The book "The Man Talking. Evolution and language "

    imageHuman language is a unique communication system that only Homo sapiens has. Why and, most importantly, why did we learn to talk? Why are any of us in our early childhood easily and naturally assimilating our native language, and learning foreign languages ​​is not easy? Did the Neanderthal language exist, did our ancestors ever talk to them? What is the hypothesis of linguistic relativity and how does it
    affect our understanding of human nature? Answers to these and many other questions can be found in the book of Noam Chomsky - the greatest, eccentric and indomitable linguist of our time - written in collaboration with Robert Berwick, an expert in artificial intelligence.

    Chapter 2. The evolution of biolinguistics


    Before discussing a language, especially in the context of biology, it should be clarified how we understand this term. Sometimes the term “language” is used to designate a human language, sometimes to refer to any symbolic system or method of communication or representation (for example, when it comes to the language of bees, programming languages ​​or the language of the heavenly bodies). We will adhere to the first definition and note that the study of the human language as an object of the biological world has been called the biolinguistic perspective.

    Among the many questions about the language of the most important - two. First, why do languages ​​exist at all, and only in humans? (In evolutionary biology, such a phenomenon is called autoapomorphy.) Secondly, why are there so many languages? These are basic questions about the origin and diversity that interested Darwin and other evolutionary thinkers and which form the basis of modern biology (why is there such a series of life forms in the world, and not any other?). From this point of view, the science of language fits perfectly into the modern biological tradition, despite the seeming abstraction of its details.

    Most paleoanthropologists and archaeologists agree that both of the voiced questions are quite fresh by the standards of evolutionary time. About 200,000 years ago, none of them would have come to mind, because there were no languages ​​yet. And about 60,000 years ago, the answers to them would be the same as now. At that time, our ancestors migrated from Africa and began to spread throughout the planet, and since, as far as is known, language ability has not changed in principle (which is not surprising for such a short period). Specifying more accurate dates will not work, but for our purposes they are not particularly important, because in general the picture looks true. Another important point: if you take a baby born in the Amazon, in an Indian tribe, which in its development is stuck at the level of the Stone Age, and transport it to Boston, then, in terms of language and other cognitive functions, you cannot distinguish it from local children, whose pedigree can be traced back to the first English colonists. The reverse is also true. The uniformity of language ability inherent in our species (the so-called language ability) convinces us that this sign of anatomically modern man should have existed by the time our ancestors left Africa and settled around the world. Erik Lenneberg (Lenneberg, 1967: 261) also drew attention to this fact. As far as we know, in addition to cases of pathology, linguistic ability is inherent in the entire human population. that this sign of anatomically modern man should have existed by the time our ancestors left Africa and settled around the world. Erik Lenneberg (Lenneberg, 1967: 261) also drew attention to this fact. As far as we know, in addition to cases of pathology, linguistic ability is inherent in the entire human population. that this sign of anatomically modern man should have existed by the time our ancestors left Africa and settled around the world. Erik Lenneberg (Lenneberg, 1967: 261) also drew attention to this fact. As far as we know, in addition to cases of pathology, linguistic ability is inherent in the entire human population.

