Language: Myth or Instinct?

A Literature Review

1. Introduction

In 2015, Vyvyan Evans launched a veritable grenade into the linguistic community with the publication of his book, The Language Myth: Why Language is Not an Instinct. This response was somewhat predictable, as his aim was to discredit and dismantle the theory of “Universal Grammar,” which had largely been dominant since it was set out by Noam Chomsky in the 1950’s and ‘60’s. While Chomsky’s theory holds that language arises from a sort of preprogrammed knowledge of one underlying “human” grammar, Evans’ theory is that language develops in tandem with, and as a result from, other cognitive abilities of the human mind (2015: 3). Because of his assertion that language is acquired incrementally along with other cognitive abilities--and largely acquired from “our parents and primary caregivers”-- he names his theory “language-as-use” to contrast Chomsky’s “language-as-instinct” (2015: 5). 

Characterized by Generativists as a polemic against Universal Grammar, this incendiary book and the reaction to it are representative of the current divide in the linguistic community. The “Cognitive Revolution” that took off in the 1980’s (largely due to the work of George Lakoff) has a culminating moment in Evans’ work, and the wave of backlash it received attests to the enduring strength of Formalism in research and academia--both in the US and abroad. But how did these warring factions become so polarized in their beliefs, and why did the book warrant such a vehement reception if it was based on previously published research? To answer these questions it is necessary to provide some background on the evolution of linguistic and otherwise cognitive theories over the last seventy years, and some clarification about what Chomsky himself (historically and currently) purports Universal Grammar to be. 

2. Background Literature Review: Generative Grammarians follow their Instincts

2.1 Noam Chomsky and the Original Four Cornerstones of Universal Grammar

The start of the 20th century marked the start of linguistic theorists’ fascination with the systematic study of grammar (Newemayer). Inspired largely by the work of Ferdinand de Saussure, the focus began shifting away from the systematic analysis of language change throughout recorded history, and toward a deep analysis of the rules and processes by which more modern languages were governed and evolving. From this climate of structural considerations the “functionalist approach” to language change arose. This approach held that languages change to serve the ultimate goal of optimization--a goal speakers were intuitively, if not consciously aware of; it recognized simplification, repair, and avoidance of merger as valid types of language change (History of Linguistics). It was this approach to the study of language to which Noam Chomsky responded with his theory, which was cited as “Transformational Grammar” in some of his earlier publications, but ultimately came to be known as “Generative Grammar” or “Universal Grammar” (History of Linguistics). 

Chomsky rose to fame in the Linguistic community in 1957 with the publication of his first book Syntactic Structures. The aim of this and his other early writings was essentially to “construct a formalized general theory of linguistic structure,” which would be comprehensive enough to provide answers to both existing linguistic inquiries, and those yet to arise (Lees and Chomsky, 1957). His was an approach to grammar that focalize grammatical rules and the computation of meaning. That is, he considered language to be a finite set of elements, which could be combined in a theoretically infinite set of sequences, and the grammar of a language to be “the device that generates all [possible] grammatical sequences” (Lees and Chomsky, 1957). 

Chomsky does acknowledge in the preface of his 1965 book Aspects of The Theory of Syntax that his theory of grammar--despite being more comprehensive than its predecessors--was not entirely new in essence.  The basic concept of applying grammatical formulae to lexical units to create meaning like a mathematical equation had been put forth in theories by several researchers over the previous century. Wilhelm Von Humboldt was the first researcher to describe grammar as the “mechanism” by which language made “infinite use of finite means” (1836); and according to Chomsky, some aspects of his work are even reminiscent of the ancient Sanskrit philologist Panini (1965: 27).  

Despite having some roots in previously published theory, “Generative Grammar” caused tides of controversy among the linguistic community, due in large part to the assumption about language acquisition that it was predicated upon. In his 1965 publication Aspects of the Theory of Syntax, Chomsky sought to “justify” the theory of generative grammar on both “external” and an “internal” grounds (1965: 27). On the “external” front, he felt the generative approach was justified since it “correctly” described language it is used and understood, matching “the linguistic intuitions of the native speaker” (1965: 27). 

But Chomsky felt that to justify his grammatical theory on a deeper, more “internal” level, it would have to be connected to an “explanatory hypothesis” about why the extant grammar is the preferred system. This endeavor, he felt, essentially necessitated the construction of a theory of language acquisition, and an account of the innate abilities that make [it] possible [to acquire and utilize a functional system of grammar]” (1965: 27). In Chomsky’s view the child acquires language by gathering raw data from observing the interactions around her, and applying these to a pre-existing, innately known “schema,” which gets refined and differentiated as the child learns its native language (1965: 27). 

