Holidays

I am going to be offline for most of the next two weeks. I don't swear not to put up an occasional post, but I am pretty sure that I won't. So, enjoy the time off. See you again in January.

A view from the front row

For those that have not seen this, here is a view of Chomsky by Bev Stohl his long time personal assistant. For what it is worth, I recognize the person that Bev describes more or less the same way that she describes him.  It is not a view that is universally shared, though it is hardly unique (see here).

A long time ago I learned something that has proven to be invaluable (even if it is hard to implement). The unity of the virtues is a myth. What is that? It's the view that people who are virtuous are also smart and that the beautiful are also moral and kind and that the kind are also smart etc. In other words, the virtues come as a package. This view, a natural one I believe, comes with the corollary that one's friends are smart and kind and beautiful and that people whose views you don't agree with are lesser beings along some dimension. As I said, this is a myth, or so I believe. Smart people can be very unpleasant (if not worse) and nice people can be wrong (if not worse). For this reason I try to distinguish people's views from their personalities and try to evaluate them separately. But, like everyone else, I like it when those I admire intellectually are also people I admire personally. Chomsky is one such person. Bev's portrait accurately represents why.

Three readables

Here are some readables.

First, for those interested in some background reading on the DMZTP paper (Nai Ding, Lucia Melloni, Hang Zhang, Xiang Tian and David Poeppel) discussed here can look at Embick & Poeppel (E&P) (part deux) here. This 2015 paper is a sequel to a much earlier 2005 paper outlining the problems of relating work on the brain bases of linguistic behavior and linguistic research (discussed here). The paper situates the discussion in DMZTP against a larger research program in the cog-neuro of language.

E&P identifies three neuroling projects of varying degrees of depth and difficulty.

1.     Correlational neuroling

2.     Integrated neuroling

3.     Explanatory neuroling

DMZTP fits neatly into the first project and makes tentative conjectures relevant to the second and third. What are these three projects. Here’s how E&P describes each (CR=computational/representational and NB=Neurobiological) (360):

Correlational neurolinguistics: CR theories of language

are used to investigate the NB foundations of language.

Knowledge of how the brain computes is gained by

capitalising on CR knowledge of language.

Integrated neurolinguistics: CR neurolinguistics plus the

NB perspective provides crucial evidence that adjudicates

among different CR theories. That is, brain data enrich our

understanding of language at the CR level.

Explanatory neurolinguistics: (Correlational + Integrated

neurolinguistics) plus something about NB structure/

function explains why the CR theory of language involves

particular computations and representations (and not

others).

The whole paper is a great read (nothing surprising here) and does a good job at identifying the kinds of questions worth answering. It’s greatest virtue, IMO, is that it treats results both in linguistics and in cog-neuro respectfully and asks how their respective insights can be integrated. This is not a program of mindless reduction, something that is unfortunately characteristic of too much current NB work on language.

Second, here’s a piece on some big methodological goings-on in physics. The question, relevant to our little part of the scientific universe, is what makes a theory scientific. It seems that many don’t like string theory or multiverses and think them and the thoeires that make use of them unscientific. Philosophers are called in to help clear the muddle (something that physicists hate even the idea of, but times are desperate it seems) and philosophers note that the muddle partly arises from mistaking the hallmarks of what makes something science. Popper and falsificationism is now understood by everyone to be a simplification at best and a sever distortion with very bad consequences at worst.

Chomsky once observed that big methodological questions of what makes something science is usefully focused on those areas of our greatest scientific success. I think that this is right. However, I also think that listening in on these discussions is instructive.  Here’s a place where eavesdropping might be fun and instructive.

Third, here’s a Tech Review article on some recent work by Tenenbaum and colleagues on handwriting recognition (funny, just as cursive is on the cusp of disappearing, we show how to get machines to recognize it. The curious twists and turns of history). The research described is quite self-consciously opposed to deep learning approaches to similar problems. Where does the difference lie? Effectively the program uses a generative procedure using “strokes of an imaginary pen” to match the incoming letters. Bayes is then used to refine these generated objects. In other words, given a set of generative procedures for constructing letters, we can generate better and better matches to input through an iterative process in a Bayes like framework. And there is real payoff. In putting this kind of generative procedure into the Bayes system, you can learn to recognize novel “letters” from a very small number of examples.

Sound familiar? It should. Think Aspects! So, it looks like the tech world is coming to appreciate the power of “innate” knowledge, i.e. how given information can be used and extended.  Good. This is just the kind of stories GGers should delight in.

How’s this different from the deep learning/big data (DL/BD) stuff? Well, by packing in prior info you can “learn” from a small number of examples. Thus, this simplifies the inductive problem. Hinton, one of the mucky-mucks in the DL/BD world notes that this stuff is “compatible with deep learning.” Yup. Nonetheless, it fits ill with the general ethos behind the enterprise. Why? Because it exploits an entirely different intuition concerning how to approach “learning.” From the few discussions I have seen, DL/BD starts from the idea that get enough data and learning will take care of itself. Why? Because learning consists in extracting the generalizations in the data. If the relevant generalizations are there in the data to be gleaned (even if lots of data is needed to glean it as there is often A LOT of noise obscuring the signal) then given enough data (hence the ‘big’ in Big Data) learning will occur. The method described here questions the utility of this premise. As Tenenbaum notes:

“The key thing about probabilistic programming—and rather different from the way most of the deep-learning stuff is working—is that it starts with a program that describes the causal processes in the world,” says Tenenbaum. “What we’re trying to learn is not a signature of features, or a pattern of features. We’re trying to learn a program that generates those characters.”

Does this make the two approaches irreconcilable? Yes and no. No, in that one can always combine two points of view that are not logically contradictory, and these views are not. But yes in that what one takes to be central to solving a learning problem (e.g. a pre-packaged program describing the causal process) is absent form the other. It’s once again the question of where one thinks what the hard problem is: describing the hypothesis space or the movements around it. DL/BD downplays the former and bets on the latter. In this case, Tenenbaum does a Chomsky and bets on the former.

As many of you know, I have had my reservations concerning many of Tenenbaum’s projects, but I am think he is making the right moves in this case. It always pays to recognize common ground. DL/BD is the current home of unreconstructed empiricism. In this particular case, Tenenbaum is making points that challenge this worldview. Good for him, and me.

Brains do linguistic hierarchy

I was going to do something grand in praise of the paper I mentioned in an earlier post by Nai Ding, Lucia Melloni, Hang Zhang, Xiang Tian and David Poeppel (DMZTP) in

Nature Neurosceince (here). However, wiser heads have beaten me to the punch (see the comments sections here). Still, as Morris Halle once noted, we here discuss not the news but the truth, and with a mitigated version of this dictum in mind, I want to throw in my 2 cents (which in Canada, where I am writing this now, would amount to exactly 0 cents, given the recent abandonment of the penny (all amounts are rounded to nearest 0)). So here is my summary judgment (recall, I AM NO EXPERT IN THESE MATTERS!!!). It is the best neurolinguistics paper I have ever read. IMO, it goes one step beyond even the best neuro-ling papers in outlining a possible (as in ‘potential’) mechanism for a linguistically relevant phenomenon. Let me explain.

The standard good neuroling paper takes linguistically motivated categories and tries to localize them in brain geography. We saw an example of this in the Frankland and Greene paper wrt “theta roles” (see here and here) and in the Pallier et. al. paper for Merge (see here). There are many other fine examples of this kind of work (see comment section here for other many good references)[1]. However, at least to me, these papers generally don’t show (and even don’t even aim to show) how brains accomplish some cognitive task but try to locate where in the brain it is being discharged. DMZTP also plays the brain geography game, but aims for more. Let me elaborate.

DMZTP accomplishes several things.

First, it uncovers brain indices of hierarchy building. How does it do this? It isolates a brain measure of on-line sentence parsing, a measure that “entrains” to (correlates with) to linguistically relevant hierarchy independently of prosodic and statistical properties of the input. DMZTP assume, as any sane person would, that if brains entrain to G relevant categories during comprehension then these brains contain knowledge of the relevant categories and structures. In other words, one cannot use knowledge that one does not have (cannot entrain to data structures that are not contained in the brain). So, the paper provides evidence that brains can track linguistically significant categories and rationally concludes that the brain does so whenever confronted with linguistic input (i.e. not only in artificial experimental conditions required to prove the claim, but reflexively does this whenever linguistic material is presented to it).

Showing this is no easy matter. It requires controlling for all other sorts of factors. The two prominent ones that DMZTP controls for are prosodic features of speech and the statistical properties of sub-sentential inputs. Now, there is little doubt that speech comprehension exploits both prosodic and statistical factors in parsing incoming linguistic input. The majority opinion in the cog-neuro of language is that such features are all that the brain uses. Indeed, many assume that brains are structurally incompatible with grammatical rules (you know, neural nets don’t do representations) that build hierarchical structures of the kind that GGers have been developing over the last 60 years. Of course, such skepticism is ridiculous. We have scads of behavioral evidence that linguistic objects are hierarchically organized and that speakers know this and use this on line.[2] And if dualism is false (and neuro types love to rail against silly Cartesians who don’t understand that there are no ghosts (at least in brains)), then this immediately and immaculately implies that brains code for such hierarchical dependencies as well.[3] DMZTP recognizes this (and does not interpret its results Falkland&Greenishly i.e. as finally establishing some weak-kneed hair brained linguistic’s conjecture). If so, the relevant question is not whether this is so, but how it is, and this resolves into a series of other related questions: (i) What are the neural indices of brain sensitivity to hierarchy? (ii) What parts of the brain generate these neural markers? (iii) How is this hierarchical information coded in neural tissue? and (iv) How do brains coordinate the various kinds of linguistic hierarchical information in online activities? These are hard question. How does DMZTP contribute to answering them?

DMZTP shows that different brain frequencies track three different linguistically relevant levels: syllables, phrases and sentences. In particular, DMZTP shows

that cortical dynamics emerge at all timescales required for the processing of different linguistic levels, including the timescales corresponding to larger linguistic structures such as phrases and sentences, and that the neural representation of each linguistic level corresponds to timescales matching the timescales of the respective linguistic level (1).

Not surprisingly, the relevant frequencies go from shorter to longer. Moreover, the paper shows that the frequency responses can only be accounted for by assuming that the brain  exploits “lexical, semantic and syntactic knowledge” and cannot be explained in terms of the brain’s simply tracking prosodic or statistical information in the signal.

The tracking is actually very sensitive. One of the nicest features of DMZTP is that it shows how “cortical responses” change as phrasal structure changes. Bigger sentences and phrases provide different (yet similar) profiles to shorter ones (see figure 4). In other words, DMZTP identifies neural correlates that track sentence and phrase structure size as well as type.

Second, DMZTP identifies the brain areas that generate the neural “entrainment” activity they identified. I am no expert in these matters, but the method used seems different from what I have seen before in such papers. They used “intracranial cranial” electrodes (i.e. inside brains!) to localize the generators of the activity. Using this technique (btw, don’t try this at home, you need hospitals with consenting brain patients (epileptics in DMZTP’s case) who are ready to allow brain invasions), DMZTP shows that the areas that generate the syllable, phrase and sentence “waves” spatially dissociate.

Furthermore, they show that some areas of the brain that respond to phrasal and sentential structure “showed no significant syllabic rate response” (5). In the words of the authors:

In other words, there are cortical circuits specifically encoding larger, abstract linguistic structures without responding to syllabic-level acoustic features of speech. (5)

The invited conclusion (and I am more than willing to accept the invitation) is that there are neural circuits tuned to tracking this kind of abstract linguistic information. Note: This does not imply that these circuits are specifically tuned to exclusively tracking this kind of information. The linguistic specificity of these brain circuits has not been established. Nor has it been established that these kinds of brain circuits are unique to humans. However, as DMZTP clearly knows, this is a good first (and necessary) step towards studying these questions in more detail (see the DMZTP discussion section). This, IMO, is a very exciting prospect.

The last important contribution of the DMZTP lies in a speculation. Here it is:

Concurrent neural tracking of hierarchical linguistic structures provides a plausible functional mechanism for temporally integrat­ing smaller linguistic units into larger structures. In this form of concurrent neural tracking, the neural representation of smaller linguistic units is embedded at different phases of the neural activity tracking a higher level structure. Thus, it provides a possible mechanism to transform the hierarchical embedding of linguistic structures into hierarchical embedding of neural dynamics, which may facilitate information integration in time. (5) [My emphasis, NH]

DMZTP relates this kind of brain wave embedding to mechanisms proposed in other parts of cog-neuro to account for how brains integrate top-down and bottom-up information and allows for the former to predict properties of the latter. Here’s DMTZP:

For language processing, it is likely that concurrent neural tracking of hierarchical linguistic structures provides mechanisms to generate predictions on multiple linguistic levels and allow interactions across linguistic levels….

Furthermore, coherent synchronization to the correlated linguistic structures in dif­ferent representational networks, for example, syntactic, semantic and phonological, provides a way to integrate multi-dimensional linguistic representations into a coherent language percept just as tempo­ral synchronization between cortical networks provides a possible solution to the binding problem in sensory processing. (5-6)

So, the DMZTP results are theoretically suggestive and fit well with other current theoretical speculations in the neural literature for addressing the binding problem and for providing a mechanism that allows for different kinds of information to talk to one another, and thereby influence online computation.

More particularly, the low frequency responses to which sentences entrain are

… more distributed than high-gamma activity [which entrain to syllables, NH], possibly reflecting the fact that the neural representations of different levels of linguistic structures serve as inputs to broad cortical areas. (5)

And this is intriguing for it provides a plausible way for the brain to use high level information to make useful predictions about the incoming input (i.e. a mechanism for how the brain uses higher level information to make useful top-down predictions).[4]

There is one last really wonderful speculation; the oscillations DMZTP has identified are “related to intrinsic, ongoing neural oscillations” (6). If they are, then this would ground this speech processing system in some fundamental properties of brain dynamics. In other words, and this is way over the top, (some of) the system’s cog-neuro properties might reflect the most general features of brain architecture and dynamics (“the timescales of larger linguistic structures fall in the timescales, or temporal receptive windows that the relevant cortical networks are sensitive to”). Wouldn’t that be amazing![5] Here is DMZTP again:

A long-lasting controversy concerns how the neural responses to sensory stimuli are related to intrinsic, ongoing neural oscillations. This question is heavily debated for the neural response entrained to the syllabic rhythm of speech and can also be asked for neural activity entrained to the time courses of larger linguistic structures. Our experiment was not designed to answer this question; however, we clearly found that cortical speech processing networks have the capacity to generate activity on very long timescales corresponding to larger linguistic structures, such as phrases and sentences. In other words, the timescales of larger linguistic structures fall in the timescales, or temporal receptive windows that the relevant cortical networks are sensitive to. Whether the capacity of generating low-frequency activity during speech processing is the same as the mechanisms generating low-frequency spontaneous neural oscilla­tions will need to be addressed in the future. (6)

Let me end this encomium with two more points.

First, a challenge: Norbert, why aren’t you critical of the hype that has been associated with this paper, as you were of the PR surrounding the Frankland & Greene (F&G) piece (see here and here)? The relevant text for this question is the NYU press release (here). The reason is that, so far as I can tell, the authors of DMZTP did not inflate their results the way F&G did. Most importantly, they did not suggest that their work vindicates Chomsky’s insights. So, in the paper, the authors note that their work “underscore the undeniable existence of hierarchical structure building operations in language comprehension” (5). These remarks then footnote standard papers in linguistics. Note the adjective ‘undeniable.’

Moreover, the press release is largely accurate. It describes DMZTP as “new support” for the “decades old” Chomsky theory that we possess an “internal grammar.” It rightly notes that “psychologists and neuroscientists predominantly reject this viewpoint” and believe that linguistic knowledge is “based on both statistical calculations between works and sound cue structures.” This, sadly, is the received wisdom in the cog-neuro and pysch world, and we know why (filthy Empiricism!!!). So, the release does not misdescribe the state of play and does not suggest that neuroscience has finally provided real evidence for a heretofore airy-fairy speculation. In fact, it seems more or less accurate, hence no criticism from me. What is sad is the noted state of play in psych and cog-neuro, and this IS sad, very very sad.

Second, the paper provides evidence for a useful methodological point: that one can do excellent brain science using G theory that is not at the cutting edge. The G knowledge explored is of Syntactic Structures (SS) vintage. No Minimalism here. And that’s fine. Minimalism does not gainsay that sentences have the kinds of structures that SS postulated. It suggests different generative mechanisms, but not ones that result in wildly different structures. So, you out there in cog-neuro land: it’s ok to use G properties that are not at the theoretical cutting edge. Of course, there is nothing wrong with hunting for Merge (go ahead), but many questions clearly do not need to exploit the latest theoretical insight. So no more excuses regarding how ling theory is always changing and so is so hard to use and is so complicated yada yada yada.

That’s it. My 2 cents. Go read the paper. It is very good, very suggestive and, oddly for a technical piece, very accessible. Also, please comment. Others may feel less enthralled than I have been. Tell us why.

---

[1] I would include some recent papers by Lyna Pylkkanen on adjectival modification in this group as well.

[2] These are two different claims: it could be that the linguistic knowledge exists but is not used online. However, we have excellent evidence for both the existence of grammatical knowledge and its on-line usage.  DMZTP provides yet more evidence that such knowledge exists and is used online.

[3] P.S. Most who rail against dualism really don’t seem to understand what the doctrine is. But, for current purposes, this really does not matter.

[4] Note, the paper does not claim to explain how hierarchical information is coded in the brain. It might be that it is actually coded in neural oscillations. But DMZTP does not claim this. It claims that these oscillations reflect the encoding (however that is done) and that they can be used to possibly convey the relevant information. David Adger makes this point in the comments section of the earlier post on the DMZTP paper. So far as I can tell, DMZTP commits no hostages as to how the G information is coded in brains. It is, for example, entirely consistent with the possibility that a Gallsitel like DNA coding of this info is correct. All the paper does is note that these oscillations are excellent indices of such structure, not that they are the neural bases of this knowledge.

[5] Here’s a completely wild thought: imagine if we could relate phases to the structure of these intrinsic oscillations? So the reason for the phases we have is that they correspond to the size of the natural oscillations which subvene language use. Now that would be something. Of course, at present there is zero reason to believe anything like this. But then again, why exactly phases exist and are the ones there are is theoretically ungrounded even within linguistics. That suggests that wild speculation is apposite.

Once we are on the topic of grad education

Here is a recent piece on grad education in Nature. The "problem" it points to is that academia is producing more PhDs than there are academic positions to fill. This is true, it appears in virtually every domain, including the hard sciences like physics and biology. I am pretty sure that this applies to linguistics as well. I put 'problem' in scare quotes for it is an interesting question what kind of problem this is. Let's stipulate that in the best of all possible worlds PhDs only go to the deserving and all the deserving get the jobs desired. WHat is the best policy when this is not what we have? One reaction is that we should not produce more PhDs than there are academic positions for. Another is that we train PhDs for several kinds of career tracks (with several kinds of PhDs) only some of which are academic and another is that we train PhDs as we have always done but we are up front about the dangers of employment disappointment.  All of these options are reviewed in the Nature piece.

For what it is worth, I am not sure what the right thing to do is. I do know that given this situation we should really be making sure that PhDs do not go into debt in order to get a degree. This requires that grad programs find ways to fully fund their students. As for the rest, I really don't know. There is something paternalistic about putting quotas on PhDs just because landing a job at the end is problematic. We must be clear about the prospects. But we should never lie to students or mislead them,  But once the employment facts are made clear, are there further responsibilities?  Maybe offering courses that make career options more flexible in the worst case? I don't know.  However, this is, sadly, once again, a very important problem that could use some discussion.

When I graduated, jobs were not thick on the ground. Grad students were told that landing a job was chancy, even if one graduated from a good place with a good PhD. This did not serve to deter many of us. We all thought that we we would be ok. Some were. Some weren't. Looking back I am not sure that I would endorse a system that pre-culled us. Why? Because doing grad work was rewarding in itself. It was fun doing this stuff and, moreover, it turns out, in retrospect, that figuring out who would be the successful and who the less successful was not antecedently obvious. There may not be a better system.

This is an important problem. How should we proceed? Grad students out there are especially invited to  chime in.

The brain does linguistic hierarchy

I hope to blog on this paper by Poeppel and friends (the authors are Ding, Melloni, Zhang, Tian and Poeppel) more in the near future. However, for now, I wanted to bring it to your attention. This is an important addition to the growing number of papers that are beginning to use linguistic notions seriously to map brain behavior. There have been earlier papers by Dehaene and Friederici and Frankland and Greene and, no doubt others. But it looks to me that the number of these kinds of papers is increasing, and I take this to be a very good thing. Let's hope it soon puts to sleep the idea,s till prevalent among influential people, that hierarchy is tough or unnatural for brains to do, despite the overwhelming behavioral evidence that brains/minds indeed do it (see here for a monumentally dumb paper).

One warning: some of the hype around the Poeppel & Co paper are reporting it as final vindication of Chomsky's views (see here). From a linguists point of view, it is rather the reverse: this is a vindication of the idea that standard neuro methods can be of utility in investigating human cog-neuro capacities. In fact, the title of the Poeppel & Co paper indicates that this is how they are thinking of it as well. However, the hype does in fact respond to a standing prejudice in the brain sciences and so advertising the results in this way makes some rhetorical sense. As the Medical Press release accurately notes:

Neuroscientists and psychologists predominantly reject this viewpoint, contending that our comprehension does not result from an internal grammar; rather, it is based on both statistical calculations between words and sound cues to structure. That is, we know from experience how sentences should be properly constructed—a reservoir of information we employ upon hearing words and phrases. Many linguists, in contrast, argue that hierarchical structure building is a central feature of language processing.

And given this background, a little corrective hype might be forgivable. By the way, it is important to understand that the result is linguistically modest. It shows that hierarchical dependencies is something the brain tracks and that stats cannot be the explanation for the results discovered. It does not tell us what specific hierarchical structures are being observed and which linguistic structures the might point to. That said, take a look. the paper is sure to be important.

How deep are typological differences?

Linguists like to put languages into groups. Some of these, as in biology, are groupings based on historical descent (Germanic vs Romance), some of long standing (Indo-European vs Ural Altaic vs Micronesian). Some categorizations show more sensitive to morpho-syntactic form (analytic vs agglutinative) and some are tied to whether they got to where they are spoken by tough guys who rode little horses over long distances (Finno-Ugaric (and Basque?)).  There is a tacit agreement that these groupings are significant typologically and hence linguistically significant as well. In what follows, I want to query the ‘hence.’ I would like to offer a line of argument that concludes that typological differences tell us nothing about FL. Or, to put this another way, the structure of FL in no way reflects the typological differences that linguists have uncovered. Or, to put this in reverse, typological distinctions among languages have no FL import. If this is correct, then typology is not a good probe into the structure of FL. And so if your interest is the structure of FL (i.e. if liming the fine structure of FL is how you measure linguistic significance), you might be well advised to study something other than typology.

Before proceeding let me confess that I am not all that confident about the argument that follows. There are several reasons for this. First, I am unsure that the premises are as solid as I would like them to be. As you will see, it relies on some semi-evolutionary speculation (and we all know how great that is, not!). Second, even given the premises, I am unsure that the logic is airtight. However, I think that the argument form is interesting and it rests on widely held minimalist premises (based on a relatively new and, IMO, very important observation regarding the evolutionary stability of FL), so even if the argument fails it might tell us something about these premises.  So, with these caveats, cavils, hedges and CYAs out of the way, here is the argument.

Big fact 1: the stability of FL. Chomsky has emphasized this point recently. It is the observation that whatever change (genetic, epi-genetic, angelic) led to the re-wiring of the human brain thus supporting the distinctive species specific nature of human linguistic facility, whatever change that was, it has remained intact and unchanged in the species since its biological entrance.  How do we know?

We know because of Big Fact 2: any kid can learn any language and any kid learning any language does so in essentially the same way. Thus, for example, a kid from NYC raised in Papua New Guinea (PNG) will acquire the local argot just like a native (and in the same way, with the same stages, making the same kinds of mistakes etc.). And vice versa for a PNGer in NYC, despite the relative biological isolation of PNGers for a pretty long period of time. If you don’t like this pair, plug in any you would like, say Piraha speakers and German speakers or Hebrew Speakers and Japanese. A child’s biological background seems irrelevant to which Gs it can acquire and how it acquires them. Thus, since humans separated about 100kya (trek out of Africa and all that), FL has remained biologically stable in the species. It has not changed. That’s the big fact of interest.

Now, observe that 100k years is more than enough time for evolution to work its magic. Think of Darwin’s finches. As soon as a niche opened up, these little critters evolved to exploit it. And quickly filling niches is not reserved just for finches. Humans do the same thing. Think of lactase persistence (here). The capacity to usefully digest milk products arose with the spread of cattle domestication (i.e. roughly 5-10kya).[1] So, humans also evolutionarily track novel “environmental” options and change to exploit them at a relatively rapid rate. If 5-10k years is enough for the evolution of the digestive system, then 100k years should be enough for FL to “evolve” should there be something there to evolve. But, as we saw above, this seems to be false. Or, more accurately, Big Fact 2 implies Big Fact 1 and Big Fact 1 denies that FL has evolved in the last 100k years. In sum, it seems that once the change allowing FL to emerge occurred nothing else happened evolution wise to differentially affect this capacity across humans. So far as we can tell, all human FLs are the same.

We can add to this a third “observation,” or, more accurately, something I believe that linguists think is likely to be true though we probably only have anecdotal evidence for it. Let’s call this Big Fact 3 (understanding the slight tendentiousness of the “fact” part): kids can learn multiple first languages simultaneously and do so in the same way despite the languages involved. [2] So, LADs can acquire English and Hebrew (a Germanic and Semitic language) as easily as German and Swedish (two Germanic languages), or Navajo and French or Basque and Spanish as easily as French and Spanish or… In fact, kids will acquire any two languages no matter how typologically distinct in effectively the same way. In short, typological difference has no discernable impact on the course of acquisition of two first languages. So, not only is there no ethnically-biologically based genetic pre-disposition among FLs for some Gs over others, there is not even a cognitive preference for acquiring Gs of the same type over Gs that are typologically radically different.

If thess “facts” are indeed facts, the conclusion seems obvious: to the degree that we understand FL as that cognitive-neural feature of humans that underlies our capacity to acquire Gs then it is the same across all humans (in the same sense that hearts or kidneys are, (i.e. abstracting from normal variation)) and this implies that it has not evolved despite apparently sufficient time for doing so.[3]

This raises an obvious question: why not? Why does the process of language acquisition not care about typological differences? Or, if typological differences run deep then why have they had no impact on the FLs of people who have lived in distinct linguistic eco-niches?

Here’s one obvious answer: typological differences are irrelevant to FL. However big these differences may seem to linguists, FL sees these typologically “different” languages as all of a piece. In other words, from the point of view of FL, typological variation is just surface fluff.

Same thing, said differently: there is a difference between variation and typology. Variation is a fact, languages appear on the surface to have different properties. Typology is a mid-level theoretical construct. It is the supposition that variation comes in family types, or, that variation is (at least in part) grammatically principled. The argument above does not question the fact of variation. It calls into question whether this variation is in any FL sense principled, whether the mid level construct is FL significant. It argues it isn’t.

Let me put this last point more positively. Variation establishes a puzzle for linguists in that kids acquire Gs that result in different surface features. So, FL plus a learning theory must be able to accommodate variation. However, if the above is on the right track, then this is not because typological cleavages reflect structural fault lines (or G-attractors) in FL or the learning theory. How exactly FL and learning theories yield distinctive Gs is currently unknown. We have good cases studies of how experience can fix different Gs with different surface properties but I think it is fair to say that there is still lots more fundamental work to be done.[4] Nonetheless, even without knowing how this happens, the argument above suggests that it does not happen in virtue of a typologically differentiated FL.

Let me end with one last observation. Say that the above is correct, it seems to me that a likely corollary is that FL has no internal parameters. What I mean is that FL does not determine a finite space of possible Gs, as GB envisioned. Why not?

Well say that acquisition consisted in fixing the values of a finite series of FL internal open parameters. Then why wouldn’t evolution have fixed the FL of speakers of typologically isolated languages so that the relevant typological parameters were no longer open. On the assumption that “closing” such a parameter would yield an acquisition advantage (fixing parameters would reduce the size of the parameter space, so the more fixed parameters the better as this would simplify the acquisition problem), why wouldn’t evolution take advantage of the eco-niche to speed up G acquisition? Thus, why wouldn’t humans be like finches with FLs quickly specializing to their typological eco-niches? Doesn’t this suggest that parameters are not internal properties of FL?

I am pretty sure that readers will find much to disagree with here. That’s great. I think that the line of reasoning above is reasonable and hangs together. Moreover, if correct, I believe that it is important for pretty obvious reasons. But, there are sure to be counter-arguments and other ways of understanding the “facts.” Can’t wait.

---

[1] The example provided by Bill Idsardi. Thanks.

[2] By two “first” languages I intend to signal the fact that this is a different process from second language acquisition. Form the little I know about this process, there is no strcit upper bound on how many first languages one can simultaneously acquire, though I am willing to bet that past 3 or 4 the process gets pretty hairy.

[3] It also strongly casts doubt on the idea that FL itself is the product of an evolutionary process. If it is, the question becomes why did it stop when it did and not continue after humans separated? Why no apparent changes in the last 100k years?

[4] Charles Yang has a forthcoming book on this topic (which I heartily recommend) and Jeff Lidz has done some exemplary work showing how to think of FL and learning theory together to deliver integrated accounts of real time language acquisition. I am sure that there is other work of this kind. Feel free to mention them in comments.