    Moreover, since ancient times, about which written evidence has been preserved, to this day, the fundamental parametric properties of the human language remain the same, the variation occurs only within the established limits. For example, when forming passive constructions such as The apple was eaten (“The apple was eaten”), no language uses the position account so that the passive indicator is placed, say, after the third position in the sentence. This fact is consistent with the findings of a recent tomographic study (Musso et al., 2003). Unlike any machine language, human languages ​​allow for dislocation (displacement): a phrase can be interpreted in one place, and pronounced in another, as in the sentence What did John guess? (“What did John guess?”). This property results from the join operation. The sounds of all human languages ​​are built from a finite, fixed inventory or a basic set of articulation gestures — such as, for example, vibrations of the vocal cords that distinguish the sound “b” from “n”, although not in all languages ​​“b” and “n” are different. Simply put, languages ​​can make different “orders” from the structural elements available to all of them “menu”, but this “menu” itself is unchanged. It is possible to adequately model the variability of such a choice * using simple models based on dynamic systems. This is demonstrated by Niyogi & Berwick (2009), modeling the transition of the English language from word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. a fixed inventory or a base set of articulation gestures — such as, for example, vibrations of the vocal cords that distinguish the sound “b” from “n”, although not in all languages ​​“b” and “n” are different. Simply put, languages ​​can make different “orders” from the structural elements available to all of them “menu”, but this “menu” itself is unchanged. It is possible to adequately model the variability of such a choice * using simple models based on dynamic systems. This is demonstrated by Niyogi & Berwick (2009), modeling the transition of the English language from word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. a fixed inventory or a base set of articulation gestures — such as, for example, vibrations of the vocal cords that distinguish the sound “b” from “n”, although not in all languages ​​“b” and “n” are different. Simply put, languages ​​can make different “orders” from the structural elements available to all of them “menu”, but this “menu” itself is unchanged. It is possible to adequately model the variability of such a choice * using simple models based on dynamic systems. This is demonstrated by Niyogi & Berwick (2009), modeling the transition of the English language from word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. although not all languages ​​"b" and "p" are different. Simply put, languages ​​can make different “orders” from the structural elements available to all of them “menu”, but this “menu” itself is unchanged. It is possible to adequately model the variability of such a choice * using simple models based on dynamic systems. This is demonstrated by Niyogi & Berwick (2009), modeling the transition of the English language from word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. although not all languages ​​"b" and "p" are different. Simply put, languages ​​can make different “orders” from the structural elements available to all of them “menu”, but this “menu” itself is unchanged. It is possible to adequately model the variability of such a choice * using simple models based on dynamic systems. This is demonstrated by Niyogi & Berwick (2009), modeling the transition of the English language from word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. Berwick, 2009), modeling the transition of the English language from the word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such. Berwick, 2009), modeling the transition of the English language from the word order as in German (with a verb at the end of a sentence) to a more modern one. However, such language changes should not be confused with the evolution of the language as such.

    Thus, in the center of our attention is a curious biological object - a language that appeared on earth not so long ago. This species-specific property without significant differences (except in cases of severe pathology) is inherent in all people. Language, in fact, is not like anything else in the organic world and has played a crucial role in human life since its inception. This is the central component of what Alfred Russell Wallace, the founder (along with Darwin) of modern evolutionary theory, called "the mental and moral nature of man" (Wallace, 1871: 334). It is about the abilities of a person to creative imagination, language and in general to the symbolism, recording and interpretation of natural phenomena, complex social practices, etc. This complex is sometimes called human capacity (human capacity). It took shape quite recently in a small group of inhabitants of East Africa, whose descendants are all of us, and distinguishes modern man from other animals, which has enormous consequences for the entire biological world. It is believed that the emergence of language played a major role in this sudden and colossal transformation (we note that this idea sounds quite plausible). In addition, language is one of the components of human abilities that is available for in-depth study. Here is another reason why even purely linguistic studies actually intersect with biolinguistics, although they seem far from biology. which had enormous consequences for the entire biological world. It is believed that the emergence of language played a major role in this sudden and colossal transformation (we note that this idea sounds quite plausible). In addition, language is one of the components of human abilities that is available for in-depth study. Here is another reason why even purely linguistic studies actually intersect with biolinguistics, although they seem far from biology. which had enormous consequences for the entire biological world. It is believed that the emergence of language played a major role in this sudden and colossal transformation (we note that this idea sounds quite plausible). In addition, language is one of the components of human abilities that is available for in-depth study. Here is another reason why even purely linguistic studies actually intersect with biolinguistics, although they seem far from biology.

