Can a metaphor solve an enigma?


Review of The Language Game: How Improvisation Created Language and Changed the World, by Morten Christiansen and Nick Chater, Basic Books, 2022. By Paul Bouissac.

Explaining the emergence and evolution of language, and the diversity of languages, remains the most challenging issue for linguists who endeavor to approach the problem in a scientific rather than ideological way. It is also a central, unresolved issue in semiotics. The contemporary linguistic landscape is mapped unto several theoretical views based on postulates that often were construed as axioms and eventually gave rise to various dogmatic – at times even cult-like —  schools of thought. These divergent beliefs are grounded in subjective phenomenological evidence that is trusted to yield reliable intuitive knowledge and heuristic models that run the risk of being confused with reality. The current existence of competing paradigms, though, should make it clear that the lack of a consensus demands an explanation beyond the accusations of ignorance or incompetence that  partisans throw at each other. After centuries, if not millennia of intense but inconclusive speculations on language, one may wonder whether language can at all be construed as a domain of scientific inquiry using its own means to investigate itself. The nature and origin of language remain a horizon of ignorance, a daunting enigma. Any approach that claims to be novel is worth examining with a critical eye. 

Morten Christiansen and Nick Chater confront anew this challenge in the wake of their earlier, ambitious Creating Language: Integrating Evolution, Acquisition, and Processing (MIT 2016), a typical work of scholarship that aims at expounding their original theoretical views on language supported by a combination of creative insights, heuristic metaphors, scientific evidence and logical reasoning. When such books are published by a reputable academic press, they have been submitted to an exacting peer-review process that usually improves the quality of their argumentation. The copy-edited style of MIT books is conventionally austere and somewhat demanding. The benefits of such publications for their authors are measured more in terms of  reputation and professional advancement than in financial gains. Christiansen’s and Chater’s  latest work, by contrast, belongs to a different genre: the trade book, a medium that targets a much wider audience by creating a different kind of expectations: a novel, revelatory narrative that promises intellectual pleasures and unrestricted enlightenment rather than scientific self-doubt and tentative, provisory conclusions. The usual channel for such trade books is mediated by a literary agent who assesses the potential interest of the topic and title, and endeavors to find, for a fee, the right publisher that is ready to meet the costs of  its production and marketing with a view to garner a likely profit, a significant part of which is expected to go to the authors who often get a substantial advance commensurate with the attractiveness of their topic and their own reputation based on previous records. In this process, the seductive potential of the title and the alluring style of the contents, including vivid anecdotes, are foregrounded. There is also a quantitative constraint as the pricing of the printed book is commensurate with the number of pages, thus potentially motivating the author(s) to introduce “fillers” in the form of supplementary or superfluous textual developments. In addition, the critical evaluation of specialists in the relevant field of inquiry is less important than the public appeal of the narrative’s rhetoric and the marshalling of supporting references verging at times on mere name dropping. It is symptomatic that, in their acknowledgement at the beginning of their MIT book,  the authors thanks the reviewer whose remarks have improved their text (and who, incidentally, agreed to write a praising foreword), while, in the concluding “Acknowledgments” of The Language Game, they express their gratitude not only for the inputs of the many sympathetic students and colleagues who offered comments on the successive drafts of the book but also, more crucially, for the help of the literary agents that brought their book to concrete existence. These remarks are not meant to disparage this genre of trade books that is crucial for effectively communicating advances in important research domains to a wider audience than the restricted circles of specialists who have access to scientific publications. However, it is important to demarcate these two genres that are not submitted to quite the same criteria. It is easy to take the measure of  the crucial difference between academic presses and trade books publishers by checking the websites of MIT and Basic Books, the respective publishers of Creating Language and The Language Game, in which some policy details are described. The peer-review process is emphasized in the former and played-down in the latter. 

