qrcode

Monday, December 25, 2006

The Evolution of Language -- IV

I began looking at the issue of the evolution of language because of a philosopher friend's claim that it is somehow obvious that human language is a natural evolutionary development of primate communication systems. As far as I can tell, this is false. What is clearer is that this thesis cannot be proved to be true. There is, in fact, not a shred of evidence for it. There have been cognitive developments by humans not found in other primates and that there have been physiological developments in humans not shared by other primates, both of which help to explain how we might have left them so far behind.

Vocalizations by nonhuman primates in the wild do occur, of course, and we may be sure that they have meaning, perhaps even conventional meaning, but these calls differ in a quite fundamental way from human vocal communication. All human languages employ discrete sounds that do not themselves have either conventional meaning nor meaning in context (meaning as significance) that are combined to form sound sequences (morphemes) that have conventional meaning. The word "ban" is comprised of three sounds, none of which have meaning, but these three sounds when combined to form "ban" do have meaning. Moreover, if we string these three sounds together in another order, as in "nab" we get a word that is entirely different in meaning. No primate vocalizations exhibit this critical property. All human languages do. This compositional property is exhibited at every level of linguistic structure.

The notion that primate communication in the wild can usefully be compared to human communication nevertheless persists. In their paper on social cognition and the origin of language, Seyfarth, Cheney, and Bergman argue that call sequences consisting of individual calls, each of which has a different meaning and which is produced by different animals can combine to form complex messages "that [are] interpreted by listeners in a manner that resembles the way we interpret sentences."1 This claim reminds me of the Huey, Dewey, and Louie sentences of cartoon fame in which each of Donald Duck's nephews contributes words in succession that form whole utterances. The key word in the claim of Seyfarth, Cheney, and Bergman's paper is, of course, the term "resembles." There are few words in the scholarly literature that are weaker than this one. I would be surprised that a paper that is this silly could make it into the scholarly literature if I didn't already know that this sort of thing goes on all the time.

So far observation of nonhuman primates in the wild and such efforts to teach language to them while in captivity as was discussed in prior blogs demonstrates that these animals do not have the capacity to learn human-like languages. The failure derives in part from the absence of the ability to create their own or learn to use signaling devices (ASL signs or computer keyboard chips) we provide. Humans learn languages spontaneously without much help and this includes the learning of oral as well as signed language. I believe that there are two reasons for this inability. Nonhuman primates don't have the cognitive ability to learn human language and, if I may channel Ludwig Wittgenstein for awhile, it seems they don't have much interest in doing so. Wittgenstein claimed that if a Lion could talk, we couldn't understand it. I would argue that nonhuman primates lack the human desire to imitate other humans.

With one avenue to the understanding of how human communication arose closed, we must turn to either the fossil record or to studies of the brains of humans and other animals to search for parallels in animals to the linguistic centers of the human brain such as Broca's area (for speech production and language understanding) and Wernicke's area (for sound processing) and connections between them. As for the fossil record, there is little to say. Language per se leaves no fossil record.

The human vocal tract is capable of producing a wide variety of sounds from such garden variety sounds as the vowel of "bad" to the much more complex clicks of some African languages. One reason we can make so wide a variety of sounds is our "descended larynx." This enlargement of the vocal tract allows humans to modulate sounds issued forth from the larynx. A descended/lowered larynx was once believed to be unique to humans and an evolutionary adaptation for language but it seems that there are other animals that have a descended larynx which do not have language. Fitch and Reby claim that there are deer that exhibit this property. They make the point that since deer do not have anything like human speech, we would not want to infer a capacity for human like speech for any species based on fossils exhibiting a descended larynx. They point out that deer exaggerate their body size by further enlarging the length of their vocal tracts which results in sounds of a lower frequency. In general, the lower the frequency of an animal's vocalizations, the larger it is. A need to seem bigger and stronger to survive may be why we developed a lowered larynx, but their are other theories.

One of the interesting ironies of research on nonhuman primate oral productions in the wild and of the teaching of human language to nonhuman primates is that various animal species can imitate human speech such as birds (parrots) and even seals it seems, but nonhuman primates cannot do so. The ability to imitate speech is, of course, of huge importance to language learning for without that capacity children could not learn language. As I noted above, we humans have a strong, innate desire to imitate other humans. Nonhuman primates do not have a desire to imitate us. That is part of the problem of teaching them language. They don't want to be humans. They want to be monkeys or bonobos or gorillas.