    From a biolinguistic point of view, language can be represented as an “organ of the body” (on a par with the visual, digestive or immune systems). Like them, language is a subcomponent of a complex organism with significant internal integrity, so it must be studied separately from its complex interactions with other systems in the organism's life cycle. In this case, the language is a cognitive organ, as well as planning systems, interpretations, reflections (reflection), etc., with characteristics that are called mental and are reduced to "organic brain structure", in the words of Josef Priestley (Josef Priestley), a scientist and the philosopher of the eighteenth century (Priestley, 1775/1968: 131) *. Priestley formulated this conclusion after Newton, to his own amazement, demonstrated that the world is not a machine, contrary to the main provisions of the scientific revolution of the XVII century **. This conclusion effectively eliminated the traditional dualism of soul and body, because the clear concept of “(physical) body” or “matter” that existed in the XVIII – XIX centuries disappeared. Language can be perceived as a mental organ, and the word “mental” simply indicates certain characteristics of the world that can be studied in the same way as chemical, optical, electrical properties, hoping to finally bring the results together. However, we note that in the listed areas of science such an association was often achieved in completely unexpected ways and not necessarily by reduction. Language can be perceived as a mental organ, and the word “mental” simply indicates certain characteristics of the world that can be studied in the same way as chemical, optical, electrical properties, hoping to finally bring the results together. However, we note that in the listed areas of science such an association was often achieved in completely unexpected ways and not necessarily by reduction. Language can be perceived as a mental organ, and the word “mental” simply indicates certain characteristics of the world that can be studied in the same way as chemical, optical, electrical properties, hoping to finally bring the results together. However, we note that in the listed areas of science such an association was often achieved in completely unexpected ways and not necessarily by reduction.

    As stated at the beginning of the chapter, two obvious questions about language arise. Why does language exist at all, and only in humans? And why are there many languages? It is also interesting, why languages ​​“differ from each other infinitely and unpredictably,” what should be approached to the study of each language “without any ready-made scheme indicating what the language should be”? We cited words more than half a century ago belonging to the outstanding linguist theorist Martin Yeos (Joos, 1957: v, 96). Jos summarized the dominant "Boasian tradition," as he successfully called it, referring to the works of one of the founders of modern anthropology and anthropological linguistics, Franz Boas. The publication Methods of Structural Linguistics (Methods in Structural Linguistics) by Zellig Harris (Harris, 1951), which laid the foundation of the American structural linguistics of the 1950s, contained in the title the word “methods” precisely because there was little that was said about language (besides the methods that made it possible to reduce the unlimited variety of language material to an organized form). European structuralism had a lot in common with the American. So, similar in concept was the classic introduction to the phonological analysis created by Nikolai Trubetskoy (Trubetskoy, 1939/1960). Generally speaking, structuralist attention was almost entirely focused on phonology and morphology — linguistic levels at which its wide and complex diversity manifests itself. This question is of great interest, and we will return to it. allowing to reduce the limitless variety of language material to an organized form). European structuralism had a lot in common with the American. So, similar in concept was the classic introduction to the phonological analysis created by Nikolai Trubetskoy (Trubetskoy, 1939/1960). Generally speaking, structuralist attention was almost entirely focused on phonology and morphology — linguistic levels at which its wide and complex diversity manifests itself. This question is of great interest, and we will return to it. allowing to reduce the limitless variety of language material to an organized form). European structuralism had a lot in common with the American. So, similar in concept was the classic introduction to the phonological analysis created by Nikolai Trubetskoy (Trubetskoy, 1939/1960). Generally speaking, structuralist attention was almost entirely focused on phonology and morphology — linguistic levels at which its wide and complex diversity manifests itself. This question is of great interest, and we will return to it. Structuralist attention was almost entirely focused on phonology and morphology — language levels at which its wide and complex diversity manifests itself. This question is of great interest, and we will return to it. Structuralist attention was almost entirely focused on phonology and morphology — language levels at which its wide and complex diversity manifests itself. This question is of great interest, and we will return to it.

    In general biology, at about the same time, a similar point of view prevailed. It is expressed, for example, by molecular biologist Gunter Stent. He notes that the variability of organisms is so free that it forms “almost an infinite number of special cases, each of which should be considered separately” (Stent, 1984: 569–570).

    In fact, both in general biology and in linguistics, the problem of compromise between unity and diversity arose constantly. Studies of the language that were conducted during the scientific revolution of the 17th century established a distinction between general (universal) and private grammar (although the meaning of this distinction was not exactly the same as in the framework of the modern biolinguistic approach). The general grammar was called the intellectual core of this discipline, and private grammars were considered as unimportant, random incarnations of the universal system. With the flourishing of anthropological linguistics, the pendulum has swung in the other direction - towards diversity, which is well reflected in the Boas definition, quoted above. Within the framework of general biology, the problem in question vividly discussed in the famous controversy between naturalists Georges Cuvier and Geoffroy St. Hilaire in 1830. Cuvier's point of view, which emphasized diversity, won out (especially in the light of the Darwinian revolution). This led to the conclusions about the “almost infinite set” of special cases that need to be considered separately. Probably the most often quoted statement by biologists is the final words of Darwinian's "Origin of Species" about how "from such a simple beginning an infinite number of the most beautiful and most amazing forms developed and continues to develop" (Darwin, 1859/1991: 419). Evolutionary biologist Sean Carroll put Darwin's expression in the title of his book (Carroll, 2005/2015) - an introduction to the “new science of evo-devo” or evolutionary developmental biology, which seeks to show

