evolution of language essay

Evolution of Language

Language allows us to share our thoughts, ideas, emotions, and intention with others. Over thousands of years, humans have developed a wide variety of systems to assign specific meaning to sounds, forming words and systems of grammar to create languages. Many languages developed written forms using symbols to visually record their meaning. Some languages, like American Sign Language (ASL), are an entirely visual language without the need for vocalizations.

Although languages are defined by rules, they are by no means static, and evolve over time. Some languages are incredibly old and have changed very little over time, such as modern Icelandic, which strongly resembles its parent, Old Norse. Other languages evolve rapidly by incorporating elements of other languages. Still other languages die out due to political oppression or social assimilation, though many dying languages live on in the vocabularies and dialects of prominent languages around the world.

Teach your students how the languages of the world have evolved over time, and how their own languages continue to evolve today with this curated collection of resources.

Social Studies, Anthropology, English Language Arts, World History, Geography

Introduction: Origin and Evolution of Language—An Interdisciplinary Perspective

• Published: 02 May 2018
• Volume 37 , pages 219–234, ( 2018 )

Cite this article

• Francesco Ferretti 1 ,
• Ines Adornetti 1 ,
• Alessandra Chiera 1 ,
• Erica Cosentino 2 &
• Serena Nicchiarelli 1  

19k Accesses

5 Citations

11 Altmetric

Explore all metrics

Avoid common mistakes on your manuscript.

1 In Search for a Paradigm: The Origin of Language from Biolinguistics to Embodied Cognition

Anyone who intends to deal with the origins of language should face what critics describe as an inescapable truth: since language does not fossilize, the investigation of its origins should not rely on empirical evidence but merely have a mainly theoretical character. These considerations are at the heart of two well-known edicts— Société de Linguistique de Paris in 1866 and the Philological Society of London in 1872—that forbade all members from presenting speeches on the topic. The arguments underlying these edicts have a strong intuitive character and seem guided by a matter of common sense: it is simply not possible to rewind tape to the starting point; therefore, the origin of language is not empirically analyzable. Everything done relies on speculation, the field (very dangerous, according to many) of philosophy rather than science.

Ostracism imposed by the two Societies have had negative consequences for a long time. However, the contemporary situation is radically different: for many years, research on language origins is in a revival and currently one of the most discussed topics in literature on human communicative capabilities. In addition, the essential feature of ongoing research is the clear prevalence of empirical over theoretical studies (Fitch 2017 ; Wacewicz and Zywiczynski 2017 ). Paradoxical though it may seem, the big issue today is opposite to that raised by the edicts: it is finding the key to the problem of too much available data.

1.1 Interdisciplinarity

Several disciplines contribute to the discussion on the origin of language: computer simulation, cognitive psychology, genetics, paleoanthropology, and comparative studies, as some examples (see Tallerman and Gibson 2012 ). Some disciplines employ sophisticated analysis techniques resulting in empirical evidence inconceivable until a few years ago. The discovery of mirror neurons—an example from neuroscience to which we will return—was a turning point for studying mechanisms underlying language origin (Rizzolatti and Arbib 1998 ; Arbib 2005 ). Further, mirror neurons are an interpretative key to promote a new research paradigm within this field (Hostetter and Alibali 2008 ; Glenberg and Gallese 2012 ). According to Fitch ( 2017 ), research on paleogenetics represents the time machine allowing us to reconstruct the starting point of human communication. In Fitch’s opinion, “If a single empirical development warrants optimism and excitement about the coming decades of language evolution research, it is these advances in genetics and genomics” (p. 18). Other technologically advanced research areas are computer simulation (Kirby et al. 2014 ) and evolutionary robotics (e.g. Cangelosi 2012 ). Among traditional disciplines, comparative studies on nonhuman species continue to play a major role, specifically comparisons with other primates (e.g., Hewes 1973 ; Pollick and de Waal 2007 ) as well as other species that are phylogenetically distant from Homo sapiens , such as birds (Pfenning et al. 2014 ). Moreover, there is a widespread confront with extinct species of hominins, especially Homo neanderthalensis (Dediu and Levinson 2013 , 2018 ; Mithen 2005 ; Lieberman and McCarthy 2007 ) but also with more ancient forefathers such as Homo ergaster / erectus (Corballis 2002 ; Donald 1991 ). Examples showing the role of empirical studies in the origin of language could multiply but, for present purposes, there is no need to delve further.

What considerations should be made starting from these studies? First, the flourishing of empirical investigations on language origins allows the overturn of common opinion about the entirely speculative character of the topic itself. Assuming a considerably different perspective compared to the past, it is possible to reflect on language origins in terms of an “ongoing transition of scientific research on language evolution from one dominated by speculation and pet hypothesis to ‘normal’ science, marked by attempts to empirically evaluate multiple plausible hypotheses” (Fitch 2017 , p. 3). That said, the huge volume of data available and the heterogeneity of disciplines have had problematic effects on research: each discipline contributing to the study of language is different from another, so the first difficulty is identification of a point of convergence among extremely heterogeneous empirical results. As pointed out by Gong et al. ( 2013 ), the first step in this direction relies on the interdisciplinary character of research. In their view, only a perspective of this kind "based on pooled knowledge from diverse disciplines to reconcile seemingly contrary positions and rule out solutions plausible only within a single discipline, can lead to a biologically plausible, computationally feasible, and behaviorally adequate understanding of language and its evolution" (p. 10).

The available “bewilderingly diverse and voluminous [data] span a set of discipline that no single scholar, however knowledgeable, could hope to individually master” (Fitch 2017 , p. 24). This interdisciplinary approach appears to be an essential step. From a methodological point of view, the interdisciplinary character of a study on language origins is the distinctive feature of current research compared to previous investigation—a feature reflecting the complex and multistratified nature of language.

However, the reference to interdisciplinarity is not enough to overcome a second difficulty—theoretically more relevant—imposed by the “bewilderingly diverse and voluminous” body of evidence. Fitch ( 2017 ) argues that the problem to face “is not with data or hypotheses, but sociological” (p. 6). In his opinion, the partition of knowledges due to multiplying of disciplines has led to a significant mutual suspicion among scholars. Fitch’s considerations should not be underestimated; equally, the issue of theoretical models should not be underrated. Different fields of research communicating with one another require common ground—a shared conceptual space in which to construct connections among the various disciplines. Only within this space could empirical studies build hypotheses, supported by strong inferences, characterizing multidisciplinary research on language origins: “the method of empirically testing the predictions of multiple scientifically plausible hypotheses simultaneously” (Fitch 2017 , p. 6). What identifies strong inferences is included in the word simultaneously , emphasizing the need for a “bird’s-eye view” and the ability to make connections and convergences that cannot be understood from inside each discipline. Construction of a shared conceptual space is possible with a double plan: the plan of the achievements of each discipline and that of building interpretative models capable of providing a unitary perspective on different data. From these considerations follows the difficulty imposed by Fitch’s “bewilderingly diverse and voluminous” body of data, strictly connected to the investigation, selection, and choice of suitable interpretative models.

The flourishing of interpretative models goes hand in hand with the flourishing of empirical disciplines. Countless theoretical proposals stress some specific aspects of language and address the issue of origins from a multidisciplinary perspective. Tomasello ( 2008 ), for example, combines different fields of study and theoretical topics—from the topic of altruism to the issue of the gestural foundation, from the function of intentionality to the role played by specific cognitive systems as mindreading—and proposes the idea that the origin of human communication is connected to specific ways of cooperation of our species. In a similar way, studying the role of systems as temporal and spatial projection in language processing, Corballis ( 2017 ) proposes a model in which studies of cognitive psychology and neuroscience integrate with data from ethology and paleoanthropology. Tomasello and Corballis represent merely two approaches for constructing models capable of proposing a unitary perspective from the contributions provided by different disciplines. That said, the number of models is just as problematic as the amount of empirical data. The effort of constructing a shared conceptual space requires further effort to build more general conceptual research program.

1.2 From Models to Paradigms

Many scholars interested in language origin research (e.g., Bickerton 2012 ; Tattersall this volume) have explicitly assumed Chomsky’s model of language, that of Universal Grammar (UG), which is still relevant in the current debate (e.g., Hauser et al. 2014 ) despite some criticism (e.g., Tomasello 2009 ; Pennisi and Falzone 2016 ; Corballis 2017 ). Biolinguistics is the multidisciplinary enterprise that assumes UG as the reference model (Chomsky 2007 ; Lennerberg 1967 ; Piattelli-Palmarini 1974 ). Referring to UG, biolinguistics has a clear theoretic advantage over competitive interpretative perspectives: a refined and well-established conception of language, the outcome of > 50 years of thinking. A conception of this kind is also valuable from an empirical point of view, such as the illuminating example of investigations aimed at corroborating the principle of structure dependency and the autonomy of syntax (Musso et al. 2003 ). Furthermore, the adhesion to UG has implications for wider conceptual issues that go beyond the reflection on language (e.g., the Cartesian nature of UG affecting the way we conceptualize human nature). For all these reasons, UG represents an ideal reference framework to construct a shared conceptual space in which to interpret data and elaborate hypotheses from different disciplines. UG is more than a conceptual model; it is a genuine interpretative paradigm in the sense used by Kuhn ( 1962 ).

The minimalist turn within this generative paradigm (Chomsky 1995 ) fortifies the idea of UG as a shared conceptual space. In fact, according to some scholars (Benítez-Burraco and Boeckx 2014 ; Boeckx and Benítez-Burraco 2014 ), the minimalist program paved the way for development of biolinguistics 2.0, a theoretical approach that overcomes some difficulties that still characterize classic biolinguistics. Boeckx and Benítez-Burraco ( 2014 ) have suggested that biolinguistics 1.0 is “unable to properly deal with the attested complexity observed at the genetic, neurological, developmental, or even evolutionary level” (p. 10). Instead, biolinguistics 2.0 is a wider and more general research program (a paradigm) that is able to encompass different interpretative models, even those promoted by scholars who do not work within the generative paradigm. Di Sciullo and Boeckx ( 2011 ) state that “biolinguistics (…) allows for the exploration of many avenue of research: formalist; functionalist; nativist and insisting on the uniqueness of the language faculty; nativist about general (human) cognition but not about language per se; etc. From Chomsky to Givon, from Lenneberg to Tomasello, all this is biolinguistics” (p. 5).

In allowing minimalism to launch a new phase within biolinguistics, there is the distinction between faculty of language in the broad sense (FLB) and in the narrow sense (FLN; Hauser et al. 2002 ). Through minimalism, Chomsky moved away from the early “isolationist” view (the idea that language is processed by a specific module) and led the way to a broader-minded perspective in which language functioning is not autonomous from other cognitive systems. The idea of considering language as the broad FLB and not as FLN is a way to explain language functioning through more complex cognitive architectures than those postulated by classic biolinguistics. Furthermore, many cognitive devices characterizing FLB are also present in nonhuman animals; the focus on FLB allows development of a continuistic view of language. In light of these considerations, it seems possible to argue that early Chomsky concerns are outdated regarding the relationship between UG and evolutionary theory (Chomsky 2010 ; Berwick and Chomsky 2015 ). Therefore, the minimalist turn has allowed to again address the taboo question of language origins. As highlighted by Wacewicz and Żywiczyński ( 2014 ), indeed, anything important regarding language origins is ascribable to FLB.

So far, so good. Biolinguistics seems to be an interpretative paradigm able to include in a unitary perspective different theoretical models. Moreover, it offers a shared conceptual space useful in dealing with heterogeneous data that, as we have argued, might represent a risk for the multidisciplinary research on language evolution.

1.3 Uniqueness

Biolinguistics presents several aspects that need to be analyzed. In our opinion, when it comes to the origins of language, the reference to a refined and well-established model of language represents both the strength and weakness of such paradigm. More specifically, the problematic issue is represented by uniqueness —an issue in the Chomskyan perspective that is linked to Cartesian tradition (see for a discussion Ferretti and Adornetti 2014 ). There is no harm in emphasizing traits that differentiate human language from other forms of communication, nor in considering language as the peculiar trait of humans. But what is true for humans must also be true for other animals. As pointed out by Pinker ( 1994 ), the elephant has a type of nose that distinguishes it from all other animals, and the bat uses a distinctive perceptual system to relate to the environment. In such cases, the concept of uniqueness is not a problematic issue because it is consistent with Darwinian continuism and gradualism. The problem arises when the reference to uniqueness is used in support of the idea that is not possible to seek the precursors of language in hominins and species different from Homo sapiens because of a qualitative difference.

There are two possible moves to curtail such a Cartesian drift in the research on language uniqueness. The first is to appeal to the idea that FLB includes some traits shared with other species. According to Fitch ( 2017 ), “Thus, despite the fact that language in toto is unique to our species, most components underlying it are shared, sometimes very broadly and sometimes only with a few other species” (p. 7). Genetic data play a key role in this regard, as the focus on “deep homologies” allows possible comparison with many nonhuman species (see also Fitch 2005 ). According to Fitch ( 2017 ), indeed, narrow comparative approaches must be extended because “there is a rising awareness that distant relatives like birds may have as much, or more, to tell us about the biology and evolution of human traits as comparisons with other primates” (p. 6).

The second move against the Cartesian notion of uniqueness is showing that language emerged outside Homo sapiens . This considers the communication systems of our species’ ancestors as possible precursors of language. Consistent with that, some authors suggest that language arose in Homo heidelbergensis (Dediu and Levinson 2013 ) or in Homo ergaster–erectus (Corballis 2002 ; McBride 2014 ), while some others consider language even more ancient (Shaw-Williams 2017 ). In such a framework, the hardest debate is focused on the linguistic abilities of Homo neanderthalensis . The view by (Lieberman 1975 ; Lieberman and McCarthy 2007 ) according to which Neanderthals had no articulatory capacity to speak has been the prevailing perspective for a long time (e.g., Bickerton 1990 ). New data from different disciplines (e.g., paleoanthropology, archaeology, genetics) are now challenging this view (for an overview, see Fitch 2010 ; Dediu and Levinson 2013 , 2018 ) paving the way to building a gradualist scenario. In fact, according to Dediu and Levinson ( 2013 ), modern humans, Denisovan and Neanderthal share a common genetic line that can be considered the basis of language.

Two considerations are worthy: the first concerns considering the specificity of language in the more general framework of continuism; claiming that modern language is in continuity with more ancient forms of communication does not mean overlooking those specificities that characterize language as we know it today. The second consideration is that, despite elements of continuity attributed to the traits of FLB, the crux of language is still represented by FNL, namely by “those traits that are unique to humans and unique, within humans, to language itself” (Fitch 2017 , p. 5; see also Fitch 2005 ; Hauser et al. 2002 ). Ascribing such traits to language means asserting that its essential elements (what makes it “language”) are related to unique features with no precursors in the animal kingdom. Thus, continuity with other animals and other species of hominins applies only to those aspects of language that are not linked to FLN; that is, those aspects not specifically linguistic.

This is where the strength of biolinguistic paradigm also turns out to be its limitation. The reference to a model of language strongly linked to UG is the crux for authors more willing to revise the notion of uniqueness. Thus, for instance, while criticizing Chomsky for his position on gradualism and continuism, Dediu and Levinson ( 2013 ) defined language “as the full suite of abilities to map sound to meaning, including the infrastructure that supports it (vocal anatomy, neurocognition, ethology of communication)—FLB or ‘faculty of language broad’ in the sense of Hauser et al. ( 2002 )” (p. 2). Considering that syntax is the mechanism underlying the mapping between sound and meaning (Jackendoff 1993 ), FLN then paves the way for the issue of uniqueness. According to Fitch ( 2017 ), “without hierarchical syntax, we would not have modern language [because] without this fundamental characteristic, our open-ended ability to map novel thoughts onto understandable signals would be impossible” (pp. 10–11). At the base of this kind of considerations there is a definition of language in terms of “a complex faculty that allows us to encode, elaborate and communicate our thoughts and experiences via words combined into hierarchical structures called sentences” (Fitch 2017 , p. 7). Although investigations on the communicative abilities of animals such as birds, distant from Homo sapiens , have shown their capacity to combine elements in a rule-governed manner (a basic phonological “syntax” of birdsong), such a capacity is considered insufficient to reach the human phrasal syntax and semantics.

From these considerations follows that the syntactic component (rather than vocal or semantic-pragmatic ones) is the distinctive trait of language when comparing it to animal communication. Distinguishing our way of communication from that of other animals is crucial to the issue of language origins. Not under discussion is the fact that the topic of emergence of syntax is a main problem in the agenda on the studies of language origins. Instead, the point at issue is whether the topic of language origins should coincide with the topic of the origin of syntax: in other words, should syntax be considered the distinctive trait that, in the early stages of language, ensured the transition from animal to human communication? In the next section, we consider this point as the main limit of the well-established model of language promoted by biolinguistics 2.0.

1.4 Compatibilism

In our view, the first attempt made by Pinker and Bloom ( 1990 ) to integrate Chomsky’s model of language in an evolutionary context is an example of the limitations of the biolinguistic paradigm. Assuming UG as the indisputable starting point of the argument, Pinker and Bloom disputed Chomsky in proving UG to be compatible with the theory of evolution. Their proposal was based on the following logical argument: the only successful account of the origin of complex biological structure is the theory of natural selection; language is a complex biological structure (subsequently refuted by minimalism); the origin of language must be explained by the theory of natural selection. This argument led to a new stage in the study of the relationship between UG and evolutionary theory—namely, a compatibilist phase in which the theoretical structure of UG represents the a priori assumption that the empirical research is called to verify. The main goal underlying this compatibilist perspective is finding a model of the theory of evolution that better fits the model of language. This compatibilist attitude, in fact, has characterized Chomsky’s thought development, from the early aversion to evolutionary theory to the recent adhesion to evo-devo approaches centered on the criticism toward Pinker and Bloom’s ( 1990 ) neo-Darwinism (Chomsky 2010 ; Berwick and Chomsky 2015 ).

Relevant to the present article is that this sort of compatibilism also affects the study of language origins. In fact, if the starting point of investigation is the model of language, the analysis of the origin of communication strongly depends on the conclusive outcome of the process. The way in which Berwick and Chomsky ( 2015 ) synthetize the evolutionary steps that in their opinion have led to language provides an example of the implications of adopting this kind of analysis:

We enter here into large and extremely interesting topics that we will have to put aside. Let us just summarize briefly what seems to be the current best guess about the unity and diversity of language and thought. In some completely unknown way, our ancestors developed human concepts. At some time in the very recent past, apparently some time before 80,000 years ago if we can judge from associated symbolic proxies, individuals in a small group of hominids in East Africa underwent a minor biological change that provided the operation Merge—an operation that takes human concepts as computational atoms and yields structured expressions that, systematically interpreted by the conceptual system, provide a rich language of thought. (…). At some later stage, the internal language of thought was connected to the sensorimotor system, a complex task that can be solved in many different ways and at different times (p. 87).

Since language relies on syntax (and since syntax depends on Merge), the evolutionary steps that have led to language should include not only an explanation of how Merge appeared but also a description of how the necessary tools (i.e., concepts) that provided operation Merge have arisen. Despite this, Berwick and Chomsky ( 2015 ) interpret the evolution of Merge in terms of a (casual) biological mutation and assume the emergence of concepts as a given without explaining it. In other words, this way of viewing language directs the evolutionary process.

As we have suggested, biolinguistics can count on a refined and well-established model of language. Because of this, the biolinguistic paradigm can foster hypotheses and predictions in different areas of research that share a common idea on the nature of language. That said, it is certainly true that interpretative paradigms are necessary to organize the large amount of heterogeneous data from different disciplines, but it is also true that choice of the paradigm strongly affects construction of a unitary model of language and its origins. What if the perspective is overturned? What if the theory of evolution (and not the model of language) is the starting point of the analysis? At the base of such change is the idea that language must meet the conformity constraint of the theory of evolution rather than the compatibilist one. This idea paves the way to construction of a new interpretative model of language inspired by Darwin (more than Descartes); for this reason, it is not only plausible but also advisable when considering the question of the origin of linguistic capacity. Following these considerations, the next sections will consider a new paradigm that, unlike biolinguistics, assumes the principles of evolutionary theory as points of reference.

2 Toward a New Paradigm for the Study of Language and Mind: The Embodied Approach

The investigation of human mind within the field of cognitive science has been influenced for a long time by what Hurley ( 1998 ) defined the “classical sandwich model”. The idea is that the external layers of the sandwich, that is, the sensory and the motor are not so important after all, because what matters is the inner layer, the cognitive processes. Furthermore, according to this metaphor, the sensory and the motor are separate, they are just input and output systems, which do not affect what happens at the level of the cognitive processes. This way of understanding the relation between sensory-motor and cognitive processes has also influenced the view on language. Considering the mind as a device that manipulates symbols, symbolic theories of meaning have assumed that linguistic meaning arises from the syntactic combination of mental symbols. This view is often conjoined with a modularist assumption that meaning is processed in an informationally encapsulated way such that these mental symbols are amodal, i.e., largely decoupled from sensory, motor and emotional processes (e.g., Fodor 1975 ; Pylyshyn 1984 ).

This traditional view has been more recently criticised and rejected by the embodied theories of mind, which assume that supposedly “high” processes such as language and thought are grounded in “low” sensory, motor and emotional processes (Barsalou 1999 ; Kemmerer 2010 ; Gallese and Lakoff 2005 ; Prinz 2005 ; Werning 2012 ; Werning et al. 2013 ). According to these theories, in fact, there is no such distinction between “low” and “high” processes, because “[c]ognitive activity takes place in the context of a real-world environment, and it inherently involves perception and action” (Wilson 2002 , p. 626). From this perspective, cognition is grounded, context-dependent, situated, anchored to experience: the organism’s environment has a central role in its behavior (the environment is part of the cognitive system); such environment is not only a rich source of constraints and opportunities for the organism, but also a context that gives meaning to its actions (Beer 2014 ). Furthermore, cognition is for action: “the function of the mind is to guide action, and cognitive mechanisms such as perception and memory must be understood in terms of their ultimate contribution to situation-appropriate behaviour” (Wilson 2002 , p. 626). Focusing on the primacy of action, embodied theories lead to a strong rejection of the traditional cartesian distinction between “knowing how” (execution) and “knowing that” (competence), because all knowledge, not only the “knowing how” is supposed to emerge from doing. Furthermore, the theoretical “knowing that” has not anymore a privileged status as the foundational way of acquiring knowledge. In the following sections we will suggest that these two aspects—grounding and the primacy of action—are important for the building of a model of language alternative to biolinguistics. Precisely, we will argue that: the action system has a crucial role in language evolution and that the acknowledgment of this allows to elaborate perspectives of language emergence in line with the conformity constraint (Sect.  3 ); the embodied approach, supposing a kind of anchoring to the external reality, is suitable to explain a crucial property of human communication, i.e. how language is appropriate to the situation (Sect.  4 ). Before going into details of these arguments, it is necessary to analyze how the embodied account has deeply influenced the way of thinking about language.

According to the embodied theories of mind, language comprehension is based on the multimodal simulation of perceptions, actions, and emotions (e.g., Barsalou 1999 , 2010 ; Gibbs 2006 ; Glenberg et al. 2013 ; Lakoff 1987 ; Pecher and Zwaan 2005 ). Therefore, understanding the meaning of a word like “chair” partially re-activates the brain regions involved in perceiving chairs and the motor areas relevant to interactions with chairs (as well as emotion circuits if the word elicits affective states). This notion of comprehension provides a way of addressing the “symbol grounding problem”, that is, the problem of explaining how symbols can grasp reality and be meaningful (cf. Harnad 1990 ). As revealed by Searle’s Chinese Room Argument (Searle 1980 ), amodal-symbolic theories seem to have difficulties explaining that, given that no matter how many abstract symbols a subject relates to one another, she is never going to determine the meaning of the sentence.

The sciences of mind and brain are currently facing the challenge of accommodating in a unifying framework (a paradigm different than biolinguistics) growing evidence in favour of this embodied model of language. Neuroimaging investigations have supported this view exploring several different domains. For example, in the domain of perception, it has been shown that perceptual brain regions that process object-related information are also activated by as words related to visual features (e.g., “brown”; Pulvermüller and Hauk 2006 ), odours (e.g. “cinnamon”; Gonzalez et al. 2006 ), sounds (e.g., “telephone”; Kiefer et al. 2008 ), and taste (“salt”; Barrós-Loscertales et al. 2012 ). The domain of emotions and pain has also been explored with a variety of methods, including EEG (Rak et al. 2013 ), fMRI (Richter et al. 2010 ) and behavioural measures (Reuter et al. 2017 ), producing converging results that suggest that language processing hinges upon processes that are active when people actually feel emotions and pain.

As for actions, it is known that somatotopic areas in the motor and premotor cortex, which are active when subjects move specific body parts (e.g., “face”, “leg”, “arm”), are also active when they understand action-related words that refer to those body parts (e.g., “lick”, “pick”, or “kick”; Pulvermüller 2005 ) and motion sentences (Tettamanti et al. 2005 ), and the processing of action-related verbs is impaired specifically in patients with degenerative brain diseases that affect the motor system, including amyotrophic lateral sclerosis (Grossmann et al. 2008 ), Parkinson’s disease (e.g., Cotelli et al. 2007 ) and other motor neuron diseases (Bak et al. 2001 ).

Starting from the assumption that people have bodies by means of which they explore the surrounding environment, the embodied theories of mind have argued that there is a relation of mutual inter-dependence between action and perception (Berthoz 1997 ). Perception is indeed action-oriented, as expressed by the notion of “affordances” which, following Gibson ( 1979 ), are defined as properties things have in virtue of being the object of certain potential actions. The neuroscientific plausibility of this notion is supported by the finding that a set of neurons in the premotor cortex called “canonical neurons” respond not only when manipulable objects are actually manipulated but also when they are simply perceived (see for a review Martin 2007 ). Moreover, canonical neurons are also active when tool-related nouns are presented (Cattaneo et al. 2010 ; Marino et al. 2011 ) and behavioural studies confirm that the processing of nouns can interact with motor activity (Tucker and Ellis 2004 ; Lindemann et al. 2006 ). Furthermore, affordances seem to be involved in the construction of sentence meaning. This has been shown by Glenberg and Robertson ( 2000 ) in behavioural studies in which sentences containing affordance violations (e.g., “After wading barefoot in the lake, Erik used his glasses to dry his feet”) were judged as less sensible than semantically and grammatically matched sentences with no violations (e.g., “After wading barefoot in the lake, Erik used his shirt to dry his feet”). Also, using the EEG method, it has been shown that the N400 component of the ERPs is sensitive to the modulation of object affordances as induced by the previous linguistic context (Cosentino et al. 2017 ).

Converging evidence from different disciplines, including psychology, neuroscience, neurolinguistics, and neuropsychology, has informed the theorizing about the embodiment of mind, leading initially some scholars to suggest that a radical shift of paradigm was happening with respect to more traditional amodal-symbolic theories. Currently, most researchers acknowledge that the correct approach lies probably in the between, but the debate is still completely open as to level of embodiment and involvement in different processing stages. Theories of language origin and evolution have been deeply influenced by this interdisciplinary discussion, making the notion of “action” a very important attractor in the theory space concerning how language emerged and evolved. There are at least three levels of involvement of the notion of “action” in explanations of language origin and evolution.

The first, more obvious, way of endorsing the link between action and language from an evolutionary perspective is constituted by the well-represented strand of gestural theories of language evolution. Second, the idea of a strong connection between language and action is the starting assumption of some models of language evolution, which assume that the mechanisms originally evolved for action control might have been exploited for language, at both the grammatical and semantic level (see Glenberg and Gallese 2012 ). This idea has been declined in different ways, but the most representative is maybe the so-called “mirror system hypothesis” (see Arbib et al. 2014 for a review). Finally, the notion of “action” connects theories of language origin and evolution to the so-called motor account of social cognition (e.g., Cosentino 2014 ). This approach addresses the issue of language evolution focusing on the pragmatics of language (see Ferretti 2013 ) and assuming that the ability to read other people’s communicative intentions crucially involves the capacity to understand their actions (Blakemore and Decety 2001 ). In the following sections, we now turn to these different approaches.

3 Evolution of Expressive System from an Interdisciplinary Action-Oriented Perspective

The embodied models discussed in the previous section have important consequences for the question of language origins. The idea that language is grounded in “low” sensory-motor processes has allowed scholars to elaborate bottom-up perspectives of language emergence, which focusing on the constituent capacities underlying larger cognitive phenomena are more in line with evolutionary biology. In other words, embracing a model of knowledge as action permits development of a language model that meets the conformity constraint of the theory of evolution rather than the compatibilist one. Furthermore, adhesion to such a model (setting aside the distinction between competence and execution typical of biolinguistics) allows us to reconsider the fundamental role of the expressive dimension of communication for both the origin and functioning of language. In fact, perspectives that embrace the idea of “language as action” deal with the issue of language origin referring to evolution of the communicative expressive modality. Acknowledgment that the action system has a crucial role in language comprehension and production has provided new views on the involvement of such a system in language evolution, bolstering the gesture-first theory of human communication, according to which human language first originated as a gestural-based communicative system (Arbib 2005 ; Arbib et al. 2008 ; Armstrong and Wilcox 2007 ; Corballis 2010a , b ; Fogassi and Ferrari 2007 ).

The gesture-first theory has taken advantage of the interdisciplinary enterprise that characterizes language evolution research (for a review, see Corballis 2010a ). The first modern effort in this direction was that of anthropologist Hewes ( 1973 ). He reopened the way to the studies on language origins, explaining the origin of human communication in a gestural theoretical framework and synthetizing evidence from primatology (i.e., the success of teaching sign-based communication systems to nonhuman apes; Gardner and Gardner 1969 ), paleoanthropology (the late emergence in human evolution of the full anatomic equipment for vocal production; Lieberman and Crelin 1971 ) and neuroscience (the relationship between handedness and lateralization; Lennerberg 1967 ). Nonetheless, since the 1990s, gestural theory has become a very influential model to account for the origin of human communication. The reason is tied to an important neuroscientific achievement: discovery of mirror neurons in the F5 area of the premotor cortex of the macaque’s brain (di Pellegrino et al. 1992 ; Gallese et al. 1996 ). These neurons are defined as mirror because they allow a kind of mirroring between perception and action. Specifically, they discharge when the monkey performs an intentional act with its hands (e.g., trying to grasp an object) and when it observes another primate (human or monkey) accomplish a similar intentional act, unlike the so-called canonical neurons, which respond only to presentation of the object. The functional role of mirror neurons is relevant to the gestural origin of language. According to several authors (e.g., Rizzolatti and Sinigaglia 2008 ), their primary function is related to an implicit and nonreflective understanding of manual actions: these neurons allow the subject to understand actions made by others through mapping those actions on acts that he or she is able to perform (Rizzolatti et al. 2001 ).

When highlighting the role of mirror neurons in manual action understanding, several scholars hypothesized that these neurons may have had a key role in a communication system based on hand gestures that paved the way to human language (e.g., Arbib 2005 ; Corballis 2010b ; Rizzolatti and Arbib 1998 ). One of the key elements of this hypothesis is that the F5 area of the ventral premotor cortex of the macaque is homologous to Broca’s area in humans, specifically to Broadman area 44 (Rizzolatti and Arbib 1998 ). In humans, this area is involved in general motor functions, such as the control of the complex hand movements (Binkofski and Buccino 2004 ), but it also plays a key role in some linguistic processes (e.g., Broca 1861 ; Embick et al. 2000 ; Fedorenko et al. 2012 ). As Broca’s area developed from a region originally involved in the processing of action, one could assume that the ability to recognize and perform manual actions provided the basis for developing the ability to perform and recognize communicative hand gestures that thereafter contributed to development of brain mechanisms that support spoken language (for a discussion, see Adornetti and Ferretti 2014 ). This provides an example of what has been called “neural reuse of action perception circuits for language” (for a review, see Pulvermüller 2018 ), the idea that mechanisms primarily carrying motor and sensory functions in animals (in our case, a system of manual action understanding) are altered and enhanced in humans to allow for their neural reuse in the service of linguistic or other higher cognitive functions (Anderson 2010 ).

This gesture-first account has been further corroborated by comparative data on monkey and ape communication that shows the existence of important differences between their vocal and gestural communicative signals. Although it is well known that nonhuman primates produce different acoustically vocal signals to communicate about different events or entities (e.g., Gouzoules et al. 1984 ; Seyfarth et al. 1980 ; Slocombe and Zuberbühler 2005 ), it is attested that gestural communication systems in these animals are more flexible than are vocal ones (Tomasello 2008 ). Moreover, neurophysiological investigations showed that nonhuman primates do not have the full neural equipment necessary for vocal control that would enable the production of novel sounds from the environment (Ploog 2002 ). The voluntary modulation, exaggeration, or inhibition of their calls can be viewed as an internal emotional state, such as the production of human emotional vocalizations (e.g., cry, laugh, scream; Meguerditchian and Vauclair 2014 ). Therefore, although the vocal mode of communication is often considered a precursor of speech (Burling 2005 ; MacNeilage 2008 ), animal vocal behavior alone does not seem to represent a starting point for the evolution of human communication.

Contrary to vocalizations, which are mostly instinctive expressions of emotions, nonhuman primate (especially apes) make manual gestures (i.e., visible movements of hands produced without using or touching objects) that can be voluntarily produced by the animal and used in a more flexible way than vocal signals. For example, apes can use gestures in different contexts to communicate different things (Pollick and de Waal 2007 ), such as modifying the behavior of a specific receiver (Roberts et al. 2013 ). Furthermore, when producing gestures, apes appraise the attentional state of the recipient: visual gestures (not accompanied by any sound) are frequently used when the receiver is paying attention to the indicator (Tomasello and Call 2007 ), while auditory and tactile gestures are produced to attract the attention of an individual who is not looking at the signaler (Tomasello et al. 1994 ). From this view, ape gestures and human language share a very important property: intentionality (for a discussion, see Roberts et al. 2013 ).

Nevertheless, the gesture-first theory of language origin is not without criticism. The most powerful argument against it is the so-called “modality transition problem” (Hewes 1973 ; Orzechowski et al. 2016 ). If language first emerged as a gestural–manual system, why should language have assumed the vocal-auditory form dominant today? This question is very relevant when considering that sign language—the communicative system used by people who are deaf—is as expressive as spoken language (Stokoe 1960 ). The gestural/manual system should have produced sign language as its natural consequence. In this respect, Wacewicz and Zywiczynski ( 2017 maintained that "the persistence of the problem, together with new sources of empirical data … was a powerful motivation for language evolution researchers to look to the multimodal alternatives whereby, from the start, the evolutionary emergence of language involved an intimate connection and interplay between the vocal-auditory and motor-visual modalities (e.g., Kendon 2011 ; McNeill 2012 ; Collins 2013 ; Sandler 2013 ; Zlatev 2014 )" (p. 4).

Some of the arguments supporting the multimodal scenario come from gesturology research, according to which both the organization of body movement and speech contribute to the process of languaging (Kendon 2004 ). The idea is that gesture and speech comprise a single multimodal system with gesture not as an ornament or accompaniment to speech but rather part of it (Goldin-Meadow 2003 ; McNeill 2012 ). Adhering to this view, Hostetter and Alibali ( 2008 ) highlighted that both gesture and speech rely on the same simulative processes. As described in the previous section, simulations are neural enactments or reenactments of interactions with the world. According to Hostetter and Alibali ( 2008 ), when a speaker engages in these reenactments, the same motor and perceptual cerebral areas are recruited that would be involved in physically performing or perceiving the scene. Forming a simulation evokes a motor plan that can be expressed alongside speech and gesture. Thus, gestures are a natural byproduct of the cognitive processes that underlie speaking; it is not possible to consider the two separately because both are expressions of the same simulation (Pouw and Hostetter 2016 ).

The multimodal theory of language evolution, the idea that bodily-visual and vocal-auditory signals were fully integrated at least from the beginning of language (e.g., McNeill 2012 ), is supported by recent comparative data. It has been shown that apes, especially chimpanzees, have a multimodal system of communication whereby the production of gestures is often associated with vocal signals and facial expressions (Liebal et al. 2013 ; Taglialatela et al. 2015 ). Furthermore, neurofunctional investigations have revealed that chimpanzees’ manual gesturing selectively activates their homologue to Broca’s area (Taglialatela et al. 2008 ) and that the pattern of activation is enhanced in subjects who simultaneously use both gestural and vocal signals (attention-getting calls; Taglialatela et al. 2011 ). These neurofunctional data support the “potential existence in chimpanzees of a multimodal intentional system that not only includes gestures but can also integrate, in some individuals, oro-facial and atypical vocal sounds” (Meguerditchian and Vauclair 2014 , p. 145).