The book under review is designed to both enlighten and entertain, and it relies on appealing metaphors – literally – to formulate its thesis and develop its arguments toward resolving the language enigma. Metaphors are invaluable poetic, rhetorical, and heuristic resources, but they are prone to generate cognitive biases, errors, and even moral abjections, such as “ethnic cleansing” (that equates some ethnic identities with dirt) or “refugee filtration” (that construes human discrimination as purification). As one of the authors, Nick Chater, warned in his earlier book published by Yale University Press, ”Metaphors are just as apt to mislead as they are to inform” (Chater 2018: 210) [emphasis in the text]. Sensitivity to homologies in dynamic patterns in the social or natural environments can trigger true discoveries with far reaching scientific consequences but can as well induce humans to adopt delusory visions of harmony or unresolvable conflict implicitly carried by the images. Metaphors are tricky and must be handled with care by researchers because they can lead to tragic pitfalls. 

The metaphors of “game” and “charade” are the two main themes that generate the “ultimate” theory of language that Christiansen and Chater confidently propose to their readers. They make indeed the strong claim that these children’s common social behaviors that are assumed to be universal, explain why language was “invented”, evolved, and has become so important that, from such trivial beginnings, it “changed the world” for the better as Chapter eight clamors.  Understandably,   trade books must display provocative titles of  the “Tail that wags the dog” kind to catch a distracted audience’s attention. One may then wonder why spend time to peruse and critically review a book that orchestrates such a thesis grounded in rather casual metaphorical thinking. In a socio-political context in which truth is the victim of fake claims and scientific knowledge is mocked as “just theories”, a book that asserts that language is (just) an improvised  game of charade — “just a game” as the authors proclaim in a self-promoting article in the popular New Scientist magazine (2022) — must be called out as vigorously as possible. Language is not only impromptu, innocuous chats designed to collaboratively share information within a Panglossian “virtuous circle” but, more decisively, it articulates propositional statements that may or may not be true or ethical, and conveys narratives that impact behavior, and, thus, are highly consequential (Gottschall 2012, 2021). Even casual spoken language, that is often more articulate and formal that the authors suggest, must be kept accountable. It more often than not carries devastating lies and prejudices, and is used for crippling calumnies, and even deadly bullying.  

Naturally, as Christiansen and Chater point out at the onset of the book, the idea of “language as game” is not new. It is a leitmotiv in Ludwig Wittgenstein’s posthumously published Philosophical Investigations (1953). The questioning of the stability of words’ meanings that change from context to context is central to his notion of language as a game. Never mind that Wittgenstein’s “thought experiments” and ad hoc examples cannot survive close critical scrutiny. Contemporary pragmatics has extensively dealt with these issues. Christiansen and Chater pick up as a telling example Wittgenstein’s choice of the term “water” that in isolation may convey an indefinite variety of meanings (except, of course, what is not water such as air or earth), but we always know what is meant when we use or hear it. As Wittgenstein specified in his earlier Tractatus Logico-philosophicus: “only propositions have meaning; and it is only in the context of a proposition that a name has a signification” (Tractatus 3.3). Did the “second” Wittgenstein denounce this view as a fallacy? Probably not since the context remains crucial for the determination of the meaning of the word “water”. We can note, though, that Wittgenstein’s semantic skepticism should  ironically apply as well to the risky use of the charade metaphor since this word  may mean in English EITHER “a game of guessing a word, a character, a quality, or an object through pantomiming clues of some of their properties” OR “an absurd pretense of respectable appearance”. In The Language Game, the word “charade” is laboriously  redefined to meet the purpose of the authors and escape its darker meaning. The same ambiguity is attached to the term “game” whether it is taken in its ludic sense (as Christiansen and Chater, following Wittgenstein, consistently do with their definition of language as a “game of charade”) or used more technically as Saussure did when it compared the dynamic structure of synchronic language to the interdependent rules of a game of chess, or, still more to the point, as social interactions leading to decisions that are formalized in Game Theory. Those latter games can be deadly serious and do not allow for “improvisations” on the fly.  Other aspects of the book will be questioned in the course of this review but its claim concerning the origin of language deserve utmost attention because it is ideologically loaded and politically consequential.