    To reconcile the observed diversity of organic forms with their obvious deep uniformity (why we observe just such a series of living organisms, and not some other one, and just such a number of languages ​​/ grammars, and not any other), the three interacting factors formulated by biologist Mono allow The book "Accident and necessity" (Le hasard et la nécessité) (Monod, 1970).

    The first factor is the historically conditioned circumstance that we are all descendants of a single tree of life and, therefore, we have a common ancestry with all other living beings, whose diversity exhausts, obviously, only an insignificant share of all kinds of biological outcomes. Therefore, it should not be surprising that we have common genes, biochemical pathways of metabolism and much more with other organisms.

    The second factor is the physical and chemical limitations of our world, which narrow the range of biological possibilities. For example, it is almost unbelievable that wheels form for our movement, because it is physically difficult to bring nerves and blood flow to a rotating object.

    The third factor is the screening effect of natural selection, which, of the previously known “menu” of possibilities given by historical circumstances and physico-chemical constraints, leaves only the number of organisms that we observe in the outside world. Note that the effect of the limited “menu” of options is extremely important. If the list of options is extremely narrow, then there is little to choose from which (it’s not surprising that a person at a fast food restaurant usually orders a hamburger and fries). As Darwin would say about this, natural selection is not the only means by which nature has acquired its present appearance. “In addition, I am convinced that natural selection was the most important, but not the only means of modification” (Darwin, 1859/1991: 24).

    Recent discoveries have breathed new life into the general approach of Darcy Thompson (D'Arcy Thompson, 1917/1942) and Alan Turing (Turing, 1952) to the principles that limit the diversity of organisms. According to Wardlow (Wardlaw, 1953: 43), true biological science should consider each "living organism as a special kind of system to which the general laws of physics and chemistry apply," which sharply limit the possible diversity of organisms and fix their fundamental properties. Such a point of view no longer looks extreme today, after the discovery of master genes, deep homology, conservation, and much more, up to such severe restrictions on the processes of evolution / development that "reproduction of the protein film of life can be surprisingly monotonous." In this quote from the review article Pulwayka et al. (Poelwijk et al., 2006) on permissible mutation paths is rethought by the famous metaphor of Stephen Gould, who believes that the film of life, if reproduced again, can follow new routes. As Michael Lynch further notes (Lynch, 2007: 67), “for many decades it was known that all eukaryotes mostly share the same genes for transcription, translation, replication, nutrient intake, basic metabolism, the structure of the cytoskeleton, etc. Why, when it comes to development, do we expect to see something else? ”

    In a review article on “evo-devo”, Gerd Muller (Müller, 2007: 947) notices how much more thoroughly we come to understanding the patterns of forming patterns like the Turing machine:

    “Generalized forms ... arise as a result of the interaction of the basic properties of a cell with various mechanisms for the formation of patterns. Differential adhesion and polarity of the cell, changing under the influence of different types of physical and chemical mechanisms of patterning, form standard sets ... The properties of differential adhesion and their polar distribution on the cell surface result in combination with a diffusion gradient to hollow spheres, and in combination with a deposition gradient - to spheres with invaginated ... The combination of differential adhesion with the reaction-diffusion mechanism generates radial-periodic structures, and its combination with chemical oscillation gives the series but-periodic structure. The organisms of ancient animals reflect in their structure the action of such standard sets of patterns of pattern formation. ”

    For example, in explaining the historically determined fact that we have five fingers and toes, it would be more correct to refer to the process of the development of the fingers than to the optimal number five for their functioning.