These latter neurofunctional findings point to the question of connections between the mouth and the hand that may have played a role in the evolution of language. The mirror system (again) is relevant to explain these connections. Evidence attests that the monkey premotor cortex is involved in the production and perception of both oro-facial and forelimb actions: some neurons in F5 area activate when the monkey makes a movement to grasp an object with either the hand or mouth (Rizzolatti et al. 1988 ). Another category of mirror neurons, called “communicative mouth mirror neurons,” is activated by the observation of both mouth-communicative gestures (i.e., lip-smacking, lip protrusion, and tongue protrusion) carried out by the experimenter standing in front of the primate using motor actions associated with eating (Ferrari et al. 2003 ). Many investigations have attested this link in humans between hand and mouth (e.g., Gentilucci et al. 2001 ). For instance, Iverson and Thelen ( 1999 ) proposed that an association between the manual system and the vocal system is present from birth and paves the way for embodied language processing later in life. Pexman and Wellsby ( 2016 ) found a connection between children’s manual dexterity—the ability to make coordinated hand and finger movements to grasp and manipulate objects—and language skills. Based on the presence of a coupling between hand and mouth, Fogassi and Ferrari ( 2007 ) hypothesized that "the ventral premotor cortex, endowed with the control of both hand and mouth actions, could have played a pivotal role in associating gestures with vocalizations, thus producing new motor representations. At this stage, the mirror-neuron system, because of its capacity to match the seen/heard gesture or vocalization with internal motor representations, allowed the observer/listener to assign a meaning to these new vocal–gesture combinations" (p. 140).

It is important to highlight that both the gesture-first account and the multimodal perspective are compliant with the action-orient paradigm. In fact, in both cases, the core idea is that language is grounded in bodily sensory-motor systems; it is not a coincidence that both perspectives assign a crucial explicative role for the rise of language to mirror neurons. The difference between the two accounts lies in the degree of importance conferred to the vocal mode of expression. According to the gesture-first account, during the early stages of language evolution, vocalizations were simply an accompaniment to gesticulation, which unfold the primary communicative function. From the view of the multimodal perspective, vocalizations and gesticulation were functionally equivalent from the beginning, both being necessary for the whole communicative process.

To conclude, the study of the evolution of the expressive modality from an action-oriented perspective offers an illuminating example of how collecting and synthesizing data from a broad range of disciplines has been possible when examining, in a scientific and systematic way, a subject that is crucial for understanding the origins of linguistic communication.

4 Deep Link Between Action, Social Cognition and Language

So far, we have discussed how the contribution of the action-oriented model of language has shifted the emphasis on the role of gesture in the origin of human communication. In this section, we will consider further crucial implications of assuming such action-oriented model. After considering the idea that action contributes to the making of the communicative expressive system, we now turn our attention to a different but strictly connected topic: the possible role of the embodied perspective in the explanation of the processes of language comprehension and production. How do the embodied models raise the issue of the interpretative level of language? In this regard, a further aspect distinguishing the embodied models from the syntactic-centric models emerges. As is well known, models that are guided by UG focus on the sentence constituent structure. This generates as consequence an “internalist” perspective in which the study of language coincides with the analysis of the rules governing the combination of symbols in the construction of the sentence. On the contrary, the embodied models are characterized by a focus on the relation between language and the external reality, specifically on how language might be grounded in the world (in this respect, the reference to grasping as basic condition for the emergence of language is illustrative; see Rizzolatti and Arbib 1998 ). Namely, whereas the biolinguistic models of language addressed the interpretative processes with reference to the combinatorial aspect of relating sound and arbitrary referents, now the embodied models shift the focus to the issue of grounding.

From this point of view, providing a comprehensive explanation of language means providing an explanation of how linguistic expressions are linked to external reality, ultimately it means understanding what meaning itself really is—the so-called “grounding problem” (Harnad 1990 ). This problem relates to what Chomsky ( 1966 ) defined as “Descartes’ problem” or the creative aspect of language use: how linguistic expressions can be appropriate to circumstances not being caused by them. Even though Chomsky ( 1964 ) states that “a theory of language that neglects this ‘creative’ aspect is of only marginal interest” (pp. 7–8), he considers the problem of appropriateness an unsolvable mystery. Indeed, the Chomskyan perspective can offer indications about how symbols relate to each other within the linguistic system but cannot account for how concepts emerge and relate to entities outside the system (see for a discussion Ferretti and Adornetti 2014 ). Much of the mystery clearly depends on the priority assigned to the syntactic plan. The grounding problem demands a link between language and reality that the syntactic-centric theories—defining language with reference to abstract, amodal, and arbitrary symbols combined by syntactic rules—are not equipped to address. The incompatibility between syntax and grounding calls for an inversion of paradigm. Can the action-oriented perspective, emphasizing the priority of grounding, embody such an inversion of paradigm?

By identifying the roots of language in the interaction between agents and environment, the framework of embodiment offers an approach to solving the symbol grounding problem, starting from a deep focus on the role of context (see Sect.  2 ). This focus strongly reframes the definition of symbol , here viewed as a structural coupling between an agent’s sensorimotor activations and its environment (Vogt 2002 ). To this extent, the question of appropriateness to the context is seen as a general problem concerning appropriateness of action to the ecological environment. The embodied account solves this latter problem of producing contextually appropriate behaviors taking into account a brain organized in terms of goal-directed motor acts (Rizzolatti et al. 2000 ). As pointed out by Glenberg and Gallese ( 2012 ), appealing to the “neural re-use” hypothesis (see Sect.  3 ), “the brain takes advantage of the solution of one different problem, namely contextually appropriate action, to solve another difficult problem, namely contextually appropriate language” (p. 911). This represents a completely different approach compared to that offered by the biolinguistic paradigm. If the issue of creativity of language use remained a mystery within that paradigm, the embodied perspective places critical constraints on the construction of meaning, starting from the primary constraints endemic to effective action. Given these constraints, the model can provide an account of how meaning emerges within a deep coupling between agent and context. Specifically, meanings are viewed as grounded through simulation in concrete experiences.

One area where the embodied approach has provided a fruitful contribution to language grounding is that of lexicon. Such a contribution has been particularly fostered by the mirror system hypothesis (for a review, see Arbib et al. 2014 ). As we have shown in Sect.  3 , according to this hypothesis, the mechanisms that originally evolved for action control have been exploited for language processing (Glenberg and Gallese 2012 ). More specifically, a “mirror for actions” system—concerned with both generating an action appropriate to the object’s affordances and recognizing the action being performed by another individual—provided the evolutionary basis for the emergence of a “mirror for words and constructions” system. In this way, the mirror system allows for complex imitation to be transferred from manual skills to a new communicative domain. The link between action-perception circuits and merging of circuits into higher-order ones makes a direct association between meaning and motor acts on one side and between meaning and object representations on the other. The master role played by the interplay between action-supporting regions of the brain to the emergence of semantics (e.g., Pulvermüller et al. 2014 ) provides a bottom-up account of language that is grounded in sensory-motor representation.

So far so good. The embodied approach presents itself as an option alternative to the biolinguistic program. The shifting in perspective towards embodiment has shown that alternative models of language referring to properties very different from those considered by biolinguistics are possible. Providing an action-based model of cognition that founds linguistic information on the grounding of symbols through the agent’s interaction with the environment, the embodied account goes in the desirable direction of building a different model of language compared to UG. Moreover, as it suggests that the supposedly high processes might be reframed in terms of a constituent contribution provided by low processes intrinsically linked to the domain of action, the embodied account offers a model of language that has a strong evolutionary plausibility.

That said, when it comes to the definition of language provided within the embodied perspective, two difficulties emerge, both involving the issue of language origins. The first difficulty concerns the priority assigned to lexicon. In terms of the origins, approaches centered on the rule of atomic elements are the most intuitive, carried out by scholars such as Bickerton ( 1990 ). However, the lexical protolanguage models are also very problematic. As emphasized by Fitch ( 2010 ), they presuppose an infrastructure for speech founded on vocal imitation and a capacity for referential communication that are not well thematized in approaches of this kind, leaving important evolutionary problems open. The second difficulty regards the fact that, after emphasizing the importance of grounding, embodied approaches (unlike biolinguistics) cannot count on a model of language fortified by > 50 years of thinking. When called upon to deal with defining what they mean by language , in fact, several scholars relating to embodied cognition have offered a definition that, quite unexpected, coincides with that of Chomsky: Glenberg and Gallese ( 2012 , p. 910), for example, defined language as “a productive system in that a finite number of words and syntactic rules can be used to generate an infinite number of sentences.” The emphasis on syntax is further underlined by the agenda for the future proposed by the embodied perspective; that is, explaining a motor theory of syntax for truly a theory of language (Caruana and Borghi 2016 ). Clearly, that of bringing together biolinguistics and embodiment is a legitimate program. However, such an option incurs the risk of leading again to a model of language that undermines the strong distinctions existing between the two paradigms. An alternative option is that, starting from the centrality of grounding, the action-oriented perspective paves the way to a different model of language. Indeed, if embodiment identifies the issue of grounding as a central tenet of its proposal and acknowledges meaning as a function of language use more than language per se (Evans 2006 ), more than focusing on the combinatorial aspects of language, it is reasonable to turn attention to models of language that stress the notion of anchoring to the context. What direction to take to combine an action-oriented perspective with a model addressing the issue of language origins, starting from notions such as that of context? It is our claim that, as a result of its features, the embodied approach has deep similarities with a pragmatic-based model of language—centered on a definition of language in terms of use more than in terms of grammatical competence.

Such a theory stating that language should be explained with reference to a strong pragmatic notion of context is the ostensive–inferential model of communication (OIMC) or relevance theory (Sperber and Wilson 1986 / 1995 ). Starting from the well-known Gricean philosophical account of communication (Grice 1975 ), OIMC has been corroborated by cognitive data that attest to the plausibility of a communicative model centered on the role of context and intentionality (Scott-Phillips 2014 ; Sperber and Wilson 1986 / 1995 ). Within the OIMC the notion of intentional communication becomes a topic of central concern. Then how is the role of such notions framed within the OIMC? According to the model, a linguistic interaction is characterized by the speaker’s meaning, a complex communicative intention aimed to achieve a certain effect on the hearer’s mind by means of the hearer’s recognition of the intention to achieve this effect. In these terms, pragmatic interpretation is ultimately an exercise of mindreading, with speakers trying to use the right sort of evidence to allow the audience to determine a contextually appropriate interpretation of linguistic expressions. This way of interpreting communication broadens the perspective on the notion of context. Pragmatics is not just elaborating the contextually appropriate actions in a physical environment; it also means solving problems related to the social context. From this point of view, the social-cognitive infrastructure underlying engagement with other minds represents a major crux in the investigation of language as well as its origins and evolution (e.g., Dunbar 1998 ; Origgi and Sperber 2000 ; Scott-Phillips 2014 ; Tomasello 2008 ). In fact, the focus on the role of mindreading as a precondition for the phylogenetic development of language is at the heart of much current work (e.g., Scott-Phillips 2014 ; Origgi and Sperber 2000 ). In this regard, we do have a robust theory of language alternative to that proposed by the biolinguistic program and conforms to evolutionary theory that takes into account the pragmatic aspect of language.

Although there are no models available allowing us to explicitly deal with the relation between embodiment and OIMC, we have some interesting indications suggesting that such a kind of relation might be fruitfully considered the agenda for the future of both the embodied perspective and the OIMC. At least two considerations are in favor of this claim. First, such a relation might provide the action-oriented perspective with a robust model of language (as robust as the Chomskyan theory) considering that to date the major limitation of this perspective is exactly the lack of a model of this kind. Further, combining embodiment with OIMC might be also fruitful for the relevance account that, although reconsidering the pragmatic aspects related to context, at least at the level of cognitive architectures is deeply tied to classical computationalism. Given the importance assigned to it within OIMC, theory of mind might be an example in this respect. The embodied perspective allows to rethink important aspects of theory of mind starting from a different angle compared to the ostensive perspective. The metapsychological device considered within the relevance account as a high-level cognitive mechanism (Sperber and Wilson 2002 ) can be reframed in terms of the more basic system of mindreading. Such a shift in perspective is founded on the claim that perception–action systems are mechanisms devoted to regulating not just the intentional action control but also the shaping of social interactions (Gallese et al. 2002 ; Fusaroli et al. 2012 ). Neurocognitive literature has provided strong evidence in support of these intersubjective bases of embodied cognition (e.g., Galantucci and Sebanz 2009 ), showing that mirror neurons can serve a self-other matching function (e.g., Gallese 2009 ); they are involved in coupling observation and execution of goal-related motor actions responding both when an action is performed and when the same action, performed by another individual, is simply observed (Gallese and Goldman 1998 ). The idea that similar intercorporeal matching mechanisms ground our connectedness to others is related to their functional role, that is, embodied simulation or motor resonance. Embodied simulation implies the mutual resonance of intentionally sensory-motor behaviors. Neuroscientific evidence shows that it is mediated by the activation of the same brain regions underlying our own sensory experiences (Gallese et al. 2004 ). In this perspective, the association of observed behaviors with a reactivation of our own body states is a way to recognize and understand those behaviors according to motor vocabulary mapping actions and representations (Rizzolatti and Luppino 2001 ). This common motor vocabulary guarantees a degree of congruence between the action repertoires of different individuals, a similarity of brain schemas when they interact. The notion of embodied simulation based on the immediate understanding of other’s actions lies at the core of the so-called motor account of social cognition. From this account follows the idea that mirroring other people’s actions is a way to directly comprehend their minds (Blakemore and Decety 2001 ). Indeed, if observing other’s actions allows for interpreting them as if they were one’s own, then it is possible to infer what motivated those actions and the intentions of the actor who performed them (Goldman 2006 ).

To this extent, the motor account of social cognition could provide a suitable alternative hypothesis about the nature of mindreading and its resulting role in language origins and evolution. As the mirror system is implicated in capturing intentions via action simulation, mindreading could be considered as playing a putative role in the pragmatic function of inferring communicative intentions (Cosentino 2014 ). Thus, the basic sensory-motor system would represent the key to pragmatics, as it implements the mental simulation of others’ actions, giving rise to expectations that provide direct comprehension of intentions, regardless of the explicit attribution of propositional attitudes (Glenberg and Gallese 2012 ). This is a simulative theory of comprehension based on the claim that the sensory-motor system provides a brain that is not only a brain that acts , but it is first of all a brain that understands (Rizzolatti and Sinigaglia 2008 ). If perceiving and acting pass through comprehension, then the building of a we-centric space underpinned by embodied simulation (Gallese 2009 ) is a bridge to meaning. From an evolutionary point of view, these considerations are particularly relevant as they pave the way to the idea that embodiment has a crucial role in meaning-making processes within a social-situated interaction (Arbib and Rizzolatti 1997 ; Gilissen 2005 ; Tettamanti et al. 2005 ). Moreover, the metapsychological complex levels of mindreading involved in communication can be framed as readjustments of the early basic sensory-motor skills. The process of reading each other’s minds is driven by the same sequence schema representation active in both communicative partners, allowing them to predict and understand each other’s actions, giving rise to meaning. Meaning is considered as a context-bound phenomenon that emerges in the context of embodied social action.

Based on the considerations made in this paper, the effort to bring together embodied theory of cognition with a pragmatic approach to communication represents the most fruitful way to build a model of language that meets the conformity constraint. Rather than stressing the necessity to hold embodiment and syntax together, future challenges might concern working on the contact points between embodiment and the OIMC. This could represent a line of investigation that, along with the challenge of providing more solid explanations of how simulation is a constituent component of language processing, might provide a paradigm of interest for understanding how the intertwined dimension of action, perception, and intersubjectivity might be involved in language origins and evolution.

5 Present Issue

This special issue provides an interdisciplinary view on contemporary language evolution research. It opens with two articles, those of Nathalie Gontier and Francesco Suman, which address epistemological issues concerning the relation between theory of evolution and language origin research. Considering the theory of evolution as the starting point of the investigation (following what, in this article, we have called the “conformity constraint”), both papers show how the conceptual assumptions of the theory of evolution affect and can fruitfully inform the study of language phylogeny. Gontier suggests how an applied evolutionary epistemological approach can help evolutionary linguistics at individuating the units, levels, and mechanisms of language evolution. Suman shows, within the framework of the Extended Evolutionary Synthesis, how factors such as niche construction, inclusive inheritance, phenotypic plasticity, and developmental constraints are relevant for conceptualizing the evolution of language: all these factors highlight the crucial role of the relationship between organisms and their environment for the origin of linguistic abilities.

The analysis of the genetic and phenotypic preconditions underlying language evolution is also the focus of the contribution from Antonio Benítez-Burraco, Constantina Theofanopoulou, and Cedric Boeckx. The authors hypothesize that genetic changes that have led to globularization of the braincase, and the processes responsible for the emergence of the (self-)domestication in our species (the fact that humans share many of the typical characteristics of domesticated species, such as reduced skeletal and cranial robusticity, changes in dentition, retention of juvenile characteristics, etc.) are closely related phenomena, and both have contributed to the appearance of natural languages.

Andrea Parravicini and Telmo Pievani present three different and historically recurrent approaches to the evolution of language: “evolution-free discontinuity” (the early version of UG model), “gradual evolutionary continuity” (the compatibilist model of UG elaborated by Pinker and Bloom 1990 ), and “punctuationist evolutionary approach” (the minimalist version of UG). The two authors present the main concepts of these approaches and discuss the limitations of each.

Ian Tattersall reconstructs the origin of linguistic capacity within a paleoanthropological and archaeological framework, assuming UG as model of language. The author hypothesizes that the neural capacity for language was acquired with the emergence of Homo sapiens 200,000 years ago, but that this new potential was not exploited until about 100,000 years later, through processes of exaptation and emergence. In proposing this hypothesis, Tattersall adheres to the idea that was a single, short-term event that led to the rise of human language.

A different scenario is outlined by Michael Corballis in his contribution. The author advocates the idea that developmental changes that led to language probably took place gradually during the Pleistocene epoch, rather than as a sudden event in the evolution of Homo sapiens . The author examines both the evolution of the cognitive capacities underlying language and the unfolding of the communicative sensory modality. Concerning the first point, Corballis focuses on two specific properties of linguistic communication: generativity and the ability to understand the thoughts of others. He suggests that such properties have precursors in the cognitive capacities of nonhuman animals. Regarding the question of the evolution of the expressive code (in line with the action-oriented paradigm we have described in this paper) he proposes a gestural account in which a crucial role is assigned to pantomime.

The role of pantomime in language evolution is also the topic of the article by Przemysław Żywiczyński, Slawomir Wacewicz, and Marta Sibierska. The authors start from the observation that there is not a commonly accepted definition of “pantomime” in language evolution research, although many scholars have hypothesized a pantomimic stage in phylogenetic processes that have led to human communication. For this, after reviewing different areas of investigations (e.g., theatre studies, semiotics, neuroscience), they offer an expanded definition of pantomime , bringing out its nonconventional, motivated, and multimodal nature.

Another contribution that focuses on communicative sensory modality from an action-oriented perspective is that of Antonella Tramacere and Richard Moore. The authors deal with a specific issue characterizing the gestural theories: the relation between imitation, mirror neurons, and social learning. Relying on data coming from neuroscience, primatology, and archaeology, Tramacere and Moore suggest that while gestural communication played a crucial role in language evolution, the grounds for thinking that manual imitation also did are currently unconvincing.

Finally, the last two articles of the special issue adhere to a pragmatic model of language and address two important issues in this regard: the evolution of cooperative communication and the emergence of informative categories, such as topic and presupposition. Richard Moore analyses the evolutionary plausibility of Gricean cooperative model of communication. Against the idea that cooperative communication presupposes intentional action and abilities of joint action, the author advocates a bottom-up hypothesis according to which the abilities and motivations for joint and intentional action might be acquired through participation in communicative interaction. Based on the debate of the nature of protolanguage, Edoardo Lombardi Vallauri and Viviana Masia suggest that presuppositions and topics, which are entrusted to a form of automatic processing, might have been developed to improve language ergonomics by sparing processing effort on some utterance contents.

Adornetti I, Ferretti F (2014) The pragmatic foundations of communication: an action-oriented model of the origin of language. Theoria et Historia Scientiarum 11:63–80

Article   Google Scholar  

Anderson ML (2010) Neural reuse: a fundamental organizational principle of the brain. Behav Brain Sci 33:245–266

Arbib MA (2005) From monkey-like action recognition to human language: an evolutionary framework for neurolinguistics. Behav Brain Sci 28(2):105–124

Google Scholar  

Arbib MA, Rizzolatti G (1997) Neural expectations: a possible evolutionary path from manual skills to language. Commun Cogn 29:393–423

Arbib MA, Liebal K, Pika S (2008) Primate vocalization, gesture, and the evolution of human language. Curr Anthropol 49(6):1053–1076

Arbib MA, Gasser B, Barrès V (2014) Language is handy but is it embodied? Neuropsychologia 55:57–70

Armstrong DF, Wilcox SE (2007) The gestural origin of language. Oxford University Press, Oxford

Book   Google Scholar  

Bak TH, O’Donovan DG, Xuereb JH, Boniface S, Hodges JR (2001) Selective impairment of verb processing associated with pathological changes in Brodmann areas 44 and 45 in the motor neuron disease–dementia–aphasia syndrome. Brain 124:103–120

Barrós-Loscertales A, González J, Pulvermüller F, Ventura-Campos N, Bustamante JC, Costumero V, Parcet MA, Avila C (2012) Reading salt activates gustatory brain regions: fMRI evidence for semantic grounding in a novel sensory modality. Cereb Cortex 22:2554–2563

Barsalou LW (1999) Perceptual symbol systems. Behav Brain Sci 22:577–660

Barsalou LW (2010) Grounded cognition: past, present, and future. Top Cogn Sci 2:716–724

Beer R (2014) Dynamical systems and embedded cognition. In: Frankish K, Ramsey WR (eds) The Cambridge handbook of artificial intelligence. Cambridge University Press, Cambridge

Benítez-Burraco A, Boeckx C (2014) Universal grammar and biological variation: an EvoDevo agenda for comparative biolinguistics. Biol Theory 9(2):122–134

Berthoz A (1997) Le Sens du movement. Odile Jacob, Paris

Berwick RC, Chomsky N (2015) Why only us: language and evolution. MIT press, Cambridge

Bickerton T (1990) Language and species. The University of Chicago Press, Chicago

Bickerton D (2012) The origin of syntactic language. In: Tallerman M, Gibson K (eds) The Oxford handbook of language evolution. Oxford University Press, New York

Binkofski F, Buccino G (2004) Motor functions of the Broca’s region. Brain Lang 89(2):362–369

Blakemore SJ, Decety J (2001) From the perception of action to the understanding of intention. Nat Rev Neurosci 2(8):561

Boeckx C, Benitez-Burraco A (2014) Biolinguistics 2.0. In: Fujita K, Fukui N, Yusa N, Ike-Uchi M (eds) The design, development and evolution of human language: biolinguistics explorations. Kaitakusha, Tokyo

Broca P (1861) Perte de la parole, ramollissement chronique et destruction partielle du lobe antérieur gauche du cerveau. Bull Soc Anthropol 2:235–238

Burling R (2005) The talking ape: how language evolved. Oxford University Press, New York

Cangelosi A (2012) Robotics and embodied agent modelling of the evolution of language. In: Tallerman M, Gibson K (eds) The Oxford handbook of language evolution. Oxford University Press, New York

Caruana F, Borghi A (2016) Il cervello in azione. Il Mulino, Bologna

Cattaneo Z, Devlin JT, Salvini F, Vecchi T, Silvanto J (2010) The causal role of category-specific neuronal representations in the left ventral premotor cortex (PMv) in semantic processing. NeuroImage 49:2728–2734

Chomsky N (1964) Current issues in linguistic theory. The Hague, Mouton

Chomsky N (1966) Cartesian linguistics: a chapter in the history of rationalist thought. Harper & Row, New York

Chomsky N (1995) The minimalist program. MIT Press, Cambridge

Chomsky N (2007) Biolinguistic explorations: design, development, evolution. Int J Philos Stud 15(1):1–21

Chomsky N (2010) Some simple evo devo theses: how true might they be for language? In: Larson R, Déprez V, Yamakido H (eds) The evolution of human language. Cambridge University Press, Cambridge

Collins C (2013) Paleopoetics: the evolution of the literary imagination. Columbia University Press, New York

Corballis MC (2002) From hand to mouth: the origins of language. Princeton University Press, Princeton

Corballis MC (2010a) The gestural origins of language. WIREs Cogn Sci 1:2–7

Corballis MC (2010b) Mirror neurons and the evolution of language. Brain Lang 112(1):25–35

Corballis MC (2017) The truth about language: what it is and where it came from. University of Chicago Press, Chicago

Cosentino E (2014) Embodied pragmatics and the evolution of language. HumanaMente J Philos Stud 27:61–78

Cosentino E, Baggio G, Kontinen J, Werning M (2017) The time-course of sentence meaning composition. N400 effects of the interaction between context-induced and lexically stored affordances. Front Psychol 8:(818)

Cotelli M, Borroni B, Manenti R, Zanetti M, Arévalo A, Cappa SF, Padovani A (2007) Action and object naming in Parkinson’s disease without dementia. Eur J Neurol 14:632–637

Dediu D, Levinson SC (2013) On the antiquity of language: the reinterpretation of Neandertal linguistic capacities and its consequences. Front Psychol 4:397

Dediu D, Levinson SC (2018) Neanderthal language revisited: not only us. Curr Opin Behav Sci 21:49–55

Di Pellegrino G, Fadiga L, Fogassi L, Gallese V, Rizzolatti G (1992) Understanding motor events: a neurophysiological study. Exp Brain Res 91(1):176–180

Di Sciullo AM, Boeckx C (2011) Introduction: contours of the biolinguistic research agenda. In: Di Sciullo M, Boeckx C (eds) The biolinguistic enterprise. New perspective on the evolution and nature on the human language faculty. Oxford University Press, Oxford

Donald M (1991) Origins of the modern mind: three stages in the evolution of culture and cognition. Harvard University Press, Cambridge

Dunbar RIM (1998) The social brain hypothesis. Brain 9(10):178–190

Embick D, Marantz A, Miyashita Y, O’Neil W, Sakai KL (2000) A syntactic specialization for Broca’s area. Proc Nat Acad Sci 97(11):6150–6154

Evans V (2006) Cognitive linguistics. Edinburgh University Press, Edinburgh

Fedorenko E, Duncan J, Kanwisher N (2012) Language-selective and domain-general regions lie side by side within Broca’s area. Curr Biol 22(21):2059–2062

Ferrari PF, Gallese V, Rizzolatti G, Fogassi L (2003) Mirror neurons responding to the observation of ingestive and communicative mouth actions in the monkey ventral premotor cortex. Eur J Neurosci 17(8):1703–1714

Ferretti F (2013) Navigation, discourse and the origin of language. In: Brambilla P, Marini A (eds) Brain evolution, language, and psychopathology in schizophrenia. Routledge, New York

Ferretti F, Adornetti I (2014) Against linguistic Cartesianism: toward a naturalistic model of human language origins and functioning. Lang Commun 37:29–39

Fitch WT (2005) The evolution of language: a comparative review. Biol Philos 20(2–3):193–203

Fitch WT (2010) The evolution of language. Cambridge University Press, Cambridge

Fitch WT (2017) Empirical approaches to the study of language evolution. Psychon Bull Rev 24(1):3–33

Fitch WT, Hauser MD, Chomsky N (2005) The evolution of the language faculty: clarifications and implications. Cognition 97(2):179–210

Fodor JA (1975) The language of thought. Harvard University Press, Cambridge

Fogassi L, Ferrari PF (2007) Mirror neurons and the evolution of embodied language. Curr Dir Psychol Sci 16(3):136–141

Fusaroli R, Demuru P, Borghi AM (2012) Introduction to the special issue. The intersubjectivity of embodiment. Cogn Semiot 4(1):1–5

Galantucci B, Sebanz N (2009) Joint action: current perspectives. Top Cogn Sci 1(2):255–259

Gallese V (2009) Mirror neurons, embodied simulation, and the neural basis of social identification. Psychoanal Dialogues 19(5):519–536

Gallese V, Goldman A (1998) Mirror neurons and the simulation theory of mind-reading. Trends Cogn Sci 2(12):493–501

Gallese V, Lakoff G (2005) The brain’s concepts: the role of the sensory-motor system in conceptual knowledge. Cogn Neuropsychol 22:455–479

Gallese V, Fadiga L, Fogassi L, Rizzolatti G (1996) Action recognition in the premotor cortex. Brain 119(2):593–610

Gallese V, Ferrari PF, Umiltà MA (2002) The mirror matching system: a shared manifold for intersubjectivity. Behav Brain Sci 25(1):35–36

Gallese V, Keysers C, Rizzolatti G (2004) A unifying view of the basis of social cognition. Trends Cogn Sci 8(9):396–403

Gardner RA, Gardner BT (1969) Teaching sign language to a chimpanzee. Science 165:664–672

Gentilucci M, Benuzzi F, Gangitano M, Grimaldi S (2001) Grasp with hand and mouth: a kinematic study on healthy subjects. J Neurophysiol 86(4):1685–1699

Gibbs RWJ (2006) Embodiment and cognitive science. Cambridge University Press, Cambridge

Gibson JJ (1979) The ecological approach to visual perception. Houghton Mifflin, Boston

Gilissen E (2005) Imitation systems, monkey vocalization, and the human language. Behav Brain Sci 28(2):133–134

Glenberg AM, Gallese V (2012) Action-based language: a theory of language acquisition, comprehension, and production. Cortex 48(7):905–922

Glenberg AM, Robertson DA (2000) Symbol grounding and meaning: a comparison of high-dimensional and embodied theories of meaning. J Mem Lang 43:379–401

Glenberg AM, Witt JK, Metcalfe J (2013) From the revolution to embodiment: 25 years of cognitive psychology. Perspect Psychol Sci 8:573–585

Goldin-Meadow S (2003) Hearing gesture: how our hands help us think. Harvard University Press, Cambridge

Goldman AI (2006) Simulating minds: the philosophy, psychology, and neuroscience of mindreading. Oxford University Press, Oxford

Gong T, Shuai L, Wu Y (2013) Multidisciplinary approaches in evolutionary linguistics. Lang Sci 37:1–13

González J, Barrós-Loscertales A, Pulvermüller F, Meseguer V, Sanjuán A, Belloch V, Ávila C (2006) Reading “cinnamon” activates olfactory brain regions. Neuroimage 32:906–912

Gouzoules S, Gouzoules H, Marler P (1984) Rhesus monkey (Macaca mulatta) screams: representational signalling in the recruitment of agonistic aid. Anim behav 32(1):182–193

Grice P (1975) Logic and conversation. In: Cole P, Morgan JL (eds) Speech acts. Academic Press, New York

Grossmann M, Anderson C, Khan A, Avants B, Elman L, McCluskey L (2008) Impaired action knowledge in amyotrophic lateral sclerosis. Neurology 71:1396–1401

Harnad S (1990) The symbol grounding problem. Physica D 42(1–3):335–346

Hauser MD, Chomsky N, Fitch WT (2002) The faculty of language: what is it, who has it, and how did it evolve? Science 298(5598):1569–1579

Hauser MD, Yang C, Berwick RC, Tattersall I, Ryan MJ, Watumull J, Chomsky N, Lewontin RC (2014) The mystery of language evolution. Front Psychol 5:401

Hewes GW (1973) Primate communication and the gestural origin of language. Curr Anthropol 14(1/2):5–24

Hostetter AB, Alibali MW (2008) Visible embodiment: gestures as simulated action. Psychon Bull Rev 15(3):495–514

Hurley S (1998) Consciousness in action. Harvard University Press, London

Iverson JM, Thelen E (1999) Hand, mouth and brain. The dynamic emergence of speech and gesture. J Conscious Stud 6(11–12):19–40

Jackendoff R (1993) Patterns in the mind: language and human mind. Harper and Collins, New York

Kemmerer D (2010) How words capture visual experience. In: Malt B, Wolff P (ed) Words and the mind. Oxford University Press, Oxford

Kendon A (2004) Gesture: visible action as utterance. Cambridge University Press, Cambridge

Kendon A (2011) Some modern considerations for thinking about language evolution: a discussion of the evolution of language by Tecumseh Fitch. Public J Semiot 3(1):79–108

Kiefer M, Sim EJ, Herrnberger B, Grothe J, Hoenig K (2008) The sound of concepts: four markers for a link between auditory and conceptual brain systems. J Neurosci 28:12224–12230

Kirby S, Griffiths T, Smith K (2014) Iterated learning and the evolution of language. Cur Opin Neurobiol 28:108–114

Kuhn T (1962) The structure of scientific revolution. University of Chicago Press, Chicago

Lakoff G (1987) Women, fire, and dangerous things. University of Chicago Press, Chicago

Lennerberg E (1967) Biological foundations of language. Wiley, New York

Liebal K, Waller B, Slocombe K, Burrows A (2013) Primate communication: a multimodal approach. Cambridge University Press, Cambridge

Lieberman P (1975) On the origins of language: an introduction to the evolution of human speech. Macmillian, New York

Lieberman P, Crelin E (1971) On the speech of Neanderthal man. Linguist Inq 2(2):203–222

Lieberman P, McCarthy R (2007) Tracking the evolution of language and speech: comparing vocal tracts to identify speech capabilities. Expedition 49(2):15–20

Lindemann O, Stenneken P, van Schie HT, Bekkering H (2006) Semantic activation in action planning. J Exp Psychol Hum Percept Perform 32:633–643

MacNeilage PF (2008) The origin of speech. Oxford University Press, Oxford

Marino BFM, Gough PM, Gallese V, Riggio L, Buccino G (2011) How the motor system handles nouns: a behavioral study. Psychol Res 77:64–73

Martin A (2007) The representation of object concepts in the brain. Ann Rev Psychol 58:25–45

McBride G (2014) Storytelling, behavior planning, and language evolution in context. Front Psychol 5:1131

McNeill D (2012) How language began: gesture and speech in human evolution. Cambridge University Press, Cambridge

Meguerditchian A, Vauclair J (2014) Communicative signaling, lateralization and brain substrate in nonhuman primates: toward a gestural or a multimodal origin of language?. Humana Mente J Philos Stud 27:135–160

Mithen S (2005) The singing Neanderthals: the origin of language, music, mind and body. Weidenfeld and Nicolson, London

Musso M, Moro A, Glauche V, Rijntjes M, Reichenbach J, Büchel C, Weiller C (2003) Broca’s area and the language instinct. Nat Neurosci 6(7):774

Origgi G, Sperber D (2000) Evolution, communication and the proper function of language. In: Carruthers P, Chamberlain P P (eds) Evolution and the human mind: language, modularity and social cognition. Cambridge University Press, Cambridge

Orzechowski S, Wacewicz S, Żywiczyński P (2016) Vocal-auditory feedback and the modality transition problem in language evolution. Reti Saperi Linguaggi 3(1):157–178

Pecher D, Zwaan RA (eds) (2005) Grounding cognition: the role of perception and action in memory, language, and thinking. Cambridge University Press, New York

Pennisi A, Falzone A (2016) Darwinian biolinguistics. Springer, Cham

Pexman PM, Wellsby M (2016) Linking hand to mouth. The relationship between manual dexterity and language skills in typically-developing children. Reti Saperi Linguaggi 3(1):81–94

Pfenning A, Hara E, Whitney O, Rivas M, Wang R, Roulhac P et al (2014) Convergent transcriptional specializations in the brains of humans and song-learning birds. Science 346(6215):1256846

Piattelli-Palmarini M (1974) A debate on bio-linguistics. Centre Royaumont pour une science de l’homme report. Conference held at Endicott House, Dedham, 20–21 May 1974

Pinker S (1994) The language istinct. Penguinn Press, New York

Pinker S, Bloom P (1990) Natural language and natural selection. Behav Brain Sci 13(4):707–727

Ploog D (2002) Is the neural basis of vocalization different in nonhuman primates and Homo sapiens ? In: Crow T (ed) The speciation of modern Homo sapiens . Oxford University Press, Oxford

Pollick A, De Waal F (2007) Ape gestures and language evolution. Proc Natl Acad Sci 104(19):8184–8189

Pouw W, Hostetter AB (2016) Gesture as predictive action. Reti Saperi Linguaggi 3(1):57–80

Prinz JJ (2005) Passionate thoughts. The emotional embodiment of moral concepts. In: Zwaan R, Pecher D (eds) The grounding of cognition: the role of perception and action in memory, language, and thinking. Cambridge University Press, Cambridge