The proposed quality of “innateness” put Chomsky’s theory at odds with the psychological cannon of B.F. Skinner’s “Behaviorism,” which was dominant at the time (Barman, 2014). In fact, Chomsky openly criticized Behaviorism, which sought to “understand behavior and language as a function of environment or setting” (2014: 111), rather than a “biological endowment” which is “physically represented in the brain” (Chomsky, 2002: 1). He published a critique of Skinner’s Verbal Behavior in 1959, denouncing the work’s emphasis on “the circumstances in which a language is learned “ (2014: 111). In addition to criticizing the experiential and external focus of the theory, Chomsky also criticized the “experimental data gathering techniques” calling them “unnecessary” ( ). The focus on a genetic predisposition for language acquisition places Universal Grammar in a philosophical camp known as “Linguistic Nativism,” which holds that “since part of language is unlearned it cannot be acquired by induction, that language acquisition draws on an ‘unlearned’ system of language, and that there is a special component of the human mind which has the development of language as the key function” (2014: 114). 

Chomsky proposes the “language acquisition device” (LAD) by which this process may occur. Though he contends that providing a full depiction of this schema is beyond the scope of his work, Chomsky does assert that the LAD necessitates the existence of certain “linguistic universals” that lie at the base of all languages (1965: 27). It is the job of the linguist, in his view, to “develop an account” of these universals (1965: 27). In defining the term, he offers two theories: “substantive” and “formal” (1965: 28). The former maintains the necessity of a “fixed class of items” (like phonemes for example), that combine to create larger items (like words)--and that the features of these seed items would be universal. (1965: 28). This theory could even be expanded to claim that certain syntactic categories are found within all languages. Formal universals by contrast, are more about the abstract conditions a grammar must meet: that is, “the character of the rules that appear in grammar,” such as a syntax having requisite transformational properties (1965: 28). 

Essentially, the original theory of Universal Grammar was predicated upon four main principles: domain-specificity (which entails that language acquisition occurs based upon linguistic representations only), universality (the idea that universal principles must underlie all languages), innateness (the idea that the universal principles that aid in language acquisition are not learned but genetically inherited), and autonomous syntax (the concept that the inherited language universals directly reference syntax, rather than a “functional correlate of syntax”) (Goldberg, Adele n.d.). Since Chomsky’s early publications, this theory has served as the basis for a wealth of studies, and not only in the field of linguistics, but in psycholinguistics, language evolution studies, language acquisition, neurocognition, parsing and speech recognition, and most branches of cognitive science (Evans and Levinson 2009). Its interdisciplinary pervasiveness exemplifies how it has come to be accepted almost as a sort of empirically proven maxim, rather than the theory that it is (2009 : 430). 

3. Challenges to Chomskyan Linguistics 

3.1 Reformations to Universal Grammar

Despite its apparent ubiquity, acceptance of the original guiding principles of Universal Grammar has become rather fragmented--even in the eyes of Chomsky himself. While there have been a significant number of papers penned by Universal Grammarians that made slight tweaks to the theory, for the sake of brevity, this paper will discuss only two of the most famous and widely impactful ones. Chomsky and his MIT alumnus Pinker have published revisions and to the original four tenets that have been widely accepted by the Generativist community as the natural evolution of the theory. Pinker would go on to refine Chomsky’s arguments about innateness and domain-specificity with the publication of his 1994 book, The Language Instinct: the new science and language of mind. In this book Pinker argues that there is a sort of crucial window during childhood for the innate language abilities to be activated, which is then disbanded after that critical moment in development. During that time, he argues, the language instinct is expressed by certain structures within the brain ( ). He stakes this assertion on his own research in children’s language acquisition, comparing it to similar ones about the visual cortex’s early development and the ability to acquire sight. Pinker also argues the genetic inheritance of syntax by citing the work of Derek Bickerton, who studied the formation of creoles and found that they too, regularly adhere to specific grammatical rules. 

Where Pinker seems to divulge from Chomskyan theory is around the issue of evolution, in which he, unlike Chomsky, firmly contends that the neural circuitry for human language is a product of Darwinian-style evolution (Coe 1994). Subsequently, he argues that chimps are not able to use human language not for lack of the physical apparatus like a vocal tract, but rather because their brains completely lack the facility to comprehend or utilize grammar (Coe 1994). Written for a lay audience, Pinker’s book was widely read, and catapulted him to fame. He effectively made linguistic theory more accessible, as Chomskyan theory is famously “opaque” in style and can be difficult for even professionals in academia to comprehend (Coe 1994).