    According to the controversial biochemist Michael Sherman (Sherman, 2007: 1873), "the universal genome encoding all major development programs in various types of animals (Metazoa) appeared in a single-cell or primitive multicellular organism shortly before the beginning of the Cambrian period" (about 500 million years ago ), when there was a sudden surge of diversity of complex animal forms. Sherman further argues that many “types of animals that have similar genomes are nevertheless so different because each of them uses its own particular combination of development programs” (Sherman, 2007: 1875). In accordance with this interpretation (if to think abstractly), there is only one type of multicellular animals. Such a point of view could adhere to, say, a Martian scientist - a representative of a highly developed civilization, contemplating events on Earth. Surface diversity may in part be the result of various combinations of the developmental-genetic toolkit, as it is sometimes called, preserved by the evolution of the genetic toolkit. If such ideas prove to be true, then the problem of unity and diversity can be reformulated in a completely unexpected way for some modern scholars. The extent to which this conservative “toolkit” can be the only explanation for the observed uniformity is a matter worthy of attention. As has been said, the observed uniformity arises in part because too little time has passed and the continuity of generations proportional to this amount of time makes it impossible for us to study the “too large” genetic-protein-morphological space (especially considering the impossibility of “returning” and starting the search from the very beginning to achieve the best results). Given these constraints imposed by nature, it should not be particularly surprising that all organisms are built according to a certain set of “drawings” (Baupläne), as Stephen Gould emphasized. Therefore, if advanced Martian scientists arrived on Earth, you would probably see only one organism with many observable surface variations. that all organisms are built according to a certain set of “drawings” (Baupläne), as Stephen Gould emphasized. Therefore, if advanced Martian scientists arrived on Earth, you would probably see only one organism with many observable surface variations. that all organisms are built according to a certain set of “drawings” (Baupläne), as Stephen Gould emphasized. Therefore, if advanced Martian scientists arrived on Earth, you would probably see only one organism with many observable surface variations.

    At the time of Darwin, such uniformity did not go unnoticed. In the course of naturalistic studies, Thomas Huxley (Thomas Huxley), an associate and popularizer of Darwin, came to the conclusion that there are probably "predetermined lines of modification", following which natural selection "produces variations in a limited number and variety" for each species (Huxley, 1878/1893: 223). Even in Darwin himself, the study of the sources and nature of possible variation constitutes a significant part of his research program after The Origin of Species, which is reflected in the work Changes in Domestic Animals and Cultural Plants (1868). Huxley's conclusion is similar to the older ideas of "rational morphology" (the famous example is Goethe's theories about the archetypal forms of plants, which were partially revived during the "evolutionary devo revolution"). Really, Darwin was interested in this area of ​​research and, as an adherent of synthesis, he studied more carefully the “laws of growth and form” (the limitations and opportunities associated with changes are due to developmental characteristics, random linkage with other signs that may be subject to strong positive or negative selection, and finally, selection by the most considered basis). Darwin pointed out that such laws of "correlation and balance" are of considerable importance for his theory, and noted as an example that "white cats with blue eyes are usually deaf" (Darwin, 1859/1991: 28). which may be subject to strong positive or negative selection, and, finally, by selection based on the most considered feature). Darwin pointed out that such laws of "correlation and balance" are of considerable importance for his theory, and noted as an example that "white cats with blue eyes are usually deaf" (Darwin, 1859/1991: 28). which may be subject to strong positive or negative selection, and, finally, by selection based on the most considered feature). Darwin pointed out that such laws of "correlation and balance" are of considerable importance for his theory, and noted as an example that "white cats with blue eyes are usually deaf" (Darwin, 1859/1991: 28).

    As noted in Chapter 1, for almost the entire second half of the 20th century, while the synthetic theory of evolution prevailed, the foundations of which were laid by Fisher, Haldane and Wright, the attention of evolutionary theory was focused on micromutational events and gradualism and emphasized the effect of natural selection, which goes in small steps. Recently, however, in general biology, the focus has shifted toward a combination of three factors identified by Monod, which made it possible to take a fresh look at old ideas.
    Let us return to the first of our two basic questions: why should languages ​​exist at all, obviously being autoapomorphy? As was said, quite recently (by the standards of evolutionary time) this question did not make sense, because there were no languages. There was, of course, a variety of animal communication systems. But they all differ radically from human language in structure and function. In standard typologies of animal communication systems, such as Mark Hauser’s typology, proposed in his comprehensive review of the evolution of communication (Hauser, 1997), it is not possible to find a suitable place for the human language. Usually, a language is considered as a system whose function is communication. This is a widespread view, characteristic of most breeding approaches to the language. However, it is erroneous for several reasons.