Pulvermüller F (2005) Brain mechanisms linking language and action. Nat Rev Neurosci 6(7):576–582

Pulvermüller F (2018) Neural reuse of action perception circuits for language, concepts and communication. Prog Neurobiol 160:1–44

Pulvermüller F, Hauk O (2006) Category-specific processing of color and form words in left fronto-temporal cortex. Cereb Cortex 16:1193–1201

Pulvermüller F, Moseley RL, Egorova N, Shebani Z, Boulenger V (2014) Motor cognition–motor semantics: action perception theory of cognition and communication. Neuropsychologia 55:71–84

Pylyshyn Z (1984) Computation and cognition: toward a theory for cognitive science. MIT Press, Cambridge

Rak N, Kontinen J, Kuchinke L, Werning M (2013) Does the semantic integration of emotion words depend on emotional empathy? N400, P600 and localization effects for intentional and proprioceptive emotion words in sentence contexts. In: Knauff M, Pauen M, Sebanz N, Wachsmuth I (eds) Proceedings of the 35th annual conference of the Cognitive Science Society, Cognitive Science Society, Austin

Reuter K, Werning M, Kuchinke L, Cosentino E (2017) Reading words hurts: the impact of pain sensitivity on people’s ratings of pain-related words. Lang Cognition 9(3):553–567

Richter M, Eck J, Straube T, Miltner WH, Weiss T (2010) Do words hurt? Brain activation during the processing of pain-related words. Pain 148(2):198–205

Rizzolatti G, Arbib M (1998) Language within our grasp. Trends Neurosci 21(5):188–194

Rizzolatti G, Luppino G (2001) The cortical motor system. Neuron 31(6):889–901

Rizzolatti G, Sinigaglia C (2008) Mirrors in the brain: how our minds share actions and emotions. Oxford University Press, Oxford

Rizzolatti G, Camarda R, Fogassi L, Gentilucci M, Luppino G, Matelli M (1988) Functional organization of inferior area 6 in the macaque monkey. Exp brain res 71(3):491–507

Rizzolatti G, Fogassi L, Gallese V (2000) Cortical mechanisms subserving object grasping and action recognition: a new view on the cortical motor functions. In: Gazzaniga MS (ed) The new cognitive neurosciences. A Bradford Book, MIT Press, Cambridge

Rizzolatti G, Fogassi L, Gallese V (2001) Neurophysiological mechanisms underlying the understanding and imitation of action. Nat Rev Neurosci 2(9):661–670

Roberts A, Vick S, Buchanan-Smith H (2013) Communicative intentions in wild chimpanzees: persistence and elaboration in gestural signaling. Anim Cogn 16(2):187–196

Sandler W (2013) Vive la différence: sign language and spoken language in language evolution. Lang Cogn 5(2–3):189–203

Scott-Phillips T (2014) Speaking our minds: why human communication is different, and how language evolved to make it special. Palgrave MacMillan, London

Searle JR (1980) Minds, brains, and programs. Behav Brain Sci 3(3):417–424

Seyfarth R, Cheney D, Marler P (1980) Vervet monkey alarm calls: semantic communication in a free-ranging primate. Anim Behav 28(4):1070–1094

Shaw-Williams K (2017) The social trackways theory of the evolution of language. Biol Theory 12(4):195–210

Slocombe K, Zuberbühler K (2005) Functionally referential communication in a chimpanzee. Curr Biol 15(19):1779–1784

Sperber D, Wilson D (1986/1995) Relevance: communication and cognition, 2nd edn. Blackwell, Oxford

Sperber D, Wilson D (2002) Pragmatics, modularity and mind-reading. Mind Lang 17(1&2):3–23

Stokoe WC (1960) Sign language structure: an outline of the communicative systems of the American deaf. Linstock Press, Silver Spring

Taglialatela J, Russell J, Schaeffer J, Hopkins W (2008) Communicative signaling activates ‘Broca’s’ homolog in chimpanzees. Curr Biol 18(5):343–348

Taglialatela J, Russell J, Pope S et al (2015) Multimodal communication in chimpanzees. Am J Primatol 77(11):1143–1148

Tallerman M, Gibson KR (eds) (2012) The Oxford handbook of language evolution. Oxford University, Oxford

Tettamanti M, Buccino G, Saccuman MC, Gallese V, Danna M, Scifo P, Fazio F, Rizzolatti G, Cappa SF, Perani D (2005) Listening to action-related sentences activates fronto-parietal motor circuits. J Cogn Neurosci 17:273–281

Tomasello M (2008) Origins of human communication. MIT Press, Cambridge

Tomasello M (2009) Universal grammar is dead. Behav Brain Sci 32(5):470–471

Tomasello M, Call J (2007) Ape gestures and the origins of language. In: Call J, Tomasello M (eds) The gestural communication of apes and monkeys. Lawrence Erlbaum, London

Tomasello M, Call J, Nagell K, Olguin R, Carpenter M (1994) The learning and use of gestural signals by young chimpanzees: a trans-generational study. Primates 37:137–154

Taglialatela JP, Russell JL, Schaeffer JA, Hopkins WD, Hausberger M (2011) Chimpanzee vocal signaling points to a multimodal origin of human language. PLoS ONE 6(4):e18852

Tucker M, Ellis R (2004) Action priming by briefly presented objects. Acta Psychol 116:185–203

Vogt P (2002) The physical symbol grounding problem. Cogn Syst Res 3(3):429–457

Wacewicz S, Zywiczynski P (2017) The multimodal origins of linguistic communication. Lang Commun 54:1–8

Wacewicz S, Żywiczyński P (2014) From the narrow to the broad. Multiple perspectives on language evolution. Theoria et Historia Scientiarum 11:5–18

Werning M (2012) Non-symbolic compositional representation and its neuronal foundation: towards a simulative semantics. In: Werning M, Hinzen W, Machery M (eds) The Oxford handbook of compositionality. Oxford University Press, Oxford

Werning M, Tacca M, Mroczko-Wasowicz A (2013) High- vs low-level cognition and the neuro-simulative theory of mental representation. In: Gähde V, Hartmann U, Wolf JH (eds) Models, simulations, and the reduction of complexity. De Gruyter, Berlin

Wilson M (2002) Six views of embodied cognition. Psychon Bull Rev 9(4):625–636

Zlatev J (2014) Bodily mimesis and the transition to speech. In: Pina M, Gontier N (eds) The evolution of social communication in primates. Springer, New York

Download references

Acknowledgements

We would like to thank Przemysław Żywiczyński and Sławomir Wacewicz for giving us the opportunity to host Protolang4 conference. We also thank the scholars who agreed to act as anonymous referees and the general editor of Topoi , Fabio Paglieri.

Author information

Authors and affiliations.

Cosmic Lab, Department of Philosophy, Communication and Performing Arts, Roma Tre University, Via Ostiense 234, 00146, Rome, Italy

Francesco Ferretti, Ines Adornetti, Alessandra Chiera & Serena Nicchiarelli

Institute of Philosophy II, Ruhr-University Bochum, Universitätstraβe 150, 44780, Bochum, Germany

Erica Cosentino

You can also search for this author in PubMed   Google Scholar

Contributions

For the specific concerns of the Italian Academy, we specify that: FF wrote Sect.  1 ; IA wrote Sects.  3 and 5 ; AC wrote Sect.  4 ; EC wrote Sect.  2 ; SN wrote Sect.  4 .

Corresponding author

Correspondence to Francesco Ferretti .

Rights and permissions

Reprints and permissions

About this article

Ferretti, F., Adornetti, I., Chiera, A. et al. Introduction: Origin and Evolution of Language—An Interdisciplinary Perspective. Topoi 37 , 219–234 (2018). https://doi.org/10.1007/s11245-018-9560-6

Download citation

Published : 02 May 2018

Issue Date : June 2018

DOI : https://doi.org/10.1007/s11245-018-9560-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Find a journal
• Publish with us
• Track your research

• Question and Answer
• Open access
• Published: 24 July 2017

Q&A: What is human language, when did it evolve and why should we care?

• Mark Pagel 1  

BMC Biology volume  15 , Article number:  64 ( 2017 ) Cite this article

139k Accesses

17 Citations

165 Altmetric

Metrics details

Human language is unique among all forms of animal communication. It is unlikely that any other species, including our close genetic cousins the Neanderthals, ever had language, and so-called sign ‘language’ in Great Apes is nothing like human language. Language evolution shares many features with biological evolution, and this has made it useful for tracing recent human history and for studying how culture evolves among groups of people with related languages. A case can be made that language has played a more important role in our species’ recent (circa last 200,000 years) evolution than have our genes.

What is special about human language?

Human language is distinct from all other known animal forms of communication in being compositional . Human language allows speakers to express thoughts in sentences comprising subjects, verbs and objects—such as ‘I kicked the ball’—and recognizing past, present and future tenses. Compositionality gives human language an endless capacity for generating new sentences as speakers combine and recombine sets of words into their subject, verb and object roles. For instance, with just 25 different words for each role, it is already possible to generate over 15,000 distinct sentences. Human language is also referential , meaning speakers use it to exchange specific information with each other about people or objects and their locations or actions.

What is animal ‘language’ like?

Animal ‘language’ is nothing like human language. Among primates, vervet monkeys ( Chlorocebus pygerythrus ) produce three distinct alarm calls in response to the presence of snakes, leopards and eagles [ 1 ]. A number of parrot species can mimic human sounds, and some Great Apes have been taught to make sign language gestures with their hands. Some dolphin species seem to have a variety of repetitive sound motifs (clicks) associated with hunting or social grouping. These forms of animal communication are symbolic in the sense of using a sound to stand in for an object or action, but there is no evidence for compositionality, or that they are truly generative and creative forms of communication in which speakers and listeners exchange information [ 2 ].

Instead non-human animal communication is principally limited to repetitive instrumental acts directed towards a specific end, lacking any formal grammatical structure, and often explainable in terms of hard-wired evolved behaviours or simple associative learning [ 2 ]. Most ape sign language, for example, is concerned with requests for food. The trained chimpanzee Nim Chimpsky’s longest recorded ‘utterance’, when translated from sign language, was ‘give orange me give eat orange me eat orange give me eat orange give me you’ [ 3 ]. Alarm calls such as observed in the vervet monkeys often evolve by kin-selection to protect one’s relatives, or even selfishly to distract predators away from the caller. Hunting and social group communications can be explained as learned coordinating signals without ‘speakers’ knowing why they are acting as they are.

When did human language evolve?

No one knows for sure when language evolved, but fossil and genetic data suggest that humanity can probably trace its ancestry back to populations of anatomically modern Homo sapiens (people who would have looked like you and me) who lived around 150,000 to 200,000 years ago in eastern or perhaps southern Africa [ 4 , 5 , 6 ]. Because all human groups have language, language itself, or at least the capacity for it, is probably at least 150,000 to 200,000 years old. This conclusion is backed up by evidence of abstract and symbolic behaviour in these early modern humans, taking the form of engravings on red-ochre [ 7 , 8 ].

The archaeological record reveals that about 40,000 years ago there was a flowering of art and other cultural artefacts at modern human sites, leading some archaeologists to suggest that a late genetic change in our lineage gave rise to language at this later time [ 9 ]. But this evidence derives mainly from European sites and so struggles to explain how the newly evolved language capacity found its way into the rest of humanity who had dispersed from Africa to other parts of the globe by around 70,000 years ago.

Could language be older than our species?

Ancient DNA reveals us to be over 99% identical in the sequences of our protein coding genes to our sister species the Neanderthals ( Homo neanderthalensis ) [ 10 ]. The Neanderthals had large brains and were able to inhabit much of Eurasia from around 350,000 years ago. If the Neanderthals had language, that would place its origin at least as far back as the time of our common ancestor with them, currently thought to be around 550,000 to 750,000 years ago [ 10 , 11 ].

However, even as recently as 40,000 years ago in Europe, the Neanderthals show almost no evidence of the symbolic thinking—no art or sculpture for example—that we often associate with language, and little evidence of the cultural attainments of Homo sapiens of the same era. By 40,000 years ago, Homo sapiens had plentiful art, musical instruments and specialized tools such as sewing needles. Neanderthals probably didn’t even have sewn clothing, instead they would have merely draped themselves with skins [ 12 ]. And, despite evidence that around 1–5% of the human genome might be derived from human–Neanderthal matings [ 13 ], the Neanderthals went extinct as a species while we flourished.

Can genetic evidence help to decide when language evolved?

Yes. Modern humans and Neanderthals share a derived version of a transcription factor gene known as FOXP2 that differs from the chimpanzee version by two amino acid replacements [ 14 ]. FOXP2 influences the fine-motor control of facial muscles required for the production of speech. Indeed, inserting this derived form into mice causes them to squeak differently [ 15 ]! However, in spite of having identical primary sequences to Neanderthals, modern humans have acquired changes to the regulation of their FOXP2 genes that seem likely to cause their FOXP2 to be expressed differently to that of the Neanderthals [ 16 ], and these expression differences are pronounced in brain neurons. Combining these genetic hints with the differences in symbolic and cultural behaviour that are evident from the fossil record suggests language arose in our lineage sometime after our split from our common ancestor with Neanderthals, and probably by no later than 150,000 to 200,000 years ago.

Was there a single origin of language?

This question has parallels in biological evolution. Did life evolve once or many times? The presence of the same RNA and DNA in all organisms and homologies in the machinery of DNA transcription and translation suggest that at least all current life on Earth has a common origin. It is possible that life evolved more than once but all descendants of these other origins went extinct and left no fossil or other traces.

With language the inference is harder to make because features such as vocabulary and grammar change too rapidly to be able to link all of the world’s languages to a common original mother tongue. On the other hand, all human languages rely on combining sounds or ‘phones’ to make words, many of those sounds are common across languages, different languages seem to structure the world semantically in similar ways [ 17 ], all human languages recognize the past, present and future and all human languages structure words into sentences [ 18 ]. All humans are also capable of learning and speaking each other’s languages (some phones are unique to some language families—such as the famous ‘click’ sound of some San languages of Southern Africa—but these are probably within the capability of all human speakers if they are exposed to learning that sound at the right time of life).

These considerations suggest that the anatomical, neurological and physiological underpinnings of language are shared among all of humanity. If the capacity for language did evolve more than once, all traces of it seem to have been lost. This conclusion is buttressed by the FOXP2 evidence (all humans share the same derived gene) and by the fact that genetic data point to all modern humans descending from a common ancestor [ 19 ].

Is language evolution like biological evolution?

Darwin observed that “The formation of different languages and of distinct species, and the proofs that both have been developed through a gradual process, are curiously the same” (page 59 in [ 20 ]). He also asserted that “The survival and preservation of certain favoured words in the struggle for existence is natural selection.” (pages 59–60 in [ 20 ]).

Darwin was right on both counts. Linguists have known from at least the late 18th century [ 21 ]—about 100 years before Darwin—that languages predominantly evolve by a process of descent with modification from earlier ancestral languages, just as biological species descend from earlier ancestral forms. An example is differences observed between the ancient Greek vocabulary in Homer’s Iliad from around 750 BCE and modern Greek vocabulary (Table  1 ) [ 22 ]: some words have merely changed their pronunciation while others have been replaced by new unrelated words.

Regarding Darwin’s assertions that certain words are favoured in the ‘struggle for existence’, it is useful to remember that there is seldom any connection between a sound (a word) and its meaning. This means that selection is reasonably free to choose among words and so features of the words we actually use might reveal its actions. The simplest example is that words that are used more often—such as I , he , she , it , the , you —tend to be shorter, and consequently easier to pronounce, than less frequently used words, such as obstreperous or catafalque [ 23 ]. This is an example of a form of natural selection except here instead of biological individuals competing in the physical environment to survive and reproduce, words compete for space in the environment of the human mind. Our minds give preference to shorter versions of the frequently used words, presumably to reduce effort [ 23 ]. This pressure is relaxed among the less frequently used words, allowing them to be longer. It might also be the case that once the frequently used words have occupied the space of possible short words, there are fewer opportunities for the less frequently used words [ 24 ].

Is it possible to reconstruct the history of a group of languages like we do with species?

Yes. Using common lists of words that are found in all or nearly all languages, linguists can identify shared sets of cognate words—words that descend from common ancestral words— just as it is possible to identify homologous genes that share a common ancestral gene. For instance the Spanish mano (‘hand’) and the French main descend from the earlier Latin manus , while the English and German words hand do not. A cognate set identifies groups of related languages. In the example here mano and main identify the so-called Romance languages (Spanish, French, Italian, Portuguese) and hand and hand identify the Germanic languages (Fig.  1 ). By combining the information in many different cognate sets with appropriate statistical models [ 25 , 26 ], it is possible to infer detailed family histories or phylogenetic trees of language families, such as has been done for the Indo-European languages (Fig.  1 ). These phylogenies are directly analogous to phylogenies of biological species.

Phylogenetic tree of a small subset of the approximately 400 or so Indo-European languages. Words that the languages use for the meaning ‘hand’ are colour-coded to identify cognate classes. Rectangles along the branches identify regions of the tree where new cognate classes might have arisen. Here the French and Spanish languages share cognate forms for ‘hand’ derived from an earlier Latin form ‘manus’. French and Spanish are part of the familiar grouping of Romance languages. By comparison, the word ‘hand’ is cognate between English and German and this cognate class identifies part of the Germanic grouping of languages. The words for ‘hand’ in Greek and in the extinct Anatolian languages Hittite and Tocharian form two additional cognate sets. Combining many different cognate sets from many different vocabulary items allows investigators to draw detailed phylogenetic trees of entire language families (see text)

What other evolutionary features do genes and language share?

Linguistic and biological evolution share features beyond descent with modification and selection, including mechanisms of mutation and replication, speciation, drift and horizontal transfer (Table  2 ). At a deeper level, both genes and languages can be represented as digital systems of inheritance, built on the transmission of discrete chunks of information—genes in the case of biological organisms, and words in the case of language. Genes in turn comprise combinations of the four bases or nucleotides (A, C, G, T) while words can be modelled as comprising combinations of discrete sounds or phones (in fact, phones or sounds vary in a continuous space but languages are commonly represented as expressing a particular set of discrete phonemes).

These similarities mean that we can—and should—think of language as a system for the transmission of information that is tantamount to ‘aural DNA’. Even the peculiar phenomenon of concerted evolution in genetics—where a nucleotide replacement at a specific site in one gene is quickly followed by the same nucleotide replacement at the same site in other, typically related, genes—is also observed in language. Known as regular sound change , a specific phone or sound changes over a relatively short period of time to the same other phone in many words in the lexicon [ 27 , 28 ]. A well-known example is the p → f sound change in the Germanic languages where an older Indo-European p sound was replaced by an f sound, such as in pater → father ; or pes, pedis → foot .

Can changes to language be used to trace human history?

There are currently about 7000 languages spoken around the world, meaning that, oddly, most of us cannot communicate with most other members of our species! Even this number is probably down from the peak of human linguistic diversity that was likely to have occurred around 10,000 years ago, just prior to the invention of agriculture [ 29 ]. Before that time, all human groups had been hunter-gatherers, living in small mobile tribal societies. Farming societies were demographically more prosperous and group sizes were larger than among hunter-gatherers, so the expansion of agriculturalists likely replaced many smaller linguistic groups. Today, there are few hunter-gatherer societies left so our linguistic diversity reflects our relatively recent agricultural past.

Phylogenies of languages can be used in combination with geographical information or information on cultural practices to investigate questions of human history, such as the spread of agriculture. Phylogenies of language families have been used to study the timing, causes and geographic spread of groups of farmers/fishing populations, including the Indo-Europeans [ 30 , 31 , 32 , 33 ]; the pace of occupation of the Pacific by the Austronesian people [ 34 ]; and the migration routes of the Bantu-speaking people through Africa [ 35 , 36 ].

Linguistic phylogenies are also used to investigate questions of human cultural evolution, including the evolution and spread of dairying [ 37 , 38 , 39 ], relationships between religious and political practices [ 40 ], changing political structures [ 41 ] and the age of fairy tales [ 42 ], and have even supplied a date for Homer’s Iliad [ 22 ].

What role has language played in our species’ success?

Language has played a prominent and possibly pre-eminent role in our species’ history. Consider that where all other species tend to be found in the environments their genes adapt them to, humans can adapt at the cultural level, acquiring the knowledge and producing the tools, shelters, clothing and other artefacts necessary for survival in diverse habitats [ 12 , 43 ]. Thus, chimpanzees are found in the dense forests of Africa but not out on the savannah or in deserts or cold regions; camels are found in dry regions but not in forests or mountaintops, and so on for other species. Humans, on the other hand, despite being a species that probably evolved on the African savannahs, have been able to occupy nearly every habitat on Earth. Our behaviour is like that of a collection of biological species [ 43 ]. Why this striking difference?

It is probably down to language. Possessing language, humans have had a high-fidelity code for transmitting detailed information down the generations. Many, if not most, of the things we make use of in our everyday lives rely on specialized knowledge or skills to produce. The information behind these was historically coded in verbal instructions, and with the advent of writing it could be stored and become increasingly complex.

Possessing language, then, is behind humans’ ability to produce sophisticated cultural adaptations that have accumulated one on top of the other throughout our history as a species. Today as a result of this capability we live in a world full of technologies that few of us even understand. Because culture, riding on the back of language, can evolve more rapidly than genes, the relative genetic homogeneity of humanity in contrast to our cultural diversity shows that our ‘aural DNA’ has probably been more important in our short history than genes.

Seyfarth RM, Cheney DL, Marler P. Vervet monkey alarm calls: semantic communication in a free-ranging primate. Anim Behav. 1980;28(4):1070–94.

Article   Google Scholar  

Fitch WT. The evolution of language. Cambridge University Press; 2010.

Terrace HS. How Nim Chimpsky changed my mind. Psychol Today. 1979;13(6):65.

Google Scholar  

Fleagle JG, Assefa Z, Brown FH, Shea JJ. Paleoanthropology of the Kibish Formation, southern Ethiopia: introduction. J Hum Evol. 2008;55(3):360–5.

Article   PubMed   Google Scholar  

Poznik GD, Henn BM, Yee M-C, Sliwerska E, Euskirchen GM, Lin AA, et al. Sequencing Y chromosomes resolves discrepancy in time to common ancestor of males versus females. Science. 2013;341(6145):562–5.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Brown FH, McDougall I, Fleagle JG. Correlation of the KHS Tuff of the Kibish Formation to volcanic ash layers at other sites, and the age of early Homo sapiens (Omo I and Omo II). J Hum Evol. 2012;63(4):577–85.

Henshilwood CS, d’Errico F, Yates R, Jacobs Z, Tribolo C, Duller GA, et al. Emergence of modern human behavior: Middle Stone Age engravings from South Africa. Science. 2002;295(5558):1278–80.

Article   CAS   PubMed   Google Scholar  

Henshilwood CS, Dubreuil B. Reading the artefacts: gleaning language skills from the Middle Stone Age in southern Africa. Cradle Language. 2009;2:61–92.

Klein RG. The human career: human biological and cultural origins. University of Chicago Press; 2009.

Prüfer K, Racimo F, Patterson N, Jay F, Sankararaman S, Sawyer S, et al. The complete genome sequence of a Neanderthal from the Altai Mountains. Nature. 2014;505(7481):43–9.

Meyer M, Arsuaga J-L, de Filippo C, Nagel S, Aximu-Petri A, Nickel B, et al. Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos hominins. Nature. 2016;531(7595):504–7.

Pagel M. Wired for culture: origins of the human social mind. WW Norton & Company; 2012.

Kuhlwilm M, Gronau I, Hubisz MJ, de Filippo C, Prado-Martinez J, Kircher M, et al. Ancient gene flow from early modern humans into Eastern Neanderthals. Nature. 2016;530(7591):429–33.

Enard W, Przeworski M, Fisher SE, Lai CS, Wiebe V, Kitano T, et al. Molecular evolution of FOXP2, a gene involved in speech and language. Nature. 2002;418(6900):869–72.

Enard W, Gehre S, Hammerschmidt K, Hölter SM, Blass T, Somel M, et al. A humanized version of Foxp2 affects cortico-basal ganglia circuits in mice. Cell. 2009;137(5):961–71.

Maricic T, Günther V, Georgiev O, Gehre S, Ćurlin M, Schreiweis C, et al. A recent evolutionary change affects a regulatory element in the human FOXP2 gene. Mol Biol Evol. 2013;30(4):844–52.

Youn H, Sutton L, Smith E, Moore C, Wilkins JF, Maddieson I, et al. On the universal structure of human lexical semantics. Proc Natl Acad Sci U S A. 2016;113(7):1766–71.

Greenberg JH. Some universals of grammar with particular reference to the order of meaningful elements. In: Greenberg JH, editor. Universals of language. Cambridge: MIT Press; 1963. p. 73–113.

Rosenberg NA, Pritchard JK, Weber JL, Cann HM, Kidd KK, Zhivotovsky LA, et al. Genetic structure of human populations. Science. 2002;298(5602):2381–5.

Darwin CR. The descent of man and selection in relation to sex: in two volumes: with illustrations. J. Murray; 1871.

Jones W. Discourses delivered before the Asiatic Society: and miscellaneous papers, on the religion, poetry, literature, etc., of the nations of India. CS Arnold; 1824.

Altschuler EL, Calude AS, Meade A, Pagel M. Linguistic evidence supports date for Homeric epics. BioEssays. 2013;35(5):417–20.

Article   PubMed   PubMed Central   Google Scholar  

Zipf GK. Human behaviour and the principle of least-effort. Cambridge MA edn. Reading: Addison-Wesley; 1949.

Pagel M, Meade A. The deep history of the number words. Phil Trans R Soc B. in press.

Pagel M. Human language as a culturally transmitted replicator. Nat Rev Genet. 2009;10(6):405–15.

CAS   PubMed   Google Scholar  

Pagel M. Darwinian perspectives on the evolution of human languages. Psychonomic Bull Rev. 2017;24(1):151.

Crowley T, Bowern C. An introduction to historical linguistics. Oxford University Press; 2010.

Hruschka DJ, Branford S, Smith ED, Wilkins J, Meade A, Pagel M, et al. Detecting regular sound changes in linguistics as events of concerted evolution. Curr Biol. 2015;25(1):1–9.

Pagel M. The history, rate and pattern of world linguistic evolution. In: Knight C, Studdert-Kennedy M, Hurford J, editors. The evolutionary emergence of language: social function and the origins of linguistic form. Cambridge University Press; 2000

Bouckaert R, Lemey P, Dunn M, Greenhill SJ, Alekseyenko AV, Drummond AJ, et al. Mapping the origins and expansion of the Indo-European language family. Science. 2012;337(6097):957–60.

Chang W, Cathcart C, Hall D, Garrett A. Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis. Language. 2015;91(1):194–244.

Gray RD, Atkinson QD. Language-tree divergence times support the Anatolian theory of Indo-European origin. Nature. 2003;426(6965):435–9.

Haak W, Lazaridis I, Patterson N, Rohland N, Mallick S, Llamas B, et al. Massive migration from the steppe was a source for Indo-European languages in Europe. Nature. 2015;522(7555):207–11.

Gray RD, Drummond AJ, Greenhill SJ. Language phylogenies reveal expansion pulses and pauses in Pacific settlement. Science. 2009;323(5913):479–83.

Currie TE, Meade A, Guillon M, Mace R. Cultural phylogeography of the Bantu Languages of sub-Saharan Africa. Proc R Soc B. 2013;280(1762):20130695.

Grollemund R, Branford S, Bostoen K, Meade A, Venditti C, Pagel M. Bantu expansion shows that habitat alters the route and pace of human dispersals. Proc Natl Acad Sci U S A. 2015;112(43):13296–301.

Holden C, Mace R. Phylogenetic analysis of the evolution of lactose digestion in adults. Hum Biol. 2009;81(5/6):597–619.

Holden CJ, Mace R. Spread of cattle led to the loss of matrilineal descent in Africa: a coevolutionary analysis. Proc R Soc Lond B Biol Sci. 2003;270(1532):2425–33.

Mace R, Jordan F, Holden C. Testing evolutionary hypotheses about human biological adaptation using cross-cultural comparison. Comp Biochem Physiol A Mol Integr Physiol. 2003;136(1):85–94.

Watts J, Greenhill SJ, Atkinson QD, Currie TE, Bulbulia J, Gray RD. Broad supernatural punishment but not moralizing high gods precede the evolution of political complexity in Austronesia. Proc Biol Sci. 2015;282(1804):20142556.

Currie TE, Greenhill SJ, Gray RD, Hasegawa T, Mace R. Rise and fall of political complexity in island South-East Asia and the Pacific. Nature. 2010;467(7317):801–4.

Da Silva SG, Tehrani JJ. Comparative phylogenetic analyses uncover the ancient roots of Indo-European folktales. R Soc Open Sci. 2016;3(1):150645.

Pagel M, Mace R. The cultural wealth of nations. Nature. 2004;428(6980):275–8.

Download references

Acknowledgements

An Advanced Investigator Award 268744 to M. Pagel from the European Research Council has supported most of my recent research on language evolution.

Author information

Authors and affiliations.

School of Biological Sciences, University of Reading, Reading, RG6 6UR, UK

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Mark Pagel .

Ethics declarations

Competing interests.

The author declares that he has no competing interests.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Pagel, M. Q&A: What is human language, when did it evolve and why should we care?. BMC Biol 15 , 64 (2017). https://doi.org/10.1186/s12915-017-0405-3

Download citation

Published : 24 July 2017

DOI : https://doi.org/10.1186/s12915-017-0405-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Neanderthal
• Animal Communication
• Formal Grammatical Structure
• Regular Sound Change
• Distinct Alarm Calls

BMC Biology

ISSN: 1741-7007

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Loading metrics

Open Access

Essays articulate a specific perspective on a topic of broad interest to scientists.

See all article types »

How Could Language Have Evolved?

* E-mail: [email protected]

Affiliations Cognitive Neurobiology and Helmholtz Institute, Departments of Psychology and Biology, Utrecht University, Utrecht, The Netherlands, Department of Zoology and Sidney Sussex College, University of Cambridge, Cambridge, United Kingdom

Affiliation Division of Anthropology, American Museum of Natural History, New York, New York, United States of America

Affiliation Department of Linguistics and Philosophy, MIT, Cambridge, Massachusetts, United States of America

Affiliation Department of Electrical Engineering & Computer Science and Brain and Cognitive Sciences, MIT, Cambridge, Massachusetts, United States of America

• Johan J. Bolhuis, 
• Ian Tattersall, 
• Noam Chomsky, 
• Robert C. Berwick

Published: August 26, 2014

• https://doi.org/10.1371/journal.pbio.1001934
• Reader Comments

The evolution of the faculty of language largely remains an enigma. In this essay, we ask why. Language's evolutionary analysis is complicated because it has no equivalent in any nonhuman species. There is also no consensus regarding the essential nature of the language “phenotype.” According to the “Strong Minimalist Thesis,” the key distinguishing feature of language (and what evolutionary theory must explain) is hierarchical syntactic structure. The faculty of language is likely to have emerged quite recently in evolutionary terms, some 70,000–100,000 years ago, and does not seem to have undergone modification since then, though individual languages do of course change over time, operating within this basic framework. The recent emergence of language and its stability are both consistent with the Strong Minimalist Thesis, which has at its core a single repeatable operation that takes exactly two syntactic elements a and b and assembles them to form the set {a, b}.

Citation: Bolhuis JJ, Tattersall I, Chomsky N, Berwick RC (2014) How Could Language Have Evolved? PLoS Biol 12(8): e1001934. https://doi.org/10.1371/journal.pbio.1001934

Copyright: © 2014 Bolhuis et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: JJB is funded by Utrecht University and by Netherlands Organization for Scientific Research (NWO) grants (ALW Open Competition and NWO Gravity and Horizon Programmes) ( http://www.nwo.nl/ ). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

It is uncontroversial that language has evolved, just like any other trait of living organisms. That is, once—not so long ago in evolutionary terms—there was no language at all, and now there is, at least in Homo sapiens . There is considerably less agreement as to how language evolved. There are a number of reasons for this lack of agreement. First, “language” is not always clearly defined, and this lack of clarity regarding the language phenotype leads to a corresponding lack of clarity regarding its evolutionary origins. Second, there is often confusion as to the nature of the evolutionary process and what it can tell us about the mechanisms of language. Here we argue that the basic principle that underlies language's hierarchical syntactic structure is consistent with a relatively recent evolutionary emergence.

Conceptualizations of Language

The language faculty is often equated with “communication”—a trait that is shared by all animal species and possibly also by plants. In our view, for the purposes of scientific understanding, language should be understood as a particular computational cognitive system, implemented neurally, that cannot be equated with an excessively expansive notion of “language as communication” [1] . Externalized language may be used for communication, but that particular function is largely irrelevant in this context. Thus, the origin of the language faculty does not generally seem to be informed by considerations of the evolution of communication. This viewpoint does not preclude the possibility that communicative considerations can play a role in accounting for the maintenance of language once it has appeared or for the historical language change that has clearly occurred within the human species, with all individuals sharing a common language faculty, as some mathematical models indicate [1] – [3] . A similar misconception is that language is coextensive with speech and that the evolution of vocalization or auditory-vocal learning can therefore inform us about the evolution of language ( Box 1 ) [1] , [4] . However, speech and speech perception, while functioning as possible external interfaces for the language system, are not identical to it. An alternative externalization of language is in the visual domain, as sign language [1] ; even haptic externalization by touch seems possible in deaf and blind individuals [5] . Thus, while the evolution of auditory-vocal learning may be relevant for the evolution of speech, it is not for the language faculty per se. We maintain that language is a computational cognitive mechanism that has hierarchical syntactic structure at its core [1] , as outlined in the next section.

Box 1. Comparative Linguistics: Not Much to Compare

A major stumbling block for the comparative analysis of language evolution is that, so far, there is no evidence for human-like language syntax in any nonhuman species [4] , [41] , [42] . There is no a priori reason why a version of such a combinatorial computational system could not have evolved in nonhuman animals, either through common descent (e.g., apes) or convergent evolution (e.g., songbirds) [1] , [18] . Although the auditory-vocal domain is just one possible external interface for language (with signing being another), it could be argued that the strongest animal candidates for human-like syntax are songbirds and parrots [1] , [41] , [42] . Not only do they have a similar brain organization underlying auditory-vocal behavior [4] , [43] , [44] , they also exhibit vocal imitation learning that proceeds in a very similar way to speech acquisition in human infants [4] , [41] , [42] . This ability is absent in our closest relatives, the great apes [1] , [4] . In addition, like human spoken language, birdsong involves patterned vocalizations that can be quite complex, with a set of rules that govern variable song element sequences known as “phonological syntax” [1] , [4] , [41] , [42] , [45] . Contrary to recent suggestions [46] , [47] , to date there is no evidence to suggest that birdsong patterns exhibit the hierarchical syntactic structure that characterizes human language [41] , [48] , [49] or any mapping to a level forming a language of thought as in humans. Avian vocal-learning species such as parrots are able to synchronize their behavior to variable rhythmic patterns [50] . Such rhythmic abilities may be involved in human prosodic processing, which is known to be an important factor in language acquisition [51] .

The Faculty of Language According to the “Strong Minimalist Thesis”

In the last few years, certain linguistic theories have arrived at a much more narrowly defined and precise phenotype characterizing human language syntax. In place of a complex rule system or accounts grounded on general notions of “culture” or “communication,” it appears that human language syntax can be defined in an extremely simple way that makes conventional evolutionary explanations much simpler. In this view, human language syntax can be characterized via a single operation that takes exactly two (syntactic) elements a and b and puts them together to form the set {a, b}. We call this basic operation “ merge ” [1] . The “Strong Minimalist Thesis” (SMT) [6] holds that merge along with a general cognitive requirement for computationally minimal or efficient search suffices to account for much of human language syntax. The SMT also requires two mappings: one to an internal conceptual interface for thought and a second to a sensory-motor interface that externalizes language as speech, sign, or other modality [1] . The basic operation itself is simple. Given merge , two items such as the and apples are assembled as the set { the, apples }. Crucially, merge can apply to the results of its own output so that a further application of merge to ate and { the, apples } yields the set { ate , { the, apples }}, in this way deriving the full range of characteristic hierarchical structure that distinguishes human language from all other known nonhuman cognitive systems.