We can be confident that language, like anything else in organic nature, has emerged through natural selection, by chance rather than design, as an efficient tool of communication from a human adaptation perspective. However, as the authors counter-intuitively, albeit timidly indicate, language itself has evolved ways of spreading itself through its own adaptive strategies under co-evolutionary constraints. This latter view is mentioned by Christiansen — an earlier proponent of the memetic theory of language — and Chater, but remains undeveloped in this latest book. They only address the current debate concerning the nature of language between those who believe that language appeared as the result of sudden mutations and those who privilege the hypothesis that language as we know it results from cumulative small changes, both anatomical and cognitive,  each being adaptive in their own right, not in view of a distant ideal advantage. Naturally, the rise of natural language in its many forms is beyond the range of our direct observation – lost in “the mist of time” as the authors felicitously write — and can only be imagined through more or less educated guesses —  and the perilous use of metaphors. 

A common mistake made by modern linguists who ponder the origins of language is that they fail to conceptually disentangle the interferences  between oral /aural languages that festered among humans for at least two million years, and their relatively recent written forms that were created by their users for special purposes over the last ten millennia at the most. In face-to face oral/aural interactions the agents have access to shared contexts and to the obvious current situations of the interactants. Oral language allow them to fill a few information gaps that are relevant to the purpose of the action, communication or negotiation at hands, information that cannot be fully supplied by the postures and gestures recruited to mimic narratives. This assumed earliest forms of language – that can also be readily observed among illiterate interactants — are rarely seriously considered by linguists who tend to view parataxic verbal communication blended with nonverbal communication in the mental space of working memory, as a form lacking the richness of syntactic resources, in other words, as “not-yet language”. It could be claimed, though, that syntax is needed only when the communicating interactants are distant from each other in time or space, and have no access to immediate shared contexts and occurring situations,  that is, when social groups disperse through fission over areas that make direct face-to-face communication impossible. In brief, while oral language is an obvious adaptation by natural selection, written language is a cultural artefact with specific functions, and it possesses a dynamic of its own not necessarily congruent with the dynamic of the oral/aural language it endeavors to represent. While it is possible to conceive spoken language as an “instinct” in the sense that it emerges spontaneously in children who are exposed to speaking caretakers, it is impossible to refer to the “writing instinct” since writing and reading must be painfully learnt and practiced in order to achieve competence in that type of visual communication modality. Orality and literacy command a different approach when we try to understand how they evolved and developed. Failing to conceptualize this crucial difference leads to confusion. This is particularly challenging in a literate society in which disentangling the two modalities of language is often impossible because they are inextricably mingled in the mind of most linguists – who are all highly literate — , a fact that has been appropriately dubbed by some “the tyranny of writing” (e.g., Weth and Juffermans, 2018).  

This general framework must be kept in mind when considering Christiansen’s and Chater’s The Language Game in which they attempt  to elucidate the nature of language in its many forms, and claim to have solve the enigma of its evolutionary origin. Their 291 pages book is divided into eight chapters introduced  by a preface and followed by an epilogue (pages 1-228). Acknowledgements, notes and resources, and an index take the remaining pages of the volume (pages 229-291). 

The preface makes a stunning claim: language is a human invention and, more expectedly, language defines what it means to be human. Moreover, “actual language is always a matter of improvisation, of finding an effective way to meet the communicative demands of the moment”. This opening theoretical stand exclusively refers, it would seem, to spoken language. However, the authors soon backtrack to explain that “today’s sophisticated linguistic systems” have spontaneously emerged from an incremental, inadvertent “bricolage” leading to systematic grammatical rules,  their characterization of which appears to mostly apply to the norms of written language as they impact orality. Surprisingly, scripts in any form, that is, artefacts that are truly human inventions, are taken for granted as a simple visual version of languages. Invoking the authority of Ludwig Wittgenstein, who, as we noted above, made the trivial remark that the meaning of a word depends on the context of its use, they come to the conclusion that “learning a language is learning to become a skillful charade player”, hence the major claim of the book that language is a game of charades. New perspectives on the nature and origin of language should always be welcome but should also be critically examined because, as we emphasized above, languages are consequential and dramatically impact social behavior and history.

Chapter One develops the postulate that may appear as shockingly counter-intuitive to mainstream linguists and more so to lay readers whose daily life experience of language is anything but a game of “collaborative improvisation”.  The bold goal of these 23 pages is to reveal the essential nature of language, thus bringing to closure an age-long quest. The whole epistemological  construction of the book hangs from the soundness and irrefutability of the evidence offered here. The challenge, of course, is to demonstrate the adequacy of the root metaphor that is claimed by the authors to explain the evolutionary origin, development, and diversity of language. However, this end is not to be achieve through empirical evidence but by means of suggestive anecdotes, thought experiments, and intuitive truths construed as postulates. The stakes are very high and the claim deserves scrutiny all the more so since the Wittgenstein’s quotation placed in exergue at the start of the chapter may raise concern in the mind of the reader if this truism – “[…] speaking a language (sic) is part of an activity” — is seriously proposed as a revelatory statement that inspired the authors’ quest.      

Any word in any language is the provisory end point of a long morphological and semantic history that cannot be traced back beyond a few attested stages to some absolute beginning. There are nevertheless some traceable terminological paths. The word  charade in English comes from the French charrade, originally the Occitan charrada that is derived from the verb charrar  meaning chat, idle conversation. It became, in the 18thcentury, a parlor game, probably originating in popular children pastime, a kind of verbal rebus, that can be acted out in a silent version. It is this latter semantic avatar that is chosen by Christiansen and Chater as the ludic model metaphor of their theory of language. Interestingly, this game is close to the games of cross-words, riddles, or scrabble as it consists of playing with the components of an pre-existing object (word, action, role, or any other identifiable elements of the contextual culture). Notwithstanding that reconstructing a target each syllable or other part of which is indicated by a puzzling definition or gesture as a clue, presupposes the existence of a language that provides the players with a nomenclature, the claim is forcefully put forward  that the game of charade is the key to explain the origin of language. In other words, it would seem that language is needed to play charade. In order  to overcome that contradiction, Christiansen and Chater conveniently restrict at first their understanding of charade to the gestural version of the game before adding vocal sounds to the evolutionary process. In doing so, they refer to examples they observed in their own children – who, incidentally, grew up in the pervasive language environment of a highly literate and educated family.

In attempts to represent early cognitive stages in the evolution of the genus Homo, it is common to imagine more or less consciously these “primitive” humans as children. No metaphor could be more misleading. Natural selection eliminated quickly infants that were affected by ill-adaptive mutations. Evolution can fashion changes only through mature organisms who can transmits their genetic mutations to their offspring if they are adaptive with respect to the state of the physical or social environment through natural selection. 

The way children progressively master the language spoken in their social environment is an important topic of research that has produced an abundant literature. It is problematic, though, to draw from these observations logical conclusions regarding the nature and origin of language simply because language has to have evolved as a means of communication — or other adaptive behaviors such as courtship and mating rituals, affiliative singing or magic mantras, and other acoustic productions in hunting and warfare — among mature individuals before it can be transmitted by parents to children. The development of language does not provide reliable clues about its origin unless, of course, one uncritically follows and generalizes Ernst Haeckel’s discredited idea that ontogeny recapitulates phylogeny, a theory that was later controversially revived by Stephen Jay Gould under the name of evolutionary developmental biology, and continues to be more or less implicitly endorsed by psychologists such as Michael Tomasello whose research (and thought experiments) are abundantly quoted in The Language Game. Considering ontogenetic development as a recapitulation of  the biological and cognitive evolution of the genus Homo leads to a fundamental misconception that more or less consciously projects on early humans some child-like, immature abilities. The implicit narrative that construes today’s humans as the matured end-result of an evolutionary “growing up” metaphor toward (western) “civilized” humans, explains the tendency of looking at the development of children as a window on the very origin of language. This  approach overlooks the fact that Homo sapiens, their ancestors and their contemporary evolutionary relatives were fully adapted  organisms that had survived and reproduced for several millions years with sufficient means of intra-specific communication as it is the case for all social species. The development of language in modern children presupposes the existence of a spoken language in the group within which they are born and brought up. Consequently, explaining language evolution on the model of language development in children amounts to presupposing language to explain its existence, an obvious self-defeating paradox.   