    Attempts to deduce the “purpose” or “function” of a biological trait from its external form are always fraught with difficulties. Levontin's remarks in the book "The Triple Helix" (Lewontin, 2001: 79) demonstrate how difficult it is to attribute to a body or a sign a certain function even in a case that at first glance seems quite simple. For example, bones do not have a single function. The bones support the body (this allows us to stand and walk), but they also store calcium and the bone marrow produces red blood cells, so the bones can in some sense be considered part of the circulatory system. The same is true of human language. Moreover, there has always been an alternative tradition expressed by, among others, Burling (Burling, 1993: 25). He claims,

    Language, of course, can be used for communication, as well as any aspect of our activity (style of clothes, gestures, etc.). But language is also widely used in many other situations. According to statistics, in most cases the language is used for the needs of thinking. Only a great effort of will can be kept from a silent conversation with oneself during wakefulness (and in a dream too, which often annoys us). A prominent neurologist Harry Jerison (Jerison, 1977: 55), along with other researchers, made a bolder statement that "language did not evolve as a communicative system ... It is more likely that the initial evolution of language intended it ... to build an image of the real world" to be "an instrument of thinking ". Not only in the functional dimension, but in all other respects - semantic, syntactic,

    But how then did this strange object appear in the biological record, moreover in the close framework of evolution? There is, of course, no exact answer, but you can sketch a couple of quite plausible assumptions that are related to the latest research in biolinguistics.

    In the fossil record, the first anatomically modern people appear several hundred thousand years ago, but the evidence of the emergence of human abilities is much later and dates back to the time just before migration from Africa. Paleo-anthropologist Ian Tattersal (Tattersall, 1998: 59) reports that “a vocal tract capable of producing articulate sounds” existed already half a million years before the earliest evidence of the use of language by our ancestors. “We are forced to conclude,” the researcher writes, “that the appearance of language and its anatomical correlates was not driven by natural selection, no matter how profitable these novelties were in retrospect” (this conclusion does not contradict the standard evolutionary biology in spite of the fallacies that can be found in the popular literature). The human brain has reached its present size not very long ago, perhaps about 100 years ago, and this gives some experts a reason to think that “human language probably developed - at least in part - as an automatic, but at the same time adaptive consequence of an increase in absolute brain sizes ”(Striedter, 2006: 10). In Chapter 1, we pointed out some differences in the genome that could lead to such an increase in brain size, and the rest will be discussed in Chapter 4.

    Tattersall writes about Tattersall (2006: 72) that “after a long - and not very understandable - period of chaotic augmentation and reorganization of the brain in human history, something happened that prepared the ground for mastering the language. This innovation should have depended on the surprise effect, when an accidental combination of ready-made elements gives something completely unexpected, supposedly a neural change ... for a certain population in the history of mankind ... relatively small in genetic terms [which] probably had nothing to do with adaptation ”, although it gave advantages and subsequently spread. Perhaps it was an automatic consequence of the growth of the absolute size of the brain, as Stridter believes *, and perhaps an accidental mutation. After some time - by the standards of evolution not very long - there were further innovations

    What was this neural change in a small group, and relatively small in genetic terms? To answer this question, one must pay attention to the specific properties of the language. The elementary property of language ability that we all possess is that it allows us to build and interpret a discretely infinite number of hierarchically structured expressions (discrete - because there are five-word sentences and a six-word sentence, but there are no five-sentence sentences half the words, and infinite - because the length of sentences is unlimited). Consequently, the basis of the language is the recursive generating procedure, which accepts elementary word-like elements from some repository (let's call it a lexicon) and acts iteratively, generating structured expressions, not limited in complexity. To explain the emergence of language ability - and therefore, the existence of at least one language - we must solve two main problems. The first is to deal with “atoms of calculations”, lexical units, the number of which usually ranges from 30 to 50 thousand. The second is to find out what the computational properties of language ability are. This task has several aspects: we need to understand the generating procedure, building "in mind" an infinite number of expressions, and the methods by which these internal mental objects are transmitted to interfaces with two systems external to the language (but internal to the organism) thinking and sensorimotor system that serves to externalize internal computing and thinking). There are a total of three components, as discussed in Chapter 1. This is one of the ways to reformulate the traditional concept, which goes back at least to Aristotle and says that language is “sound, meaning something”. All of these tasks contain problems, and much more serious than previously thought.

