper-building in America, a lot of the construction workers were Mohawks, because they were free from giddiness. And although the modern urban environment was completely different from that of their own homes, as were the affordances of steel construction work, they constantly used their native language to communicate about their work, had absolutely no problem to adapt their linguistic output to this new world, and even taught some of their co-workers their language to facilitate communication.

In all, Ladd et al. (2008) propose to move beyond dualistic nature or nature-view and take seriously the fact that

"there is a fundamentally complex and irreducible interaction between one’s genes and one’s language and culture — between nature and nurture” (Ladd et al. 2008: 120).
As an example they cite the recent finding that breastfeeding can indeed have positive effects on a baby’s IQ, but only if she possesses a certain variant of the FADS2 gene (Caspi et al. 2007, Ladd et al. 2008: 120f.).

In their view this also raises problems with static generativist views of language capacities as a perfect system, because from an evolutionary point of view, it makes much more sense to treat language competency as a complex dynamic system, which instead is “diverse and dynamic” and constantly changing on the timeline of both cultural transmission as well as individual development (Ladd et al. 2008: 121).

Ladd et al. see their results on the genetic influences “on typological linguistic features” as being in favor of the claim that there are innate linguistic ‘biases’, “deep cognitive, and ultimately genetic, causes” (Ladd et al. 2008: 121) of some properties of language, thus supporting the Chomskyan research paradigm in some ways. However, they also remark that

Linguistic theorizing in general, and biolinguistics in particular, must take into account and integrate the idea that human linguistic capacities are variable and probably still evolving" (Ladd et al. 2008: 121).

Moreover, they are also opposed to Chomsky’s stance on the evolution of language, which he renewed in the first issue of Biolinguistics (Chomsky 2007), in which he wrote that:

"The core principle of language, unbounded Merge, must have arisen from some rewiring of the brain, presumably the effect of some small mutation” (Chomsky 2007: 22).
And that this property yielded
“a language of thought, later externalized and used in many ways” (Chomsky 2007: 24).

In contrast, Ladd et al. hold that

language has evolved through the standard mechanisms of evolutionary biology, in a gradual manner, and that it continues to do so” (Ladd et al. 2008: 122).

In my next post I’ll come back to Chomsky’s position on these issues and some neuroscientific evidence that argues against it.

References:

Arsenijević, Boban (2008): From Spatial Cognition to Language Biolinguistics 2.1, 3–23.

Bickerton, Derek (2008): Bastard Tongues: A Trailblazing Linguist Finds Clues to Our Common Humanity in the World's Lowliest Languages. New York: Hill and Wang

Bühler, Karl (1934): Sprachtheorie: Die Darstellungsfunktion der Sprache. Stuttgart: Fischer.

Caspi, Avshalom, Benjamin Williams, Julia Kim–Cohen, Ian W. Craig, Barry J. Milne, Richie Poulton, Leonard C. Schalkwyk, Alan Taylor, Helen Werts & Terrie E. Moffitt. (2007) Moderation of breastfeeding effects on the IQ by genetic variation in fatty acid metabolism. Proceedings of the National Academy of Sciences of the USA 104, 18860-18865.

Chomsky, Noam (2007): Of Minds and Language. In: Biolinguistics 1, 9-27.

Dediu, Dan & D. Robert Ladd. (2007) Linguistic tone is related to the population frequency of the adaptive haplogroups of two brain size genes, ASPM and Microcephalin. Proceedings of the National Academy of Sciences of the USA 104, 10944-10949.

Eilan, Naomi (2005): Joint Attention, Communication, and Mind. In: Naomi Eilan, Christoph Hoerl, Teresa McCormack und Johannes Roessler (Eds.): Joint Attention: Communicationd and Other Minds. Oxford: Clarendon Press (Issues in Philosophy and Psychology), 1-33.

Flavell, John H. (1988): The Development of Children’s Knowledge about the Mind: From Cognitive Connections to Mental Representations. In: Janet W. Astington, Paul L. Harris und David R. Olson (Eds..): Developing Theories of Mind. Cambridge: Cambridge University Press, 141-172.

Kuhn, Thomas S. (1962): The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

Ladd, D.Robert, Dan Dediu & Anna R. Kinsella (2008): Languages and Genes: Reflections on Biolinguistics and the Nature–Nurture Question. In: Biolinguistics 2.1, 114–126

MacWhinney, Brian (2005): The Emergence of Grammar from Perspective. In: Diane Pecher und Rolf A. Zwaan (Eds.), The Grounding of Cognition: The Role of Perception and Action in Memory, Language, and Thinking. Cambridge: Cambridge University Press, 199-223.

Sterelny, Kim (2003): Thought in a Hostile World: The Evolution of Human Cognition. Malden: Blackwell.


2 comments:

Anonymous said...

You quote Chomsky:
"The core principle of language, unbounded Merge, must have arisen from some rewiring of the brain, presumably the effect of some small mutation” (Chomsky 2007: 22).

- and then quote Ladd:
“language has evolved through the standard mechanisms of evolutionary biology, in a gradual manner, and that it continues to do so” (Ladd et al. 2008: 122).

- and claim there is some contrast or conflict here. How so? The "standard mechanism of evolutionary biology" is diversity winnowed by natural selection, and diversity mostly arises from... well, from mutation, no?

As for "syntactocentrism", this is a nifty term of abuse invented by Jackendoff to sound like "racism" or "sexism". A neat academic dirty trick, but nothing more than that.

in reality, to the extent an intelligent linguist might hold that that semantics or phonology interprets the syntax rather than the other way round, it's because that's a claimed research result, not a "belief". It's what most detailed, serious attempts to model the interaction of syntax with other components seem to find. And, by the way, it is also a view that is occasionally challenged in a serious way too, on specific factual grounds. See, for example, Danny Fox's arguments that syntactic operations may be blocked when they fail to produce semantic distinctions. Or suggestions going back as far as work of Ken Hale (Chomsky's colleague) that phonological ill-formedness can block syntactic operations. All this work faces the challenge, however, of explaining the many ways in which semantics and phonology fail to influence syntax. This poses a research problem, and people (never cited by Jackendoff) actually work on these issues. It's complex stuff, but fascinating. And I doubt you hear about it at evolang08, because how many people there actually know real linguistics?

The fact is, real linguistics is a field that discovers empirical puzzles about language and tries to make sense of them. The vague, high-level discussion of linguistics that goes on at places like evolang08 is almost entirely beside the point, since it fails to come to grips with the content of the field it claims to be demolishing.

And yes, alas, much the same can be said of the small group that claims to be Chomskyan and indulges in their own fuzzy high-level speculation - the Biolinguistics crowd, for example. For every bad action, I guess, there is an equal and opposite bad reaction. But that is a story for another time.

Michael said...

Thanks for your interesting post, I want to adress your objections in two parts:
1.
I'm sorry if I haven't made this clear enough, but people like Ladd et al., Simon Kirby, and Morten Christiansen, think of our propensity for language as possibly consisting of a variety of small 'biases' that attune us to certain characteristics of linguistic input, without necessarily prespecifiying any mechanisms.
I thought of their quote as implying that language is matter of descent with modification of gradual accumulation of small changes which in effect then lead to phenotypic consequences. But with Chomsky you get the idea that he wants to have just one mutation which immediately sparks an icnredibly changed phenotype, which sounds a lot like a saltational 'hopeful monster' to me, which , as Pinker & Bloom (1990) have forcefully argued. Also, Chomsky doesn't mention any of the rich work on non-human primates regarding the likely precursors of complex categorization, (e.g. (Cheney & Seyfarth 2007) or on the other foundations of language, which in my opinion, which makes his "small mutation" really sound a bit like "abdracadra" as Derek Bickerton called it. I'm also not sure whether unbounded merge is 'the' core principle of language. It might be more apt to describe one of the key cognitive innovations in human cognition not as unbounded merged yielding a language of thought, but as a gradual expansion of prefrontal dominance over other brain processes among with other neural chances, yielding further and further 'large scale neural integration'(Donald 2006) and the ability of 'conceptual blending', of unified and overlapping conceptual representations coming from different domains, what I have called the ability to integrate various conceptual strucutres into a 'shared systemic space'.
i.e. : http://sharedsymbolicstorage.blogspot.com/2008/01/cognitive-foundations-of-perspective.html

Given that there is a "Continuity of the conceptual system across species" (Barsalou 2005), I'd say that the key components of language are much more likely to have evolved via gradual mutational processes.
(please see also this post by Edmuin Blair Bolles:)

http://ebbolles.typepad.com/babels_dawn/2008/02/chomskys-theory.html

This is, of course, not really incompatible with Chomsky's ideas, especially if there really is some combination of evo-devo and Chomskian linguistics, and you could see this all as just a matter
of emphasis. The final mutation could well be a small one, but I think such a statement highlights some aspects of language evolution, i.e. its exceptionality, and uniqueness, and hides others, i.e. the continuity of conceptual structures and social cognitive foundations of language. (see, e.g. my posts on Baboon Metaphysics and Brain Evolution). I still think that things like the ability to establish joint attentional and intentional frames, creating shared cognitive spaces, and negoiating meaning, are more fundamental than say, unbounded merge per se, and these abilities have been shown to exist in a rudimentary, context-dependent form in chimpanzees and orangutans.

e.g. http://sharedsymbolicstorage.blogspot.com/2008/03/shared-intentionality-and-linguistic.html

Tomasello et al. 2005, Cartmill & Byrne 2007

2. I know that syntactocentrism is a polemic term, as is the term 'believer' (which is why I pointed out that it came from Derek Bickerton: it was meant as a collouqial aside not as an insult). I know that finding most minimal tree representations of is a taxing formal enterprise, and that its methodology is rooted in deep scientific commitments.
I am sure that that generativist researchers in general don't advertise the kind of straw man Jackendoff attacks (i.e. in Jackendoff 2007 he claims that mainstream generative grammar doesn't have interface rules, something which Adger 2003 clearly makes use of.)
However, comments such as "“as if syntax carved the path interpretation must blindly follow.” (quoted in Chomsky 2007) clearly sound syntactocentric to me. Jackendoff also says that there is no agreed upon view of the lexicon in mainstram generative theory, without for example mentioning the work of his former Brandeis colleague James Pustejovsky.
However, as I said in my other comment, I feel that in the approaches I know of the conceptual-intentional system, and the sensori-motor system is poorly defined, and makes it hard to combine other cognitive approaches with generativism presented this way. If you know of generativist approaches other than that of Jackendoff who treat these systems with the same well-informednes, precision and respect to data from other disciplines as he does, I'd be happy to look at them.

I think that approaches present at evolang08 and generativist approaches may have to offer much to one another, but for this both have to go beyond claims that either syntacticians have no idea of biology, only speculate from their ivory tower, and therefore have nothiung to contribute, or that researchers interested in the mulitdisciplinary facets of language evolution don't "actually know real linguistics" anyway.

Thanks again for your post,

kind regards

Michael

References:


Barsalou, Lawrence W. 2005. “Continuity of the conceptual system across species.” Trends. Cog. Sc. 9.7: 309-311.

Cartmill, Erica A. and Richard W. Byrne (2007): Orangutans Modify Their Gestural Signaling According to Their Audience’s Comprehension. In: Current Biology 17: 1345-1348.

Cheney, Dorothy L. and Robert M. Seyfarth. 2007. Baboon Metaphysics: The Evolution of a Social Mind. University of Chicago Press, Chicago.

Chomsky, Noam (2007): Of Minds and Language. In: Biolinguistics 1, 9-27.

Donald, Merlin (2006) Art and Cognitive Evolution. In M. Turner, (Ed.) The Artful Mind: Cognitive Science and the Riddle of Human Creativity. Oxford University Press.

Pinker, Steven & Paul Bloom 1990. “Natural Language and Natural Selection.” In: Behavioral and Brain Sciences 13.4: 707-726.

Tomasello, Michael Malinda Carpenter, Josep Call, Tanya Behne, und Henrike Moll (2005): Understanding and Sharing Intentions: The Origins of Cultural Cognition. In: Behavioral and Brain Sciences 28:5, 675-691

Turner, Mark. 2003. “Double-Scope Stories.” Narrative Theory and the Cognitive Sciences. Ed. David Herman