It is obviously from that unsustainable assumption that Christiansen and Chater endorse another equally controversial hypothesis that explains the origin of language as having evolved from the use of gestures to communicate. Here too, they abundantly refer to Michael Tomasello who is one of the contenders, albeit with qualifications, of Haeckel’s theory. The gesture hypothesis is appealing as it fits the ideology of progressive evolution leading humans from simple concrete, “animalistic”, corporeal behavior toward the proclaimed perfection of linguistic communication and articulate cognition. However  paleontology and fossil evidence show that the common ancestors of primates were rather small tree-dwelling organisms. It is therefore difficult to imagine that arboreal life would have been compatible with primarily gestural communication beyond grooming and proximal facial expressions. In such an environment, limbs are adapted to maintain balance and grip on branches, and can hardly be fully coopted for gesturing toward distant targets. Moreover, in an mostly opaque, leafy milieu, visual communication is of limited functionality whereas acoustic signaling is the only medium able to transfer vital information over functional distances. Gestures as a productive semiotic resource could emerge only after the genus Homo became bipedal (erectus) and found themselves endowed with upper limbs  that could be exapted to other uses than the ones for which natural selection in the arboreal environment had fashioned them. It can be very confidently assumed that acoustic communication greatly predates gestural signaling and provides the most plausible evolutionary continuum from vocal signals to articulate language. 

Christiansen’s and Chater’s reliance on Tomasello’s theoretical views is further indicated by their bringing into focus one of his thought experiments that they consider “compelling” (p. 13), although with ethical and plausibility qualifications, according to which two groups of infants are  brought up in separate islands, without any adults around (sic!). One group cannot rely on vocalization to communicate and is forced to use gestures alone; the other group must only use vocalization. Never mind that we are not told who makes and enforces the rules. The expected result, Tomasello argues, would be that only the children from the gesture-alone group “would have a chance  of evolving language-like communication” (P.14). Whether at the Max-Planck-Institute or elsewhere, such a reasoning should raise alarm.    

Thought experiments may be productive heuristic devices to explore ideas by virtually playing out their consequences, but if they are used as rhetorical tropes for arguments in the service of unprovable (or unfalsifiable) hypotheses, they must be denounced as fake logic abusing the bona fide quest for truth of inquisitive minds who might be so impressed that they become insensitive to these thought experiments’ absurdity.  Fortunately, Christiansen and Chater invoke another experiment, to complement Tomasello’s by the result of empirical investigations of the capacity of humans to vocally mimic entities to which they refer, thus casting their charade-game  metaphor as the likely origin of verbal languages.

 We can confidently assume that the common ancestors and their descendants lived in an environment perfused with sounds produced by other species that were eminently relevant to the survival of those who could relate them to adaptive behaviors as well as to a host of physical acoustic phenomena. Human brains are prone to pick up the algorithms that enable them to reproduce both precise actions and specific sounds. Mark Dingemanse, for instance, has shown that natural sounds such as dripping water or rolling stones can give rise to original elements of the lexicon through a very productive vocal process called “ideophones”. The evidence do not come from thought experiments but from fieldwork observation of the language of a population in Ghana (Siwu). (Dingemanse 2012, 2017, 2018). Humans and some other species easily reproduce vocally what they hear. This is so much the case that natural selection has equipped the brain with inhibitory circuits making it possible to keep imitation under relative control in mature brains (Brass et al. 2009; Darda and Ramsey 2019; Nishimura et al. 2018). Impairments of these inhibitory capacities lead to echolalia, the automatic, dysfunctional repetition of acoustic inputs or other dynamic behaviors. In altricial species, the offspring are hard-wired to imitate their immediate caretakers. This forms a robust basis for a continuum between the relevant acoustic information picked up in the environment and intra-specific communication, both gestural and vocal. Since Christiansen and Chater make a frequent use of anecdotal evidence in their argumentation, I  would like to contribute one myself to the debate: I was once in Russia for a conference in St. Petersburg. On my way to the venue, I entered a pastry shop and bought a coffee and a piece of cake. With both hands busy holding the cup and the plate, I realized that I also wanted some milk in my coffee. I did not know how to say milk in Russian and my hands were not free to imitate the milking of a cow’s udders. I spontaneously uttered a “mooing” sound and immediately received the desired supplement. This illustrates the fragility of the gesture origin argument since using gestures for communicating assumes that the hands are idle, an unlikely feature of situations in which vital information must be conveyed to group members who are within hearing distance.

Another important source of multimodal social behavior is grooming and phatic communication. The latter is the constant production of affiliative species-specific differentiated vocal sounds that maintains the highly adaptive assurance that a group is in contact, even if they are not mutually visible, lest some sudden danger could jeopardize individual members’ survival. All social species have evolved distinctive alarm cries, that, it has been often noted, can be used by some individuals to mislead their con-specifics in order to remove them from desirable sources of food. As some theoreticians of signs have claimed, the hallmark of the semiotic competence is the possibility to lie, hence to manipulate acoustic output in view of their effects on conspecifics and beyond.

Moreover, vocal sounds that are disproportionate with the size of the organisms that produce them or that mimic the cries of much larger species are highly adaptive. Many animals such as primates have tissue folding in their vocal track that contribute to these acoustic effects but interfere with the possibility of producing the better defined vibrations that enable  language articulation. However, it has been recently shown that Homo lost this anatomical feature in the course of evolution with the result of gaining an ability to utter clearly distinct sounds thanks to this chance mutation (Gouzoules 2022; Nishimura et al. 2022).      

It is all too often overlooked that no adaptation by natural selection is absolute since the source of the selection process is the environment, both natural and social. All adaptations have a cost in the sense that they always imply the possibility of giving rise to automatisms triggered by irrelevant contextual features or situations that mimic, by design or by chance, vital information. This is why they open the possibility of  deadly manipulations. In altricial species, infants are hard wired to imitate the caretakers lest they perish. This life-saving process may also happen to be a fatal one. The human brain was bound to be under strong pressure to select ways of discriminating among broadly similar patterns although the time gap necessary for learning differences implies the risk of missing this opportunity. Even the mature brains of altricial species, including humans, can be tricked into falling for lures or fake cues. All adaptations are necessarily, it would seem, vulnerable to errors and manipulations, hence the ambiguity of metaphors that may enrich as well as wreck our lives (Chater 2018: 210).     

In their first chapter, after having implicitly dismissed Chomsky’s theory as a fundamental misconception, the authors lay out their main argument by contrasting it with other models of linguistic communication that they consider to be reductive and lacking the actual richness and dynamic creativity of verbal, multimodal interactions. In so doing, they merge Shannon and Saussure in a caricatural one-way stream of encoding and decoding of information. Unfortunately, they skip, for the sake of legitimizing their own approach, the considerable work that has been accomplished in functional linguistics and pragmatics since Shannon and Saussure proposed their early models. Roman Jakobson (1960), for instance, developed the organon model of communication that had been devised earlier by Carl Buehler (1934), by adding dimensions and functions that encompass all the components that are claimed to be missing in the models of linguistic communication that Christiansen and Chater denounce as incomplete and flawed. Moreover, during the last five decades at least, numerous studies in socio-linguistics, pragmatics, and nonverbal communication research have applied these earlier models to a range of cultural domains including verbal, textual, and performance  productions, and comprehensive methods to study human interactions as multimodal processes have been elaborated.

Another issue that can be raised concerning Christiansen’s and Chater’s theory is their “bottle neck” metaphor, that is, the here and now instants through which the language flow runs like through a very narrow opening. How, they ask, can we get “more than just a trickle of language through this tight bottleneck?” (38). If we were to represent metaphorically language as a liquid, like the word “trickle” indicates that they do, the compelling physics of the image might make sense. Humans, though, as many other animals, benefit from the cognitive capacity of holding present to their awareness the various elements of tasks that require their attention if they are to be successfully completed. Research on “working memory” has shown that, although it can involve very short spans of time in some species, it affords more than an instant “trickle”. It is rather similar to a “catchment” that makes complex plans possible without the help of a graphic representation. Castellucci et al. (2022) have shown how various brain regions are activated in speech planning during natural conversations. 

A commonplace found in books tackling some aspects of cognitive evolution is the comparison of humans’ remarkable language competencies with the actual lack of such a perfect means of communication and tool of cognition in animals. While some similarities may be noticed and the idea of a continuum may not be excluded off hand, the underlying assumption is that there exists an unbridgeable gap, verging on ontological divide, between animal communication and human language. It makes little evolutionary sense to compare human competencies with the whole range of signaling behavior that has been observed in the whole spectrum of other species. The only plausible biological and cognitive continuum must be looked for within the primates. Far too much attention is given to the last century’s experiments that tried to make chimpanzees and gorillas “speak” like humans, that is, communicate human-relevant information through signs of appropriate kinds. Of course, these primates had to be controlled and trained, a monumental challenge given their strength and temperamental behavior even for those that were raised exclusively in human terms. It required a great deal of interpretation to construe their mediated “language-like” production as revealing truly linguistic competencies. Only data coming from primates’ behavior in their natural settings can offer a chance of true discovery. However, direct observations of the dense networks of information exchanges in a troupe of baboons on the move, for instance, or the constant monitoring of signals among gibbons high in the trees during their life span are out of reach for humans whose recording technologies still have drastic limits. Moreover, as far as the evolution of language is concerned, only acoustic communication and its physiological apparatus are relevant, and variations in the thresholds of perception among species add to the difficulties of reliable investigations without the help of further trans-modal technological mediations. The danger of being excessively assertive about the uniqueness of human articulate language, as Christiansen and Chater are, is that the scientific inquiry about primate acoustic communication is far from being completed. During the last couple of decades, the social life of marmosets (   ) have been intensely investigated. The advantage of studying common marmosets is that they are very small, congenial, and easy to breed in natural conditions. A galore of recent studies have revealed surprising features of their acoustic communication competencies that may not be construed as the equivalent of human language but can definitely be considered to be significant steps of the kind in which evolution proceeds. It has been shown, for instance, that marmosets initially babble like human infants until they narrow down their sound repertory through imitating the well-defined vocal repertory of adults who take care of them (Gultekin et al. 2021, Agamaite et al. 2015, Yirka 2015). Other features include turn-taking in conversation-like exchanges that may imply a meaningful structuration of utterances (Chow et al. 2015, Brugger et al. 2021) and the possible equivalent of speech planning similar to what has been observed in macaque (Weth 2022).

The second commonplace that runs through works dealing with cognitive evolution is the celebration of language as the supreme adaptation that sets humans apart for their greater good. The genus Homo is more or less explicitly placed at the top of a ladder of organisms that were lacking something until modern humans emerged from the natural selection process, endowed with articulate langage. However, as we noted above, all adaptations have a cost. For example, the so-called “theory of mind” that equipped humans with the possibility to figure out how other organisms interpret situations and to anticipate strategically what their next move is likely to be, can be indiscriminately applied to natural phenomena that do not involve intentions and purposes such as volcanoes, flooding rivers, or thunderstorms. Over their relatively long cultural history, humans have wasted considerable resources to placate these blind forces through costly sacrifices. Another problematic trade-off for the advantage of benefitting from articulate language communication systems, whether through charades or not, is the spreading of narratives that can not only transfer vital information and entertain, but also, more often  than not, manipulate humans’ behavior and drive them to mindless and endless mutual massacres. This dark side of language as narratives has been amply demonstrated, lately in the works of Jonathan Gottschall (2012, 2021).  

The third commonplace is the assumption that the adaptive function of a language is the possibility to communicate information large and wide among the members of the community that share that language, and possibly beyond as more individuals and groups become conversant with its lexical and syntactical resources. Vital information, though, is often a too precious commodity that must be used only selectively. The users of a language are more prone to disguise or hide information than to indiscriminately spread it beyond their own group. In any natural language there are constant changes whose effects are to favor some forms of crypto linguistic currency, a tendency that takes radical forms in the creation of military secret code. 

Conclusion: ranking, mating, arguing life/death issues not ludic games    


Agamaite, James A., Chia-Jung Chang, Michael S. Osmanski, and Xiaoqin Wang. (2015). A quantitative acoustic analysis of the vocal repertoire of the common marmoset (Callithrix jacchus). The Journal of te Acoustical Society of America, vol. 138, issue 5 (2906-2928). 

doi: 10.1121/1.4934268 

Boles, David B. (2019). Cognitive Evolution. Abingdon: Routledge.