    Let us turn to the basic elements of the language and begin with the generating procedure, which originated approximately 80,000 years ago (by the standards of evolutionary time - in the blink of an eye). Probably, while in the brain, there was some rerouting (change in neural connections). Here the “evolution of the evolution of devo” in biology is important for us. She provided a decent amount of data so that two conclusions could be drawn. The first is that the genetic fund, even at the level of regulatory systems, is distinguished by deep conservation (very stable). And the second is that very small changes may entail huge differences in the observed result, although the variation of the phenotype is limited due to the deep conservation of genetic systems and the operation of the laws of nature (those that interested Thompson and Turing). Let's give a simple example: There are sticky fish with and without a spiny ventral fin. About 10,000 years ago, a mutation in the genetic “switch” near the gene involved in the formation of the fin, delimited these two forms - with and without spines. The first form has adapted to the oceans, and the second to the lakes (Colosimo et al., 2004, 2005; Orr, 2005a).

    Much more ambitious results were obtained in works on the evolution of the eyes (we discussed this actively studied topic in Chapter 1). It turns out that the number of types of eyes is very small - partly due to limitations imposed by the physics of light, and partly because only one category of proteins (opsins) can perform the necessary functions (and the events leading to the "capture" of opsin molecules by the cells, had (apparently, stochastic nature). The genes encoding opsin are of ancient origin and are constantly used, but only in a limited set of ways (again, due to physical limitations). The same is true for lens proteins. As noted in Chapter 1, eye evolution is an example of the complex interaction of the laws of physics, stochastic processes, and the role of natural selection in choosing a path within a narrow “corridor” of physical capabilities (Gehring, 2005).

    The work of Jacob and Monod (1961), during which the operon was discovered in E. coli (E. coli) and for which the authors later won the Nobel Prize, allowed Mono to formulate his famous aphorism, quoted in (Jacob, 1982: 290): “What true for E. coli, then true for an elephant. " Although it is sometimes said that this statement anticipated the modern “evolutional devo” approach, but most likely Monod had in mind that the theory of generalized negative regulation created by him together with François Jacob should be suitable for describing all cases of gene regulation. This summary seems to have been overly bold. In fact, to create negative feedback, it is sometimes possible to get by with far less means, because a single gene can be negatively regulated or self-regulated. Moreover, it is now known

    The discovery of more complex methods of gene regulation and development used by eukaryotes, just became the most important contribution to the current "evolution of the evolution of the devo." Nevertheless, Mono's main idea that small differences in the sequence and combination of regulatory mechanisms that activate genes can lead to different results turned out to be true, although the very principle of action was not thought out. It was Jacob (1977: 26) who had to construct a convincing model for the development of other organisms, based on the notion that “due to complex regulatory circuits” everything that “is responsible for the difference between a butterfly and a lion, a chicken and a fly ... is the result of mutations that have changed more the regulatory contours of the body than its chemical structure. " Jacob’s model, in turn, became the basis for the emergence of the theory of principles and parameters,

    The theory of principles and parameters is based on the assumption that languages ​​are characterized by immutable principles tied to a parameter switching unit. Parameters can be compared with questions that the child must answer, relying on the data he has in order to select a specific language from a limited set of languages ​​possible in principle. For example, the child must determine where the language is with the initial position of the vertices (head initial), for example, English (in which substantive elements precede the additions with them; cf. read books), and where is the language with the final position of the vertices ( head final), for example, Japanese (it has the same meaning of the phrase as hon-o yomimasu (literally: “books to read”)). As in the case of reordering of regulatory mechanisms, within this approach one can understand

    The theory of principles and parameters has borne fruit: data from a wide typological range of languages ​​have been rethought, questions have been raised that have never been raised before, and in some cases answers have been given. It would not be an exaggeration to say that in the past 25 years more has become known about languages ​​than in the preceding millennia. Answering the two fundamental questions with which we began the conversation, we note: this approach assumes that the novelty, which arose almost suddenly (by the standards of evolutionary time), was the generating procedure, which led to the emergence of principles. A variety of languages ​​follows from the fact that the principles do not define the answers to all possible questions about the language, and even some questions are left open in the form of parameters. Note that the only example that we gave above is associated with linear order. Although this is a controversial topic, it seems that quite a lot of linguistic data have accumulated by now, indicating that the order is subject to externalization of internal calculations through the sensorimotor system and does not play any role in the core (syntax) core and semantics. The correctness of this conclusion is confirmed, among other things, by biological data provided by both little-known and eminent biologists (we will return to this question a little later).