Almost a decade after Pinker published the book that would constitute a modification to one of the core tenets of Universal Grammar, In 2002, Hauser, Chomsky, and Fitch published a paper that greatly minimized one of its core principles in claiming that “recursion” is the only real property in the domain-specific faculty of language (2002: 1569). Chomsky et. al begin by defining the intended sense of “language” in the discussion, and subsequently redefining the term “faculty of language” (2002: 1570). Rather than what Chomsky calls an external or “E-language” like English or Navajo, “language” is defined instead as something that describes the processes internal to the mind, or “I-language” (2002: 1570). Chomsky asserts that even this definition however, leaves room for “misinterpretation,” citing an example from neuroscience in which “most” of the nervous system can be considered involved in some part of language in the “broadest sense.” For this reason, the distinction is made between the “faculty of language in the broad sense” (FLB), and the “faculty of language in the narrow sense” (FLN). FLB is taken to encompass “sensory motor” and “conceptual intentional” systems more broadly, and may be shared by animals, but certainly exhibits unique and exclusive features in humans, while FLN is regarded as the “abstract linguistic computational system alone,” ie the innate device preprogrammed to facilitate the acquisition of human grammar (2002: 1571). 

According to the researchers, it is agreed upon among proponents of the FLN theory that “recursion,” or the ability to embed clauses within other clauses, is a necessary and defining feature of FLN. It is in this paper, that Chomsky specifies it is perhaps the only property of a specifically human FLN (other than the mapping of the universal grammar program onto the sensory motor and conceptual-intentional interfaces, as critics of cognitive linguistics are apt to note [Hasplemath, 2018]). What Chomsky et al. see as the uniquely human property of recursion is a device which can create a “discrete infinity,” meaning there is no inherent cap on the possible length of an utterance, except organism-specific limitations such as lung capacity and working memory (2002: 1571). 

Having established these definitions, Chomsky et al. (2002) goes on to examine data from a series of experiments dealing with “animal speech perception” and “animal speech production”, with the goal of proving that while the animals in the experiments demonstrate impressive aptitude in recognizing and even producing (some forms of) human speech, the hypothesis of a “continuity” between human speech and animal communication is a null one (2002: 1574).  They stake this claim on a few central assumptions, which are then backed with research: research showing that most non-human animals are not able to imitate vocal or gestural utterances (though the latter is possible with diligent training)

3.3 Trends in Objections to Universal Grammar Pre-”Language Myth”

Chomsky’s 2002 redefinition of the “human language faculty” and the emphasis on the concept of “recursion” is representative of the terminological vein in which much of the linguistic debate has thus far been carried out. Following Chomsky’s admissions that universal linguistic properties may be relegated mainly to that feature, “recursion” came to be seen as a sort of last line of defense for Universal Grammar, and the veracity of the claim that it is a uniquely human property was attacked from a few different angles. Everett for example, famously published a study of the Amazonian language Piraha, claiming it to be a human language that does not exhibit the property of recursion, and that these findings falsified Chomsky’s. Reception to this study was mixed, consisting of a predictable amount of criticism from the Generativists, and a fair amount of support from cognitive linguists. 

The existence of a non-European root language that does not exhibit recursion is one of the claims backing accusations that the folly of Universal Grammar stems in part from a certain “ethnocentrism,” a by-product of the fact that most cognitive linguists speak European languages (Evans and Levinson 430). From the other side of the argument, researchers like Antunes et. al (2011) and have argued that animal vocalizations may be seen as a rudimentary form of language, exhibiting basic syntax and in some cases, even recursion. The rejection of the idea that any such vocalizations could be considered “language” in the same manner as human language seems to come down to semantic hair-splitting.

Criticisms of the theory of course proliferated within the academic sphere as well. Evans’ 2015 book may have compiled the research against Universal Grammar and presented it in a digestible format for a lay audience, but the work hardly represents the first research-based publication to expose some of the cracks in the theory. As Maher and Groves noted in 1998, Chomsky’s internal, biologically-based view that intentionally excludes any role of community or environmental factors as being relevant to language acquisition put him at odds on “five fronts: anthropological linguistics, sociology, political economy, philosophy, and humanistic’ (Barman, 2014: 117).  The general defense put forth by Universal Grammarians seems to be that cognitive linguists “misunderstand” or “misinterpret” Chomsky’s use of the word “recursion,” as well as the implications of “domain-specific.” These terminological “quibbles” have been carried out amongst academics in both formal and informal settings like the “blogosphere,” but do not represent as holistic a takedown of Universal Grammar as research-based papers like those cited in Evans’ 2014 book.

3.2 The Language Myth 

Evans’ book is a compilation of research from a range of linguistic sub-disciplines. The work debunks Universal Grammar both by addressing each of its core principles and the “myths” they propogate, and by a sort of Gestalt shift in its conception of language. Rather than an “innately programmed knowledge of human grammar,” Evans argues that language “reflects and builds upon general properties and abilities of the human mind…[particularly] our species-specific cultural subjective communication” (2015: 3). He argues that it is exactly the external, social influences that Chomsky was quick to dismiss (2002: 1571), that make language possible. Because of his contention that acquisition occurs through trial and error within social contexts, he calls the theory “language-as-use.”

Evans does contend that recursion is “not a uniquely human trait,” citing a study in which groups of starlings were able to learn to recognize recursion, but his broader and more emphasized contention was that focusing on and defending the delineation between human language and animal languages is ill-founded, and detracts from the study of communication as a whole (2014: 34).  Rather than zero-ing in on “recursion” as the last stake in the “human faculty of language” theory, Evans espouses a view that human and animal communication exist on a sort of continuum in which animal language constitutes a sort of “proto-language” (Evans, 2014, 26). To provide alternate criteria for identifying what hallmark traits constitute languages, he offers a schema laid out by Hockett, which examines language in terms of a plethora of criteria including “semanticity,” “learnability,” “discreteness,” “cultural transmission,” and “vocal-auditory channel” to name a few (2014: 62). Hockett’s system has canonically been highly influential in the “language-as-instinct” worldview (Evans, 2014, 70), but Evans seems to consider it more as a selection of potentialities for language than a comprehensive list of equally mandatory, defining features. The existence of sign language, for example, invalidates the “vocal-auditory channel” feature (2014: 70). Evans then presents a collection of studies in which animal communication exhibits these features to varying degrees.

Notably, Evans cites a study that was part of the early 1900’s trend of researchers attempting to teach non-human primates how to speak, in which a chimp named Washoe learned to recognize around 200 signs in American Sign Language (2014: 41). She was also able to comprehend and produce sequences of signs with the subject and object in the correct position, such as You tickle me, I tickle you (2014: 42). Evans emphasizes that though Washoe and other non-human primates that learned to sign were limited by the amount of working memory they were able to access, they were still able to grasp a produce a rudimentary form of syntax. In his 2002 paper Chomsky specifies “working memory” as one of the “organism-internal factors” that limits the human language function as well, capping the potential length of a comprehensible utterance (2002: 1571). By even Chomsky’s logic, the example of Washoe seems to point to non-human primates communicating in the same sense as people; they are able to grasp a sort of “proto-language” that while less complex than that of humans, is clearly on the other end of the same continuum.

The other key point in Evans’ argument is that all communication, both human and animal, is rooted in sociocultural learning and interaction. To support this contention, Evans cites other research in which primates’ natural gestural communication serves social, rather than simply survival purposes (2014: 38), and a study in which sperm whales were found to “name” themselves with a sequence of whistles so they could be identified by the other members of their pod (2014: 37). This social dimension is one in which Evans’ and other cognitive linguists theory diverges most from Chomsky. Universal Grammar holds that input from the mother and father activates the child’s innate sense of syntax, but there is no recognition of language evolving in tandem with or as an effect of pro-social skills and cooperative behaviors. 

If language does indeed develop as part and parcel of sociocultural conditioning, and in tandem with other cognitive functions, the Chomskyan arguments of “autonomous syntax,” and “innateness” are effectively nullified. So what of “universality’? Evans and various other cognitive researchers such as Nicholas Evans and Stephen C. Levinson contend that the belief in language universals is ill-founded, potentially ethnocentric, and ultimately a result of a lack of communication between experts with knowledge of the unique intricacies of the less-studied languages, and researchers in the other cognitive disciplines (Evans and Levinson 430). Nicholas Evans and Stephen C. Levinson provide a robust contention in their 2009 paper that “languages differ so fundamentally from one another...that it is very hard to find any single structural property that they share” (2009: 429).

To support this theory, Evans offers examples of linguistic diversity in terms of sound systems, morphology, parts of speech, and syntax (2014: 70-77). In terms of sound systems for example, Evans presents the example of the Australian aboriginal language “Arrernte,” which organizes its syllables in a “vowel-consonant” structure rather than “consonant-vowel,” as it was assumed until recently that all the languages of the world did (2014: 71). This discovery is emblematic of an important distinction that nullifies Universal Grammar: while Chomskyan theory would assume that there are a finite number of “universal phonetic categories,” a cognitive perspective emphasizes a more physiologically-based system of “universal phonetic resources” which can be “organized and harnessed by the cognitive system” (2014: 71).

In addition to these examples of linguistic diversity, Evans also presents some reasoning around why and how such diversity has evolved, in terms of linguistic “usage pressures” (2014: 65). He compares language evolution to biological evolution in neo-Darwinian theory, contending that language change can be compared to the concept of gene mutation, in which a base “replicator” (like a gene) is passed down through generations, but necessarily undergoes changes and variations over time as a result of interacting with its environment (2014: 80). These changes may be subtle and internal, such as phonetic changes to the Queen’s English between 1952 and 1982, or dramatic and external, as in the famous example of the Normans conquering Saxony. This perspective literally compares language to a living entity, in stark contrast to the Chomskyan model in which the basis of language is assumed to function like a preprogrammed software.

Universal Grammar can be seen as a product of the “mind-as-computer” metaphor, which has been pervasive in cognitive science since the 1950’s (2014: 133). It was a convenient analogy with the recent development of the computer, a man-made entity capable of abstract reasoning. From this viewpoint, it seems reasonable to assume that the human mind, like the computer, would be “modular” or rather, a “system” made up of “component parts,” each “associated, almost exclusively, with very specific functions,” and capable of functioning independently without “qualitatively affecting the performance of the whole” (2014: 133). This theory of mind modularity was popularized by the philosopher Jerry Fodoer in his 1983 book, in which he asserted that language is a prototypical example of a “mental module” (Evans, 2014, 134). The theory of “mind-modularity” provides the modern basis for Chomsky’s original assertion of language as a “domain-specific” phenomenon, localized in the brain by either a physical region, neural system, or perhaps a set of identifiable genes. Pinker himself admits that the precise manifestation of this module has not been found (1994: 49), but the arguments defending it, defining it, and redefining it have been waged at length by the linguistic community (see section 3.3). 

According to Evans, for language to be a distinct module in the mind, three suppositions must be true: “there must be an identifiable location or system within the brain associated only with grammar,” “there must be a dissociation between grammar and other types of cognitive behavior” including other types of non-innate linguistic knowledge such as “vocabulary,” and there must be a defined “maturational trajectory” for grammar that shows it developing on an independent timeline (2014: 136). Evans systematically provides a wealth of counter-evidence against each supposition, claiming that the theory as a whole provides an overly-simplistic account of “the relationship between language and other aspects of cognition” (2014: 135). 

In terms of localization, Evans presents the examples of Broca’s area and Wernicke’s area; these have both been long associated with the function of language via studies of aphasia, a disorder in which sufferers are unable to either produce or comprehend language respectively, as a result of those areas being damaged (2014: 137). These associations however well-documented, are not substantial proof that those areas are “dedicated to language at the exclusion of all else” (2014: 142). Firstly, the exact location of Broca’s area remains a topic of disagreement among various laboratories (2014: 143). Among studies that claim a physical region for Broca’s area, there are documented cases of its being activated during types of cognition other than language, such as “the planning of various nonverbal motor tasks” and the “ability to recognize discordant sequences in music” (2014: 142). Relevantly, Broca’s area has also been documented as being involved with allowing people to “perceive and even interpret the action of others” and is even the site of specific “mirror neurons which fire in response to specific actions of others, especially gestures” (2014: 143). These findings are not only damaging to the theory that Broca’s area is a language-specific functional unit, they also support Evans’ larger contention that language develops concurrently with social knowledge.

The relevant research also rules out the possibility that the “language module” is a genetic one. Universal Grammarians are quick to claim a certain gene, “FOXP2” as evidence of such a module, citing a famous study in which several generations of a London family suffered difficulty in learning and producing language, with other aspects of cognition seemingly unaffected (2014: 138). This type of condition is known as Specific Language Impairment, or SLI (139), and would also appear to be evidence of the “dissociation between grammar and other types of cognition” (142), but Evans contends that neither of these claims are very solid. For one, the affected members of the London family had an overall lower IQ than the average by eighteen points. Like Broca’s area, FOXP2 is also associated with cognitive functions beyond language, including the “ability to sequence muscles of the face and mouth that are not involved in language,” “development of the heart, gut, and lungs,” and also “general intelligence” (2014: 143-144). 