Brass, Marcel, Perrine Ruby, and Stephanie Spengler. (2009). Inhibition of imitative behaviour and social cognition. The Royal Society Philosophical Transactions B. 364 (1528): 2359-2367.

Brugger, R.K., E.P. Willems, and J.M. Burkart. (2021). Do marmosets understand other’s conversations? A thermographic approach. Science Advances, 3 February 2021. 7: eabc8790 (1-11). 

Castellucci, Greg A,, Christopher K. Kovach, Matthew A. Howard III, Jeremy D. W.Greenlee, and Michael A. Long. 2022. A speech planning network for interactive language use. Nature, vol. 602 (3 February 2022), 117-122.  

Castellucci, Gregg A., Frank H. Guenther, and Michael A. Long. 2022. A theoretical framework for human and nonhuman vocal interaction. Annual Review of Neuroscience, 45: 295-316.

Chater, Nick. (2018). The Mind is Flat: The Remarkable Shallowness of the Improvising Brain. New Haven: Yale University Press.

Chow, Cecilia P., Jude F. Mitchell, and Cory T.Miller. (2015). Vocal turn-taking in a non-human primate is learned during ontogeny. Proceedings of the Royal Society B. 

Christiansen, Morten H. and Nick Chater (2022). How language evolved: A new idea suggests it’s all just a game. New Scientist (Humans 23 March 2022). 

Darda, Kohinoor M. and Richard Ramsey. (2019). The inhibition of automatic imitation: a meta-analysis and synthesis of fMRI studies. Neuroimage 

Dingemanse, Mark. (2017). On the margins of language: ideophones, interjections, and dependencies in linguistics theories. In Dependencies in Language: On the Causal Ontology of Linguistic Systems, ed N. J. Enfield. Berlin: Language Science Press (195-202).

Gottschall, Jonathan. (2012). The Storytelling Animal: How Stories Make us Human. Boston: Houghton Mifflin Harcourt.

Gottschall, Jonathan. (2021). The Story Paradox: How our Love of Storytelling Builds Societies and Tears them down. New York: Basic Books.

Gouzoules, Harold. (2022). When less is more in the evolution of language: Did loss of vocal fold membranes typical of nonhuman primates enable human speech? Science, vol. 377, issue 6607 (706-707).

Gultekin, Yasemin B., David G C. Hildebrand, Kurt Hammerschmidt, and Steffen R. Hage. (2021). High plasticity in marmoset monkey vocal development from infancy to adulthood. Science Advances, vol. 7, issue 27.

Nishimura, Takeshi, Isao T. Tokuda, Shigehiro Miyachi, Jacob C. Dunn, Christian T. Herbst, Kasuyoshi Ishimura, Akihisa Kaneko, Yuki Kinoshita, Hiroki Koda, Jaap P. P. Saers, Hirohiko Imai, Tetsuya Matsuda, Ole Naesbye Larsen, Uwe Jurgens, Hideki Hirabayashi, Shozo Kojima, and W. Tecumseh Fitch. (2022). Evolutionary loss of complexity in human vocal anatomy as an adaptation for speech. Science, vol. 377 issue 6607 (760-763).

Nishimura, Yuki, Yuki Ikeda and Shigekazu Higushi (2018). The relationship of automatic imitation and personal cognitive styles. Journal of Physiological Anthropology, 37:24.

Weth, Constanze and Juffermans, Kasper (Eds.), (2018). The Tyranny of Writing: Ideologies of the Written World. London: Bloomsbury.

Yang Xie, Peiyao Hu, Junru Ii, Jinhwen Chen, Weibin Song, Xiao-jing Wang, Tianming Yang, Stanislas Dehaene, Shiming Tang, Bin Min, and Liping Wang. (2022). Geometry of sequence working memory in macaque prefrontal cortex. Science, vol. 375, issue 6581 (11 February 2022): 632-639.  

Yirka, Bob. (2015). New evidencesuggests marmosets learn vocalizations from parents. 

Be the first to comment

Leave a Reply

Your email address will not be published.