    The simplest assumption (from which we will proceed, unless otherwise proved) is that the generating procedure arose at once as a result of a small mutation. In this case, one should expect that this generating procedure is very simple. Over the past half century, many types of generative procedures have been studied. One of their family, familiar to linguists and applied mathematicians, is the grammar of the components (phrase structure grammar). They were introduced into scientific use in the mid-1950s and have since been widely used. At one time, this approach was popular. He naturally fit into the framework of one (of several equivalent) formulations of the mathematical theory of recursive procedures (we are talking about the canonical systems of Emil Post) and covered some basic properties of the language, for example, the hierarchical structure and embedding of groups (embedding). Nevertheless, it soon became clear that the grammar components are not suitable for describing the language, moreover, they are very complex and contain many arbitrary assumptions (in general, we did not count on such systems, and they could hardly arise at once).

    Over the years, researchers have found ways to reduce the complexity of these systems and finally completely abandon them in favor of the simplest possible recursive generation method — an operation that takes as input two already constructed objects (X and Y) and forms a new object, including them (set with elements X and Y). We call this optimal operation a merge. Having access to the conceptual atoms of the lexicon, a join operation repeated an unlimited number of times generates an infinite number of discrete hierarchically structured expressions. If these expressions can be consistently interpreted on the interface with the conceptual system, this represents the internal “language of thought”.

    A strong minimalist thesis (Strong Minimalist Thesis, SMT) states that the spawning process is optimal, that is, the principles of the language are determined by the efficiency of the calculations and the language uses the simplest possible recursive operation that satisfies the conditions of the interfaces and is consistent with the principles of the efficiency of the calculations. A language acquires a specific form under the action of the laws of nature (in this case, the principles of efficiency of calculations), when the basic mode of construction is available, and satisfies the conditions of the interfaces. The main thesis is formulated in the title of the collection of scientific and technical articles “Interfaces + recursion = language?” (Sauerland & Gärtner, 2007).

    The best solution would be to reduce recursion to a join operation. Note that the question mark in the heading to the place, because the questions that arise directly relate to the current study. Further we will try to show that there is a significant inequality between the two interfaces. The semantic-pragmatic interface that connects the language with the systems of thinking and action is primary. How rich the above-mentioned external conditions is is a serious research question, and a very difficult one, since quite a little is known about systems of thinking and action that are independent of language. A very strong thesis proposed by Wolfram Hinzen (Hinzen, 2006) states that the central components of thinking, such as propositions, are generated by an optimally arranged generating procedure. If these considerations could be empirically tested,

    SMT cannot be called a generally accepted approach, but now it looks more believable than a few years ago. If the SMT is correct, the evolution of the language can be reduced to the emergence of a join operation, an evolution of conceptual atoms of the lexicon, connections with conceptual systems, and an externalization regime. For all the other principles of the language, not reducible to the operation of connection and optimality of calculations, some other evolutionary process should be responsible. And it is hardly possible to learn much about him, at least with the help of current methods, as Lewontin pointed out (Lewontin, 1998).

    Note that in this picture there is no place for the predecessors of a language, say, a language-like system, which would contain only short sentences. There is no reason to assume the existence of such a system, because in order to move from sentences of seven words to the discrete infinity of the human language, the same recursive procedure that is required to go from zero to infinity should arise. In addition, there is no direct evidence of the existence of such proto-languages. A similar picture is observed in the assimilation of the language (even if it seems that it is not), but we will leave this question beyond the framework of this book.

    »More information about the book can be found on the publisher's website
    » Table of contents
    » Fragment

    For Habrozhiteley 20% discount on the coupon -Chomsky

    Also popular now: