Friday, September 21, 2007

Thoughts on Evolution of Language Readings

So, I've read the Hauser/Chomsky/Fitch paper, and a few things struck me. For one, they claim that the language faculty, okay, syntax, is a (near) optimal mechanism for connecting the sensory motor system to the conceptual one. If true, why should displacement be a part of it?Why should recursion be a part of it? Ideas may be expressed in discrete units; the Piraha, in fact, may lack recursion altogether. Of course, at the end of the article, the authors claim that recursion may have evolved independently in a domain-general fashion, becoming specialized only later. I'm willing to accept this conclusion, but it says nothing on how our little group characterizes language. All of that is to say, then, supposed feature-checking mechanisms are optimal? For all our spouting of 'elegant rules', I would think a system without movement and D-features would be a better exemplar of the cleanliness of mental operations. I believe there are theories out there that do without movement, so maybe being an evolutionary linguist means taking a hard, difficult look at the level of theoretical complexity that we've devised.

Furthermore, if it turns out that recursion isn't a part of the language faculty (per the suggestion that it's used for other mental concerns, does this mean that birds also have language, albeit with a much reduced vocabulary? I believe the article said said that was a limiting factor to the classification of their abilities. Along this line, there is a professor at Duke who is studying songbird neurolgy in order to decode how language is learned, eventually applying it to human nuerobiology:

Department of Neurobiology, Duke University Medical Center, Durham, North Carolina 27710, USA
Address for correspondence: Eric D. Jarvis, Department of Neurology, Duke University Medical Center, Box 3209, Durham, NC 27710, USA. Voice: 919-681-1680; fax: 919-681-08772. jarvis@neuro.duke.edu

http://www.jarvislab.net/
Vocal learning, the substrate for human language, is a rare trait found to date in only three distantly related groups of mammals (humans, bats, and cetaceans) and three distantly related groups of birds (parrots, hummingbirds, and songbirds). Brain pathways for vocal learning have been studied in the three bird groups and in humans. Here I present a hypothesis on the relationships and evolution of brain pathways for vocal learning among birds and humans. The three vocal learning bird groups each appear to have seven similar but not identical cerebral vocal nuclei distributed into two vocal pathways, one posterior and one anterior. Humans also appear to have a posterior vocal pathway, which includes projections from the face motor cortex to brainstem vocal lower motor neurons, and an anterior vocal pathway, which includes a strip of premotor cortex, the anterior basal ganglia, and the anterior thalamus. These vocal pathways are not found in vocal non-learning birds or mammals, but are similar to brain pathways used for other types of learning. Thus, I argue that if vocal learning evolved independently among birds and humans, then it did so under strong genetic constraints of a pre-existing basic neural network of the vertebrate brain.

7 comments:

sarah a. goodman said...

In the second Hauser/Chomsky/Fitch paper, the writers deny categorizing cognitive capacities as part of the FLN simply because other species seem to share them. Isn't this an awfully derelict peice of methodology, seemingly trumpting the truth of a hypothesis without evidence? It could be that FLN is shared in some way with other specicies, and the Jackendoff paper simply hints at how. Rejecting it simply because it's incongrous with one's own opinion seems like poor science.

Returning to my original comment for this post, HCF place recursion in the FLN and cite comparative data for why it is language-based and unique. But it certainly doesn't need to be there to express thought. Granted, this might be touching issues of adaptive rationale (what is language for), which HCF agreed is a silly thing to discuss at the moment, agreeing either more should be learned about language and that perhaps the question should be rephrased to incorporate initial and historical adaptive pressures. But they do say that once the FLB/FLN proved utile in communication, their characters further evolved as constrained by other, presumably interface, constraints. So is this an admission that recursion is utile for communication at some point? If so, why would this be? Can't we simply use non-embedding structures tied together by co-reference to communicate complex scenarios equally well?

In the appendix, Chomsky mentions that movement is an optimal device to satisfy interface conditions of topics and agency, etc. because it is computationally efficient. I'm not buying this argument completely. It seems that the most efficient way of specifying linguistic details wouldn't involve phrases racing for features to be checked -- why not have these features already checked to begin with, hence obviating movement for a system that just merges? One can then protest that "that's just the way things are", or not -- the syntactic stories we tell could be just that, and movement the wrong description.

sarah a. goodman said...

Oops. Should have changed that first paragraph. The Jackendoff paper certainly wouldn't have hinted how the FLN is present in other animals. Quite the converse, actually.

Alex Drummond said...

Just to pick up on the question of whether movement can be considered an optimal device: Chomsky has emphasized that the sense of optimality he has in mind isn't something that can be determined a priori (i.e. it doesn't correspond to any general, philosophical notion of optimality, whatever that might be). Rather, it's something like, "given that the PF and LF interfaces work like this, what would be the optimal way of relating them?".

As I understand it, the current best argument for why movement is an optimal device goes like this. First, assume that DPs need to be marked both for their thematic role (e.g. Agent, Theme) and for their information status (e.g. Topic, Focus, old information). The mental systems interested in thematic structure look at a different portion of the tree from the systems that look at information structure. This raises a problem, since in order for a DP to be interpreted by both systems, it has to be in two positions at once.

We can imagine a number of different ways to relate one DP to two positions — movement isn't the only option. However, Chomsky has argued that movement comes for free, since it's just the result of remerging an item which has already been merged. In other words, any system which has Merge will also have Move, unless it is explicitly stipulated that each item can only be merged once.

From one point of view, Minimalism is about shifting all apparent imperfections to the interfaces, leaving core syntax as a perfect system which mediates the idiosyncratic requirements of LF and PF in an optimal fashion. One thing that's always puzzled me about Minimalism is the question of whether the interfaces themselves are supposed to be optimal in any sense. As I read him, Chomsky thinks that the PF interface imposes all sorts of arbitrary requirements, but it's less clear whether or not he expects to find any kind of perfection in the design of the LF interface.

So-One said...

I have added to the locker Pinker & Jackendoff’s reply to HCF so that we can see the complete dialogue and see what Fitch-Hauser-Chomsky were responding to. If people are interested, I can also add Gould & Lewontin’s 1979 article “The spandrels of San Marco and the panglossian paradigm: a critique of the adaptationist programme” since it’s a good representation of the view on evolution that HCF uphold.

FHC point out that a lot of the misunderstandings between them and PJ are from the conflation of terminologies. This back-and-forth debate truly highlights the importance of being clear about the assumptions on what is being studied.

Given what HCF say about FLN, I would like see this semester a discussion about what the computation of the FLN would look like. In our dialogues about “language,” we’ve always paired meaning and sound, but this is the incorporated domain of FLB. What are the input-out pairs of FLN?

Last week, some history of science on Galileo, gravity, and friction, helped us realize the methodological advantage of assuming simpler systems as the starting hypothesis. However, Galileo did not support this idea as just a WAY of doing science but rather as a belief on how nature works, on the way nature is. The Galilean method is based on this premise and has been applied at all levels of advances in physics – and is certainly not unique to linguistics’s Minimalist Program.

sarah a. goodman said...

Responding to the dialogue of optimality: good point, Alex, that Move is essentially re-Merge. I had forgotten that. However, something in Minimalism's explaination of optimality strikes me of desperation or arbitrariness, I'm not sure which. How do we know there are two such mental systems that interpret DP information? Is there any evidence for this whatsoever? Or is it just an answer proffered to simply fit the theory, which is itself one way to fit the data? I don't know the background on this. I just suspect it's the above, simply because I've never heard of this justification before, and, were it based on something independently, like fMRI data which shows different areas alight when a DP is processed, in which those areas cannot be ascribed to activity from any other source in the sentence, a tall order, I imagine.

Also, Move might come for free, but the number of applications certainly does not. Wouldn't one application of Merge be more optimal than, say, two or twelve?

sarah a. goodman said...

After reading the the Pinker and Jackendoff proposal "the FL, what's special about it", I have to say their argument seems more promising than what was rendered in the HCF paper. The notion that language evolved for inner thought rather than for communication but we yet have a descended larynx for articulating something that has to be learned externally is more striking than the notion of the vocal tract's size enlargement trickery that HCF propose. I'm also struck by the degree to which the MP influenced HCF's findings and formulations regarding optimality, perfection, and the lonely inhabitant of the FLN, recursion. The MP is far from vindicated, as perhaps both sets of authors have at least once claimed, and yet we see at least little heaps of evidence culled from a variety of sciences that must otherwise be minimized or tenuously re-concluded in order to maintain its evolutionary implications. Just because the MP places pride in proslytizing parsimony shouldn't necessarily mean that the only thing special to language is recursion. Is it me, or does that seem to exclude a good chunk of what language is, to include morphology and phonology, as PJ indicated.

Asad Sayeed said...

Why not exclude them? The way you put it, it sounds like little more than a trivial methodological argument to me.