The dissociation of language from other cognitive function can be considered from one other perspective as well: cases in which the linguistic abilities are average or above, but the IQ is significantly lower than average--a phenomenon known as “Chatterbox Syndrome”. These cases are what Pinker would call “linguistic idiot savants” (136), and frequently cited cases are largely comprised of people with Williams Syndrome, who exhibit “surprisingly good linguistic abilities...relevant to their diminished visual-spatial and reasoning capabilities” (2014: 144). Despite Universal Grammarians latching onto this example, it is again not conclusive evidence that these individuals are linguistic savants, given that their linguistic abilities fall below those of their average peers in the same age ranges (2014: 144). Additionally, there is evidence that Williams Syndrome subjects develop language by a cognitively different process than their average peers (145). Since Williams Syndrome patients cannot be considered linguistic savants, it is not feasible to assume that grammar can be dissociated from other forms of cognition.

In terms of the requirement for a clear trajectory of independently grammatical development, Evans investigates a supposition of the “language-as-instinct” thesis. Universal Grammarians hold that one of the strongest arguments for the existence of a human language module, or even Chomsky’s “language acquisition device,” is that children seem to make a sort of binary “jump” from producing isolated instances of vocabulary to suddenly being able to produce grammatical sequences around the age of two (2014: 141). According to this theory, while grammar is innate, vocabulary must be learned (146). However, recent research in the field of child language acquisition has shown that this concept of a mysterious and “discontinuous jump” is not factual--rather, the acquisition of grammar is dependent upon the acquisition of vocabulary. It seems that of children within this crucial “sixteen to thirty month age range, “performance on grammar seems to closely correlate with vocabulary size” (147). Essentially, there is no discrete developmental trajectory for grammar, instead it is a coevolutant of vocabulary.

Evans takes down the modularity fallacy in one final aspect. Some scholars (see Adger in section 3.3) hold steadfast to the modularity concept with the assertion that the brain is “functionally not physically modular,” encompassing “neurological systems, which might overlap with others [but nonetheless] have evolved to perform specific functions” (151). There are a few issues with this theory. Firstly, that if the brain is comprised of a series of informationally discrete modules with a de-emphasis on “cross-talk” or collaboration, that implies the necessity of a “meta-module” to oversee the order and operations of the smaller modules, and creates a separate issue of defining and finding evidence for such a “central intelligence” (153). Secondly, if brains were truly modular, Darwinian theory would require the modules to evolve independently over time, while recent research suggest just the opposite: that it developed in a coordinated fashion (154).

In dispelling the myth of Fodorian “modularity” and subsequently Chomskyan “domain-specificity,” Evans is left with the opportunity to suggest an alternative. To this end he offers theories in line with the studies presented throughout the section: Terrence Deacon’s theory of co-evolution, in which “changes in one part of the brain entail changes in another part,” such that language developed concurrently with other mental faculties (157). He does however highlight the necessity of accounting for instances of specialization within the human brain, despite the frequency of “cross-talk” in these instances (158). His suggestion, based on findings by the psychologist Annette Kamiloff-Smith, are that children’s brains at birth respond to stimuli with the activation occurring in an evenly-distributed fashion, but “specific circuits become increasingly specialized as the brain becomes more efficient” (159). In this view, specialization is a normal developmental feature that develops as a result of interaction with the environment, rather than a pre-programmed, immutable, innate fact. 

Ultimately, Evans refuted the theory of Universal Grammar in each of its core arguments, and provided a thorough overview of the available research in support of Cognitive Linguistics, even anticipating the objections of Universal Grammarians and addressing those. He also effectively conveyed the dissonance between the dominant dogma in the field of linguistic research, and the attitudes of modern-day cognitive research in other disciplines such as neuroscience, neurolinguistics, and psychology. He did not shy away from divulging the political aspect of this academic world, emphasizing (like many other researchers) that the propagation of the Universal Grammar Theory is not only wrong-minded and lacking of empirical base--it is staked on the professional reputations of Chomskyan followers, who tout its veracity as a matter of vested personal interest, rather than considering the prospect that it could be wrong (141).

3.3 Response to The Language Myth 

Predictably, the reception of Evans’ book within the Chomskyan community was not a warm one. Directly following its publication, a Universal Grammarian named David Adger published a heated response in the academic journal Lingua. Because the exchange that took place between Adger and Evans (who was joined by fellow cognitive linguist Behme for the papers written in defense of Evans’ book) is in some ways emblematic of the dialogue between formal and cognitive linguists in the field today, it will be reviewed here. Adger defended Generative Grammar on the grounds that Evans did not demonstrate an understanding of Chomsky’s theory sophisticated enough to properly critique it. He claimed that Evans misrepresented the theories of generative linguistics on a few counts. 

Firstly, Adger asserted that Evans’ book suffered a misapplication of fundamental terminology by misrepresenting the phrase “language instinct” as being literal rather than its intended metaphorical definition. Secondly, it conflated “typological” (surface level) universals” with Chomsky’s theory of Universal Grammar (which intends to represent the “capacity” to form and understand language as a mathematical function), and similarly conflates “category recursion” (the potential for limitless grammatical embedding to occur in a sentence) with “Chomsky’s Recursive Function / Computability theory” the idea that humans use a “finite set of bits of language” to create a “potentially infinite set of structures.” On a more biological level, Adger says that Evans’ book misunderstands the generative linguists’ claim that language is a distinct Fodorian “module” of the human mind, by asserting (as Evans predicted one would) that the module may be “functional” rather than “physical”. 

Having spent most of the article attacking Evans “understanding” and “representation” of generative grammar, Adger dedicates a proportionately small amount of the article to refuting the theory Evans proposes to replace it. He briefly summarizes Evans’ “language-as-use” theory by saying it asserts that children acquire language by learning “particular forms, not abstract categories” and thereafter inductively acquire the “categories and rules” of language through painstaking trial and error. In his argument against this proposal, Adger cites Yang (2013), whose research--which like most theories of generative grammar was founded on mathematical formulae, in this case Zipf’s law--proves that there could never be “enough data” for children to generalize grammatical rules from it, and in any case, what children know about language is not always reflected in what they say (Valian et al 2009), so studies of their production are not an accurate means for gauging their knowledge of language structure. This argument falls flat in that it conveniently negates an entire category of research, namely any study done using childrens’ speech production as data.

Behme and Evans refuted Adger’s criticisms in a paper of their own, published in the next issue of Lingua. They demonstrated how his critiques are too narrowly focused on “terminological quibbles,” rather than addressing the more pressing issues raised by not only The Language Myth but by the research presented within it from a vast array of scientific disciplines. It systematically countered each “mistake” Adger claimed was made by the original text, and pointed out, at every turn, the book’s location in the conversation in the field of linguistics at large. For example, since one of Adger’s main criticisms was Evans’ misunderstanding of the term “recursion,” the response article provided a detailed history of the term “recursion” and how its meaning has been hotly debated even among Chomskyan Linguistics since Everett (2005) refuted the definition laid out in Chomsky, Hauser, and Fitch (2002). Ultimately the paper offered a call to action for Adgers and the Chomskyan community at large, to define “which specific theories Chomskyans are currently committed to, [ii] which concrete findings from developmental psychology and neurobiology support the Chomskyan framework, and [iii] how...the Chomskyan paradigm...can overcome the familiar, long standing challenges stated in the technical literature” (2).

Adger’s response, also a published article, opened with claims of a largely different nature than his previous one, in which he asserted himself to be a theoretical pluralist, who supports collaboration between fields and theories of linguistics. He criticized the argument of Behme and Evans (2015) saying it was built more upon rhetoric than content. This criticism, not unlike Adger’s initial criticisms of the book, is too narrowly focused in scope. He ignored Behme and Evans’ advice to address specific counterarguments to Generative Grammar and dedicated a few more paragraphs to re-defining the terms “instinct” and “recursion.”Finally, Adger concluded by repeating that though collaboration between fields and theories of linguistics is necessary, The Language Myth egregiously misrepresented Generative Grammar, and it seemed at best unlikely that any variety of Construction Grammar could replace Generative Grammar as it has made great contributions to the field and would therefore always be necessary.

4. Conclusion

The great debate waging within the field of linguistics today is one steeped both in history and bouts of semantic sparring. It is equally rooted in empirical scientific study as it is shrouded by egoic interest, and misdirected by a lack of communication among specialists. Despite Evans’ book being widely read, and the attention garnered by alternative approaches to grammar, Universal Grammar remains “the single-largest ‘school’ in Anglo-American linguistics” (Evans, 2014, 140). If the academic landscape is to be assessed based on the quantity of departments focused on Cognitive Linguistics vs. the amount following a Generative curriculum, the comparison is fairly abysmal. The only doctoral program in the western hemisphere dedicated to the integrated study of Linguistics and Cognitive Science exists at the University of Birmingham in the UK, while only a handful of other programs throughout the US and Europe offer departments focusing on Cognitive Linguistics / Semiotics, or in some cases, a single faculty member. 

Whether this imbalance is a result of old-guard academics upholding the status quo, unable to acknowledge their own blindspots, remains to be seen. The publication of Evans’ book could inspire the next generation of linguists to seek more intra-disciplinary perspectives to inform the study of language--and if the attitudes of current neuroscientists are as shocked by the dissonance between the two fields as Evans purports them to be, the impetus for this shift is already gaining inertia throughout other areas of Cognitive Science. If the academic conversation to date is any indication of its future progression, Universal Grammarians are hindered by the impulse to continually defend, re-define, and semantically repackage an old theory that they are not focused on advancing the field of linguistics. When confronted with matters of language and the mind, Universal Grammarians study linguistics internally, and look to the past. Cognitive Linguists seek out external collaboration, and look to the future.

Works Cited

Adger, D. (2015). Mythical myths: Comments on Vyvyan Evans’ “The Language Myth.” Lingua, 158, 76–80. doi: 10.1016/j.lingua.2015.02.006

Adger, D. (2015). More misrepresentation: A response to Behme and Evans 2015. Lingua, 162, 160–166. doi: 10.1016/j.lingua.2015.05.005

Allott, N., & Rey, G. (2017). The many errors of Vyvyan Evans’ The Language Myth. The Linguistic Review, 34(3), 1–20. doi: 10.1515/tlr-2017-001

Barman, B. (2014). The Linguistic Philosophy of Noam Chomsky. Philosophy and Progress, 103–122. doi: 10.3329/pp.v51i1-2.17681

Behme, C., & Evans, V. (2015). Leaving the myth behind: A reply to Adger (2015). Lingua, 162, 149–159. doi: 10.1016/j.lingua.2015.05.004

Chomsky, N. (1964). Aspects Of The Theory Of Syntax. The MIT Press. doi: 10.21236/ad0616323

Evans, N., & Levinson, S. C. (2009). The myth of language universals: Language diversity and its importance for cognitive science. Behavioral and Brain Sciences, 32(5), 429–448. doi: 10.1017/s0140525x0999094x

Evans, V. (2014, December 4). Real Talk: There is No Language Instinct. (E. Lake, Ed.)Aeon. Retrieved from https://aeon.co/essays/the-evidence-is-in-there-is-no-language-instinct

Evans, V. (2015). The language myth: why language is not an instinct. Cambridge: Cambridge University Press.

Everett, D. L. (2005). Cultural Constraints on Grammar and Cognition in Pirahã. Current Anthropology, 46(4), 621–646. doi: 10.1086/431525

Everett, D. L. (2007). Challenging Chomskyan Linguistics: The Case of Pirahã. Human Development, 50(6), 297–299. doi: 10.1159/00010983

John Lyons. (1966). The Philosophical Quarterly (1950-), 16(65), 393-395. doi:10.2307/2218520

Haspelmath, M. (2018, September 13). Retrieved from https://dlc.hypotheses.org/1269

Hauser, Marc D., Noam Chomsky & W. Tecumseh Fitch. 2002. The faculty of language: What is it, who has it, and how did It evolve? Science 298. 1569–1579. doi:10.1126/science.298.5598.1569.

History of Linguistics. (n.d.). Retrieved December 1, 2019, from https://www.uni-due.de/ELE/HistoryOfLinguistics.htm.

Lees, R. B., & Chomsky, N. (1957). Syntactic Structures. Language, 33(3), 375. doi: 10.2307/411160

Levinson, S. C., & Evans, N. (2010). Time for a sea-change in linguistics: Response to comments on ‘The Myth of Language Universals.’ Lingua, 120(12), 2733–2758. doi: 10.1016/j.lingua.2010.08.001

Newmeyer, F. J. (n.d.). The History of Modern Linguistics. Retrieved December 1, 2019, from https://www.linguisticsociety.org/resource/history-modern-linguistics.

Penn, D. C., Holyoak, K. J., & Povinelli, D. J. (2008). Darwins mistake: Explaining the discontinuity between human and nonhuman minds. Behavioral and Brain Sciences, 31(2), 109–130. doi: 10.1017/s0140525x08003543

Pinker, Steven. 1994. On language. Journal of Cognitive Neuroscience 6(1): 92-98.

Pinker, S. (1994). The language instinct: the new science of language and mind. London: Alle Lane.

Toolan, M. (2014). Book Review. Journal of Literary Semantics, 43(1). doi: 10.1515/jls-2014-0005