Fitch and Reby in the two papers referenced note that a great number of adaptations would have had to be acquired for language as we know it to have evolved. We required not just a descended larynx but a very talented tongue, and quite talented lips, as well as the motor control to coordinate the various physiological events required to create human sounds. In addition, our sound perception abilities had to emerge to allow us to discriminate linguistic sounds from others. This ability to discriminate speech sounds from others occurs very early in life, but as infants become familiar with the sounds of a particular language they, like adults, become less facile perceptually. See Maye, Werker, and Gerken's paper if you can get access to it.

The development of our talented vocal tracts and the motor skills to employ it and the development of our perceptual abilities have purely physiological sides but also required significant cognitive development. Of course, the ability to hear and the ability to discriminate sounds by animals of the same species is hardly restricted to humans. Essentially all hearing animals have that ability and we would expect such animals to develop homologues to our Wernicke's area for sound perception and to our Broca's area for sound production. There is no evidence that these areas are as richly developed in other animals as they are in humans.

What we may be quite sure of is that the different physiological developments required for the production and understanding of speech, did not occur simultaneously. Rather, they must have evolved over a long period of time. Moreover, the purely cognitive side of language will not have developed full blown over night but incrementally and we can imagine that as our vocal abilities increased our computational linguistic-cum-cognitive abilities will have increased as well. From a linguistic point of view, the problem of understanding of the evolution of language is not that there is a "missing link" but that there are a multiplicity of them. So, the claim with which we began -- that the evolution of human language is somehow linked to the evolution of animal modes of communication is false, in my view, but is clearly unprovable. There is no evidence supporting this view whatever beyond the fact that animals, like humans, can communicate. And that doesn't make what they do linguistic.


1Robert M. Seyfarth, Dorothy L. Cheney and Thore J. Bergman, "Primate social cognition and the origins of language" in Trends in Cognitive Sciences Vol.9 No.6 June 2005.


Tweet This!

Sunday, December 10, 2006

The Evolution of Language -- III

I recall reading with fascination how frogs go about catching flies so easily. The claim was made that the neural mechanisms associated with frog eyes preprocess visual data in such a way that"the eye speaks to the brain in a language already highly organized and interpreted, instead of transmitting some more or less accurate copy of the distribution of light on the receptors."1 The result is a rapid determination of what is food, what might be a danger, and what is of no interest. This paper, despite its age, still seems to be one of the most commonly cited papers in the Science Citation Index, Wikipedia says.

It would be surprising if we humans did not have specialized neural structures that facilitate language learning and the use of language given that children learn languages relatively quickly without much teaching and adults use language effortlessly and with great speed. The turns of conversation commonly overlap or occur with minimal or even no gaps between them. This means that we go from hearing and understanding someone else's utterance to planning and then producing one of our own at warp speed.

We humans, like frogs, have specialized neural structures, including several structures that are associated with linguistic processing. Most of you will have heard of Broca's Area and Wernike's Area, both of which are located in the left hemispheres of the brains of most people and, though these areas are in different locations in the left hemisphere, they are connected via the arcuate fasciculus. Wernicke's Area is located toward the back of the left side of the brain and is dedicated to auditory processing. Broca's Area is located forward of Wernicke's Area and is concerned with language understanding and speech production. It is clear from all the research done over the years on language deficits due to brain damage that these separate areas exist and more recent studies employing noninvasive magnetic resonance imaging have provided further evidence. However, it is not yet possible to associate specific linguistic functions associated with understanding and producing speech with specific locations in the brain. One of the reasons is that MRI studies of the brain require the subject to be still but one cannot speak without movement. Having subjects whisper is one way to get around this.

It is no accident that the parts of the brain dedicated to language are on the same side of the brain as those parts that are related to thinking. I have argued that cognitive development is independent of language development, one of the reasons being that children evidence an ability to think before they acquire the ability to use language. Children learn to reach out for things as a way of requesting/demanding them, for instances, before they can make linguistic requests. And, they evidence the ability to make offers nonverbally (by holding an object out for someone to take it). But, the use of language and our ability to think are very intimately connected and one can argue that thinking about abstract concepts depend critically on the ability to define or characterize them in language. Though there is no way to prove this, I suspect that our brains were preadapted for language before our larynx descended in such a way as to favor the production of linguistic sounds. We had reasons to be able to think and communicate using sound before full blown language evolved.

I think that efficiency in language understanding and production has favored the development of languages that are computationally efficient. Chomsky put us on a path of studying grammar from a mathematical perspective, with grammars being represented by some kind of formal system. He also took great pains to disassociate his notion of "sentence generation" from the notion of "computation." I would like to suggest that we take precisely the opposite view. Those (like me) who worked within Chomsky's original paradigm also took a modular view of language wherein each different kind of linguistic phenomenon was relegated to a different module and this, of course, resulted in the need to study the interconnections between "adjacent" modules. So, we had phonology in one module and morphology in another and also had the field of morphophonology to account for their interactions.

In working with a computer scientist,Terry Patten and his student Barbara Becker,1 I was introduced to the notion of expert systems in which compiled knowledge, rather than reasoning from "first principles," is employed in inferencing. This widely accepted approach within computer science in the study of artificial intelligence has proved useful in understanding, among other things, the reasoning of physicians who on noting a few symptoms of an illness and facts about a patient seem very quickly to come to a diagnosis and possible treatment. In each case, the doctor could presumably have reasoned from basic principles of medicine and pharmacology to the same result but, instead, the reasoning is in some sense "short-circuited," thanks to the existence of his or her compiled medical and pharmacological knowledge.

The problem that interested Patten, Becker, and me is that linguistic processing -- from hearing someone say something to replying -- proceeds so rapidly that it is difficult to see how this very rapid processing could happen if one were to assume a modular approach to grammar, conceived as a formal system, which our utterance understanding and production routines must somehow "consult." This would be a quintessential "reasoning from basic principles" approach to the problem. Instead, we opted for the view that a proper study of grammar must be embedded withing a theory of language understanding and production using some computational paradigm.

Linguists are language analysts -- language parsers, if you will -- and it has been very natural to us to assume that we hear linguistic sounds, map these into sound sequences constituting morphemes and map these into words and map these into phrases and map these into syntactic structures of some type, associate conventional meanings with these syntactic structures, and then determine the meaning of the utterance in context (i. e., its significance). This approach tacks pragmatics on at the end of language parsing/understanding as a kind of afterthought (pun intended).

The problem with this approach is that it fails to appreciate two facts. The first is that language understanding is driven in part by contextual cues, i. e., by socio-pragmatic considerations) and that socio-pragmatic considerations drive utterance production, the other side of the language processing coin. Once a "ride request" is initiated (see the prior blogs in this thread) and recognized as such, addressees come to have expectations as to what others will say before they speak -- not the exact words, of course, but the gist of what they might say. Many misunderstandings in conversation2 reflect the fact that we commonly have such expectations.

That socio-pragmatics drives language production cannot seriously be doubted. Speakers do not start a reply to what someone has said with a "logical form" or whatever else passes for a representation of conventional meanings these days. In a context-rich interaction, where it can be assumed that someone who owns a car and is willing in principle to provide one a ride somewhere and seems also able to do so, we might say any one of a very large number of things to request the ride ranging from, 'Could you give me a ride to school?,." "Mind giving me a ride to school?," Hey, I need you to take me to school," etc., the specific choice reflecting style (formal, informal, intimate, etc.), politeness (reflecting our social relationship), and register considerations. Each of these utterances has a different conventional meaning, so, obviously, we cannot assume that we start off our utterance with selection of some "logical form." Instead, we start with what we might call the "gist" of what we mean to communicate.

We are all familiar with locutions like "the gist of what she said was..." and I take that notion of "gist" quite seriously as the (pragmatic) meaning component of conversational interactions. Going much further into this would entail replicating a chapter of of my book,3 so let me just say here that the gist reflects what the speaker means to contribute to the work of an interaction. At its most elemental level, the gist of any ride request is that the initiator of the interaction wants a ride somewhere. Patten, Becker, and I argue (a view elaborated in my Cambridge Press book), that we combine the "gist" of what we mean to say with choices of appropriate style, politeness, and register features and those somehow get mapped into an utterance with a particular linguistic form. Our claim is that there exists a pragmatic stratum, a kind of hierarchical network, that interfaces with some logico-grammatical system which itself is a kind of hierarchical network with integrated, rather than modularized, linguistic features. The logico-grammatical system we employed was a systemic grammar, something that would have been anathema to me during most of my pre-Patten days.

So, what does this have to do with the evolution of language? My last blog ended with the claim that "highly specialized perceptual and sound production and linguistic-cum-cognitive abilities would have evolved." I somehow doubt that these highly specialized "linguistic-cum-cognitive" abilities are as specialized as the perceptual mechanisms found in a frog's eye but I would like to suggest that this is the right direction to look for answers.

In my view this dictates a view of the evolution of language in which our brains were preadapted for language via our development of an ability to correlate and compile contextual information, and practical knowledge in a way that allowed us to draw inferences from stimuli and respond very rapidly to them. As our ability to speak and engage in linguistic hearing emerged, it is hardly surprising that we developed specialized areas of the brain for dealing with both linguistic hearing and language processing. And it is not at all surprising that linguistic hearing and language processing be located in different areas of the brain given the fact that hearing per se would have come very much earlier than the evolution of language.



1Terry Patten, Michael L. Geis, and Barbara D. Becker:
Toward a Theory of Compilation for Natural Language Generation.
Pages 77-101
2An extreme case of the use of expectations in language understanding occurred in an exchange I had with my wife before we married. Having visited my wife's house to let her dog out when she was away from town, I noticed that her grass had grown. I also noted that she had seeded an small patch of her lawn. When she returned, she came over to my place and said something like, "The grass has sprouted and I need to take the netting off," but I heard, "The grass has grown and I need to mow it." After I realized that I couldn't follow her next couple of utterances at all, I realized where I had gone wrong and backtracked to her initial utterance. This is what I mean by socio-pragmatic expectations (which include background assumptions and contextual information) driving utterance understanding, or, in this case, misunderstanding.
3Check out my blog The Meaning of Meaning. The concept of "gist" is essentially the same as what I call utterance significance in this blog.


Tweet This!

Wednesday, December 06, 2006

How Did Language Evolve? -- II

In my last blog, I noted that efforts to teach primates human language demonstrate that we are unable to teach them human or comparable languages or they are unable to learn them beyond what a 2 or, perhaps, a 3 year old human can learn. This research clearly adds no support to the view that human communication is somehow directly linked to advances in primate communication before we split off from the other primates, which was the contention that got me thinking about this issue..

In order to understand the evolution of language it will be useful to step back and take a look at what it is we ultimately want to account for. In the simplest of cases, I, as a speaker will have some goal I wish to achieve -- inform someone of something I believe they need or may want to know, or request information from them, or request them to do something, or invite them to join with me in doing something, or make a bet with them, among many other possible things. What I say will reflect my goals, the relationship I have with my addressee, background knowledge I believe we share that is salient and relevant to my achieving my goal, and aspects of the immediate context that are salient and relevant to my achieving my goal. There are surely other considerations. At the very least, human communication is a very complex business.

If I want you to give me a ride home from work and I am your supervisor, then I will go about this in much different way than if you were my supervisor. If we are personally close that fact will affect how I go about this. My knowledge of when you usually leave work would be relevant. My knowledge of how much you would have to drive out of your way to give me this ride would be relevant. If I have never made a request like this of you and don't know where you live exactly, I will likely approach this quite indirectly, perhaps saying something like "When are you leaving work today?," or "Do you live anywhere near Green Street and Elm Street?," expecting or hoping you can infer my goal from what I say. The former is so indirect the point might be missed. The latter will usually be inferred to be the start of a ride request. Utterances like these get at your ability to give me a ride home (our wanting to leave at much the same time) and your willingness to do so (you will likely not want to drive me home if this will add a great deal to your driving time.) And, in general for any goal I may have in addressing you, there will be conditions that must be satisfied before I can achieve my goal and what I say will reflect the nature of these conditions and how I want to go about addressing them. The linguistic choices I make will reflect social concerns as well, for I would want to avoid endangering our relationship by burdening you or disrespecting you.

The failure to understand the complexity of human communication and how language fits in in our efforts to communicate with each other lies at the heart of the failure of syntacticians to understand what syntax is about. The machine that drives communication is pragmatics and syntax must be as much about how how social choices influence the syntax and morphology of utterances ("Would you mind giving me a ride home tonight?" vs. Please give me a ride home tonight," both of which have different conventional meanings but can be used to achieve the same goal) as well as conventional meaning, but most remarkably the utterances we find in actual communication rather than the made up sentences syntacticians focus on can be quite ungrammatical by any normal linguistic standard.

In my blog on The Use of Simplification in Scientific Research, I describe a ride request in which the notion of a ride is never referred to directly, a request for a ride is never made directly, and the rejection of this request is never made directly. If you were to listen to this conversation (which I did in Italy of all places at a conference) you would find it quite unremarkable. On the other hand, a transcript of this conversation would drive any conventional syntactician crazy for the critical utterances are highly elliptical and hence ungrammatical by most standards.

In my book on Speech Acts and Conversational Interaction, now out in paper back but sill very expensive (go to your library if you want to read it), I cite a telephone conversation that, after a greetings sequence, begins with a "Guess what?" followed by the reply, "What?," as is normal, followed by
My car is stalled...and I'm up in the Glen...
followed by the addressee offering up an unhelpful utterance of "Oh", followed by the caller saying
And I don't know if its possible ... but see ... I haveta open up the bank in Brentwood.
The sequences of periods indicate pauses. I have omitted audible breathing noises.

The first of these examples is grammatical if you ignore the pause. The second is wildly ungrammatical. But, the addressee immediately grasps the meaning (significance) of what is said and replies and does so in an equally elliptical and ungrammatical way that is also perfectly clear in meaning (significance). As I said earlier, this conversation sounds perfectly normal despite bing highly elliptical and officially "ungrammatical."

Collapsing what the caller says that bears directly on the ride request into just one long sequence, we have (with the elided material put in in bold face):
My car is stalled...and I'm up in the Glen...And I don't know if its possible for you to pick me up and give me a ride to my bank but see I need for you to pick me up and give me a ride to my bank because I haveta open up the bank in Brentwood.
In my book, I show how seeing conversation as driven by pragmatics, not semantics, facilitates an account of how we understand such elliptical utterances. The basic idea is that the elided material addresses the conditions on a ride request being successful and is recoverable by the listener's understanding of the structure of ride requests.

Clearly, some inferencing was required by both parties for them to understand each other's highly elliptical utterances, which is to say, that speaking and thinking and hearing and thinking are so intimately intertwined that we cannot seriously discuss the evolution of language without also considering the evolution of cognition. Put somewhat differently, any account of the evolution of language must be embedded in an account of the evolution of communication and both must be embedded in an account of the evolution of cognition. To study the evolution of language as if it were some free-standing human acquisition would be a serious mistake in my view. When simplifying scientific problems to make them easier to understand one may oversimplify them to the point that one cannot make any real progress at all.

So, to account for the evolution of language we have three problems at the very least. They are
1. We must account for the development of the physiological abilities required for speaking and for what I will call "linguistic hearing," i.e, our ability to discriminate linguistic sounds from other sounds, something that happens at a very early age.
2. We must account for the cognitive abilities required to learn and use languages.
3. We must account for the development of any language specific abilities that are independent of our other cognitive abilities.
By 3 I mean to refer to the development of language-specific 'language centers" in the brain among other things.

At this point, I believe we can conclude that accounting for the evolution of language will likely prove to be impossible since we can infer little from the communicative abilities of our primate cousins and will likely be unable to infer much about the structure of our brains from the fossil record. That's why I have always backed away from the issue. But even if we are going to be unable to answer the question constituting the title of this blog, it may be useful to understand why.

One thing is pretty clear, I believe, and that is that in the course of the evolution of language, it only makes sense to assume that mutations that favored linguistic hearing or making linguistic sounds or learning languages or using languages will tend to have survived and that over time some highly specialized perceptual and sound production and linguistic-cum-cognitive abilities would have evolved. A demonstration of this would help us to understand why studying other primates is a waste of time if it is done by way of trying to understand the development of human communication.



Tweet This!