As the text below and Figure 1 shows, merge also accounts for the characteristic appearance of displacement in human language—the apparent “movement” of phrases from one position to another. Displacement is not found in artificially constructed languages like computer programming languages and raises difficulties for parsing as well as communication. On the SMT account, however, displacement arises naturally and is to be expected, rather than exceptional, as seems true in every human language that has been examined carefully. Furthermore, hierarchical language structure is demonstrably present in humans, as shown, for instance, by online brain imaging experiments [7] , but absent in nonhuman species, e.g., chimpanzees taught sign language demonstrably lack this combinatorial ability [8] . Thus, before the appearance of merge , there was no faculty of language as such, because this requires merge along with the conceptual atoms of the lexicon. Absent this, there is no way to arrive at the essentially infinite number of syntactic language structures, e.g., “the brown cow,” “a black cat behind the mat” [9] – [11] , etc. This view leaves room for the possibility that some conceptual atoms were present antecedent to merge itself, though at present this remains entirely speculative. Even if true, there seems to be no evidence for an antecedent combinatorial and hierarchical syntax. Furthermore, merge itself is uniform in the contemporary human population as well as in the historical record, in contrast to human group differences such as the adult ability to digest lactose or skin pigmentation [12] . There is no doubt that a normal child from England raised in northern Alaska would readily learn Eskimo-Aleut, or vice versa; there have been no confirmed group differences in the ability of children to learn their first language, despite one or two marginal, indirect, and as yet unsubstantiated correlative indications [13] . This uniformity and stability points to the absence of major evolutionary change since the emergence of the language faculty. Taken together, these facts provide good evidence that merge was indeed the key evolutionary innovation for the language faculty.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

Left: The circled structure Y, corresponding to what , the object of the verb eat , is a subset of the circled structure X, corresponding to boys eat what . Right: The free application of merge to X, Y in this case automatically leads to what occupying two syntactic positions, as required for proper semantic interpretation. The original what remains as the object of the verb so that it can serve as an argument to this predicate, and a copy of what , “displaced,” is now in the position of a quantificational operator so that the form can be interpreted as “for what x , boys eat x .” Typically, only the higher what is actually pronounced, as indicated by the line drawn through the lower what .

https://doi.org/10.1371/journal.pbio.1001934.g001

It is sometimes suggested that external motor sequences are “hierarchical” in this sense and so provide an antecedent platform for language [14] . However, as has been argued [15] , motor sequences resemble more the “sequence of letters in the alphabet than the sequences of words in a sentence” ( [15] , p. 221). (For expository purposes, we omit here several technical linguistic details about the labelling of these words; see [16] .) Along with the conceptual atoms of the lexicon, the SMT holds that merge , plus the internal interface mappings to the conceptual system, yields what has been called the “language of thought” [17] .

More narrowly, the SMT also suffices to automatically derive some of the most central properties of human language syntax. For example, one of the most distinctive properties of human language syntax is that of “displacement,” along with what is sometimes called “duality of semantic patterning.” For example, in the sentence “(Guess) what boys eat,” “what” takes on a dual role and is interpreted in two places: first, as a question “operator” at the front of the sentence, where it is pronounced; and second, as a variable that serves as the argument of the verb eat , the thing eaten, where it is not pronounced ( Figure 1 ). (There are marginal exceptions to the nonpronunciation of the second “what” that, when analyzed carefully, support the picture outlined here.) Given the free application of merge , we expect human languages to exhibit this phenomenon of displacement without any further stipulation. This is simply because operating freely, without any further constraints, merge derives this possibility. In our example “(Guess) what boys eat,” we assume that successive applications of merge as in our earlier example will first derive { boys , { eat, what }}—analogous to { boys , { eat, apples }}. Now we note that one can simply apply merge to the two syntactic objects { boys ,{ eat, what }} and { what }, in which { what } is a subcomponent (a subset) of the first syntactic object rather than some external set. This yields something like { what , { boys , { eat, what }}}, in this way marking out the two required operator and variable positions for what .

The Nature of Evolution

Evolutionary analysis might be brought to bear on language in two different ways. First, evolutionary considerations could be used to explain the mechanisms of human language. For instance, principles derived from studying the evolution of communication might be used to predict, or even explain, the structural organization of language. This approach is fraught with difficulties. Questions of evolution or function are fundamentally different from those relating to mechanism, so evolution can never “explain” mechanisms [18] . For a start, the evolution of a particular trait may have proceeded in different ways, such as via common descent, convergence, or exaptation, and it is not easy to establish which of these possibilities (or combination of them) is relevant [18] , [19] . More importantly, evolution by natural selection is not a causal factor of either cognitive or neural mechanisms [18] . Natural selection can be seen as one causal factor for the historical process of evolutionary change, but that is merely stating the essence of the theory of evolution. As we have argued, communication cannot be equated with language, so its evolution cannot inform the mechanisms of language syntax. However, evolutionary considerations—in particular, reconstructing the evolutionary history of relevant traits—might provide clues or hypotheses as to mechanisms, even though such hypotheses have frequently been shown to be false or misleading [18] . One such evolutionary clue is that, contrary to received wisdom, recent analyses suggest that significant genetic change may occur in human populations over the course of a few hundred years [19] . Such rapid change could also have occurred in the case of language, as we will argue below. In addition, as detailed in the next section, paleoanthropological evidence suggests that the appearance of symbolic thought, our most accurate proxy for language, was a recent evolutionary event. For instance, the first evidence of putatively symbolic artifacts dates back to only around 100,000 years ago, significantly after the appearance on the planet of anatomically distinctive Homo sapiens around 200,000 years ago [20] , [21] ,

The second, more traditional way of applying evolutionary analysis to language is to attempt to reconstruct its evolutionary history. Here, too, we are confronted with major explanatory obstacles. For starters, language appears to be unique to the species H. sapiens . That eliminates one of the cornerstones of evolutionary analysis, the comparative method, which generally relies on features that are shared by virtue of common descent ( Box 1 ) [1] , [4] , [18] . Alternatively, analysis can appeal to convergent evolution, in which similar features, such as birds' wings and bats' wings, arise independently to “solve” functionally analogous problems. Both situations help constrain and guide evolutionary explanation. Lacking both, as in the case of language, makes the explanatory search more difficult. In addition, evolutionary analysis of language is often plagued by popular, naïve, or antiquated conceptions of how evolution proceeds [19] , [22] . That is, evolution is often seen as necessarily a slow, incremental process that unfolds gradually over the eons. Such a view of evolutionary change is not consistent with current evidence and our current understanding, in which evolutionary change can be swift, operating within just a few generations, whether it be in relation to finches' beaks on the Galapagos, insect resistance to pesticides following WWII, or human development of lactose tolerance within dairy culture societies, to name a few cases out of many [19] , [22] – [24] .

Paleoanthropology

Language leaves no direct imprint in the fossil record, and the signals imparted by putative morphological proxies are highly mixed. Most of these involve speech production and detection, neither of which by itself is sufficient for inferring language (see Box 2 ). After all, while the anatomical potential to produce the frequencies used in modern speech may be necessary for the expression of language, it provides no proof that language itself was actually employed. What is more, it is not even necessary for language, as the visual and haptic externalization routes make clear. Moreover, even granting that speech is a requirement for language, it has been argued convincingly [25] , [26] that equal proportions of the horizontal and vertical portions of the vocal tract are necessary for producing speech. This conformation is uniquely seen in our own species Homo sapiens . In a similar vein, the aural ability of nonhuman primates like chimpanzees or extinct hominid species such as H. neanderthalensis to perceive the sound frequencies associated with speech [26] , [27] says nothing about the ability of these relatives to understand or produce language. Finally, neither the absolute size of the brain nor its external morphology as seen in endocasts has been shown to be relevant to the possession of language in an extinct hominid ( Figure 2 ) [28] . Recent research has determined that Neanderthals possessed the modern version of the FOXP2 gene [29] , malfunctions in which produce speech deficits in modern people [4] , [30] . However, FOXP2 cannot be regarded as “the” gene “for” language, since it is only one of many that have to be functioning properly to permit its normal expression.

Although after an initial flatlining this plot appears to show consistent enlargement of hominid brains over the last 2 million years, it is essential to note that these brain volumes are averaged across a number of independent lineages within the genus Homo and likely represent the preferential success of larger-brained species. From [20] . Image credit: Gisselle Garcia, artist (brain images) .

https://doi.org/10.1371/journal.pbio.1001934.g002

Box 2. The Infamous Hyoid Bone

A putative relationship between basicranial flexion, laryngeal descent, and the ability to produce sounds essential to speech was suggested [52] before any fossil hyoid bones, the sole hard-tissue components of the laryngeal apparatus, were known. It was speculated that fossil hyoids would indicate when speech, and by extension language, originated. A Neanderthal hyoid from Kebara in Israel eventually proved very similar to its H. sapiens homologue, prompting the declaration that speech capacity was fully developed in adult H. neanderthalensis [53] . This was soon contested on the grounds that the morphology of the hyoid is both subsidiary [25] and unrelated [26] to its still-controversial [36] position in the neck. A recent study [54] focuses on the biomechanics, internal architecture, and function of the Kebara fossil. The authors conclude that their results “add support for the proposition that the Kebara 2 Neanderthal engaged in speech” ( [54] , p. 6). However, they wisely add that the issue of Neanderthal language will be fully resolved only on the basis of fuller comparative material. While the peripheral ability to produce speech is undoubtedly a necessary condition for the expression of vocally externalized language, it is not a sufficient one, and hyoid morphology, like most other lines of evidence, is evidently no silver bullet for determining when human language originated.

In terms of historically calibrated records, this leaves us only with archaeology, the archive of ancient human behaviors—although we have once again to seek indirect proxies for language. To the extent that language is interdependent with symbolic thought [20] , the best proxies in this domain are objects that are explicitly symbolic in nature. Opinions have varied greatly as to what constitutes a symbolic object, but if one excludes stone and other Paleolithic implements from this category on the fairly firm grounds that they are pragmatic and that the techniques for making them can be passed along strictly by imitation [31] , we are left with objects from the African Middle Stone Age (MSA) such as pierced shell beads from various ∼100,000-year-old sites (e.g., [32] ) and the ∼80,000-year-old geometrically engraved plaques from South Africa's Blombos Cave [33] as the earliest undisputed symbolic objects. Such objects began to be made only substantially after the appearance, around 200,000 years ago, of anatomically recognizable H. sapiens , also in Africa [34] . To be sure, this inference from the symbolic record, like much else in paleontology, rests on evidence that is necessarily quite indirect. Nevertheless, the conclusion lines up with what is known from genomics.

Our species was born in a technologically archaic context [35] , and significantly, the tempo of change only began picking up after the point at which symbolic objects appeared. Evidently, a new potential for symbolic thought was born with our anatomically distinctive species, but it was only expressed after a necessary cultural stimulus had exerted itself. This stimulus was most plausibly the appearance of language in members of a species that demonstrably already possessed the peripheral vocal apparatus required to externalize it [20] , [22] . Then, within a remarkably short space of time, art was invented, cities were born, and people had reached the moon. By this reckoning, the language faculty is an extremely recent acquisition in our lineage, and it was acquired not in the context of slow, gradual modification of preexisting systems under natural selection but in a single, rapid, emergent event that built upon those prior systems but was not predicted by them. It may be relevant to note that the anatomical ability to express language through speech was acquired at a considerable cost, namely the not-insignificant risk of adults choking to death [25] , [36] , as simultaneous breathing and swallowing became impossible with the descent of the larynx. However, since this conformation was already in place before language had demonstrably been acquired (see Box 2 ), the ability to express language cannot by itself have been the countervailing advantage. Finally, there has been no detectable evolution of the language faculty since it emerged, with no known group differences. This is another signature of relatively recent and rapid origin. For reasons like these, the relatively sudden origin of language poses difficulties that may be called “Darwin's problem.”

The Minimalist Account of Language—Progress towards Resolving “Darwin's Problem”

The Strong Minimalist Thesis (SMT) [6] , as discussed above, greatly eases the explanatory burden for evolutionary analysis, since virtually all of the antecedent “machinery” for language is presumed to have been present long before the human species appeared. For instance, it appears that the ability to perceive “distinctive features” such as the difference between the sound b , as in bat , as opposed to p , as in pat , might be present in the mammalian lineage generally [37] , [38] . The same holds for audition. Both comprise part of the externalization system for language. Furthermore, the general constraint of efficient computation would also seem plausibly antecedent in the cognitive computation of ancestral species. The only thing lacking for language would be merge , some specific way to externalize the internal computations and, importantly, the “atomic conceptual elements” that we have identified with words. Without merge , there would be no way to assemble the arbitrarily large, hierarchically structured objects with their specific interpretations in the language of thought that distinguish human language from other animal cognitive systems—just as Darwin insisted: “A complex train of thought can be no more carried out without the use of words, whether spoken or silent, than a long calculation without the use of figures or algebra” ( [39] , p. 88). With merge , however, the basic properties of human language emerge. Evolutionary analysis can thus be focused on this quite narrowly defined phenotypic property, merge itself, as the chief bridge between the ancestral and modern states for language. Since this change is relatively minor, it accords with what we know about the apparent rapidity of language's emergence.

Conclusions

The Strong Minimalist Thesis that we have sketched here is consistent with a recent and rapid evolutionary emergence of language. Although this thesis is far from being established and contains many open questions, it offers an account that is compatible with the known empirical evolutionary evidence. Such an account also aligns with what we currently know about the relatively few genomic differences between our species and other ancestral Homo species—e.g., only about 100 coding gene differences between Homo sapiens and H. neanderthalensis , the majority of them in nonlanguage areas such as the olfactory and immune systems [40] . Furthermore, as far as we can tell from direct historical evidence, the capacity that emerged, namely the ability of any child to learn any human language, has remained frozen for 10,000 years or more. To be sure, such observations must be interpreted with great care and can remain only suggestive as long as we lack the knowledge to even crudely connect genomic changes to the relevant phenotypes. Even given these caveats, it appears that there has simply not been enough time for large-scale evolutionary changes, as indicated by the SMT. Clearly, such a novel computational system could have led to a large competitive advantage among the early H. sapiens who possessed it, particularly when linked to possibly preexisting perceptual and motor mechanisms.

• View Article
• Google Scholar
• 6. Chomsky N (2000) Minimalist inquiries: The framework. In: Martin R, Michaels D, Uriagereka J, editors. Step by step: Essays on minimalist syntax in honor of Howard Lasnik. Cambridge (Massachusetts): MIT Press. pp. 89–155.
• 9. Hurford J (2012) Language in the light of evolution. Oxford: Oxford University Press.
• 10. Jackendoff R (2003) Foundations of language: Brain, meaning, grammar, evolution. Oxford: Oxford University Press.
• 11. Bickerton D (1981) Roots of language. Ann Arbor (Michigan): Karoma Publishers.
• 13. Ladd DR, Dediu D (2013) Genes and linguistic tone. In: Pashler H, editor. Encyclopedia of the mind. London: Sage Publications. pp. 372–373.
• 17. Berwick R, Chomsky N (2011) The biolinguistic program: The current state of its development. In: Di Sciullo AM, Boeckx C, editors. The biolinguistic enterprise. Oxford: Oxford University Press. pp. 19–41.
• 21. Dubreuil B, Henshilwood CS (2013) Material culture and language. In: Lefebvre C, Comrie B, Cohen H, editors. New perspectives on the origin of language. Amsterdam: John Benjamins. pp. 147–170.
• 22. Tattersall I (2012) Masters of the planet. The search for our human origins. New York: Palgrave Macmillan.
• 23. Thompson JH (2013) Relentless evolution. Chicago: University of Chicago Press.
• 24. Nei M (2013) Mutation-driven evolution. Oxford: Oxford University Press.
• 28. DeSalle R, Tattersall I (2012) The brain: Big bangs, behaviors, and beliefs. New Haven (Connecticut): Yale University Press.
• 36. Lieberman DE (2011) The evolution of the human head. Cambridge, MA: Harvard University Press.
• 38. Hauser M, Fitch T (2003) What are the uniquely human components of the language faculty? In: Christiansen MH, Kirby S, editors. Language Evolution. Oxford: Oxford University Press. pp. 158–181.
• 39. Darwin C (1871) The Descent of man, and selection in relation to sex. London: John Murray.
• 42. Bolhuis JJ, Everaert M, editors (2013) Birdsong, speech, and language. Exploring the evolution of mind and brain. Cambridge (Massachusetts): MIT Press.
• 49. Everaert M, Huybregts R (2013) The design principles of natural language. In: Bolhuis JJ, Everaert M, editors. Birdsong, speech, and language. Exploring the evolution of mind and brain. Cambridge (Massachusetts): MIT Press. pp. 1–26.

English Language Evolution Essay

The evolution of species is a natural process, and so is the language evolution. Indeed, being a part of people’s culture, which is prone to changes, language also shapes together with people’s development; some words become dated, and new ones appear instead, adding new shades of meaning to already existing notions. In addition, new objects and phenomena appear, making people come up with the words which the given thing can be named with; thus, the language shapes together with the nation, growing and getting increasingly more complicated. Political processes which a country undergoes must be taken into account as well; shaping the history of a country, certain political events can spawn the language evolution or, on the contrary, cause its regression, for politics shapes culture and, therefore, is inseparable from the latter. Because of the consolidation processes which England was experiencing in the course of the sixteenth century and the following strengthening of the empire, as well as the establishment of the relationships with other states of power, the English language was shaped greatly, including its grammar, syntax and especially vocabulary, influenced by both the communication with the foreign countries and the consolidation of the land, which meant the unification of the English language standards and the establishment of an official English language, shifting the existing dialects into the shadow.

When it comes to discussing the Middle Ages, the first work of literature that comes to people’s minds is the famous Beowulf . Even though its origins can hardly be traced and that no one can say with certainty who the actual author of the poem was, it still is and will be a decent specimen of the Old English language. In every single line, one can track the details which are characteristic of the Old English period. To start with, the influence of the Celtic languages becomes especially distinct when reading the poem. For example, the word stock of the poem indicates clearly that Beowulf belongs to the Old English epoch; the few Celtic words which are used in Beowulf are the graphic evidence of the above-mentioned fact: “Hwæt”, “Syððan” (Greenblatt et al. 29).

In addition, the fact that the Old English language has borrowed a lot from the Celts becomes obvious when considering the structure of the sentences in the poem: in “We have heard of those princes’ heroic campaigns” (Greenblatt et al. 29), for example, the sentence structure is fixed, with the verb taking the second place in the sentence, which is typical for a Celtic language. However, the Celtic elements are not the only indicators of the Old English language in Beowulf . Since the Norman Conquest has had a huge impact on England and the word stock of the English language since 1066, in Beowulf , one can track the elements of the Norman language as well, such as “Gardena” (Greenblatt et al. 29). However, the Norman features of the Old English language concern mainly the vocabulary; the grammar is still shaped by the Celtic traditions. Finally, when redefining Beowulf ’s impact on the development of the English language in the Middle Ages, the introduction of the Anglo-Saxon culture into the English one must be mentioned. The Anglo-Saxon impact can be seen in such lines of the poem as “in mægþa gehwære man geþeon” and “Swa sceal geong guma gode gewyrcean” (Greenblatt et al. 29).

However, when the Sixteenth Century with its Early Modern English norms came, the language rules started taking a new direction. When considering Thomas More’s work, one can see distinctly that the English language was abandoning the inflected forms and shaped towards being a more analytical language. Keeping one change at a time, one should consider first the drop of the “e” vowel at the end of the words – “e” becomes muted or even disappears completely in some cases. However, the disappearance or, in most cases, muting of the “e” vowel at the end of the words was not the only characteristics of the Early Modern English. It is also worth noting that in More’s novel, nouns are no longer inflected – the inflected forms have dropped together with the muted “e”: “between” instead of “abitweonum” (Greenblatt et al. 521) and “above” ousting “abufan” (Greenblatt et al. 521).

Moreover, the word stock underwent a number of changes as well – instead of Celtic and Anglo-Saxon words, their Norman equivalents were used in Utopia : the word “glengan” was no longer of any use, and “adorn” was used in the text to make it sound more solid (Greenblatt et al. 521), not to mention the obvious fact that the verb “to be” took the place of “wesan” in Moor’s book. When reading Moor’s novel, one can easily realize that the language used in it is already one inch away from the English language which is used in the XXI century; even though the choice of words sometimes sounds rather weird for the XXI-century people, the vocabulary and the basic structure of the English language is already there, which can be indicated by the fact that in complex sentences, the word order is no longer fixed: “This Raphael, who from his family carries the name of Hythloday, is not ignorant of the Latin tongue” (Greenblatt et al. 522). Finally, the elements of the Latin and Greek can be seen in Moor’s work: “discoursing” (Greenblatt et al. 522), “church” (Greenblatt et al. 560). Unfortunately, the consequences of the Great Vowel Shift cannot be traced in the work, since it is written in prose and, therefore, does not have any rhymes, assonances or any other phenomena which could help show the dynamics of the “a” and “e” vowels, etc.

Finally, the changes which happened to the English language in the seventeenth century must be addressed. To analyze the issues which led to the further development of the English language, the political situation of England must be addressed. Because of the consolidation of the future British Empire, the language standard was supposed to be provided for all the cities and towns of England. The given measure was considered to be the driving force which could bring the country even closer. Hence, the rules for the English language were first introduced by Samuel Johnson, which was the final straw that the numerous English dialects could bear. The fact that the language rules finally came into force in the seventeenth century can be easily proven by offering a couple of examples from An Essay Concerning Human Understanding , one of the major works of John Locke, a prominent philosopher of the seventeenth century. Even though the new rules introduced by Johnson still felt a little bit rough around the edges, Locke obviously followed them when writing his work – the pronouns “it”, “my” (Greenblatt et al. 2152) make it clear that the English language has been finally shaped into its modern form.

Unlike one might have thought, the development of the English language is linked directly to the political processes within the state, which the authors of The Norton Anthology of English Literature show in a very graphic way. To be more exact, the more consolidated and powerful England became, the more the English language flourished, with words obtaining additional meanings and metaphors becoming an integral part of the language. The above-mentioned process was quite expected, since the country was finally strengthening its relationships with the rest of the world, therefore, allowing that the rest of the cultures were introduced to the British one and, hence, the language could be enriched with the words possessed from other languages, such as the French one. Moreover, the fact that England was finally making steps towards consolidation meant that the empire was going to improve its score on all aspects, including social, political and cultural ones; hence, language, as an integrate part of culture, was also to undergo a train of changes. Offering good premises for the further evolution of the English language and introducing a number of changes to the language, including the changes in phonetics, grammar, vocabulary and syntax, English poets and narrators have changed the English language greatly within a matter of several centuries, which on the scale of the world history is an incredibly small amount of time.

Works Cited

Greenblatt, Stephen et al. The Norton Anthology of English Literature. Vol. A. 9th ed. New York, NY: Norton & Company, 2012. Print.

• Chicago (A-D)
• Chicago (N-B)

IvyPanda. (2020, May 19). English Language Evolution. https://ivypanda.com/essays/english-language-evolution/

"English Language Evolution." IvyPanda , 19 May 2020, ivypanda.com/essays/english-language-evolution/.

IvyPanda . (2020) 'English Language Evolution'. 19 May.

IvyPanda . 2020. "English Language Evolution." May 19, 2020. https://ivypanda.com/essays/english-language-evolution/.

1. IvyPanda . "English Language Evolution." May 19, 2020. https://ivypanda.com/essays/english-language-evolution/.

Bibliography

IvyPanda . "English Language Evolution." May 19, 2020. https://ivypanda.com/essays/english-language-evolution/.

• Code switching and millennial generation
• "The Poem of Beowulf" in the Anglo-Saxon Culture
• Celtic Christianity Evolution
• Heroism in Beowulf
• Anglo-Saxon Britain: The Battle of Hastings
• The Image of the Celtic Warrior
• The Mead-Hall in “Beowulf” and Anglo-Saxon Society
• Vowel Length Contrasts in Arabic and Japanese
• Beowulf: Role of Women
• Morphology in English Learning and Instruction
• Armenian Linguistic Community in Southern California
• African Scam Letters' Linguistic Anthropology
• Language Borrowing Definition
• Language Evolution in Human Being
• Language Intervention - Accents and Dialects

REVIEW article

The mystery of language evolution.

• 1 Risk-Eraser, LLC, West Falmouth, MA, USA
• 2 Department of Linguistics and Computer and Information Sciences, University of Pennsylvania, Philadelphia, PA, USA
• 3 Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA
• 4 Division of Anthropology, American Museum of Natural History, New York, NY, USA
• 5 Department of Integrative Biology, University of Texas, Austin, TX, USA
• 6 Department of Theoretical and Applied Linguistics, Cambridge University, Cambridge, UK
• 7 Department of Linguistics and Philosophy, Massachusetts Institute of Technology, Cambridge, MA, USA
• 8 Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, MA, USA

Understanding the evolution of language requires evidence regarding origins and processes that led to change. In the last 40 years, there has been an explosion of research on this problem as well as a sense that considerable progress has been made. We argue instead that the richness of ideas is accompanied by a poverty of evidence, with essentially no explanation of how and why our linguistic computations and representations evolved. We show that, to date, (1) studies of nonhuman animals provide virtually no relevant parallels to human linguistic communication, and none to the underlying biological capacity; (2) the fossil and archaeological evidence does not inform our understanding of the computations and representations of our earliest ancestors, leaving details of origins and selective pressure unresolved; (3) our understanding of the genetics of language is so impoverished that there is little hope of connecting genes to linguistic processes any time soon; (4) all modeling attempts have made unfounded assumptions, and have provided no empirical tests, thus leaving any insights into language's origins unverifiable. Based on the current state of evidence, we submit that the most fundamental questions about the origins and evolution of our linguistic capacity remain as mysterious as ever, with considerable uncertainty about the discovery of either relevant or conclusive evidence that can adjudicate among the many open hypotheses. We conclude by presenting some suggestions about possible paths forward.

Introduction

Inquiry into the origins of language was banned by the Société de Linguistique de Paris in 1866 because speculative flourishes far outpaced hard evidence. Within the past 40 or so years, however, writings on this subject have exploded ( Lieberman, 1984 ; Bickerton, 1990 ; Pinker and Bloom, 1990 ; Jackendoff, 1999 ; Fitch, 2010 ; Hurford, 2011 ), implying that hard evidence has outpaced speculation. This perspective, shared by many, is due in part to the emergence of new techniques to study animal social behavior, decipher the fossil record, map genomes, and model evolutionary processes. The sheer abundance and public visibility of such studies, including claims of human-like cognition in birds and primates, along with talking Neanderthals, might suggest that important strides have been made in understanding the origins of human language, its precursors in other animals, the selective pressures that led to its design features and adaptive significance, as well as its genetic underpinnings. We argue instead that both scientists and journalists have rushed to premature conclusions based on woefully incomplete or absent evidence.

We begin with a brief case study to illustrate how biologists typically study the evolution of a behavioral phenotype. We then turn to a discussion of the language phenotype, including its core biological computations and representations. Next, we discuss four approaches to the evolution of language: comparative animal behavior, paleontology and archaeology, molecular biology, and mathematical modeling. In each section, we state why we consider the evidence inconclusive or irrelevant. We conclude with a brief set of empirical desiderata for moving forward, noting the limitations that lie ahead, at least for the foreseeable future.

How to Study the Evolution of a Trait

Understanding biological evolution requires distinguishing patterns and processes, dissecting potential contributions from both random and non-random mechanisms including genetic drift, migration, selection, developmental unfolding, and genetic constraints. Rarely do biologists have access to all of the relevant evidence, and this is especially true for higher vertebrates and the complicated social behaviors they exhibit. When it comes to human language evolution, the paucity of relevant evidence is significant and, as we discuss below, the potential for acquiring such evidence is completely closed off in some relevant areas of inquiry (e.g., no options for living sister species). This point is brought into focus by looking at far simpler systems, such as the túngara frog ( Ryan, 1985 ; Ryan and Rand, 2003 ), which we turn to next.

Male túngara frogs sexually advertise with simple (whine only) or complex (whine plus chucks) mating calls. Playback experiments show that males add chucks in response to the calls of other males and that females prefer calls with chucks. Thus, males gain a reproductive benefit by making complex calls, while females gain a reproductive advantage from mating with larger males who fertilize more eggs. The reproductive gain to males is, however, partially offset by the attractiveness of chucks to frog-eating bats.

The chucks are generated by large larynges with pendulous masses extending from the vocal folds. This morphology and the chucks they produce are restricted to túngara frogs and their sister species. The tuning of the female's two inner ear organs, the AP (amphibian papilla) and BP (basilar papilla), match the dominant frequencies of the whine and chuck, respectively, and the BP tuning is better matched to and thus more stimulated by the lower-frequency chucks of larger males than their higher-pitched conspecifics ( Ryan et al., 1990 ).

In sum, we know how these frogs communicate, the fitness costs and benefits of communication, the phylogenetic distribution of key traits, and details of the mechanisms underlying signal production, perception, and behavioral response. But how did this system evolve?

The obvious adaptive hypothesis is that females evolved their BP tuning because of the reproductive advantage of mating with larger males making lower-frequency calls. Alternatively, BP tuning represents an ancestral trait, whereby males evolved calls to exploit the female's preexisting sensory biases. Comparative evidence resolved this issue ( Wilczynski et al., 2001 ). Only túngara frog males and their sister species add chucks to their whines. The other related species only produce whines, their whines all stimulate their AP organs, and they all have BP organs that are not recruited into the communication system. For most of these species, the tuning is statistically indistinguishable from the BP tuning of the túngara frog. These findings reject the adaptationist hypothesis in favor of the alternative: tuning existed in this clade for millions of years prior to the evolution of the chuck, and was poised to be stimulated when males eventually evolved the larynx allowing them to produce chucks. These results highlight the point that arguments based on fit and sound adaptive logic need not be correct.

In sum, evolutionary analyses demand a clear specification of the target phenotype, empirical evidence linking details of trait design features to fitness consequences, an understanding of the comparative landscape in terms of homologous and analogous traits, and tests that distinguish adaptive from non-adaptive explanations for trait diversification. This recipe for successful evolutionary analysis has rarely been followed in the case of language, and given the limited evidence available, the current prospects are not strong, especially in some domains of analysis. For example, unlike the túngara frog, there are no living sister species to test out phylogenetic hypotheses, and for both methodological and ethical reasons, no ability to manipulate particular characteristics of the language faculty to assess the impact on individual fitness. Nonetheless, the most productive way forward, we believe, is to define important details of the language phenotype, recognize generally accepted methods and evidence in evolutionary biology, and work within this framework to assess what we may learn about the evolution of language.

The Language Phenotype

In this paper, we are interested in biological as opposed to cultural evolution. Given this focus, we ask: what are the core biological mechanisms that enable the capacity for language? As we and many other language scientists see it, the core competence for language is a biological capacity shared by all humans and distinguished by the central feature of discrete infinity—the capacity for unbounded composition of various linguistic objects into complex structures. These structures are generated by a recursive procedure that mediates the mapping between speech- or sign-based forms and meanings, including semantics of words and sentences and how they are situated and interpreted in discourse. This approach distinguishes the biological capacity for language from its many possible functions, such as communication or internal thought.

One approach to exploring the core biological competence for language, and thus its evolution, was set out by Hauser et al. (2002) . A central focus of this paper was on conceptual and methodological issues that might help distinguish capacities that are shared with other animals as opposed to being uniquely human, as well as capacities that are uniquely human and unique to language, as opposed to shared with other domains of knowledge. These distinctions were mapped, respectively, to the Faculty of Language in the Broad (FLB) sense and the Faculty of Language in the Narrow (FLN) sense. FLB designates processes that are shared with other animals, and thus, are involved in language and other sensory-motor and conceptual-intentional processes. FLN, in contrast, describes processes that are uniquely human and unique to language. As a hypothesis, Hauser et al. proposed that FLN consists of the recursive mechanisms for discrete infinity along with mappings to the interfaces with the conceptual-intentional and sensory-motor systems. Note that this proposal is not just about the recursive operations, but about how such procedures connect to other mind-internal representations, often discussed as the systems of semantics and phonology. This is an example of an evolutionary hypothesis that focuses on the core biological competence, and lays out a proposal for empirical research. In particular, if FLN is as described, there should be no homologs or analogs in other animals and no comparable processes in other domains of human thought. To help sharpen this proposal and others, we turn next to some of the core components of the language phenotype.

Language, in all aspects, consists of abstract units of information that are organized and combined following specific computational procedures. The phoneme, originally held to be the basic unit of phonology, has been shown to further decompose into combinations of features that characterize articulatory actions, which shape the sound patterns of languages. For instance, the English past tense “ed” is sometimes pronounced as /t/ as in “kissed” but /d/ as in “hugged.” This seemingly arbitrary fact is predicted on the basis of articulatory features. If the final phoneme of the verb ends in the feature “voiced,” which involves the vibration of the vocal cords during articulation (e.g., /g/ in “hug”), the past tense “ed” is pronounced as /d/, which is also voiced. By contrast, /s/ in “kiss” is unvoiced—no vocal cord vibration—which automatically results in the similarly unvoiced /t/ for “kissed.” These phonological rules can be described as familiar IF-THEN statements in computational systems. Children unconsciously and spontaneously follow these rules that may be generalized to novel words (e.g., twerked, Googled). Importantly, all of these phonological processes are paralleled in sign language (except for those directly linked to voice articulation), highlighting the fact that the computations of language are not tied to a particular sensory modality.

Word formation also takes place by combining informational units in a stepwise process familiar in computational systems. For instance, the word “unlockable” is doubly ambiguous—meaning usable either as a functional lock or a broken lock. Such duality can be captured by differences in the logical sequence of morpheme combinations: the functional lock results from combining “un” and “lock” together to be followed by “able”—something that is possible to unlock—whereas the broken lock is derived from combining “lock” and “able” first, before “un” imposes a negative sense. Such ambiguity can be represented in a form similar to arithmetic expressions: [un-[lock-able]] vs. [[un-lock]-able], which encodes the combinatorial process in word formation.

The syntactic system of human language offers the clearest demonstration of discrete infinity, but with important constraints on the range of variation. Consider the sentence “Instinctively, eagles that eat swim.” The adverb “instinctively” describes the capacity of swimming rather than eating, even though “instinctively” is linearly closer to “eat” than “swim.” In fact, no matter how far apart “instinctively” and “swim” are from each other in terms of the number of words or linear distance, the association remains: “Instinctively, eagles that eat in the Rocky Mountains swim.” The correct semantic interpretation can be derived if we consider the logical sequence in which sentence structures are built, as in the “unlockable” example above. The relative clause “eat in the Rocky Mountains” combines with and modifies the noun phrase “eagles” first, which together as a unit combines with the verb “swim.” The resulting nested structure—a clause—is then combined with the adverb “instinctively;” or, in the arithmetic representation, [Instinctively, [eagles [that eat in the Rocky Mountains] swim]]. These results, together with those involving hundreds of other languages, reveal both that the capacity for language takes hierarchical/nested structures as the basic building blocks of syntax, and that hierarchical rather than linear distance between elements is central to syntactic computation ( Berwick et al., 2011 ; Moro, 2013 ).

Lastly, the recursive mechanism is typified by three properties (see, for example, Watumull et al., 2014 ): computability, definition by induction, and mathematical induction. Note that recursion characterizes the mechanism, not patterns (e.g., embedding) in its outputs. Computability is reflected in a procedure that generates new and complex representations by combining and manipulating discrete symbols. The computable function must be defined by a form of induction: outputs must be carried forward and returned as inputs to generate a hierarchical structure over which can be defined complex relations (e.g., syntactic, semantic, phonological, etc.). Finally, mathematical induction is realized in the jump from finite to infinite, as in the projection from a finite set of words to an infinite set of sentences. Thus, the recursive mechanisms generate an infinite set of hierarchically structured expressions that yield interpretations at the interfaces.

The structural constraints of language, observed at all levels of linguistic representation and derived from a few familiar languages, have proven remarkably successful in the description and explanation of linguistic diversity ( Chomsky, 1981 ; Bresnan, 2001 ; Roberts, 2012 ). From these formal systems it is possible to deduce linguistic universals as consequences, thereby generating empirical predictions. For instance, a cross-linguistic survey of 500 languages shows that every language consists of sentences based on a verb phrase surrounded by modifiers in predictable, non-varying patterns ( Cinque, 1999 ). The aboriginal languages of Australia, once believed to be very different from other languages, can be described as a constellation of properties, each identifiable in the more familiar languages ( Hale, 1992 ; Legate, 2001 ). The distribution of the structural properties of language, such as word order and agreement, do not seem to follow any cultural or historical patterns ( Baker, 2001 ). Rather, they exhibit the same limited range of variation, a result that is consistent with a species-specific linguistic capacity.

Language ontogeny provides another important source of evidence for the biological basis of language and its particular phenotype. Language acquisition often reveals a pattern of winnowing, where a child makes use of non-target but biologically possible grammars, ultimately narrowing down to the target grammar. For instance, English-learning children may omit subjects or objects (“tickles me” instead of “he tickles me”): these forms are ungrammatical for English and do not appear in the input data, but are consistent with grammars such as Mandarin Chinese that omit discourse topics ( Yang, 2002 ). These non-target forms are gradually eliminated, in a manner similar to the winnowing process in birdsong acquisition ( Marler, 1997 ); see (Comparative Animal Behavior). This reveals that the child is endowed with a capacity to acquire a wide range of possible grammars, which are then selected by the linguistic data in the specific environment.

Recently, steps have been taken toward the unification of linguistic theory with the genetic, neurobiological, and cognitive underpinnings of language. These studies provide rich accounts of the computations and representations underlying the language phenotype and its acquisition, but with poor understanding of its evolution. An account of language evolution is highly deficient if it cannot account for these specific empirical results.

Comparative Animal Behavior

Talking birds and signing apes rank among the most fantastic claims in the literature on language evolution, but examination of the evidence shows fundamental differences between child language acquisition and nonhuman species' use of language and language-like systems. For instance, dogs can respond to a few hundred words, but only after thousands of hours of training; children acquire words rapidly and spontaneously generalize their usage in a wide ranges of contexts ( Kaminski et al., 2004 ; Pilley and Reid, 2011 ). Similarly, Nim Chimpsky, the chimpanzee that produced the only public corpus of data in all animal language studies, produced signs considerably below the expected degree of combinatorial diversity seen in two-year old children ( Yang, 2013 ), and with no understanding of syntactic structure or semantic interpretation. Though these studies are of potential interest to understanding the acquisition of specialized, artificial skills—akin to our learning a computer language—they do not inform understanding of language evolution. Hence, we focus on two potentially more promising lines of empirical inquiry: (i) observations and experiments of naturally communicating animals and (ii) experiments assessing the computational and perceptual capacities of animals, focusing on abilities necessary for human language processing.

Researchers claim that songbirds and nonhuman primates exhibit features of communication that parallel human linguistic communication. Like human infants, some songbirds acquire their species-typical song, constrained by an innately-specified template, but requiring particular acoustic input and auditory feedback during a sensitive period of development. Songbirds and babies also progress through a babbling phase on their way to developing the adult form Doupe and Kuhl (1999) , Petkov and Jarvis (2012) , and both learn to string syllables together in an analogous manner prior to full articulation ( Lipkind et al., 2013 ). These observations, generated from behavioral as well as neurobiological evidence, are interesting, but do not guide understanding of language acquisition in humans for at least two reasons: unlike human language, (i) song is a highly specialized and finite system, with the underlying neurobiology linked to one sensory channel (acoustic), and the signal itself is linked to a narrow function and hardly changes once acquired; (ii) when song syllables are combined to create longer structures, there are only limited combinatorial operations and new creations have no impact on the function or “meaning” of the song. Students of child language acquisition thus rarely turn to work on songbirds for insights, except to make the very general point that there are analogous learning processes in early development.

Research on nonhuman primates has focused more on how sounds are produced than how they are acquired because our closest relatives exhibit no parallels (genetically, neurobiologically, and behaviorally) with child language acquisition: there is no vocal learning, no babbling, no sensitive period, no inductive leaps. Nonhuman primates do, however, have a vocal apparatus that is closer to our own species than that of songbirds. Nonhuman primates produce rhythmic articulations, generating sounds with formant structures (resonances that reflect the filtering properties of the vocal tract), to which they are perceptually sensitive ( Fitch, 2006 ; Yip, 2006 ). Together, these observations have led to the conclusion that we share with other primates several key mechanisms for vocal articulation and sound perception, and thus, that the origins of human speech production can be traced to ancestral primates. But there is no evidence that monkeys and apes configure their repertoire on the basis of distinctive features that map to distinct articulatory gestures that, in turn, generate acoustic or visual signals. Further, both neurobiological and behavioral studies show that nonhuman primates have extremely poor voluntary control over the structure of their vocalizations, as shown by studies that have failed to operantly condition monkeys to alter spectral properties of their species-specific calls ( Hauser, 1996 ; Jurgens, 2002 ; Fitch, 2010 ). These points are critical to understanding the nature of phonology and its externalization in speech or sign ( Yip, 2006 ).

Other studies have explored the possibility that nonhuman animals produce vocalizations or gestures that are like our words—that is, symbolic or referential—and with the capacity for combination based on some syntactic principles ( von Frisch, 1967 ; Seyfarth et al., 1980 ; Gould and Towne, 1987 ; Zuberbuhler et al., 1999 ; Manser, 2013 ). The classic work in this area focused on honeybee dances and vervet monkey alarm calls, with other taxonomic groups added over time. The general evidence is that when animals confront particular situations, say a predator, a dominant attacker, or the discovery of highly coveted food, they produce distinctive vocalizations or visual gestures while others respond as if the triggering context were present, causing flight, submission, or movement to food.

The question of interest is whether these seemingly modest claims about animal signals help us understand the evolution of our capacity to represent words, including not only their referentiality but their abstractness, their composition via phonology and morphology, and their syntactic roles. Our simple answer is No, for five specific reasons: for animals, (i) acquisition of the entire lexicon is complete by the end of the early juvenile period, and for most species, the sounds or gestures are innately specified; (ii) those sounds and gestures refer, at best, to directly observable objects or events, with great uncertainty about the precise meaning, and no evidence for signals that map to abstract concepts that are detached from sensory experiences; (iii) with a few rare exceptions, individuals only produce single utterances or gestures, never combining signals to create new meaning based on new structures; (iv) utterances are holistic, with no evidence of complex syntactic composition derived from an inventory of discrete morphological elements; (v) the utterances or gestures are not marked by anything remotely resembling grammatical classes, agreement, etc. Given these differences, it is not possible to empirically support a continuity thesis whereby a nonhuman animal form served as a precursor to the modern human form.

The second approach to exploring possible evolutionary precursors to language focuses on the capacity to process patterned sequences that map to different generative rules that capture some elements of human syntax. The first experiment ( Fitch and Hauser, 2004 ) along these lines was based on the Chomsky (1959) hierarchy of formal grammars, a perspective that reveals how different procedures can generate different levels of structural complexity. This experiment tested cotton-top tamarin monkeys, comparing the two lowest levels of the Chomsky hierarchy, each necessary but insufficient to explain the full richness of linguistic competence. Using a habituation-discrimination method that is common in studies of child language acquisition, results showed that monkeys spontaneously (no training) discriminated the lowest level grammar, but not the next level up. Where they failed suggested that monkeys (at least tamarins) may not be able to spontaneously compute grammars that generate embedded patterns of output, a feature of human syntactic competence. Although embedding is certainly part of our linguistic competence, it is a computation that is far too limited to explain the richness of our syntactic capacity. Consequently, the authors concluded, some of our closest living relatives are far too impoverished, computationally, to provide insights into our own evolutionary history.

Subsequent studies ( Gentner et al., 2006 ; van Heijningen et al., 2009 ; Abe and Watanabe, 2011 ; Rey et al., 2012 ) focused on the problem of embedding, virtually all used methods of extensive training, and all mistakenly equated embedding with both recursion and the claim that any evidence of embedding would rule out earlier claims of human uniqueness. For example, in one study of starlings and one on baboons, subjects were trained for months in tens of thousands of reinforced trials to learn a pattern of embedding that was comparable to that tested on tamarins. Both species learned this pattern, with limited generalization to novel patterns. The researchers concluded that recursive computations are not unique to humans, and so our competence can be explained by non-linguistic processes. For at least four reasons, however, these results do not inform our understanding of human language competence: (i) recursion, as realized in the language faculty, is a set of properties defining the generative procedure, not its output, and so should not be equated with embedding (see section The Language Phenotype); (ii) human language acquisition does not involve training with reinforcement; thus, even if the results showed parallel competences, the acquisition process and underlying computations would be entirely different; (iii) even if animals can process embedded structures, the generalization results show that the capacity is limited to the training level, and pales relative to human competence, especially if one removes some of the working memory constraints; (iv) as the Chomsky hierarchy perspective reveals, embedding is far too weak to explain human language competence, and thus, even strong evidence in animals would contribute little to our understanding of human language evolution.

What would be interesting—as a step toward addressing the definition of recursion sketched in section The Language Phenotype—would be to develop a robust and spontaneous method showing that animals can extract a generative procedure that underpins a pattern of structured inputs, and use this procedure to generalize far beyond the input. We return to this possibility in our final section.

For now, the evidence from comparative animal behavior provides little insight into how our language phenotype evolved. The gap between us and them is simply too great to provide any understanding of evolutionary precursors or the evolutionary processes (e.g., selection) that led to change over time.

Paleontology and Archaeology

Given the phenotypic characterization of language (II), it is no surprise that direct prehistoric traces of language, spoken or signed, are lacking. Consequently, those interested in language origins have often tried to document putative proxies for language in the fossil and archaeological records ( Tattersall, 2012 ; Dediu and Levinson, 2013 ; Johansson, 2013 ). Such proxies have been of two kinds: anatomical (including genomic) and behavioral. Since neither the generation of language nor its lack leaves any identifiable imprint on brain endocasts, proxies of the first sort are limited to the preserved bony structures associated with the production and reception of articulate speech, and to alleles purportedly associated with language. The gestural origins theory of language ( Corballis, 2003 ; Studdert-Kennedy and Goldstein, 2003 ), in which signed expressions predated spoken ones, is on even less stable ground: though the use of the hand in this context was possible far back in hominid evolution, the hand and its motor correlates did not evolve for this purpose, and there are no relevant comparative or fossil findings to illuminate mental representations and computations.

Due to its relative recency and the completeness of its fossil and genomic evidence, we know more about Homo neanderthalensis than most extinct hominids; and because of their evolutionary proximity to humans, Neanderthals provide one of the most intriguing test cases for exploring the antiquity of the language phenotype. The ability of H. neanderthalensis and other extinct hominids to produce the sounds that Homo sapiens uses in speech today has both been denied from the bony structure of the roof of the upper vocal tract, and affirmed from the occasionally preserved anatomy of the hard portion of the hyoid apparatus ( Laitman et al., 1979 ; Lieberman, 1984 ; Arensburg et al., 1989 ; Martinez et al., 2004 ). Other suggestions include that Neanderthals probably spoke because they had the aural ability to process the sound frequencies associated with speech, and that they might have had language because their genome included the modern human variant of the FOXP2 gene (see Molecular Biology), known to play a role in speech articulation, among other things ( Martinez et al., 2004 ; Krause et al., 2007 ; Dediu and Levinson, 2013 ). However, while the modern versions of each of these attributes may be necessary for speech production and comprehension, none can be regarded as a sufficient condition for inferring speech, let alone language ( Tattersall, 2012 ).

Two other observations highlight the limitations of this research. First, acoustic perception is a highly conserved trait within primates, such that chimpanzee hearing is basically identical to ours. In contrast, vocal tract anatomy has changed significantly. Consequently, though other primates can hear what we hear, they can't produce many of our essential articulatory gestures. This demonstrates that perception and production did not coevolve, leaving claims about Neanderthal capacity completely uncertain. Second, recent studies suggest that approximately equal proportions of the horizontal and vertical sectors of the vocal tract are necessary for speech production ( Lieberman, 2011 ). This conformation is present in Homo sapiens alone, as a result of the autapomorphic retraction of its face below the neurocranium. This points to a critical change after divergence from the Neanderthals. Additionally, Neanderthals (and Denisovans) appear to have lacked other alleles (CNTAP2, ASPM, MCPH1, PCDH11YandX) allegedly associated with language (see section Molecular Biology), pointing to significant molecular changes, and presumably, different selective pressures. The fossil evidence thus stands mute on the issue of central language capacity, and ambiguous at best on the question of its externalization in speech.

Another line of paleontological evidence comes from cranial endocast size and shape. Approximately 2 million years ago, following the emergence of the genus Homo , cranial capacity began to expand. Eventually, Neanderthal brain size was larger than ours today. Many have interpreted this consistent expansion as indicative not only of substantive changes in cognitive ability, but of the capacity for language. In particular, given the comparable cranial capacities of Homo neanderthalensis and H. sapiens , many have concluded that both had language. As such, the antiquity of the language phenotype can be traced back at least as far as the Neanderthals ( Dediu and Levinson, 2013 ; Johansson, 2013 ). But as Lenneberg (1967) long ago noted, and as many more recent neurological and genetic studies have affirmed, raw brain size provides little to no insights into the computations and representations of language, either in terms of deficiencies or advantages ( Price et al., 2010 ; Schoeneman, 2012 ). For example, autistics have significant problems in both the acquisition and expression of language, and yet early in development often have larger brains than healthy children. Similarly, children with one hemisphere removed prior to the full acquisition of language often display normal language expression and comprehension. These findings emphasize that a large brain is no predictor at all of language capacity or competence.

Turning to archaeology, the relevant record starts at about 2.5 million years ago—well after the origin of the hominid family—with the deliberate production of stone tools ( Semaw et al., 1997 ). These do not in themselves tell us anything about syntax, semantics, phonology, or their interfaces, as it is abundantly clear that the manufacture of even quite complex stone tools is not necessarily associated with modern cognition. The same applies to other cognitively complex expressions such as the controlled use of fire, the manufacture of compound tools, and even the simple burial of the dead. What does appear to be significant for early cognitive style, however, is the pattern of innovation. Following the initial invention of stone tools, refinements in technology emerged sporadically. There is no hint until very recently of the pattern of continual enhancement typical of modern linguistic Homo sapiens . Although technologies became more complex over the history of the genus Homo ( Tattersall, 2012 ), indications of modern-style iconic and representational activities ( Henshilwood et al., 2002 , 2004 ) begin only significantly after the first anatomically recognizable H. sapiens appears at a little under 200 thousand years ago ( White et al., 2003 ; McDougall et al., 2005 ). Indeed, the sketchy archaeological traces of the earliest Homo sapiens in Africa are remarkably archaic. Further, despite recurrent claims to the contrary, there is no firm evidence for “modern” behaviors on the part of Homo neanderthalensis or any other extinct hominid species ( Klein, 2009 ; Bar-Yosef and Bordes, 2010 ; Higham et al., 2010 ).

Whether or not language is principally an instrument for thought, we have no substantive reason to suspect language use by the hominids that preceded us. In striking contrast to the Cro-Magnon Homo sapiens populations that replaced it in Europe, even the highly encephalized Homo neanderthalensis failed to leave any unequivocal evidence for the symbolic behavior patterns (including painted and engraved imagery, the use of musical instruments and symbolic and notational systems) that characterize modern, linguistic, human beings. Neanderthal material productions represented at best an incremental increase in complexity relative to those of their predecessors. The artifactual record of contemporaneous Middle Stone Age sapiens in Africa after about 100k year ago tells a very different story, a qualitative transformation in behavior that was reflected in the earliest symbolic objects, complex planning, multi-stage technologies, and other anticipations of Cro-Magnon cognitive prowess.

As we know from the acquisition of language by small-brained babies and even individuals with pathologically small brains ( Lenneberg, 1967 ), language is clearly independent of crude brain mass. It is presumably the product of a complex and specific internal wiring, and not simply some slowly-evolved gross by-product of increasing encephalization. Indeed, greater encephalization characterizes several independent lineages within the genus Homo , without substantive archaeological indications of symbolic and putatively linguistic behaviors except in our own case.

In summary, the paleontological evidence is silent with respect to the capacity for both the internal computations and representations of language and its externalization in linguistic expression and communication. As we note in our final section, it is conceivable that methodological advances will enable a more fine-grained understanding of internal neurobiological structure from details of skull structure, but we are nowhere near such discoveries at present. Archaeological evidence, in contrast, points to the emergence of a language of thought in early Homo sapiens , replete with symbolic representations that were externalized in iconic form. We know nothing, however, about when the relevant syntactic and semantic machinery evolved, what selective pressures—if any—were responsible for its emergence, and when such internal computations were externalized in spoken or signed language. Whenever this occurred, present evidence suggests it was after our divergence with Neanderthals, and thus, a very recent event.

Molecular Biology

The comparative method provides an important approach to identifying genetic mechanisms and evolutionary change, but runs into significant challenges in the case of language. To illustrate, consider FOXP2, an autosomal dominant transcription factor that has been linked to the Mendelian disorder verbal dyspraxia SPCH1, and so a gene apparently necessary, but crucially not sufficient, for human speech, let alone language. The human variant of this gene differs from chimpanzees and gorillas in just two amino acid coding positions ( Enard et al., 2002 ), while humans and Neanderthals do not differ at all. However, as noted in our section Paleontology and Archeology, it is not possible to draw firm inferences from the Neanderthal genome since we lack evidence for the relevant behavioral signals. Moreover, it is now unclear whether the putative adaptive evolutionary changes in FOXP2—perhaps related to speech—were in fact centered on these two protein coding regions as opposed to non-coding regions of FOXP2 which are arguably different between humans and Neanderthals ( Maricic et al., 2012 ). Comparative work with birds and mice has provided some insight regarding the functioning of the FOXP2 transcription factor in zebra finch vocal learning ( Haesler et al., 2007 ; Scharff and Petri, 2011 ) as well sound production in mice ( Enard et al., 2009 ), and more recently, neural growth in mice ( Tsui et al., 2013 ). However, these results are still far removed from the computations and representations that underlie the language phenotype.

As predicted by King and Wilson (1975) , the fact that there are remarkably few protein differences between humans and chimpanzees implies that the uniquely distinguishing differences between our species might be better attributed to regulatory changes (both cis and trans) along with other non-protein coding differences [e.g., non-coding RNA, microRNAs, methylation patterning, chromatin epigenetic effects ( Carroll, 2005 ; Somel et al., 2013 )]. The results to date regarding specific human-language related genes have generally confirmed this expectation, as emphasized in a recent, comprehensive review by Geschwind and Rakic (2013 p. 638) noting that there are “only about two dozen genes estimated to be present in human… and not in chimpanzee.”

To be sure, steady progress is being made in building a more complete “parts list” of the genome-implicated changes in the hominid lineage that ultimately resulted in the substrate for human cognitive abilities, including regulatory genomic changes. The catalog of differences has grown to include the possible genomic underpinnings for neoteny (MEF2A, Somel et al., 2013 ); distinct brain architecture and development, such as novel cortical layer architecture and gyri ( Bae et al., 2014 ); novel neuronal cell types, evolutionary duplication of developmental proteins (SRGAP2) resulting in novel dendritic spine density and form ( Dennis et al., 2012 ; Geschwind and Rakic, 2013 ), and so forth. Thus the possible avenues for explanation are expanding, an indication of positive paths forward (e.g., see Boeckx and Benítez-Burraco, 2014 ). Nonetheless, even in the best understood cases, the genotype-phenotype gap remains large. Constructing bridge theories becomes more difficult because we must unravel regulatory networks not directly “wired” to phenotypes. Konopka et al. (2012) have begun unraveling this structure by a functional, modular analysis of the human vs. nonhuman primate transcriptome, but much remains unknown. Further, language is not like the examples of body plan segmentation or eye formation where functional and developmental processes are well understood in numerous species, both closely and distantly related. There are simply no precise analogs or homologs of human language in other species.

Given current limitations, genomic-driven analysis of language has resorted to roughly four research strategies: (i) locating candidate genes based on the genomic signatures of rapid, recent evolutionary change or uniqueness in the Homo lineage as compared to the closest nonhuman primates; (ii) exploiting language-related aberrant phenotypes, generally at the “input-output” interfaces of human language (e.g., speech and hearing), proceeding as in the genomic analysis of complex human pathologies; (iii) using twin studies and/or SNP analysis to search for genes correlated with normal human variation; (iv) developing nonhuman animal models, as in the work with mice, while setting aside the absence of a full nonhuman language phenotype to examine plausible subcomponents such as transcriptional neural plasticity and motor control.

The FOXP2 case exemplifies research strategies (i), (ii), and (iv). FOXP2 is a transcription factor that up- or down-regulates DNA in many different tissue types (brain, lung, gut lining) at different times during development as well as throughout life. This broad functional effect makes evolutionary analysis difficult. In particular, the exact mechanisms by which FOXP2 mutations disrupt speech remain uncertain, variously posited as disruptions in motor articulation/serialization in speech, vocal learning generally, or broader difficulties with procedural serialization. This is critical because FOXP2 mutations may disrupt only the input/output systems of language, sparing the more internal computations of human language syntax or semantics; or it may be that FOXP2 affects general cognitive processing, such as general serial ordering of procedures. Second, it is not clear whether the amino acid changes distinguishing FOXP2 in humans and nonhumans represent adaptations “for” language, since their functional effects remain unclear. One of the two protein-coding changes along the lineage to modern humans is also associated with the order Carnivora. Since FOXP2 also targets the gut lining, this evolutionary step may have had little to do directly with language but instead with digestion modifications driven by forest-to-savannah habit and so dietary change ( Zhang et al., 2002 ), as in the well-established case of lactose tolerance and dairy culture ( Bersaglieri et al., 2004 ).

FOXP2's many downstream regulatory targets have also been associated with other language disorders; for example, CNTNAP2, a neurexin-family neural growth factor, appears linked to autism and specific language impairment (SLI) ( Vernes and Fisher, 2011 ; Vernes et al., 2011 ). Note that CNTNAP2 itself is also regulatory, in line with the King and Wilson prediction, but unlike FOXP2, is different in humans and Neanderthals. Moreover, its link to SLI is not clear-cut. Rice (2012) presents evidence that SLI might best be pictured as a disruption of a growth timing mechanism, implicating a different set of genetic components, some that are non-regulatory and involved in neuronal migration. The story is not yet complete; as Geschwind and Rakic (2013) note, comparison of human vs. mouse FOXP2 by Tsui et al. (2013) points to a role beyond neural circuit construction, to neural cell proliferation itself.

Together, these observations underscore the fact that we lack a connect-the-dots account of any gene to language phenotype. Furthermore, to the extent that our account of the language phenotype is diffuse (some general system of cognition rather than a precisely delimited and narrow computational module), the genotype-phenotype mapping will be correspondingly more challenging to address. Given our currently impoverished understanding of such mappings for far less complicated phenotypes, in far simpler organisms, molecular biology has a long way to go before it can illuminate the evolution of language.

Biological models of language evolution often start with a population of individuals communicating by means of their particular languages, broadly defined as mappings between forms and meanings. A certain measure of fitness is introduced, which in turn differentially affects the transmission of the languages to the next generation of individuals. This evolutionary dynamic is believed to shed light on the emergence of human language and its associated properties. A notable result ( Nowak et al., 2001 ), congruent with the comparative linguistics and mathematical learning theory, suggests that if the reliability and efficiency of language learning is the fitness metric, the space of possible languages must be limited in its size and complexity for language to emerge.

The vast majority of modeling efforts, like those above, presuppose the existence of a language phenotype equipped with compositionality and discrete infinity. This assumption is directly built into the mathematical models ( Nowak and Komarova, 2001 ; Kirby and Hurford, 2002 ) or enabled by human subjects in behavioral studies, who may impose linguistic structures upon the materials presented ( Kirby et al., 2008 ). But this presupposition regarding the language phenotype offers no insight into how it arose in the first place, nor does it illuminate the fundamental distinction between the emergence of the core biological competence and its adaptive or non-adaptive functions. Lastly, the underlying assumptions of these models, including their commitment to an adaptationist program, are often made without empirical verification and in some cases, are contrary to known facts about languages. As noted in our section How to Study the Evolution of a Trait, it is essential for proposals of adaptive function to be tested against non-adaptive hypotheses.

A leading proposal in evolutionary modeling is to identify language fitness with communicative success ( Nowak and Komarova, 2001 ; Baronchelli et al., 2012 ): individuals who communicate with and learn languages more successfully have greater reproductive success. While our pre-linguistic past is not accessible for direct investigation, the uniformitarian principle of historical science does enable us to test these assumptions: If communication has played a significant role in the evolution of language, its force should be observable in the process of language transmission. The history of language change provides the only testable case for the predictions of this communication optimization thesis, and the evidence points in the opposite direction.

Language change generally proceeds mechanically irrespective of communicative purpose, a perspective held by traditional historical linguistics and strengthened by the quantitative study of ongoing language variation and change ( Labov, 1994 ). For instance, one of the most robustly attested linguistic changes is phonemic merger, whereby the distinction between two consonants or vowels is lost. In many dialectal regions of North American English, the vowels in “cot” and “caught” are the same, while the distinction is retained in other regions. Mergers, by definition, obliterate the distinction between words, which increases the ambiguity of communication; unsurprisingly, therefore, information theoretic accounts of phonemic change have been unsuccessful ( King, 1967 ; Surendran and Niyogi, 2006 ). Yet mergers can spread rapidly across dialect boundaries and are rarely, if ever, reversed ( Labov, 1994 ). Another major difficulty with the communication as adaptation thesis can be observed in the redundancy and reduction of linguistic information. For example, the word final consonants /t/ and /d/ in English are sometimes omitted in speech so “walked” is pronounced as “walk.” In a sentence such as “I have walked home”, the perfective meaning is doubly expressed by the auxiliary “have” and “by” the verb. In the simple past “I walked home”, only the final “d” on the verb conveys the temporal information. A communication-based approach to language use would predict a higher rate of deletion for the past participle than for past tense on grounds of communicative efficiency, yet the deletion rates do not differ in these contexts ( Guy, 1991 ).

Under the adaptationist assumption in language evolution modeling, languages that facilitate more efficient communication are more successful in transmission to the next generation. But there is no evidence of a communicative advantage for typologically more common, and thus more successfully transmitted languages such as those with the Subject Verb Object order (e.g., English) over those with the rare word order of Verb Object Subject (e.g., Malagasy). Likewise, the cross-linguistic studies of language acquisition show a largely uniform developmental trajectory, with no evidence to suggest that some languages are easier to learn than others ( Slobin, 1985 ). While the differences between individuals' language learning abilities may have a genetic basis, there is currently no evidence to support a higher biological fitness for the more proficient learners, except in extreme cases of neurological impairment. That communication is intimately related to language is too obvious to dismiss entirely, but its lack of theoretical constraint and repeated failures, long recognized in the empirical study of language, cast serious doubt on its utility in models of language evolution.

The success of evolutionary population genetics lies in the mutually constraining connection between idealized models and empirically grounded work in the laboratory and in the wild. As noted in section How to Study the Evolution of a Trait, many tools and methodologies available to biological investigations cannot be applied to the study of language. But conceptual confusion and detachment from the empirical research of language, as we have seen in the modeling work on language evolution, unsurprisingly provides scant insights into language origins or its subsequent evolution. Moving forward, modeling work must focus on the computations and representations of the core competence for language, recognize the distinction between these internal processes and their potential externalization in communication, and lay out models that can be empirically tested in our own and other species. This is a tall order, but a necessary one if the fruits of evolutionary modeling that have been reaped from studies of, for example, mating behavior and cooperation, can be obtained for the language phenotype.

Answering evolutionary questions is of profound interest largely because of our deep-seated curiosity about the past, about how things were, and how they have become what they are. Thanks in part to the revolution that Darwin sparked, including his ideas and methods, we now have many fine examples in which theoretical predictions about the origins and subsequent evolution of a phenotype have been described in great detail, including analyses of genomes, anatomy, and behavior. And yet some phenotypes remain poorly understood, and may remain so due to inadequate methods and impoverished evidence.

The evolution of our language phenotype may remain stubbornly resistant to empirical inquiry, and yet, as indicated in Table 1 , there are potential empirical prospects, some near term, others quite remote. We conclude with a brief discussion of potential paths forward.

Table 1. Some prospects for future empirical work on language evolution .

Animal communication systems have thus far failed to demonstrate anything remotely like our systems of phonology, semantics, and syntax, and the capacity to process even artificially created stimuli is highly limited, often requiring Herculean training efforts. Should new methods reveal more richly structured systems of communication or more powerful, spontaneous abilities to process strongly generated stimuli, then comparative data would gain greater interest and relevance to evolutionary understanding. For example, we can imagine that in the not so distant future, it will be possible to non-invasively obtain neural recordings from free-ranging animals, and thus, to provide a more fine grained and quantitative measure of spontaneous processing of different stimuli. This would solve the methodological desiderata of creating a technique that reveals a capacity in the absence of reinforced training. With this tool, future work on artificial language processing might develop a set of stimuli that are generated from a recursive operation such as Merge (a recursive operation that combines two objects, such as two lexical items, to construct a new object, such as a phrase, in a process that can be iterated indefinitely), expose animals to a subset of these, and then test them on a wide range of alternatives that extend beyond the initial set in ways that can reveal substantial generalization, and thus comprehension of the underlying generative operation. As in all such studies, it would be necessary to show that simpler, finite mechanisms, cannot account for the patterns of generalization.

With respect to paleontology, it is difficult to imagine how any kind of fossil evidence could shed light on the computations and representations of language: as noted, peripheral anatomy without soft tissue says little about either the output or the phonological representations, and endocasts say even less about potential computations and representations. Nonetheless, it is not inconceivable that finer-grained analyses of endocasts from modern humans might be linked to more fine-grained neurobiological structures at the surface, and that these in turn might reveal details of the internal circuitry. That said, it is important to note here that our current understanding of how neurobiological systems link to even “language-like” communication in animals is, at best primitive, and is absent when it comes to the core competences of language in humans. For example, despite the relative simplicity of the honeybee's brain, we know nothing about how neurons encode the perception of the waggle dance, or how neurons generate motor sequences for dancing. For our own species, we know nothing about the neurobiology of our recursive procedures, and even for such seemingly simpler systems, such as phonology, our understanding is very poor ( Poeppel, 2012 ). Needless to say, this makes comparative work virtually impossible as the target circuitry for modern humans is unclear. As advances in neuroimaging and other cellular techniques improve, so too perhaps will understanding.

In terms of the archaeological record, we can certainly imagine the discovery of richer symbolic artifacts, perhaps even non-iconic strings of symbols, dating before the emergence of Homo sapiens . Such findings would push back the origins of symbolic capacities, and provide greater traction into questions of both origin and subsequent evolution.

Should such discoveries from comparative animal behavior, paleontology, neurobiology, and archaeology be made, along with greater depth of understanding of gene-phenotype mapping, it would open the door to more relevant genomics and modeling. These are all big IFs about the nature and possibility of future evidence. Until such evidence is brought forward, understanding of language evolution will remain one of the great mysteries of our species.

Author Contributions

All authors contributed to the writing of this review.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Abe, K., and Watanabe, D. (2011). Songbirds possess the spontaneous ability to discriminate syntactic rules. Nat. Neurosci . 14, 1067–1074. doi: 10.1038/nn.2869

Pubmed Abstract | Pubmed Full Text | CrossRef Full Text

Arensburg, B., Tillier, A. M., Vandermeersch, B., Duday, H., Schepartz, L. A., and Rak, Y. (1989). A middle paleolithic human hyoid bone. Nature 338, 758–760. doi: 10.1038/338758a0

CrossRef Full Text

Bae, B., Tietjen, I., Atabay, K. D., Evrony, G. D., Johnson, M. B., Asare, E., et al. (2014). Evolutionarily dynamic alternative splicing of GPR56 regulates regional cerebral cortical patterning. Science 343, 764–768. doi: 10.1126/science.1244392

Baker, M. (2001). Atoms of Language . New York, NY: Basic Books.

Baronchelli, A., Chater, N., Pasto-Satorras, R., and Christiansen, M. H. (2012). The biological origins of linguistic diversity. PLoS ONE 7:e48029. doi: 10.1371/journal.pone.0048029

Bar-Yosef, O., and Bordes, J.-G. (2010). Who were the makers of the Chatelperronian culture? J. Hum. Evol . 59, 586. doi: 10.1016/j.jhevol.2010.06.009

Bersaglieri, T., Sabeti, P. S., Patterson, N., Vanderploeg, T., Schaffner, S. F., Drake, D. A., et al. (2004). Genetic signatures of strong recent positive selection at the lactase gene. Am. J. Hum. Genet . 74, 1111–1120. doi: 10.1086/421051

Berwick, R., Pietroski, P., Yankama, B., and Chomsky, N. (2011). The poverty of the stimulus revisited. Cogn. Sci . 35, 1207. doi: 10.1111/j.1551-6709.2011.01189.x

Bickerton, D. (1990). Species and Language . Chicago, IL: Chicago University Press.

Boeckx, C., and Benítez-Burraco, A. (2014). The shape of the human language-ready brain. Front. Psychol . 5:282. doi: 10.3389/fpsyg.2014.00282

Bresnan, J. (2001). Lexical-Functional Syntax . Oxford: Blackwell.

Carroll, S. B. (2005). Evolution at two levels: on genes and form. PLoS Biol . 3:e245. doi: 10.1371/journal.pbio.0030245

Chomsky, N. (1959). On certain formal properties of grammars. Inf. Control 2, 137–167. doi: 10.1016/S0019-9958(59)90362-6

Chomsky, N. (1981). Lectures on Government and Binding . New York, NY: Foris.

Cinque, G. (1999). Adverbs and Functional Heads: a Cross-Linguistic Perspective . Oxford: Oxford University Press.

Corballis, M. (2003). “From hand to mouth: the gestural origins of language,” in Language Evolution , eds M. H. Christiansen and S. Kirby (Oxford: Oxford University Press), 201–218.

Dediu, D., and Levinson, S. C. (2013). On the antiquity of language; the reinterpretation of neandertal linguistic capacities and its consequences. Front. Lang. Sci . 4:397. doi: 10.3389/fpsyg.2013.00397

Dennis, M. Y., Nuttle, X., Sudmant, P. H., Antonacci, F., Graves, T. A., Nefedov, M., et al. (2012). Evolution of human-specific neural SRGAP2 genes by incomplete segmental duplication. Cell 149, 919–922. doi: 10.1016/j.cell.2012.03.033

Doupe, A., and Kuhl, P. (1999). Birdsong and human speech: common themes and mechanisms. Ann. Rev. Neurosci . 22, 567–631. doi: 10.1146/annurev.neuro.22.1.567

Enard, W., Khaitovich, P., Klose, J., Zollner, S., Heissig, F., Giavalisco, P., et al. (2002). Intra- and interspecific variation in primate gene expression patterns. Science 296, 340–343. doi: 10.1126/science.1068996

Enard, W., Gehre, S., Hammerschmidt, K., Hölter, S., Blass, T., Somel, M., et al. (2009). A humanized version of Foxp2 affects cortico-basal ganglia circuits in mice. Cell 137, 961–971. doi: 10.1016/j.cell.2009.03.041

Fitch, W. T. (2006). “Production of vocalizations in mammals,” in Encyclopedia of Language and Linguistics , ed K. Brown (Oxford: Elsevier), 115–121.

Fitch, W. T. (2010). The Evoluiton of Language . Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511817779

Fitch, W. T., and Hauser, M. D. (2004). Computational constraints on syntactic processing in a nonhuman primate. Science 303, 377–380. doi: 10.1126/science.1089401

Gentner, T. Q., Fenn, K. M., Margoliash, D., and Nusbaum, H. C. (2006). Recursive syntactic pattern learning by songbirds. Nature 440, 1204–1207. doi: 10.1038/nature04675

Geschwind, D. H., and Rakic, P. (2013). Cortial evolution: judge the brain by its cover. Neuron 80, 633–647. doi: 10.1016/j.neuron.2013.10.045

Gould, J. L., and Towne, W. F. (1987). Evolution of the dance language. Am. Nat . 130, 317–338. doi: 10.1086/284713

Guy, G. R. (1991). Explanation in variable phonology: an exponential model of morphological constraints. Lang. Variat. Change 3, 1–22. doi: 10.1017/S0954394500000429

Haesler, S., Rochefort, C., Georgi, B., Licznerski, P., Osten, P., and Scharff, C. (2007). Incomplete and inaccurate vocal imitation after knockdown of FoxP2 in songbird basal ganglia nucleus Area X. PLoS Biol . 5:e321. doi: 10.1371/journal.pbio.0050321

Hale, K. (1992). “Basic word order in two ‘Free word order’ languages,” in Pragmatics of Word Order Flexibility , ed D. L. Payne (Amsterdam: John Benjamins), 63. doi: 10.1075/tsl.22.03hal

Hauser, M. D. (1996). The Evolution of Communication . Cambridge: MIT Press.

Hauser, M. D., Chomsky, N., and Fitch, W. T. (2002). The faculty of language: what is it, who has it, and how did it evolve? Science 298, 1569–1570. doi: 10.1126/science.298.5598.1569

Henshilwood, C., d'Errico, F., Vanhaeren, M., van Niekerk, K., and Jacobs, Z. (2004). Middle stone age shell beads from South Africa. Science 304, 404. doi: 10.1126/science.1095905

Henshilwood, C., d'Errico, F., Yates, R., and Jacobs, Z. (2002). Emergence of modern human behavior: middle stone age engravings from South Africa. Science 295, 1278–1280. doi: 10.1126/science.1067575

Higham, T., Jacobi, R., Julien, M., David, F., Basell, L., Wood, R., and Ramsey, C. B. (2010). Chronology of the grotte du renne (France) and implications for the context of ornaments and human remains within the chatelperronian. Proc. Nat. Acad. Sci. U.S.A . 107, 20234–20239. doi: 10.1073/pnas.1007963107

Hurford, J. (2011). The Origins of Grammar: Language in the Light of Evolution . Oxford: Oxford University Press.

Jackendoff, R. (1999). Possible stages in the evolution of the language capacity. Trends Cogn. Sci . 3, 272–279. doi: 10.1016/S1364-6613(99)01333-9

Johansson, S. (2013). The talking Neandertals: what do fossils, genetics, and archaeology say? Biolinguistics 7, 35–74.

Jurgens, U. (2002). Neural pathways underlying vocal control. Neurosci. Biobehav. Rev . 26, 235–258. doi: 10.1016/S0149-7634(01)00068-9

Kaminski, J., Call, J., and Fischer, J. (2004). Word learning in a domestic dog: evidence for “fast mapping.” Science 304, 1682–1683. doi: 10.1126/science.1097859

King, M., and Wilson, A. (1975). Evolution at two levels in humans and chimpanzees. Science 188, 107–116. doi: 10.1126/science.1090005

King, R. D. (1967). Functional load and language change. Language 43, 831. doi: 10.2307/411969

Kirby, S., Cornish, H., and Smith, K. (2008). Cumulative cultural evolution in the laboratory: an experimental approach to the origins of structure in human language. Proc. Natl. Acad. Sci . 105, 10681–10686. doi: 10.1073/pnas.0707835105

Kirby, S., and Hurford, J. (2002). “The emergence of linguistic structure: an overview of the iterated learning model,” in Simulating the Evolution of Language , eds A. Cangelosi and D. Parisi (London: Springer Verlag), 121–148.

Klein, R. (2009). The Human Career . Chicago, IL: Chicago University Press. doi: 10.7208/chicago/9780226027524.001.0001

Konopka, G., Friedrich, T., Davis-Turak, J., Winden, K., Oldham, M. C., Gao, F., et al. (2012). Human-specific transcriptional networks in the brain. Neuron 75, 601–617. doi: 10.1016/j.neuron.2012.05.034

Krause, J., Lalueza-Fox, C., Orlando, L., Enard, W., Green, R. E., Burbano, H. A., et al. (2007). The derived FOXP2 variant of modern humans was shared with Neanderthals. Curr. Biol . 17, 1908–1912. doi: 10.1016/j.cub.2007.10.008

Labov, W. (1994). Principles of Linguistic Change: Internal Factors . New York, NY: Blackwell.

Laitman, R. C., Heimbuch, E. S., and Crelin, E. S. (1979). The basicranium of fossil hominids as an indicator of their upper respiratory systems. Am. J. Phys. Anth . 51, 15–34. doi: 10.1002/ajpa.1330510103

Legate, J. (2001). Walpiri: Theoretical Implications . Ph.D thesis, Cambridge: MIT.

Lenneberg, E. H. (1967). Biological Foundations of Language . New York, NY: Wiley.

Lieberman, D. E. (2011). The Evolution of the Human Head . Cambridge: Harvard University Press.

Lieberman, P. (1984). The Biology and Evolution of Language . Cambridge: Harvard University Press.

Lipkind, D., Marcus, G., Bemis, D. K., Sasahara, K., Jacoby, N., Takahasi, M., et al. (2013). Stepwise acquisition of vocal combinatorial capacity in songbirds and human infants. Nature 498, 104–108. doi: 10.1038/nature12173

Manser, M. B. (2013). Semantic communication in vervet monkeys and other animals. Anim. Behav . 86, 491–496. doi: 10.1016/j.anbehav.2013.07.006

Maricic, T., Gunther, V., Goergiev, O., Gehre, S., Curlin, M., Schreiweis, C., et al. (2012). A recent evolutionary change affects a regulatory element in the human FOXP2 gene. Mol. Biol. Evol . 30, 844–852. doi: 10.1093/molbev/mss271

Marler, P. (1997). Three models of song learning: evidence from behavior. J. Neurobiol . 33, 1–16. doi:10.1002/(SICI)1097-4695(19971105)33:5<501::AID-NEU2>3.0.CO;2-8

Martinez, I., Rosa, M., Arsuaga, J.-L., Jarabo, P., Quam, R., Lorenzo, C., et al. (2004). Auditory capacities in middle pleistocene humans from the sierra de atapuerca in spain. Proc. Nat. Acad. Sci. U.S.A . 101, 9976–9981. doi: 10.1073/pnas.0403595101

McDougall, I., Brown, F. H., and Fleagle, J. G. (2005). Stratigraphic placement and age of modern humans from kibish, ethiopia. Nature 433, 733. doi: 10.1038/nature03258

Moro, A. (2013). On the similarity between syntax and actions. Trends Cogn. Sci . 18, 109–110. doi: 10.1016/j.tics.2013.11.006

Nowak, M. A., and Komarova, N. L. (2001). Towards an evolutionary theory of language. Trends Cogn. Sci . 5, 288–295. doi: 10.1016/S1364-6613(00)01683-1

Nowak, M. A., Komarova, N. L., and Niyogi, P. (2001). Evolution of universal grammar. Science 291, 114–118. doi: 10.1126/science.291.5501.114

Petkov, C. I., and Jarvis, E. D. (2012). Birds, primates, and spoken language origins: behavioral phenotypes and neurobiological substrates. Front. Evol. Neurosci . 4:12. doi: 10.3389/fnevo.2012.00012

Pilley, J. W., and Reid, A. K. (2011). Border collie comprehends object names as verbal referents. Behav. Processes 86, 184–195. doi: 10.1016/j.beproc.2010.11.007

Pinker, S., and Bloom, P. (1990). Natural language and natural selection. Behav. Brain Sci . 13, 707–784. doi: 10.1017/S0140525X00081061

Poeppel, D. (2012). The maps problem and the mapping problem: two challenges for a cognitive neuroscience of speech and language. Cogn. Neuropsychol . 29, 34–55. doi: 10.1080/02643294.2012.710600

Price, C. J., Seghier, M. L., and Leff, A. P. (2010). Predicting language outcome and recovery after stroke: the PLORAS system. Nat. Rev. Neurol . 6, 202–210. doi: 10.1038/nrneurol.2010.15

Rey, A., Perruche, P., and Fagot, J. (2012). Centre-embedded structures are a by-product of associatve learning and working memory constraints: evidence from baboons ( Papio papio ). Cognition 123, 180–184. doi: 10.1016/j.cognition.2011.12.005

Rice, M. L. (2012). Toward epigenetic and gene regulation models of specific language impairment: looking for links among growth, genes, and impairments. J. Neurodev. Disord . 4, 1–14. doi: 10.1186/1866-1955-4-27.

Roberts, I. G. (2012). “Macroparameters and minimalism: a programme for comparative research,” in Parameter Theory and Linguistic Change , eds C.Galves, S. Cyrino, R. Lopes, F. Sandalo, and J. Avelar (Oxford: Oxford University Press), 320–335. doi: 10.1093/acprof:oso/9780199659203.003.0017

Ryan, M. J. (1985). The Túngara Frog, A Study in Sexual Selection and Communication . Chicago: University of Chicago Press.

Ryan, M. J., Fox, J. H., Wilczynski, W., and Rand, A. S. (1990). Sexual selection for sensory exploitation in the frog, physalaemus pustulosus. Nature 343, 66–67. doi: 10.1038/343066a0

Ryan, M. J., and Rand, A. S. (2003). Mate recognition in tungara frogs: a review of some studies of brain, behavior, and evolution. Acta Zool. Sinica 49:713.

Scharff, C., and Petri, J. (2011). Evo-devo, deep homology and FoxP2: implications for the evolution of speech and language. Phil. Trans. R. Soc. Lond. B Biol. Sci . 366, 2124–2140. doi: 10.1098/rstb.2011.0001

Schoeneman, P. T. (2012). Evolution of brain and language. Prog. Brain Res . 195, 443–459. doi: 10.1016/B978-0-444-53860-4.00022-2

Semaw, S., Renne, P., Harris, J. W. K., Feibel, C. S., Bernor, R. L., Fesseha, N., et al. (1997). 2.5 million-year-old stone tools from Gona, Ethiopia. Nature 385, 333–336. doi: 10.1038/385333a0

Seyfarth, R. M., Cheney, D. L., and Marler, P. (1980). Monkey responses to three different alarm calls: evidence of predator classification and semantic communication. Science 210, 801–803. doi: 10.1126/science.7433999

Slobin, D. I. (1985). The Crosslinguistic Study of Language Acquisition . Hillsdale, NJ: Lawrence Erlbaum Associates.

Somel, M., Liu, X., and Khaitovich, P. (2013). Human brain evolution: transcripts, metabolites and their regulators. Nat. Rev. Neurosci . 14, 112–127. doi: 10.1038/nrn3372

Studdert-Kennedy, M., and Goldstein, L. (2003). “Launching language: the gestural origin of discrete infinity,” in Language Evolution , eds M. Christiansen and S. Kirby (Oxford: Oxford University Press), 235–254.

Surendran, D., and Niyogi, P. (2006). “Quantifying the functional load of phonemic oppositions, distinctive features, and suprasegmentals,” in Competing Models of Linguistic Change: Evolution and Beyond , ed O. N. Thomsen (Amsterdam; Philadelphia, PA: John Benjamins), 43–58.

Tattersall, I. (2012). Masters of the Planet . New York, NY: Palgrave Macmillan.

Tsui, D., Vessey, J. P., Tomita, H., Kaplan, D. R., and Miller, F. D. (2013). FoxP2 regulates neurogenesis during embryonic cortical development. J. Neurosci . 33, 244–258. doi: 10.1523/JNEUROSCI.1665-12.2013

van Heijningen, C. A. A., de Visser, J., Zuidema, W., and ten Cate, C. (2009). Simple rules can explain discrimination of putative recursive syntactic structures by a songbird species. Proc. Nat. Acad. Sci. U.S.A . 106, 20538–20543. doi: 10.1073/pnas.0908113106

Vernes, S. C., Oliver, P. L., Spiteri, E., Lockstone, H. E., Puliyadi, R., Taylor, J. M., et al. (2011). Foxp2 regulates gene networks implicated in neurite outgrowth in the developing brain. PLoS Genet . 7:e1002145. doi: 10.1371/journal.pgen.1002145

Vernes, S., and Fisher, S. (2011). “Functional genomic dissection of speech and language disorders,” in Genomics, Proteomics, and the Nervous System , ed J. D. Clelland (New York, NY: Springer), 253–278. doi: 10.1007/978-1-4419-7197-5_10

von Frisch, K. (1967). The Dance Language and Orientation of Bees . Cambridge, MA: Belknap Press of Harvard University Press.

Watumull, J., Hauser, M. D., Roberts, I. G., and Hornstein, N. (2014). On recursion. Front. Psych . 4:1017. doi: 10.3389/fpsyg.2013.01017

White, T. D., Asfaw, B., DeGusta, D., Gilbert, H., Richards, G. D., Suwa, G., et al. (2003). Pleistocene Homo sapiens from Middle Awash, Ethiopia. Nature 423, 742. doi: 10.1038/nature01669

Wilczynski, W., Rand, A. S., and Ryan, M. J. (2001). Evolution of calls and auditory tuning in the Physalaemus pustulosus species group. Brain Behav. Evol . 58, 137–151. doi: 10.1159/000047268

Yang, C. (2002). Knowledge and Learning in Natural Language . New York, NY: Oxford University Press.

Yang, C. (2013). Ontogeny and phylogeny of language. Proc. Nat. Acad. Sci. U.S.A . 10, 6324–6327. doi: 10.1073/pnas.1216803110

Yip, M. (2006). The search for phonology in other species. Trends Cogn. Sci . 10, 442–446. doi: 10.1016/j.tics.2006.08.001

Zhang, J., Zhang, Y. P., and Rosenberg, H. F. (2002). Adaptive evolution of a duplicated pancreatic ribonuclease gene in a leaf-eating monkey. Nat. Genet . 30, 411–415. doi: 10.1038/ng852

Zuberbuhler, K., Cheney, D. L., and Seyfarth, R. M. (1999). Conceptual semantics in a nonhuman primate. J. Comp. Psych . 113, 33–42. doi: 10.1037/0735-7036.113.1.33

Keywords: language evolution, computation, animal behavior, genetics, paleoarcheology, modelling

Citation: Hauser MD, Yang C, Berwick RC, Tattersall I, Ryan MJ, Watumull J, Chomsky N and Lewontin RC (2014) The mystery of language evolution. Front. Psychol . 5 :401. doi: 10.3389/fpsyg.2014.00401

Received: 03 March 2014; Accepted: 16 April 2014; Published online: 07 May 2014.

Reviewed by:

Copyright © 2014 Hauser, Yang, Berwick, Tattersall, Ryan, Watumull, Chomsky and Lewontin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Marc D. Hauser, Risk-Eraser, LLC, PO Box 376, 410 West Falmouth Hwy #376, MA 02574, USA e-mail: [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Where Did Language Come From? (Theories)

Theories on the Origin and Evolution of Language

• An Introduction to Punctuation
• Ph.D., Rhetoric and English, University of Georgia
• M.A., Modern English and American Literature, University of Leicester
• B.A., English, State University of New York

The expression language origins refers to theories pertaining to the emergence and development of language in human societies.

Over the centuries, many theories have been put forward—and almost all of them have been challenged, discounted, and ridiculed. (See Where Does Language Come From? ) In 1866, the Linguistic Society of Paris banned any discussion of the topic: "The Society will accept no communication concerning either the origin of language or the creation of a universal language ." Contemporary linguist Robbins Burling says that "anyone who has read widely in the literature on language origins cannot escape a sneaking sympathy with the Paris linguists. Reams of nonsense have been written about the subject" ( The Talking Ape , 2005).

In recent decades, however, scholars from such diverse fields as genetics, anthropology, and cognitive science have been engaged, as Christine Kenneally says, in "a cross-discipline, multidimensional treasure hunt" to find out how language began. It is, she says, "the hardest problem in science today" ( The First Word , 2007).

Observations on the Origins of Language

" Divine origin [is the] conjecture that human language originated as a gift from God. No scholar takes this idea seriously today."

(R.L. Trask, A Student's Dictionary of Language and Linguistics , 1997; rpt. Routledge, 2014)

"Numerous and varied explanations have been put forth to explain how humans acquired language—many of which date back to the time of the Paris ban. Some of the more fanciful explanations have been given nicknames , mainly to the effect of dismissal by ridicule. The scenario by which language evolved in humans to assist the coordination of working together (as on the pre-historic equivalent of a loading dock) has been nicknamed the 'yo-heave-ho' model. There's the 'bow-wow' model in which language originated as imitations of animal cries. In the 'poo-poo' model, language started from emotional interjections .

"During the twentieth century, and particularly its last few decades, discussion of language origins has become respectable and even fashionable. One major problem remains, however; most models about language origins do not readily lend themselves to the formation of testable hypotheses, or rigorous testing of any sort. What data will allow us to conclude that one model or another best explains how language arose?"

(Norman A. Johnson, Darwinian Detectives: Revealing the Natural History of Genes and Genomes . Oxford University Press, 2007)

Physical Adaptations

- "Instead of looking at types of sounds as the source of human speech, we can look at the types of physical features humans possess, especially those that are distinct from other creatures, which may have been able to support speech production. . . .

"Human teeth are upright, not slanting outwards like those of apes, and they are roughly even in height. Such characteristics are . . . very helpful in making sounds such as f or v . Human lips have much more intricate muscle lacing than is found in other primates and their resulting flexibility certainly helps in making sounds like p , b , and m . In fact, the b and m sounds are the most widely attested in the vocalizations made by human infants during their first year, no matter which language their parents are using."

(George Yule, The Study of Language , 5th ed. Cambridge University Press, 2014)

-  "In the evolution of the human vocal tract since the split with other apes, the adult larynx descended to its lower position. Phonetician Philip Lieberman has persuasively argued that the ultimate cause of the human lowered larynx is its function in producing different vowels . This is a case of natural selection for more effective communication. . . .

"Babies are born with their larynxes in a high position, like monkeys. This is functional, as there is a reduced risk of choking, and babies are not yet talking. . . . By about the end of the first year, the human larynx descends to its near-adult lowered position. This is a case of ontogeny recapitulating phylogeny, the growth of the individual reflecting the evolution of the species."

(James R. Hurford, The Origins of Language . Oxford University Press, 2014)

From Words to Syntax

"Language-ready modern children learn vocabulary voraciously before they begin to make grammatical utterances several words long. So we presume that in the origins of language a one-word stage preceded our remote ancestors' first steps into grammar . The term 'protolanguage' has been widely used to describe this one-word stage, where there is vocabulary but no grammar."

The Gesture Theory of Language Origin

- "Speculation about how languages originate and evolve has had an important place in the history of ideas, and it has been intimately linked to questions about the nature of the signed languages of the deaf and human gestural behavior in general. It can be argued, from a phylogenetic perspective, the origin of human sign languages is coincident with the origin of human languages; sign languages, that is, are likely to have been the first true languages. This is not a new perspective--it is perhaps as old as nonreligious speculation about the way human language may have begun."

(David F. Armstrong and Sherman E. Wilcox, The Gestural Origin of Language . Oxford University Press, 2007)

- "[A]n analysis of the physical structure of visible gesture provides insights into the origins of syntax , perhaps the most difficult question facing students of the origin and evolution of language . . .. It is the origin of syntax that transforms naming into language, by enabling human beings to comment on and think about the relationships between things and events, that is, by enabling them to articulate complex thoughts and, most important, share them with others. . . .

"We are not the first to suggest a gestural origin of language. [Gordon] Hewes (1973; 1974; 1976) was one of the first modern proponents of a gestural origins theory. [Adam] Kendon (1991: 215) also suggests that 'the first kind of behaviour that could be said to be functioning in anything like a linguistic fashion would have had to have been gestural.' For Kendon, as for most others who consider gestural origins of language, gestures are placed in opposition to speech and vocalization. . . .

"While we would agree with Kendon's strategy of examining the relationships among spoken and signed languages, pantomime, graphic depiction, and other modes of human representation, we are not convinced that placing gesture in opposition to speech leads to a productive framework for understanding the emergence of cognition and language. For us, the answer to the question, 'If language began as gesture, why did it not stay that way?' is that it did. . . .

"All language, in the words of Ulrich Neisser (1976), is 'articulatory gesturing.'

"We are not proposing that language began as gesture and became vocal. Language has been and always will be gestural (at least until we evolve a reliable and universal capacity for mental telepathy)."

(David F. Armstrong, William C. Stokoe, and Sherman E. Wilcox, Gesture and the Nature of Language . Cambridge University Press, 1995)

- "If, with [Dwight] Whitney, we think of 'language' as a complex of instrumentalities which serve in the expression of 'thought' (as he would say--one might not wish to put it quite like this today), then gesture is part of 'language.' For those of us with an interest in language conceived of in this way, our task must include working out all the intricate ways in which gesture is used in relation to speech and of showing the circumstances in which the organization of each is differentiated from the other as well as the ways in which they overlap. This can only enrich our understanding of how these instrumentalities function. If, on the other hand, we define 'language' in structural terms, thus excluding from consideration most, if not all, of the kinds of gestural usages I have illustrated today, we may be in danger of missing important features of how language, so defined, actually succeeds as an instrument of communication. Such a structural definition is valuable as a matter of convenience, as a way of delimiting a field of concern. On the other hand, from the point of view of a comprehensive theory of how humans do all the things they do by means of utterances, it cannot be sufficient."

(Adam Kendon, "Language and Gesture: Unity or Duality?" Language and Gesture , ed. by David McNeill. Cambridge University Press, 2000)

Language as a Device for Bonding

"[T]he size of human social groups gives rise to a serious problem: grooming is the mechanism that is used to bond social groups among primates, but human groups are so large that it would be impossible to invest enough time in grooming to bond groups of this size effectively. The alternative suggestion, then, is that language evolved as a device for bonding large social groups--in other words, as a form of grooming-at-a-distance. The kind of information that language was designed to carry was not about the physical world, but rather about the social world. Note that the issue here is not the evolution of grammar as such, but the evolution of language. Grammar would have been equally useful whether language evolved to subserve a social or a technological function."

(Robin I.A. Dunbar, "The Origin and Subsequent Evolution of Language." Language Evolution , ed. by Morten H. Christiansen and Simon Kirby. Oxford University Press, 2003)

Otto Jespersen on Language as Play (1922)

- "[P]rimitive speakers were not reticent and reserved beings, but youthful men and women babbling merrily on, without being so particular about the meaning of each word. . . . They chattered away for the mere pleasure of chattering . . ..  [P]rimitive speech . . . resembles the speech of little baby himself, before he begins to frame his own language after the pattern of the grownups; the language of our remote forefathers was like that ceaseless humming and crooning with which no thoughts are as yet connected, which merely amuses and delights the little one. Language originated as play, and the organs of speech were first trained in this singing sport of idle hours."

(Otto Jespersen, Language: Its Nature, Development and Origin , 1922)

- "It is quite interesting to note that these modern views [on the commonality of language and music and of language and dance] were anticipated in great detail by Jespersen (1922: 392-442). In his speculations about the origin of language, he arrived at the view that referential language must have been preceded by singing, which in its turn was functional in fulfilling the need for sex (or love), on the one hand, and the need for coordinating collective work, on the other. These speculations have, in turn, their origins in [Charles] Darwin's 1871 book The Descent of Man :

we may conclude from a widely-spread analogy that this power would have been especially exerted during the courtship of the sexes, serving to express various emotions. . . . The imitation by articulate sounds of musical cries might have given rise to words expressive of various complex emotions.

(quoted from Howard 1982: 70)

The modern scholars mentioned above agree in rejecting the well-known scenario according to which language originated as a system of monosyllabic grunt-like sounds that had the (referential) function of pointing at things. Instead, they propose a scenario according to which referential meaning was slowly grafted upon nearly autonomous melodious sound."

(Esa Itkonen, Analogy as Structure and Process: Approaches in Linguistics, Cognitive Psychology and Philosophy of Science . John Benjamins, 2005)

Divided Views on the Origins of Language (2016)

"Today, opinion on the matter of language origins is still deeply divided. On the one hand, there are those who feel that language is so complex, and so deeply ingrained in the human condition, that it must have evolved slowly over immense periods of time. Indeed, some believe that its roots go all the way back to  Homo habilis , a tiny-brained hominid that lived in Africa not far short of two million years ago. On the other, there are those like [Robert] Berwick and [Noam] Chomsky who believe that humans acquired language quite recently, in an abrupt event. Nobody is in the middle on this one, except to the extent that different extinct hominid species are seen as the inaugurators of language’s slow evolutionary trajectory.

"That this deep dichotomy of viewpoint has been able to persist (not only among linguists, but among paleoanthropologists, archaeologists, cognitive scientists, and others) for as long as anyone can remember is due to one simple fact: at least until the very recent advent of writing systems , language has left no trace in any durable record. Whether any early humans possessed language, or didn’t, has had to be inferred from indirect proxy indicators. And views have diverged greatly on the matter of what is an acceptable proxy."

(Ian Tattersall, "At the Birth of Language."   The New York Review of Books , August 18, 2016)

• Where Does Language Come From?: Five Theories on the Origins of Language
• Cognitive Linguistics  and  Neurolinguistics
• Five Theories on the Origins of Language
• Displacement in Language
• Holophrase in Language Acquisition
• Duality of Patterning in Language
• The Sound 'Schwa' With Definition and Examples in English
• Observations on What Is Language
• An Introduction to Semantics
• What Is Psycholinguistics?
• The Cultural Transmission of Language
• Universal Grammar (UG)
• The Theory of Poverty of the Stimulus in Language Development
• Definition and Examples of Linguists
• Indeterminacy (Language)
• 10 Types of Grammar (and Counting)
• Pragmatics Gives Context to Language
• Defining Philology

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• PMC10142271

Language: Its Origin and Ongoing Evolution

Ilia markov.

1 Department of Psychology, University of Houston, Houston, TX 77204, USA

2 Texas Institute for Measurement, Evaluation, and Statistics (TIMES), The University of Houston, Houston, TX 77204, USA

3 Center for Cognitive Sciences, Sirius University for Science and Technology, Sochi 354340, Russia

Kseniia Kharitonova

Elena l. grigorenko.

4 Baylor College of Medicine, Houston, TX 77030, USA

5 Child Study Center and Haskins Laboratories, Yale University, New Haven, CT 06520, USA

6 Rector’s Office, Moscow State University for Psychology and Education, Moscow 127051, Russia

Associated Data

Not applicable.

With the present paper, we sought to use research findings to illustrate the following thesis: the evolution of language follows the principles of human evolution. We argued that language does not exist for its own sake, it is one of a multitude of skills that developed to achieve a shared communicative goal, and all its features are reflective of this. Ongoing emerging language adaptations strive to better fit the present state of the human species. Theories of language have evolved from a single-modality to multimodal, from human-specific to usage-based and goal-driven. We proposed that language should be viewed as a multitude of communication techniques that have developed and are developing in response to selective pressure. The precise nature of language is shaped by the needs of the species (arguably, uniquely H. sapiens ) utilizing it, and the emergence of new situational adaptations, as well as new forms and types of human language, demonstrates that language includes an act driven by a communicative goal. This article serves as an overview of the current state of psycholinguistic research on the topic of language evolution.

1. Introduction

Research on language origin and evolution may be viewed as two of the most prominent research directions of the past few years. While these questions have been at the forefront of language science since its inception, only recently have we seen methodologies and techniques being developed that can provide answers backed with sufficient empirical evidence. The landscape of theoretical frameworks of language origin, the form in which it originated, and its worldwide dispersal has also been shifting in response to newly obtained evidence. The field of language evolution research can be described as currently coming of age while already equipped with a rich toolkit of methods for pursuits such as comparative research, investigating commonalities and differences between human language and animal communication systems, and studying cumulative cultural evolution of communication systems in experimental settings ( Dediu and de Boer 2016 ).

The aim of this article was to give a brief overview of the evolution of language, as well as to demonstrate how theoretical frameworks of language origin and evolution evolved with it. We sought to utilize the latest findings to argue that evolution is one of the central driving forces of the existence and development of human language. Language is a human skill, the nature and features of which are shaped in accordance with the needs of the species through continuous usage and adherence to communicative goals ( Grigorenko 2023 ). That nature is reflected through newly emergent language origin theories that move away from the innateness of language and provide plausible explanations of gradual language emergence from a multitude of other subsystems of communication. In regard to language modality, the main informational channel of origin, we intended to provide theoretical reasoning and evidence for the multimodal approach.

To illustrate that language is a skill that constantly undergoes changes due to various selective pressures, we aimed to explore three groups of factors that seem to transform language in the most significant ways: factors of the physical environment (such as aridity, vegetation, ambient temperatures, precipitation, latitude), socio-demographic factors (number of language users, geographic spread, degree of language contact, and the role of communicative situations), and technological advances (Internet, smartphones, and instant messaging). The latter group of factors is of most interest to us due to the changes that online communication is bringing about at a rate that has never been witnessed before. We propose that the evolution of language follows a similar pattern to the one outlined by recent research into general intelligence, promoting a more context-dependent and dynamic view of intelligence that is focused more radically than ever before on niche construction as a result of the cultural evolution ( Preiss 2022 ). The factors of the environment that influence language now reflect the changes brought along with the Anthropocene ( Anthropocene Working Group 2019 ), which further alters the ecological niche of our species, necessitating further adaptation.

While the argumentation provided in the present article is split into two main sections, covering the theories of origin and the evidence of evolution, all of it follows the main thesis of language emerging and evolving through usage; we deem this thesis to be the most prominent new direction in language research.

2. Origin of Language

There are more than 7000 living languages across the globe today ( Lewis 2009 ). To approach the topic of the ongoing evolution of language that spans millennia, we first need to determine its origin. However, the question by itself poses a challenge.

An important philosophical distinction collects two separate topics under the label “origin of language”: the origin of language faculty and the origin of languages ( Formigari 2013 ). The latter leads to a further question, namely, whether all world languages derive from one single “protolanguage”—i.e., monogenesis—or do several language families (e.g., Indo-European, Afro-Asiatic, Altaic) each derive from a different protolanguage—i.e., polygenesis ( Graffi 2019 ).

The theories relating to the emergence of language have been extensively debated for centuries, once leading to the infamous ban of the germane discussions by the Société de Linguistique de Paris in 1861. However, the field has seen a surge in this theorizing since the late 20th century and its considerable evolution over the past two decades ( Nölle et al. 2020b ). The focus of the debate has been on the innateness of language for many years. The central line of research that incorporates the traditional generative view ( Chomsky 1988 ), notions of innateness, universal grammar (UG), and the poverty of stimulus argument is largely classified as biolinguistics (for a more modern iteration of this framework, see ( Boeckx 2021 ; Bolhuis et al. 2014 ; Hauser et al. 2002 )). Such approaches argue that knowledge of language structures is impossible to extract from linguistic input; hence, it is suggested to be innate. An opposing point of view, referred to as “usage-based” or “emergentist” approaches, postulates that language emerged from its usage: “meaning is use—structure emerges from use” ( Tomasello 2009, p. 69 ); linguistic knowledge in these approaches proceeds via the abstraction and schematization of actual language use into fixed chunks, as well as more abstract linguistic patterns that become cognitively entrenched. These approaches, therefore, reject the notion of UG ( Pleyer and Hartmann 2019 ), instead utilizing the notion of the common communicative goal to explain language commonalities ( Arbib 2012 ). Notably, attempts to synthesize the opposing theories were few and far between. Such an attempt was undertaken by Pinker and Bloom ( 1990 ), where the authors put forth a compelling case against viewing language as a “spandrel”—an architectural allegory of a space formed at the intersection of other spaces, its shape therefore not being a significant trait on its own—arguing that certain constraints would not allow us to assume a non-adaptationist point of view, as “no adaptive organ can be adaptive in every aspect” (1990, p. 19).

In these controversies, one of the most contested topics is the question of causality, which is influenced by several factors: the problem of spurious correlation ( Roberts and Winters 2012 ), the universally concerning “replication crisis” ( Open Science Collaboration 2015 ), and a tendency to rely on indirect evidence ( Nölle et al. 2020b ). A novel solution for this contest is termed the “maximum robustness approach,” which, instead of focusing on simple causal relationships, aims to systematically construct more complex and coherent causal graphs ( Pearl 2000 ), incorporating all available evidence to form links between multiple variables. The CHIELD database, which is a repository of linguistic hypotheses produced in literature, was created to explore such graphs in order to find gaps or conflicting relationships, which can help to design empirical research addressing these issues and uncovering actual causal mechanisms ( Roberts et al. 2020 ). The database is public and functional: it is designed to be extendable by future researchers to ultimately become comprehensive and inclusive of as many languages as possible (that exist). Universal acceptance of a database such as this is the first step toward a realistic implementation of the maximum robustness approach.

To summarize, the current general trend for the linguistic field seems to move further and further away from notions of language innateness, although significant support for these viewpoints remains. In the scope of language evolution, usage-based frameworks allow for significantly more detailed and insightful investigations into the emergence of language. The same appears to be true for research into the modality of origin: this landscape of theories is also changing.

2.1. Modality of Origin

The debate on the modality of origin, which is the initial main channel carrying verbal information, also has a few contesting theories: according to the “gesture-first” view, “language evolved initially from manual gestures with vocal elements gradually added” ( Corballis 2011, p. 383 ). The “speech-first” view ( Dunbar 1997 ; MacNeilage 2008 ) argues for the pre-emergence of a vocal–auditory modality given its present-day dominance (for a full historical overview, see Fitch 2010 ). Modern theories argue for a multimodal emergence theory, incorporating complex interplay between auditory and visual channels ( Perlman 2017 ). Among newer ideas, “pantomime-first” was put forward as a distinct theoretical proposal ( Zlatev et al. 2017 ), which intrigues but does not provide much empirical evidence for its support. Another supporting usage-based account on multimodality comes from Levinson and Holler ( 2014 ), who propose that language normally occurs while embedded in a layered structure of multiple other channels of information. This view enables different phylogenetic and evolutionary origins to be assigned to each layer. Such holistic representation helps to bridge the gulf between the species, allowing us to recognize precursor adaptations such as turn-taking in current primates and the gestural skills of great apes as the first steps toward language formation, while the whole ensemble of language continues to be distinctively human.

One of the novel multimodal hypotheses is the mirror system hypothesis developed by Rizzolatti and Arbib ( 1998 ), which postulates that the mechanisms that support language in the human brain evolved atop a basic mechanism not originally related to communication: the mirror system, as the evolutionary basis for language, possesses a capacity to generate and recognize a set of actions. Arbib argues that the evolution of language is rooted in the execution and observation of hand movements, leading to the emergence of sign language, which was thereafter extended to speech. Complex imitation for hand movements evolved adaptively because of its utility in the social sharing of practical and manual skills. Skill sharing through imitation, such as grasping objects and using simple tools, existed long before language, being “more powerful than the call and gesture systems of nonhuman primates but lacking the full richness of modern human languages” ( Arbib 2012, p. 157 ).

Importantly, the origin theories based on the writing modality are characteristically absent, which is understandable given its (mostly) secondary nature to spoken language. This, however, is all too indicative of the attitudes to writing in language research prior to modern studies. The linguistic views on the emergence of writing were varied and controversial, echoing many general issues of the evolution of spoken language. The traditional outlook on writing systems since Aristotle was superficial, ostensibly viewing these systems as an optional, supplemental representation of spoken language. Moreover, writing was deemed a “wandering outcast of linguistics” ( Derrida 1976 ), leading to a suppression of research on writing. Similarly, for Saussure, written language was an object of suspicion, presenting a confounding and contaminating influence on language, going so far as to state that “to let go of the letter means a first step in the direction of truth” ( Saussure et al. 1986, p. 32 ). The views that were expressed during that time in the field were later ascribed to the “written language bias” ( Linell 2004 ).

Views that contested that bias started emerging in the mid-20th century from the historical ( Goody 1986 ) and anthropological ( McLuhan 1962 ) fields that, in turn, influenced studies of language to ascribe a more fundamental meaning to writing. One of the modern points of view from D.S. Olson proposed a special relationship of writing to the general machinery of language, which was influenced by those accounts and driven by developmental evidence ( Robinson et al. 1983 ). Olson’s most recent account postulated that reading and writing create a system of meta-representation concepts that contribute to consciousness, the formation of systematic thought, and rationality. In addition, some theories suggest that writing did not emerge as a secondary representation of spoken language but as the evolution of the token system for the purposes of goods exchanged or accounting ( Schmandt-Besserat 2012 ). The role of writing systems is similarly far from secondary according to the literacy hypothesis ( Goody and Watt 1963 ), and while it has received a lot of criticism on the matter of most aspects of civilized society preexisting and assimilating literacy at its advent, some scholars define influential “biases” ( Olson 2012 ) that may have contributed greatly to the cognitive and social development of the species. Such an impact of writing may be evidenced by tests of intelligence, including items that deal with vocabulary and the relationship between words, which test our capacity to participate in a literate environment ( Olson 2005 , as cited in Preiss and Sternberg 2006 ). Additionally, writing is essential to consider if adopting the adaptationist point of view, as the emergence of writing seems to possess several features characteristic of a Darwinian process ( Lock and Gers 2012 ).

Thus, while the dominating role of the vocal–auditory modality remains indisputable, progress in the field was made toward developing multimodal theories of language origin, which aid in unifying disparate evidence in support of different single-modality theories under a single governing principle.

2.2. Origin of Languages

The second question out of the pair laid out at the beginning of the section, namely, regarding “the origin of languages,” was mostly inquired upon in neighboring fields of inquiry and tied to the spread of human populations. A link between the human genome and the spread of languages has been debated ever since Darwin proposed that “a perfect pedigree of mankind… would afford the best classification of the various languages now spoken throughout the world” ( Darwin 1871 ). While some argued that the spread of languages is a good proxy for the dispersion of human populations ( Gray and Atkinson 2003 ; Mace and Holden 2005 ), opposition to this assumption was also persistent (e.g., Donohue and Denham 2010 ). Quantitative evidence supports both a general gene–language dispersion correspondence but also substantial (~20%) mismatches between 10 language families and corresponding populations ( Barbieri et al. 2022 ).

Globally, a consensus around the serial founder effect (SFE) process playing an important role in shaping global patterns of neutral genetic diversity is currently forming. This process entails a series of population splits, movements into an unoccupied territory, and subsequent isolation: beginning in Africa and proceeding through Eurasia into the Americas and Oceania. At the within-population level, it led to a steady decay in genetic diversity with increasing geographic distance from East Africa; at the between-population level, it led to a steady increase in genetic distance with increasing geographic distance ( Prugnolle et al. 2005 ; Ramachandran et al. 2005 ). The debate on the topic of language dispersion was later reignited by Atkinson ( 2011 ), who proposed that phoneme inventories in human languages had undergone a parallel SFE process based on the finding that the number of phonemes in 504 widespread languages decreased linearly with increasing geographic distance from Africa. Alternative assumptions for worldwide phonemic cline were tested using numerical simulations, showing that this pattern may be due to a repeated bottleneck effect and phonemic loss: low-density populations lost phonemes during the out-of-Africa dispersal of modern humans ( Pérez-Losada and Fort 2018 ). Creanza et al. ( 2015 ) further delved into this issue by performing joint and parallel analyses of phoneme counts in 2,082 languages and DNA microsatellite polymorphisms, which were used as signatures of human demographic history to calculate genetic distances between 246 populations. The results decisively vindicated Darwin’s proposal of human races and languages evolving in concert following a tree-like history of splits and isolation ( Darwin 1860 ) at the global level; however, it did not align with the SFE model with Africa as the center of origin, with it instead being more inclined toward a Eurasian-centered model. A more recent and novel analysis, which covered a cultural layer adjacent to language, namely, music, was carried out on a dataset of 152 societies (containing 1,054 songs from the public database The Global Jukebox in the form of raw coded Cantometrics data, 1,719 genomic profiles, and 152 languages); the analysis demonstrated weak links between music and language (R 2 <= 0.05), as well as with genetic distance and geographic proximity, in contrast to the much stronger relationships found between genes and geography: the results suggest that genes and culture are surprisingly decoupled ( Passmore et al. 2022 ). For the Indo-European family, Bouckaert et al. ( 2012 ) used Bayesian phylogeographic approaches with a dataset of basic vocabulary term lists from 103 ancient and contemporary Indo-European languages to model the expansion of the family, finding decisive support for an Anatolian origin over a steppe origin, with both the inferred timing and root location of the Indo-European language trees fitting with an agricultural expansion from Anatolia beginning 8000 to 9500 years ago. Certain linguistic methods, such as Bayesian phylogeographic approaches, that emerged from recent studies provided tentative answers to general questions of human prehistory: a recent study using lexical data and Bayesian phylogenetic methods placed the Austronesian origin in Taiwan approximately 5230 years ago and supported the hypothesis of “pulse-pause” expansion from Taiwan on the origin of the Austronesian settlers of the Pacific ( Gray et al. 2009 ). While being fairly recent for the field, the aforementioned techniques succeeded in helping linguists reclaim the issue of the origin of language as the viable research aim for future investigations aside from the origin of humanity research in neighboring fields.

To summarize, new technological and methodological advances have led to the most drastic changes in language evolution research. Large-scale investigations do require substantial resources, and interdisciplinary collaboration poses a challenge, but the results obtained contribute to significant advancements in the linguistic and neighboring fields in regard to the origin of language dispersal.

2.3. Neural Correlates of Language

The final important aspect of language origin studies, tangential to linguistics but central to psycholinguistics, is of utmost importance for the present essay: neural correlates of language. Evidence obtained from numerous previous studies that attempted to localize language within the brain is well established: the clinical studies of Broca ( Broca 1861 ) in the 19th century and Wernicke ( Wernicke [1874] 1994 ) in the 20th century, although contested now, served as an initial impulse for this. Similar to the debate on the language faculty origin in linguistic circles, we can note that the initial research findings that focused on narrow specificity of function were later extended to cover contesting evidence and new theoretical frameworks and have evolved into a multidimensional system. Through further publications on the topic emphasizing the importance of previously unaccounted-for brain regions (e.g., insula, Dronkers 1996 ), the model of choice for the end of the 20th century became the aphasia model (e.g., Obler and Gjerlow 1999 ). At the beginning of the 21st century, the model was further expanded into Broca-type and Wernicke-type aphasias in accordance with the impairment in one of two language comprehension axes ( Ardila 2011 , 2012 ), further solidifying the trend toward system complexity.

Modern models of language include numerous areas of the brain organized in multiple circuits within clusters of activation ( Ardila et al. 2016 ). One such model was constructed by Peter Hagoort ( 2005 ), who argued that the operation of distributed neural networks in Broca’s area and the left inferior frontal gyrus (LIFG) involves parallel processing of semantic, syntactic, and phonological information through three functional components: memory (long-term memory retrieval), unification (integrating information), and control (selecting a language “action”). Evidence from EEG and MEG studies helped to identify the specific temporal features of unification and memory retrieval components, arguing for neuronal synchronization that supports functional interrelatedness rather than strict domain specificity ( Bastiaansen and Hagoort 2006 ). These considerations are far from a theoretical conjecture nowadays, as they have been translated into presurgical planning ( Alemi et al. 2018 ). Additionally, there is evidence of neural multifunctionality for language networks, in particular, several frontal networks being linked to non-linguistic functions, such as mental rotation ( Jordan et al. 2001 ), musical syntax processing ( Maess et al. 2001 ), and arithmetic comprehension ( Baldo and Dronkers 2007 ). Such findings have driven researchers toward frameworks of multifunctional modularity and are instrumental for scholars developing usage-based approaches to language evolution.

Thus, despite initial findings that focused on the narrow specificity of function, the field has evolved to include numerous areas of the brain organized in multiple circuits within clusters of activation, in addition to the emerging evidence of neural multifunctionality for language networks.

3. Language Adaptation

Lupyan and Dale ( 2016 ) argued that observed linguistic differences arise not only from the accumulation of random changes due to the languages drifting apart but also may be reflective of the environment in which the language was developing. These environmental aspects that pressure languages into continuous diversification are social, physical, and technological in nature ( Lupyan and Dale 2016 ).

Just like birds develop different beaks adapting to different environments, languages and cultures might be undergoing similar changes ( Lupyan and Dale 2016 ). Charles Darwin, in “The Descent of Man, and Selection in Relation to Sex,” cited Max Müller to make a case for the evolution of language: “A struggle for life is constantly going on amongst the words and grammatical forms in each language. The better, the shorter, the easier forms are constantly gaining the upper hand, and they owe their success to their own inherent virtue.” ( Darwin 1871, p. 58 ). However, this idea of progress in linguistic evolution is considered dysfunctional by some ( Labov 1991 ; Mendívil-Giró 2018 ) due to its inability to explain the main patterns of linguistic structural diversity; a growing body of research asserts the contrary. The process of language diversification cannot be understood without considering the pressures that several factors (physical, ecological, and social) put on language users in different environments ( Bentz et al. 2018 ).

3.1. Ecological Adaptations

Similar to the communication systems of other species, language may be affected by ecological factors. Physiologically based predictions demonstrate that languages with complex tonality have generally not developed in very cold or otherwise desiccated climates, as air dryness decreases the control of the vocal folds and pitch production, and this, in turn, results in the absence of a (complex) tone system. The geographic–linguistic association operates within continents, major language families, and across language isolates ( Everett et al. 2015 ). However, replication of the study on a different dataset found it was not robust ( Roberts 2018 ). An analysis of over 4,000 language varieties showed a positive association between the language’s degree of reliance on vowels and the typical ambient humidity of a language’s native locale, which is consistent with other studies that focus on the link between aridity (i.e., the lack of effective moisture in a climate) and tonality of language ( Everett et al. 2015 ; Everett 2017 ), but the robustness was later found to be limited ( Roberts 2018 ).

Environments in which higher sound frequencies are less faithfully transmitted due to denser vegetation or higher ambient temperatures seem to be related to the greater use of sounds of lower frequencies (“more sonorous” languages). The results of Maddieson and Coupé ( 2015 ) point to a significant relationship between the “consonant-heaviness” of languages and several environmental factors, including tree cover and precipitation ( Maddieson and Coupé 2015 ). Further analysis of spoken samples did not find the relationship significant but identified that the percentage of sonorous material is correlated with the mean annual temperature in the area of the language ( Maddieson 2018 ). Studies that focused on the influence of temperature on languages find that languages spoken in cold, small regions tend to be more complex across a range of linguistic features, such as morphosyntactic complexity, linguistic diversity, word length, and consonant inventory ( Lewis and Frank 2016 ).

Another striking example is the observed partial correlation between latitude and the absence or presence of the word for the color blue ( Brown and Lindsey 2004 ; Lindsey and Brown 2002 ) due to the negative impact of ultraviolet light (UV-B) on the perception of the blue/green distinction (phototoxicity). In high-UV areas, languages without the word for blue prevail, which also correlates with the rates of blue-yellow color vision deficiency in these areas suggesting an evolutionary, physiological cause for both phenomena ( Brown and Lindsey 2004 ; Dediu et al. 2017 ).

A common criticism of the abovementioned studies is that they are correlational in nature, thus, do not contribute to the understanding of possible mechanisms that underlie linguistic evolutionary processes. In order to improve the methodological robustness of the studies, additional approaches, such as iterated learning, a historical case study, corpus studies, and studying individual speech, was suggested ( Roberts 2018 ). For this reason, several studies tried to experimentally investigate how environmental factors drive the emergence of linguistic conventions. Nölle et al. ( 2020a ) adapted the classical maze game task to confirm that subtle environmental motivations cause the emergence of different communicative conventions in an otherwise identical task, pointing to linguistic adaptations being highly sensitive to factors of the shared task environment. The authors speculated that these kinds of mechanisms identified at a local interactional level might contribute to the systematic global variation observed between different languages.

One of the most striking examples of linguistic adaptation to the environment is whistled languages. The main purpose of whistled languages is to facilitate spoken communication at great distances, but it is also used in other circumstances, such as secrecy, courtship, singing, and communication in noisy environments. Although they are always referred to as languages, they are considered a mode of speech because whistled languages are always based on a spoken language ( Meyer 2015 ).

Several hypotheses were put forward to explain the current existence of whistled languages. One of them posits that whistled languages are simply the vestigial remains of a widespread ancient phenomenon. This mode of speech could have been used by prehistoric hunter-gatherers for hunting in groups or signaling a danger in any type of environment ( Nettle and Romaine 2000 ). Another possible explanation is that the actual whistled languages are found only in a small minority of languages due to the erosion of traditional lifestyles and the relative ease of resorting to shouting because whistled speech would generally require more pressure to develop. This argument would be in favor of a key role played by significant environmental constraints in the emergence of whistled speech, which is supported by the observed systematic adaptation of whistled speech to typically constraining and geographically scattered ecological milieux ( Meyer 2015 ).

It is estimated that approximately 70–80 languages actively use their whistled mode of speech, but the number is rapidly declining due to modern technologies of communication, and most of them are endangered ( Meyer 2018 ). Evidently, whistled speech plays a strong functional role by complementing regular speech under unusual circumstances. Around the world, whistled forms of languages are associated with traditional activities, such as hunting, hill agriculture, or shepherding, in which individuals are relatively isolated and scattered across substantial areas of densely vegetated landscapes. In this type of environment, whistling has a clear advantage over speaking or shouting: acoustic signals can easily overcome ambient conditions and can travel longer distances. For example, La Gomera, one of Spain’s Canary Islands, holds the record for the longest distance of whistled conversations of approximately 1 km ( Meyer 2015 ); others have observed communications at approximately 8 km ( Busnel and Classe 1976 ).

The principle of whistled speech is straightforward: people articulate words while whistling, which involves acoustic reduction at the produced frequency level and selection of key salient phonetic cues for the corresponding spoken utterances. The resulting signal’s linguistic structure is identical to standard speech. Interestingly, even though the acoustic channel is reduced, whistled sentences remain highly intelligible to trained speakers ( Meyer 2015 ).

It was suggested that human whistled languages can serve as a model for understanding the coding of information in dolphin whistle communication. Comparing human and dolphin whistles could become a complementary test bench for the development of new methodologies for decoding whistled communication signals by providing new perspectives on structural and organizational aspects of encoding information ( Meyer et al. 2021 ).

Overall, exploring the connection between the structure of languages and the environment in which they are utilized is complicated by several issues. If the ecology of the area was able to influence the language at the stage of its emergence, the amount of information needed to make a conclusive statement about it is scarce. Additionally, most of the studies that focus on the link between environment and language are correlational in nature. Although the overall structural diversity of languages has not been linked with other types of diversity, some aspects, such as morphosyntactic complexity or consonant/vowel inventory, may be affected.

3.2. Socio-Demographic Adaptations

Apart from the effects of the physical environment and location, languages may be shaped by social and demographic factors. A statistical analysis of 2000 languages revealed strong relationships between the morphological complexity of a language and demographic/socio-historical factors, including the number of language users, geographic spread, and degree of language contact ( Lupyan and Dale 2010 ). It was suggested that languages spoken by large groups have a simpler inflectional morphology than languages spoken by smaller groups. Additionally, languages spoken by large groups are more likely to utilize lexical strategies in place of inflectional morphology when encoding evidentiality, negation, aspect, and possession ( Lupyan and Dale 2010 ). Based on these findings, Dale and Lupyan proposed the linguistic niche hypothesis , which describes the esoteric and exoteric niches for languages. The exoteric linguistic niche includes languages with large numbers of speakers (e.g., English, Swahili, and Hindi), which forces these languages to serve as a means of communication between strangers. Speakers of languages in the exoteric niche, compared with the esoteric niche, are more likely to be non-native speakers or have learned the language from non-native speakers and use the language to speak to individuals from different ethnic and/or linguistic backgrounds. The esoteric niche includes languages like Tatar, Elfdalian, and Algonquin ( Dale and Lupyan 2012 ; Lupyan and Dale 2010 , 2016 ). Linguistically, esoteric languages are more likely to be classified as isolating rather than fusional, have fewer grammatical categories marked on the verb, are more likely to encode negation via analytical strategies than using inflections, are less likely to have indefinite and definite articles, and are less likely to communicate distance distinction demonstratives ( Lupyan and Dale 2010 ). Further studies found only limited support for this hypothesis ( Lewis and Frank 2016 ) or did not find a strong relationship between the grammatical or statistical structure of language and the proportion of non-native speakers ( Koplenig 2019 ).

Winters et al. ( 2015 ) experimentally investigated the role of the communicative situation in which an utterance is produced and how it influences the emergence of three types of linguistic systems: underspecified languages, holistic systems, and systematic languages. Using a discrimination task in a communication game and manipulating whether the feature dimension shape was relevant or not in discriminating between two referents, it was established that different linguistic systems emerged. Furthermore, experimental languages gradually developed to encode information relevant to the communicative task in a given situational context. These results suggest that language systems adapt to their contextual niche over iterated learning.

Another interesting observation was made about the influence of population size on rates of language evolution. The rates of gain and loss of cognate words for basic vocabulary were analyzed in Polynesian languages. Larger populations were observed to have higher rates of gain of new words, while smaller populations had higher rates of word loss, which suggests that demographic factors may affect rates of language evolution and that rates of gain and loss are affected in different ways. However, the authors found that the results were strikingly consistent with general predictions of evolutionary models, paralleling positive selection in the case of greater rates of word gain in larger populations, and loss of diversity in small populations and greater rates of word loss ( Bromham et al. 2015 ).

An inquiry into language evolution that was made using estimates of cognate replacement for 200 concepts on an Indo-European language tree spanning 6–10 millennia to measure lexical evolution rates demonstrated that negative valence correlates with faster cognate replacement, even while controlling for frequency of use. Follow-up analyses showed that it is most robust for adjectives, does not consistently reach statistical significance for verbs, and never reaches significance for nouns ( Jackson et al. 2023 ).

Socio-demographic pressures are also known to lead to the emergence of new languages, the investigation of which can shed light on the process of language evolution. Among the newest languages of the world, Afrikaans is the youngest nationally recognized language, whose origin is tied to the establishment of Cape Colony in the Cape of Good Hope on the land of indigenous Khoekhoen people in the mid-16th century. The starting point of the Cape Colony is the landing of Jan van Riebeeck with three vessels to the Cape on 6th April 1652, which can also be thought of as the starting point of the Afrikaans language ( De Villiers 2012 ). The language developed through to the 18th century, becoming the language with the widest geographical, demographic, and racial distribution of all official languages of South Africa ( Webb 2003 ), and the debate on its origin was live and ongoing until the 20th century, mainly due the clash of political and ideological views that it instigated. Certain scholars at the time denied any indigenous influences on the language, while others insisted that the language was as much creolized as it was a product of West-Germanic sources. It has its roots in 17th-century Dutch but has been influenced by English, French, and German ( Hamans 2021 ), with traces of, amongst others, Malay and Portuguese ( Conradie and Groenewald 2014 ), and influenced by the pidgin talk of the indigenous Khoi and the San ( Hamans 2021 ).

While not as widely recognized as Afrikaans, Light Warlpiri would constitute the newest example of an emerging language. This language was discovered and documented by Carmel O’Shannessy ( 2005 ) and is thought to have originated sometime at the end of the previous century. It was spoken in the Warlpiri community of Lajamanu, in the Northern Territory of Australia, by children and young adults who are now mostly approximately 40 years of age ( O’Shannessy and Brown 2021 ). The language systematically combines elements of Warlpiri (a Pama-Nyungan language), Kriol (an English-based creole), and English, and was derived from the code-switched speech of parents to children following a particular pattern, where a Kriol pronoun and verb were inserted into a Warlpiri string, as part of a baby talk register ( O’Shannessy 2012 ). A new language emerged when young children internalized this pattern of speech as a single system, distinct from Warlpiri ( O’Shannessy 2020 ). In its formation, a speech pattern was further innovated in the verbal auxiliary system, namely, the =m “NONFUTURE” suffix ( O’Shannessy 2013 ). Light Warlpiri has enough systematic evidence to distinguish it as a separate language, for which its precursors act as lexifiers ( O’Shannessy 2005 ), and is nowadays the language of everyday interaction of the adult generation in the Lajamanu community.

To conclude, most typological studies suggest that linguistic diversity may be affected by several demographical and sociolinguistic factors. Social contexts can influence the way language is acquired and used, leading to linguistic structures that are specific to certain groups. Over time, this can result in variations in language usage that are reflected in typological patterns. This shows that language is not just a set of fixed rules and structures but rather a dynamic and adaptive skill that is shaped by the social context in which it is used. This ability to adapt to a social environment is a key feature of a language, and it helps to explain how language has evolved to meet the changing needs and contexts of different communities.

3.3. Technological Adaptations

Technology has drastically affected language and given rise to what is now commonly referred to as “text-speak” ( Al-Sharqi and Abbasi 2020 ). One of the biggest changes technology brought about is the speed of communication. The human brain is forced to process an unending stream of linguistic input and respond to it immediately. Christiansen and Charter called this the “Now-or-Never bottleneck,” which describes the immediacy with which the brain must compress and recode linguistic input ( Christiansen and Chater 2016 ). This bottleneck acts as a strong selection pressure against words and grammatical construction parsing, which, in real-time, is nearly impossible, especially when pressure is being put on the written language to communicate subtle nuances of face-to-face communication. The multi-faceted pressure inevitably influences the language that now has to undergo significant adaptive processes in order to fit the requirements of modern times, turning into what some deem “a natural experiment in the development of written communication” ( Varnhagen et al. 2010 ). Some of the ways that this adaptation manifests in the language are novel conventions of online communication, including acronyms, the modified use of typographic marks, and the use of emojis ( Lupyan and Dale 2016 ).

Emojis and emoticons are a group of symbolic combinations or pictures that are characteristic of online communication. An emoji (“☹”) is a graphic symbol that represents a wide variety of different things, ranging from complex facial expressions to concepts and ideas. It is thought to have developed from emoticons, i.e., representations of facial expressions usually comprised of various combinations of keyboard characters (“:)”). These symbols usually augment a message with non-verbal elements ( Novak et al. 2015 ).

Due to their growing popularity, emojis are used not only in online communications but are becoming integrated into an increasingly wider variety of contexts. Specifically, research is conducted to understand how emoji-enriched interfaces affect performance in the classroom ( Aliannejadi et al. 2021 ), marketing and advertisement ( Lee et al. 2021 ), and even their implications in law ( Goldman 2018 ).

Emojis (or “smileys”) are a unique phenomenon in terms of their nature and diverse functions in communication. On the one hand, emojis produce effects that are functionally similar to the response observed for facial expressions of emotion in face-to-face communication. They seem to affect the perceived emotional intensity of a message and accentuate its perceived valence by acting as nonverbal cues in digital communication ( Erle et al. 2021 ). On the other hand, emojis are closely connected with words. It was shown that the time course of semantic congruency effects on eye movements for emojis is similar to effects that were previously shown for words ( Barach et al. 2021 ). In the online public context, emojis alter the lexical diversity of text, which may point to a compensatory relationship between emojis and words in communication ( Feldman et al. 2021 ). Additionally, there is a link between emojis and gestures, with emojis denoting objects and activities interacting with logical operators in a text in a similar way as gestures do with speech ( Pierini 2021 ).

Lupyan and Dale argued that the divergence between conventional written languages (as well as online written communication) differs in many ways from the divergence between conventional spoken languages, for example, Dutch and Afrikaans. Both of these phenomena represent how languages (or language registers) adapt to the environments in which they are being used ( Lupyan and Dale 2016 ). Similar to Afrikaans and Dutch, it is feasible to assume that the written form of language diverges from the spoken form as an adaptation to this new environment, and the online form combines the features of the two forms into something that linguist John McWhorter described as a pure “linguistic miracle happening right under our noses” ( McWhorter 2013 ).

Shortcuts are one of the most prevalent features of the new “netspeak” ( Varnhagen et al. 2010 ). Common shortcuts in the context of instant messaging and online communication include abbreviations (prof—professor), initialisms and acronyms (LOL—laughing out loud, ASAP—as soon as possible), and logograms or “alphanumeronyms” (CUL8R—see you later).

One of the explanations of this process posits that users intuitively ignore uneconomical language rules and strive for cost-effectiveness, increasing the efficiency of the language orthography ( Lančarič 2016 ) and enriching it with new words and phrases that express complex feelings, emotions, or reactions (for example “wowzy” usually stands for extreme amazement or awe of a situation, thing, person, or place). Interestingly, some of the recently introduced units started to undergo a process of pragmaticalization, a subclass of grammaticalization, which possesses many similar features of grammaticalization processes but is distinguished from other subtypes by specific functions, domains, and syntactic integration ( Diewald 2011 ).

Pragmaticalized units may partially or completely lose their semantic meaning and move into a new pragmatic domain of function and meaning. In this sense, online communication is not only enriched by the spoken form of language with its abundance of discourse (or pragmatic) markers but gives rise to new netspeak-specific pragmatic features that slowly pave their way back into spoken language, making this interaction bidirectional and mutually enriching. This is only one of many aspects that is indicative of the emergence of a unique new hybrid register that fuses the full range of variants from the language use, namely, written, spoken, formal, informal, and vernacular variants ( Tagliamonte and Denis 2008 ).

One particular field able to inform the shortcut trend is quantitative linguistics, one of the aims of which is the development of statistical laws about language usage. Such laws can tell us a lot about speech and language efficiency principles, the most established among them being Zipf’s law of word frequency, which quantifies the frequency of occurrence of words, demonstrating that there is no unarbitrary way to distinguish between rare and common words ( Zipf 1949 ). This law is also rather common in complex systems where discrete units self-organize into groups or types ( Corral et al. 2019 ). Zipf’s law of brevity is sufficiently easier to observe through personal experience, stating that more frequent words tend to be shorter, and rarer words tend to be longer ( Bentz and Ferrer-i-Cancho 2016 ). A functional explanation for this law suggested by Zipf is the law of least effort, stating that it is human nature to want the greatest outcome at the least amount of work. Closely related to it is the Menzerath–Altmann law, which postulates that the size of the constituents (e.g., phonemes) of a construction (e.g., morpheme) decreases with the increasing size of the construction ( Altmann 1980 ). These two laws suggest that in human vocal communication, the maximization of coding efficiency and minimization of code length act as selective pressures to compress the elements supporting information ( Favaro et al. 2020 ). In addition to spoken language, there is evidence that laws of brevity hold in most writing systems (in a sample of 1262 texts and 986 different languages, see Bentz and Ferrer-i-Cancho 2016 ). However, there are exceptions, as with figurative signals, the frequency of which is shown to be positively correlated with complexity ( Miton and Morin 2019 ). The suggestion that these laws might extend beyond human language is substantivized by recent studies into vocal sequences of non-human primates ( Semple et al. 2010 ) and penguins ( Favaro et al. 2020 ), while some studies challenge that notion ( Bezerra et al. 2011 ).

Hence, technology has had a significant impact on language in recent years, giving rise to new forms of communication and new ways of using language. One of the most notable changes brought about by technology is the emergence of “text-speak,” which is a form of language that is characterized by the use of shorthand, abbreviations, and symbols in text-based communication. In this way, digital communication can be seen as a distinct form of language with unique features and conventions that are specific to the digital environment. It is not a replacement for spoken or written language, but rather, it is an additional way of communicating and conveying meaning, which is a skill that emerged in humans as a response to a new and all-encompassing technological environment with its constraints and possibilities.

4. Conclusions

The questions of the origin and evolution of language, apart from gaining new evidence for their resolution through the use of novel linguistic and psycholinguistic methods and interdisciplinary inquiries, also changed in their nature. Modern language research strays away from portraying human language as a unique anthropoid phenomenon, the expression of which has not been modified by selection pressure, instead viewing it through the lens of ongoing human evolution and strict adherence to communicative goals. In this light, language can be viewed as a toolbox, the contents of which change in accordance with the needs of the human species, which is the statement that we attempted to demonstrate through the review of the literature on language evolution and adaptation.

Manifold examples of languages adapting to and reflecting different aspects of the environment illustrate that linguistic diversification is not simply an accumulation of random changes over time. The most recent example of strong selective pressure affecting languages all over the world is the integration of instant communication technologies. This new phenomenon allows us to witness language modification in real time and better understand the underlying processes. It can already be confidently stated that human language is undergoing one of its most massive changes at this very moment while following the essential principles that governed its existence and development before the onset of the digital age.

Multimodal and usage-based emergence theories, in addition to more robust correlation and causation links, provide a framework that is apt to incorporate the majority of scientific knowledge about language. Modern neural correlate models of language processing further serve to illustrate the interconnectedness of language to other domains of cognition. Further steps in that direction of inquiry may only serve to elucidate how thoroughly integrated language is with all other types of human behavior.

We propose that the study of languages should not be confined to properties of particular languages and language in general but should incorporate a wider array of contributing factors that inevitably shape the way different species, including humans, communicate. This all-encompassing approach will provide more insights into the nature, structure, and functions of language in diverse environments and demographic contexts, as well as help to explain the way human communication adapts to and transforms in response to the pressures put forward by technological breakthroughs and societal transformations, along with the alterations in our species’ ecological niche in the Anthropocene era.

Funding Statement

The preparation of this essay was supported, in part, by a grant (R01HD109307) from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), USA, to the University of Houston (principal investigator: Elena L. Grigorenko) and by funds from Sirius University, Russia. The content is solely our responsibility and does not necessarily reflect the views of the funders.

Author Contributions

Conceptualization, I.M., K.K. and E.L.G.; investigation, I.M. and K.K.; resources, E.L.G.; writing—original draft preparation, I.M. and K.K.; writing—review and editing, I.M., K.K. and E.L.G.; supervision, E.L.G.; funding acquisition, E.L.G. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Informed consent statement, data availability statement, conflicts of interest.

The authors declare no conflict of interest.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

• Alemi Razieh, Batouli Seyed Amir Hossein, Behzad Ebrahim, Ebrahimpoor Mitra, Oghabian Mohammad Ali. Not single brain areas but a network is involved in language: Applications in presurgical planning. Clinical Neurology and Neurosurgery. 2018; 165 :116–28. doi: 10.1016/j.clineuro.2018.01.009. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Aliannejadi Mohammad, Landoni Monica, Huibers Theo, Murgia Emiliana, Pera Maria Soledad. Children’s Perspective on How Emojis Help Them to Recognise Relevant Results: Do Actions Speak Louder Than Words?; Paper presented at 2021 Conference on Human Information Interaction and Retrieval; Canberra, Australia. March 19; 2021. pp. 301–5. [ CrossRef ] [ Google Scholar ]
• Al-Sharqi Laila, Abbasi Irum Saeed. The Influence of Technology on English Language and Literature. English Language Teaching. 2020; 13 :7. doi: 10.5539/elt.v13n7p1. [ CrossRef ] [ Google Scholar ]
• Altmann Gabriel. Prolegomena to Menzerath’s law. Glottometrika. 1980; 2 :1–10. [ Google Scholar ]
• Anthropocene Working Group What Is the Anthropocene?—Current Definition and Status. 2019. [(accessed on 16 March 2023)]. Available online: http://quaternary.stratigraphy.org/working-groups/anthropocene/
• Arbib Michael A. How the Brain Got Language: The Mirror System Hypothesis. Oxford University Press; Oxford: 2012. [ Google Scholar ]
• Ardila Alfredo, Bernal Byron, Rosselli Monica. How Localized are Language Brain Areas? A Review of Brodmann Areas Involvement in Oral Language. Archives of Clinical Neuropsychology. 2016; 31 :112–22. doi: 10.1093/arclin/acv081. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ardila Alfredo. There are two different language systems in the brain. Journal of Behavioral and Brain Science. 2011; 1 :23. doi: 10.4236/jbbs.2011.12005. [ CrossRef ] [ Google Scholar ]
• Ardila Alfredo. Interaction between lexical and grammatical language systems in the brain. Physics of Life Reviews. 2012; 9 :198–214. doi: 10.1016/j.plrev.2012.05.001. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Atkinson Quentin D. Phonemic Diversity Supports a Serial Founder Effect Model of Language Expansion from Africa. Science. 2011; 332 :346–49. doi: 10.1126/science.1199295. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Baldo Juliana V., Dronkers Nina F. Neural correlates of arithmetic and language comprehension: A common substrate? Neuropsychologia. 2007; 45 :229–35. doi: 10.1016/j.neuropsychologia.2006.07.014. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barach Elisa, Feldman Laurie Beth, Sheridan Heather. Are emojis processed like words? Eye movements reveal the time course of semantic processing for emojified text. Psychonomic Bulletin & Review. 2021; 28 :978–91. doi: 10.3758/s13423-020-01864-y. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barbieri Chiara, Blasi Damián E., Arango-Isaza Epifanía, Sotiropoulos Alexandros G., Hammarström Harald, Wichmann Søren, Greenhill Simon J., Gray Russell D., Forkel Robert, Bickel Balthasar, et al. A global analysis of matches and mismatches between human genetic and linguistic histories. Proceedings of the National Academy of Sciences of the United States of America. 2022; 119 :e2122084119. doi: 10.1073/pnas.2122084119. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bastiaansen Marcel, Hagoort Peter. Oscillatory neuronal dynamics during language comprehension. In: Neuper Christa, Klimesch Wolfgang., editors. Progress in Brain Research. vol. 159. Elsevier; Amsterdam: 2006. pp. 179–96. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bentz Christian, Ferrer-i-Cancho Ramon. Zipf’s Law of Abbreviation as a Language Universal. University of Tübingen; Tübingen: 2016. pp. 1–4. [ CrossRef ] [ Google Scholar ]
• Bentz Christian, Dediu Dan, Verkerk Annemarie, Jäger Gerhard. The evolution of language families is shaped by the environment beyond neutral drift. Nature Human Behaviour. 2018; 2 :11. doi: 10.1038/s41562-018-0457-6. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bezerra Bruna M., Souto Antonio S., Radford Andrew N., Jones Gareth. Brevity Is Not Always a Virtue in Primate Communication. Biology Letters. 2011; 7 :23–25. doi: 10.1098/rsbl.2010.0455. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Boeckx Cedric. Language Science Press. Language Science Press; Berlin: 2021. Reflections on language evolution. [ CrossRef ] [ Google Scholar ]
• Bolhuis Johan J., Tattersall Ian, Chomsky Noam, Berwick Robert C. How Could Language Have Evolved? PLoS Biology. 2014; 12 :e1001934. doi: 10.1371/journal.pbio.1001934. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bouckaert Remco R., Lemey Philippe, Dunn Michael D., Greenhill Simon J., Alekseyenko Alexander V., Drummond Alexei J., Gray Russell D., Suchard Marc A., Atkinson Quentin D. Mapping the origins and expansion of the Indo-European language family. Science. 2012; 337 :957–60. doi: 10.1126/science.1219669. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Broca Paul. Remarks on the seat of the faculty of articulated language, following an observation of aphemia (loss of speech) Bulletin de La Société Anatomique. 1861; 6 :330–57. [ Google Scholar ]
• Bromham Lindell, Hua Xia, Fitzpatrick Thomas G., Greenhill Simon J. Rate of language evolution is affected by population size. Proceedings of the National Academy of Sciences of the United States of America. 2015; 112 :2097–102. doi: 10.1073/pnas.1419704112. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Brown Angela M., Lindsey Delvin T. Color and language: Worldwide distribution of Daltonism and distinct words for “blue” Visual Neuroscience. 2004; 21 :409–12. doi: 10.1017/S0952523804213098. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Busnel Rene-Guy, Classe André. Whistled Languages. Springer; Berlin and Heidelberg: 1976. [ Google Scholar ]
• Chomsky Noam. Language and Problems of Knowledge: The Managua Lectures. MIT Press; Cambridge: 1988. [ Google Scholar ]
• Christiansen Morten H., Chater Nick. The Now-or-Never bottleneck: A fundamental constraint on language. Behavioral and Brain Sciences. 2016; 39 :e62. doi: 10.1017/S0140525X1500031X. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Conradie Jac, Groenewald Gerald. In: Die ontstaan en vestiging van Afrikaans. Carstens W. A. M., Bosman N., editors. Van Schaik; Pretoria: 2014. pp. 27–60. Kontemporêre Afrikaanse Taalkunde. [ Google Scholar ]
• Corballis Michael C. The Origins of Language in Manual Gestures. Oxford University Press; Oxford: 2011. [ CrossRef ] [ Google Scholar ]
• Corral Álvaro, Serra Isabel, Ferrer-i-Cancho Ramon. Distinct flavors of Zipf’s law and its maximum likelihood fitting: Rank-size and size-distribution representations. ArXiv: Data Analysis, Statistics and Probability. 2019; 102 :052113. doi: 10.1103/PhysRevE.102.052113. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Creanza Nicole, Ruhlen Merritt, Pemberton Trevor J., Rosenberg Noah A., Feldman Marcus W., Ramachandran Sohini. A comparison of worldwide phonemic and genetic variation in human populations. Proceedings of the National Academy of Sciences of the United States of America. 2015; 112 :1265–72. doi: 10.1073/pnas.1424033112. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dale Rick, Lupyan Gary. Understanding the origins of morphological diversity: The linguistic niche hypothesis. Advances in Complex Systems. 2012; 15 :1150017. doi: 10.1142/S0219525911500172. [ CrossRef ] [ Google Scholar ]
• Darwin Charles. On the Origin of Species by Means of Natural Selection: Or the Preservation of the Favoured Races in the Struggle for Life. John Murray; London: 1860. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Darwin Charles. The Descent of Man, and Selection in Relation to Sex, Vol. 1. John Murray; London: 1871. [ CrossRef ] [ Google Scholar ]
• De Villiers Johan. Die Nederlandse era aan die Kaap, 1652–1806. In: Pretorius Fransjohan., editor. Geskiedenis van Suid-Afrika. Van Voortye Tot Vandag Bl. Tafelberg Publishers Ltd.; Cape Town: 2012. pp. 39–62. [ Google Scholar ]
• Dediu Dan, Boer Bart de. Language evolution needs its own journal. Journal of Language Evolution. 2016; 1 :1–6. doi: 10.1093/jole/lzv001. [ CrossRef ] [ Google Scholar ]
• Dediu Dan, Janssen Rick, Moisik Scott R. Language is not isolated from its wider environment: Vocal tract influences on the evolution of speech and language. Language & Communication. 2017; 54 :9–20. doi: 10.1016/j.langcom.2016.10.002. [ CrossRef ] [ Google Scholar ]
• Derrida Jacques. Of Grammatology, Trans. Gayatri Chakravorty Spivak. Johns Hopkins University Press; Baltimore: 1976. [ Google Scholar ]
• Diewald Gabriele. Pragmaticalization (Defined) as Grammaticalization of Discourse Functions. vol. 49. De Gruyter Mouton; Berlin: 2011. pp. 365–90. [ CrossRef ] [ Google Scholar ]
• Donohue Mark, Denham Tim. Farming and Language in Island Southeast Asia Reframing Austronesian History. Current Anthropology. 2010; 51 :223–56. doi: 10.1086/650991. [ CrossRef ] [ Google Scholar ]
• Dronkers Nina F. A new brain region for coordinating speech articulation. Nature. 1996; 384 :159–61. doi: 10.1038/384159a0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dunbar Robin I. M. Groups, Gossip, and the Evolution of Language. In: Schmitt Alain, Atzwanger Klaus, Grammer Karl, Schäfer Katrin., editors. New Aspects of Human Ethology. Springer; New York: 1997. pp. 77–89. [ CrossRef ] [ Google Scholar ]
• Erle Thorsten M., Schmid Karoline, Goslar Simon H., Martin Jared D. Emojis as social information in digital communication. Emotion. 2021; 22 :1529–1543. doi: 10.1037/emo0000992. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Everett Caleb, Blasi Damián E., Roberts Seán G. Climate, vocal folds, and tonal languages: Connecting the physiological and geographic dots. Proceedings of the National Academy of Sciences of the United States of America. 2015; 112 :1322–27. doi: 10.1073/pnas.1417413112. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Everett Caleb. Languages in Drier Climates Use Fewer Vowels. Frontiers in Psychology. 2017; 8 :1285. doi: 10.3389/fpsyg.2017.01285. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Favaro Livio, Gamba Marco, Cresta Eleonora, Fumagalli Elena, Bandoli Francesca, Pilenga Cristina, Isaja Valentina, Mathevon Nicolas, Reby David. Do penguins’ vocal sequences conform to linguistic laws? Biology Letters. 2020; 16 :20190589. doi: 10.1098/rsbl.2019.0589. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Feldman Laurie, Barach Eliza, Srinivasan Vidhushini, Shaikh Samira. Emojis and Words Work Together in the Service of Communication. Association for the Advancement of Artificial Intelligence; Palo Alto: 2021. [ Google Scholar ]
• Fitch William T. The Evolution of Language. Cambridge University Press; Cambridge: 2010. [ Google Scholar ]
• Formigari Lia. L’origine del linguaggio. Ricognizioni storiche e valenze epistemologiche. In: Emanuele B., editor. Sull’origine del Linguaggio e delle lingue storico-naturali. Un confronto tra linguisti e non linguisti. Bulzoni; Roma: 2013. pp. 13–22. [ Google Scholar ]
• Goldman Eric. Emojis and the Law. Washington Law Review. 2018; 93 :1227. [ Google Scholar ]
• Goody Jack, Watt Ian. The Consequences of Literacy. Comparative Studies in Society and History. 1963; 5 :304–45. doi: 10.1017/S0010417500001730. [ CrossRef ] [ Google Scholar ]
• Goody Jack. The Logic of Writing and the Organization of Society , 1st ed. Cambridge: Cambridge University Press. 1986 doi: 10.1017/CBO9780511621598. [ CrossRef ] [ Google Scholar ]
• Graffi Giorgio. Origin of language and origin of languages. Evolutionary Linguistic Theory. 2019; 1 :6–23. doi: 10.1075/elt.00002.gra. [ CrossRef ] [ Google Scholar ]
• Gray Russell D., Drummond Alexei J., Greenhill Simon J. Language Phylogenies Reveal Expansion Pulses and Pauses in Pacific Settlement. Science. 2009; 323 :479–83. doi: 10.1126/science.1166858. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gray Russell D., Atkinson Quentin D. Language-tree divergence times support the Anatolian theory of Indo-European origin. Nature. 2003; 426 :435–39. doi: 10.1038/nature02029. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Grigorenko Elena. The Never-Ending Innovativeness of the Wise Man. In: Preiss David D., Kaufman James C., Singer Marcos., editors. Innovation, Creativity and Change Across Cultures. Palgrave-Macmillan; London: 2023. [ Google Scholar ]
• Hagoort Peter. On Broca, brain, and binding: A new framework. Trends in Cognitive Sciences. 2005; 9 :416–23. doi: 10.1016/j.tics.2005.07.004. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hamans Camiel. Afrikaans: A language where ideology and linguistics meet. Scripta Neophilologica Posnaniensia. 2021; 21 :15–92. doi: 10.14746/snp.2021.21.02. [ CrossRef ] [ Google Scholar ]
• Hauser Mark D., Chomsky Noam, Fitch William T. The faculty of language: What is it, who has it, and how did it evolve? Science. 2002; 298 :1569–79. doi: 10.1126/science.298.5598.1569. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jackson Joshua Conrad, Lindquist Kristen, Drabble Ryan, Atkinson Quentin, Watts Joseph. Valence-dependent mutation in lexical evolution. Nature Human Behaviour. 2023; 7 :190–99. doi: 10.1038/s41562-022-01483-8. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Winters James, Kirby Simon, Smith Kenny. Languages adapt to their contextual niche. Language and Cognition. 2015; 7 :415–49. doi: 10.1017/langcog.2014.35. [ CrossRef ] [ Google Scholar ]
• Jordan Kirsten, Heinze H. -J., Lutz Kai, Kanowski Martin, Jäncke Lutz. Cortical activations during the mental rotation of different visual objects. Neuroimage. 2001; 13 :143–52. doi: 10.1006/nimg.2000.0677. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Koplenig Alexander. Language structure is influenced by the number of speakers but seemingly not by the proportion of non-native speakers. Royal Society Open Science. 2019; 6 :181274. doi: 10.1098/rsos.181274. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Labov William. Sociolinguistic Patterns. University of Philadelphia Press; Philadelphia: 1991. [ Google Scholar ]
• Lančarič Daniel. Sentential Acronyms in Informal Online Communication. Reviewed Conference Proceedings from an International Scientific Conference. 2016. [(accessed on 2 December 2022)]. pp. 12–17. Available online: https://faj.euba.sk/www_write/files/veda-vyskum/konferencie/zborniky/Cudzie_jazyky_v_premenach_casu_7_2016.pdf#page=10
• Lee Jungwoo, Kim Cheong, Lee Kun Chang. Investigating the Negative Effects of Emojis in Facebook Sponsored Ads for Establishing Sustainable Marketing in Social Media. Sustainability. 2021; 13 :4864. doi: 10.3390/su13094864. [ CrossRef ] [ Google Scholar ]
• Levinson Stephen C., Holler Judith. The origin of human multi-modal communication. Philosophical Transactions of the Royal Society B. 2014; 369 :20130302. doi: 10.1098/rstb.2013.0302. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lewis Molly, Frank Michael C. Linguistic niches emerge from pressures at multiple timescales. Cognitive Science. 2016. [(accessed on 25 November 2022)]. pp. 1385–90. Available online: https://langcog.stanford.edu/papers_new/lewis-2016-cogsci.pdf
• Lewis Paul M. Ethnologue: Languages of the World. Linguistic Society of America; New York: 2009. [ CrossRef ] [ Google Scholar ]
• Lindsey Delvin T., Brown Angela M. Color Naming and the Phototoxic Effects of Sunlight on the Eye. Psychological Science. 2002; 13 :506–12. doi: 10.1111/1467-9280.00489. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Linell Per. The Written Language Bias in Linguistics: Its Nature, Origins and Transformations. Routledge; London: 2004. [ Google Scholar ]
• Lock Andy, Gers Matt. Writing: A Mosaic of New Perspectives. Psychology Press; New York: 2012. The cultural evolution of written language and its effects: A Darwinian process from prehistory to the modern day; pp. 11–35. [ Google Scholar ]
• Lupyan Gary, Dale Rick. Language Structure Is Partly Determined by Social Structure. PLoS ONE. 2010; 5 :e8559. doi: 10.1371/journal.pone.0008559. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lupyan Gary, Dale Rick. Why Are There Different Languages? The Role of Adaptation in Linguistic Diversity. Trends in Cognitive Sciences. 2016; 20 :649–60. doi: 10.1016/j.tics.2016.07.005. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Mace Ruth, Holden Clare. A phylogenetic approach to cultural evolution. Trends in Ecology and Evolution. 2005; 20 :116–21. doi: 10.1016/j.tree.2004.12.002. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• MacNeilage Peter F. The Origin of Speech. Oxford University Press; Oxford: 2008. [ Google Scholar ]
• Maddieson I. Language Adapts to Environment: Sonority and Temperature. Frontiers in Communication. 2018; 3 :28. doi: 10.3389/fcomm.2018.00028. [ CrossRef ] [ Google Scholar ]
• Maddieson Ian, Coupé Christophe. Human spoken language diversity and the acoustic adaptation hypothesis. The Journal of the Acoustical Society of America. 2015; 138 :1838–38. doi: 10.1121/1.4933848. [ CrossRef ] [ Google Scholar ]
• Maess Burkhard, Koelsch Stefan, Gunter Thomas C, Friederici Angela D. Musical syntax is processed in Broca’s area: An MEG study. Nature Neuroscience. 2001; 4 :540–45. doi: 10.1038/87502. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McLuhan Marshall. The Gutenberg Galaxy: The Making of Typographic Man. University of Toronto Press; Toronto: 1962. [ Google Scholar ]
• McWhorter John. John McWhorter: Txtng is killing language. JK!!!|TED Talk. 2013. [(accessed on 28 November 2022)]. Available online: https://www.ted.com/talks/john_mcwhorter_txtng_is_killing_language_jk
• Mendívil-Giró José-Luis. Why Don’t Languages Adapt to Their Environment? Frontiers in Communication. 2018; 3 :24. doi: 10.3389/fcomm.2018.00024. [ CrossRef ] [ Google Scholar ]
• Meyer Julien, Magnasco Marcelo O., Reiss Diana. The Relevance of Human Whistled Languages for the Analysis and Decoding of Dolphin Communication. Frontiers in Psychology. 2021; 12 :689501. doi: 10.3389/fpsyg.2021.689501. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Meyer Julien. Whistled Languages. Springer; Berlin: 2015. [ CrossRef ] [ Google Scholar ]
• Meyer Julien. The Routledge Handbook of Language Revitalization. Routledge; London: 2018. Revitalization of Whistled Languages; p. 552. [ CrossRef ] [ Google Scholar ]
• Miton Helen, Morin Olivier. When iconicity stands in the way of abbreviation: No Zipfian effect for figurative signals. PLoS ONE. 2019; 14 :e0220793. doi: 10.1371/journal.pone.0220793. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Nettle Daniel, Romaine Suzanne. Vanishing Voices: The Extinction of the World’s Languages. Oxford University Press; Oxford: 2000. [ Google Scholar ]
• Nölle Jonas, Fusaroli Riccardo, Mills Gregory J., Tylén Kristian. Language as shaped by the environment: Linguistic construal in a collaborative spatial task. Palgrave Communications. 2020a; 6 :27. doi: 10.1057/s41599-020-0404-9. [ CrossRef ] [ Google Scholar ]
• Nölle Jonas, Hartmann Stefan, Tinits Peeter. Language Evolution Research in the Year 2020: A Survey of New Directions. Language Dynamics and Change. 2020b; 10 :3–26. doi: 10.1163/22105832-bja10005. [ CrossRef ] [ Google Scholar ]
• O’Shannessy Carmel, Brown Connor. Reflexive and Reciprocal Encoding in the Australian Mixed Language, Light Warlpiri. Langages. 2021; 6 :105. doi: 10.3390/languages6020105. [ CrossRef ] [ Google Scholar ]
• O’Shannessy Carmel. Light Warlpiri: A new language. Australian Journal of Linguistics. 2005; 25 :31–57. doi: 10.1080/07268600500110472. [ CrossRef ] [ Google Scholar ]
• O’Shannessy Carmel. The role of codeswitched input to children in the origin of a new mixed language. Linguistics. 2012; 50 :305–40. doi: 10.1515/ling-2012-0011. [ CrossRef ] [ Google Scholar ]
• O’Shannessy Carmel. The role of multiple sources in the formation of an innovative auxiliary category in Light Warlpiri, a new Australian mixed language. Language. 2013; 89 :328–53. doi: 10.1353/lan.2013.0025. [ CrossRef ] [ Google Scholar ]
• O’Shannessy Carmel. How ordinary child language acquisition processes can lead to the unusual outcome of a mixed language. International Journal of Bilingualism. 2020; 25 :458–80. doi: 10.1177/1367006920924957. [ CrossRef ] [ Google Scholar ]
• Obler Loraine K., Gjerlow Kris. Language and the Brain. Cambridge University Press; Cambridge: 1999. [ Google Scholar ]
• Olson David R. Intelligence and Technology. Routledge; London: 2005. Technology and intelligence in a literate society; pp. 79–92. [ Google Scholar ]
• Olson David R. Writing: A Mosaic of New Perspectives. Psychology Press; New York: 2012. Language, literacy and mind: The literacy hypothesis; pp. 3–10. [ Google Scholar ]
• Open Science Collaboration Estimating the reproducibility of psychological science. Science. 2015; 349 :aac4716. doi: 10.1126/science.aac4716. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Passmore Sam, Wood Anna L. C., Barbieri Chiara, Barbieri Chiara, Shilton Dor, Daikoku Hideo, Atkinson Quentin, Savage Patrick E. Global relationships between musical, linguistic, and genetic diversity. PsyArXiv. 2022 doi: 10.31234/osf.io/mdrsn. [ CrossRef ] [ Google Scholar ]
• Pearl Judea. Causality: Models, Reasoning, and Inference. Cambridge University Press; Cambridge: 2000. [ Google Scholar ]
• Pérez-Losada Joaquim, Fort Joaquim. A serial founder effect model of phonemic diversity based on phonemic loss in low-density populations. PLoS ONE. 2018; 13 :e0198346. doi: 10.1371/journal.pone.0198346. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perlman Marcus. Debunking two myths against vocal origins of language: Language is iconic and multimodal to the core. Interaction Studies. Social Behaviour and Communication in Biological and Artificial Systems. 2017; 18 :376–401. doi: 10.1075/is.18.3.05per. [ CrossRef ] [ Google Scholar ]
• Novak Petra Kralj, Smailović Jasmina, Sluban Borut, Mozetič Igor. Sentiment of Emojis. PLoS ONE. 2015; 10 :e0144296. doi: 10.1371/journal.pone.0144296. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pierini Francesco. Emojis and gestures: A new typology. Proceedings of Sinn Und Bedeutung. 2021; 25 :720–32. doi: 10.18148/sub/2021.v25i0.963. [ CrossRef ] [ Google Scholar ]
• Pinker Steven, Bloom Paul. Natural language and natural selection. Behavioral and Brain Sciences. 1990; 13 :707–27. doi: 10.1017/S0140525X00081061. [ CrossRef ] [ Google Scholar ]
• Pleyer Michael, Hartmann Stefen. Constructing a Consensus on Language Evolution? Convergences and Differences Between Biolinguistic and Usage-Based Approaches. Frontiers in Psychology. 2019; 10 :2537–37. doi: 10.3389/fpsyg.2019.02537. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Preiss David D. Human Intelligence in the Time of the Anthropocene. In: Sternberg Robert J., Preiss David D., editors. Intelligence in Context: The Cultural and Historical Foundations of Human Intelligence. Springer International Publishing; Berlin: 2022. pp. 361–91. [ CrossRef ] [ Google Scholar ]
• Preiss David D., Sternberg Robert J. Effects of technology on verbal and visual-spatial abilities. International Journal of Cognitive Technology. 2006; 11 :14–22. [ Google Scholar ]
• Prugnolle Franck, Manica Andrea, Balloux Francois. Geography predicts neutral genetic diversity of human populations. Current Biology. 2005; 15 :R159–R160. doi: 10.1016/j.cub.2005.02.038. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ramachandran Sohini, Deshpande Omkar, Roseman Charles C., Rosenberg Noah A., Feldman Marcus W., Cavalli-Sforza Luigi Luca. Support from the relationship of genetic and geographic distance in human populations for a serial founder effect originating in Africa. Proceedings of the National Academy of Sciences of the United States of America. 2005; 102 :15942–47. doi: 10.1073/pnas.0507611102. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rizzolatti Giacomo, Arbib Michael A. Language within our grasp. Trends in Neurosciences. 1998; 21 :188–94. doi: 10.1016/S0166-2236(98)01260-0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Roberts Sean G. Robust, Causal, and Incremental Approaches to Investigating Linguistic Adaptation. Frontiers in Psychology. 2018; 9 :166. doi: 10.3389/fpsyg.2018.00166. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Roberts Seán G., Killin Anton, Deb Angarika, Sheard Catherine, Greenhill Simon J., Sinnemäki Kaius, Segovia-Martín José, Nölle Jonas, Berdicevskis Aleksandrs, Humphreys-Balkwill Archie. CHIELD: The causal hypotheses in evolutionary linguistics database. Journal of Language Evolution. 2020; 5 :101–20. doi: 10.1093/jole/lzaa001. [ CrossRef ] [ Google Scholar ]
• Roberts Sean, Winters James. Social Structure and Language Structure: The New Nomothetic Approach. Psychology of Language and Communication. 2012; 16 :89–112. doi: 10.2478/v10057-012-0008-6. [ CrossRef ] [ Google Scholar ]
• Robinson Elizabeth, Goelman Hillel, Olson David R. Children’s understanding of the relation between expressions (what was said) and intentions (what was meant) British Journal of Developmental Psychology. 1983; 1 :75–86. doi: 10.1111/j.2044-835X.1983.tb00545.x. [ CrossRef ] [ Google Scholar ]
• Tagliamonte Sali A., Denis Derek. Linguistic ruin? Lol! Instant messaging and teen language. American Speech. 2008; 83 :3–34. doi: 10.1215/00031283-2008-001. [ CrossRef ] [ Google Scholar ]
• Saussure Ferdinand de, Bally Charles, Sechehaye Albert, Riedlinger Albert. Course in General Linguistics. Open Court Publishing; LaSalle: 1986. [ Google Scholar ]
• Schmandt-Besserat Denise. Writing: A Mosaic of New Perspectives. Psychology Press; New York: 2012. Tokens as precursors of writing; pp. 3–10. [ Google Scholar ]
• Semple Stuart, Hsu Minna J., Agoramoorthy Govindasamy. Efficiency of coding in macaque vocal communication. Biology Letters. 2010; 6 :469–71. doi: 10.1098/rsbl.2009.1062. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tomasello Michael. The usage-based theory of language acquisition. In: Bavin Edith L., editor. The Cambridge Handbook of Child Language. Cambridge University Press; Cambridge: 2009. pp. 69–88. [ CrossRef ] [ Google Scholar ]
• Varnhagen Connie K., McFall G. Peggy, Pugh Nicole, Routledge Lisa, Sumida-MacDonald Heather, Kwong Trudy E. lol: New language and spelling in instant messaging. Reading and Writing. 2010; 23 :719–33. doi: 10.1007/s11145-009-9181-y. [ CrossRef ] [ Google Scholar ]
• Webb Vic. Medium of Instruction Policies. Routledge; New York: 2003. Language Policy in Post-Apartheid South Africa. [ Google Scholar ]
• Wernicke Carl. A Psychological Study on an Anatomical Basis: The Aphasia Symptom-Complex. In: Eling Paul., editor. Reader in the History of Aphasia: From Franz Gall to Norman Geschwind. vol. 4. John Benjamins Publishing Company; Amsterdam: 1994. [(accessed on 17 March 2023)]. pp. 69–90. Classics in Psycholinguistics. First published 1874. Available online: https://benjamins.com/catalog/cipl.4.11the [ Google Scholar ]
• Zipf George Kingsley. Human Behavior and the Principle of Least Effort. Wiley; Hoboken: 1949. [ CrossRef ] [ Google Scholar ]
• Zlatev Jordan, Wacewicz Sławomir, Zywiczynski Przemyslaw, Weijer Joost van de. Multimodal-first or pantomime-first? Communicating events through pantomime with and without vocalization. Interaction Studies. Social Behaviour and Communication in Biological and Artificial Systems. 2017; 18 :465–88. doi: 10.1075/is.18.3.08zla. [ CrossRef ] [ Google Scholar ]

• Search Menu
• Browse content in Arts and Humanities
• Browse content in Archaeology
• Anglo-Saxon and Medieval Archaeology
• Archaeological Methodology and Techniques
• Archaeology by Region
• Archaeology of Religion
• Archaeology of Trade and Exchange
• Biblical Archaeology
• Contemporary and Public Archaeology
• Environmental Archaeology
• Historical Archaeology
• History and Theory of Archaeology
• Industrial Archaeology
• Landscape Archaeology
• Mortuary Archaeology
• Prehistoric Archaeology
• Underwater Archaeology
• Urban Archaeology
• Zooarchaeology
• Browse content in Architecture
• Architectural Structure and Design
• History of Architecture
• Residential and Domestic Buildings
• Theory of Architecture
• Browse content in Art
• Art Subjects and Themes
• History of Art
• Industrial and Commercial Art
• Theory of Art
• Biographical Studies
• Byzantine Studies
• Browse content in Classical Studies
• Classical History
• Classical Philosophy
• Classical Mythology
• Classical Literature
• Classical Reception
• Classical Art and Architecture
• Classical Oratory and Rhetoric
• Greek and Roman Epigraphy
• Greek and Roman Law
• Greek and Roman Papyrology
• Greek and Roman Archaeology
• Late Antiquity
• Religion in the Ancient World
• Digital Humanities
• Browse content in History
• Colonialism and Imperialism
• Diplomatic History
• Environmental History
• Genealogy, Heraldry, Names, and Honours
• Genocide and Ethnic Cleansing
• Historical Geography
• History by Period
• History of Emotions
• History of Agriculture
• History of Education
• History of Gender and Sexuality
• Industrial History
• Intellectual History
• International History
• Labour History
• Legal and Constitutional History
• Local and Family History
• Maritime History
• Military History
• National Liberation and Post-Colonialism
• Oral History
• Political History
• Public History
• Regional and National History
• Revolutions and Rebellions
• Slavery and Abolition of Slavery
• Social and Cultural History
• Theory, Methods, and Historiography
• Urban History
• World History
• Browse content in Language Teaching and Learning
• Language Learning (Specific Skills)
• Language Teaching Theory and Methods
• Browse content in Linguistics
• Applied Linguistics
• Cognitive Linguistics
• Computational Linguistics
• Forensic Linguistics
• Grammar, Syntax and Morphology
• Historical and Diachronic Linguistics
• History of English
• Language Acquisition
• Language Evolution
• Language Reference
• Language Variation
• Language Families
• Lexicography
• Linguistic Anthropology
• Linguistic Theories
• Linguistic Typology
• Phonetics and Phonology
• Psycholinguistics
• Sociolinguistics
• Translation and Interpretation
• Writing Systems
• Browse content in Literature
• Bibliography
• Children's Literature Studies
• Literary Studies (Asian)
• Literary Studies (European)
• Literary Studies (Eco-criticism)
• Literary Studies (Romanticism)
• Literary Studies (American)
• Literary Studies (Modernism)
• Literary Studies - World
• Literary Studies (1500 to 1800)
• Literary Studies (19th Century)
• Literary Studies (20th Century onwards)
• Literary Studies (African American Literature)
• Literary Studies (British and Irish)
• Literary Studies (Early and Medieval)
• Literary Studies (Fiction, Novelists, and Prose Writers)
• Literary Studies (Gender Studies)
• Literary Studies (Graphic Novels)
• Literary Studies (History of the Book)
• Literary Studies (Plays and Playwrights)
• Literary Studies (Poetry and Poets)
• Literary Studies (Postcolonial Literature)
• Literary Studies (Queer Studies)
• Literary Studies (Science Fiction)
• Literary Studies (Travel Literature)
• Literary Studies (War Literature)
• Literary Studies (Women's Writing)
• Literary Theory and Cultural Studies
• Mythology and Folklore
• Shakespeare Studies and Criticism
• Browse content in Media Studies
• Browse content in Music
• Applied Music
• Dance and Music
• Ethics in Music
• Ethnomusicology
• Gender and Sexuality in Music
• Medicine and Music
• Music Cultures
• Music and Religion
• Music and Media
• Music and Culture
• Music Education and Pedagogy
• Music Theory and Analysis
• Musical Scores, Lyrics, and Libretti
• Musical Structures, Styles, and Techniques
• Musicology and Music History
• Performance Practice and Studies
• Race and Ethnicity in Music
• Sound Studies
• Browse content in Performing Arts
• Browse content in Philosophy
• Aesthetics and Philosophy of Art
• Epistemology
• Feminist Philosophy
• History of Western Philosophy
• Metaphysics
• Moral Philosophy
• Non-Western Philosophy
• Philosophy of Science
• Philosophy of Language
• Philosophy of Mind
• Philosophy of Perception
• Philosophy of Action
• Philosophy of Law
• Philosophy of Religion
• Philosophy of Mathematics and Logic
• Practical Ethics
• Social and Political Philosophy
• Browse content in Religion
• Biblical Studies
• Christianity
• East Asian Religions
• History of Religion
• Judaism and Jewish Studies
• Qumran Studies
• Religion and Education
• Religion and Health
• Religion and Politics
• Religion and Science
• Religion and Law
• Religion and Art, Literature, and Music
• Religious Studies
• Browse content in Society and Culture
• Cookery, Food, and Drink
• Cultural Studies
• Customs and Traditions
• Ethical Issues and Debates
• Hobbies, Games, Arts and Crafts
• Lifestyle, Home, and Garden
• Natural world, Country Life, and Pets
• Popular Beliefs and Controversial Knowledge
• Sports and Outdoor Recreation
• Technology and Society
• Travel and Holiday
• Visual Culture
• Browse content in Law
• Arbitration
• Browse content in Company and Commercial Law
• Commercial Law
• Company Law
• Browse content in Comparative Law
• Systems of Law
• Competition Law
• Browse content in Constitutional and Administrative Law
• Government Powers
• Judicial Review
• Local Government Law
• Military and Defence Law
• Parliamentary and Legislative Practice
• Construction Law
• Contract Law
• Browse content in Criminal Law
• Criminal Procedure
• Criminal Evidence Law
• Sentencing and Punishment
• Employment and Labour Law
• Environment and Energy Law
• Browse content in Financial Law
• Banking Law
• Insolvency Law
• History of Law
• Human Rights and Immigration
• Intellectual Property Law
• Browse content in International Law
• Private International Law and Conflict of Laws
• Public International Law
• IT and Communications Law
• Jurisprudence and Philosophy of Law
• Law and Politics
• Law and Society
• Browse content in Legal System and Practice
• Courts and Procedure
• Legal Skills and Practice
• Primary Sources of Law
• Regulation of Legal Profession
• Medical and Healthcare Law
• Browse content in Policing
• Criminal Investigation and Detection
• Police and Security Services
• Police Procedure and Law
• Police Regional Planning
• Browse content in Property Law
• Personal Property Law
• Study and Revision
• Terrorism and National Security Law
• Browse content in Trusts Law
• Wills and Probate or Succession
• Browse content in Medicine and Health
• Browse content in Allied Health Professions
• Arts Therapies
• Clinical Science
• Dietetics and Nutrition
• Occupational Therapy
• Operating Department Practice
• Physiotherapy
• Radiography
• Speech and Language Therapy
• Browse content in Anaesthetics
• General Anaesthesia
• Neuroanaesthesia
• Browse content in Clinical Medicine
• Acute Medicine
• Cardiovascular Medicine
• Clinical Genetics
• Clinical Pharmacology and Therapeutics
• Dermatology
• Endocrinology and Diabetes
• Gastroenterology
• Genito-urinary Medicine
• Geriatric Medicine
• Infectious Diseases
• Medical Toxicology
• Medical Oncology
• Pain Medicine
• Palliative Medicine
• Rehabilitation Medicine
• Respiratory Medicine and Pulmonology
• Rheumatology
• Sleep Medicine
• Sports and Exercise Medicine
• Clinical Neuroscience
• Community Medical Services
• Critical Care
• Emergency Medicine
• Forensic Medicine
• Haematology
• History of Medicine
• Browse content in Medical Dentistry
• Oral and Maxillofacial Surgery
• Paediatric Dentistry
• Restorative Dentistry and Orthodontics
• Surgical Dentistry
• Browse content in Medical Skills
• Clinical Skills
• Communication Skills
• Nursing Skills
• Surgical Skills
• Medical Ethics
• Medical Statistics and Methodology
• Browse content in Neurology
• Clinical Neurophysiology
• Neuropathology
• Nursing Studies
• Browse content in Obstetrics and Gynaecology
• Gynaecology
• Occupational Medicine
• Ophthalmology
• Otolaryngology (ENT)
• Browse content in Paediatrics
• Neonatology
• Browse content in Pathology
• Chemical Pathology
• Clinical Cytogenetics and Molecular Genetics
• Histopathology
• Medical Microbiology and Virology
• Patient Education and Information
• Browse content in Pharmacology
• Psychopharmacology
• Browse content in Popular Health
• Caring for Others
• Complementary and Alternative Medicine
• Self-help and Personal Development
• Browse content in Preclinical Medicine
• Cell Biology
• Molecular Biology and Genetics
• Reproduction, Growth and Development
• Primary Care
• Professional Development in Medicine
• Browse content in Psychiatry
• Addiction Medicine
• Child and Adolescent Psychiatry
• Forensic Psychiatry
• Learning Disabilities
• Old Age Psychiatry
• Psychotherapy
• Browse content in Public Health and Epidemiology
• Epidemiology
• Public Health
• Browse content in Radiology
• Clinical Radiology
• Interventional Radiology
• Nuclear Medicine
• Radiation Oncology
• Reproductive Medicine
• Browse content in Surgery
• Cardiothoracic Surgery
• Gastro-intestinal and Colorectal Surgery
• General Surgery
• Neurosurgery
• Paediatric Surgery
• Peri-operative Care
• Plastic and Reconstructive Surgery
• Surgical Oncology
• Transplant Surgery
• Trauma and Orthopaedic Surgery
• Vascular Surgery
• Browse content in Science and Mathematics
• Browse content in Biological Sciences
• Aquatic Biology
• Biochemistry
• Bioinformatics and Computational Biology
• Developmental Biology
• Ecology and Conservation
• Evolutionary Biology
• Genetics and Genomics
• Microbiology
• Molecular and Cell Biology
• Natural History
• Plant Sciences and Forestry
• Research Methods in Life Sciences
• Structural Biology
• Systems Biology
• Zoology and Animal Sciences
• Browse content in Chemistry
• Analytical Chemistry
• Computational Chemistry
• Crystallography
• Environmental Chemistry
• Industrial Chemistry
• Inorganic Chemistry
• Materials Chemistry
• Medicinal Chemistry
• Mineralogy and Gems
• Organic Chemistry
• Physical Chemistry
• Polymer Chemistry
• Study and Communication Skills in Chemistry
• Theoretical Chemistry
• Browse content in Computer Science
• Artificial Intelligence
• Computer Architecture and Logic Design
• Game Studies
• Human-Computer Interaction
• Mathematical Theory of Computation
• Programming Languages
• Software Engineering
• Systems Analysis and Design
• Virtual Reality
• Browse content in Computing
• Business Applications
• Computer Security
• Computer Games
• Computer Networking and Communications
• Digital Lifestyle
• Graphical and Digital Media Applications
• Operating Systems
• Browse content in Earth Sciences and Geography
• Atmospheric Sciences
• Environmental Geography
• Geology and the Lithosphere
• Maps and Map-making
• Meteorology and Climatology
• Oceanography and Hydrology
• Palaeontology
• Physical Geography and Topography
• Regional Geography
• Soil Science
• Urban Geography
• Browse content in Engineering and Technology
• Agriculture and Farming
• Biological Engineering
• Civil Engineering, Surveying, and Building
• Electronics and Communications Engineering
• Energy Technology
• Engineering (General)
• Environmental Science, Engineering, and Technology
• History of Engineering and Technology
• Mechanical Engineering and Materials
• Technology of Industrial Chemistry
• Transport Technology and Trades
• Browse content in Environmental Science
• Applied Ecology (Environmental Science)
• Conservation of the Environment (Environmental Science)
• Environmental Sustainability
• Environmentalist Thought and Ideology (Environmental Science)
• Management of Land and Natural Resources (Environmental Science)
• Natural Disasters (Environmental Science)
• Nuclear Issues (Environmental Science)
• Pollution and Threats to the Environment (Environmental Science)
• Social Impact of Environmental Issues (Environmental Science)
• History of Science and Technology
• Browse content in Materials Science
• Ceramics and Glasses
• Composite Materials
• Metals, Alloying, and Corrosion
• Nanotechnology
• Browse content in Mathematics
• Applied Mathematics
• Biomathematics and Statistics
• History of Mathematics
• Mathematical Education
• Mathematical Finance
• Mathematical Analysis
• Numerical and Computational Mathematics
• Probability and Statistics
• Pure Mathematics
• Browse content in Neuroscience
• Cognition and Behavioural Neuroscience
• Development of the Nervous System
• Disorders of the Nervous System
• History of Neuroscience
• Invertebrate Neurobiology
• Molecular and Cellular Systems
• Neuroendocrinology and Autonomic Nervous System
• Neuroscientific Techniques
• Sensory and Motor Systems
• Browse content in Physics
• Astronomy and Astrophysics
• Atomic, Molecular, and Optical Physics
• Biological and Medical Physics
• Classical Mechanics
• Computational Physics
• Condensed Matter Physics
• Electromagnetism, Optics, and Acoustics
• History of Physics
• Mathematical and Statistical Physics
• Measurement Science
• Nuclear Physics
• Particles and Fields
• Plasma Physics
• Quantum Physics
• Relativity and Gravitation
• Semiconductor and Mesoscopic Physics
• Browse content in Psychology
• Affective Sciences
• Clinical Psychology
• Cognitive Psychology
• Cognitive Neuroscience
• Criminal and Forensic Psychology
• Developmental Psychology
• Educational Psychology
• Evolutionary Psychology
• Health Psychology
• History and Systems in Psychology
• Music Psychology
• Neuropsychology
• Organizational Psychology
• Psychological Assessment and Testing
• Psychology of Human-Technology Interaction
• Psychology Professional Development and Training
• Research Methods in Psychology
• Social Psychology
• Browse content in Social Sciences
• Browse content in Anthropology
• Anthropology of Religion
• Human Evolution
• Medical Anthropology
• Physical Anthropology
• Regional Anthropology
• Social and Cultural Anthropology
• Theory and Practice of Anthropology
• Browse content in Business and Management
• Business Strategy
• Business Ethics
• Business History
• Business and Government
• Business and Technology
• Business and the Environment
• Comparative Management
• Corporate Governance
• Corporate Social Responsibility
• Entrepreneurship
• Health Management
• Human Resource Management
• Industrial and Employment Relations
• Industry Studies
• Information and Communication Technologies
• International Business
• Knowledge Management
• Management and Management Techniques
• Operations Management
• Organizational Theory and Behaviour
• Pensions and Pension Management
• Public and Nonprofit Management
• Strategic Management
• Supply Chain Management
• Browse content in Criminology and Criminal Justice
• Criminal Justice
• Criminology
• Forms of Crime
• International and Comparative Criminology
• Youth Violence and Juvenile Justice
• Development Studies
• Browse content in Economics
• Agricultural, Environmental, and Natural Resource Economics
• Asian Economics
• Behavioural Finance
• Behavioural Economics and Neuroeconomics
• Econometrics and Mathematical Economics
• Economic Systems
• Economic History
• Economic Methodology
• Economic Development and Growth
• Financial Markets
• Financial Institutions and Services
• General Economics and Teaching
• Health, Education, and Welfare
• History of Economic Thought
• International Economics
• Labour and Demographic Economics
• Law and Economics
• Macroeconomics and Monetary Economics
• Microeconomics
• Public Economics
• Urban, Rural, and Regional Economics
• Welfare Economics
• Browse content in Education
• Adult Education and Continuous Learning
• Care and Counselling of Students
• Early Childhood and Elementary Education
• Educational Equipment and Technology
• Educational Strategies and Policy
• Higher and Further Education
• Organization and Management of Education
• Philosophy and Theory of Education
• Schools Studies
• Secondary Education
• Teaching of a Specific Subject
• Teaching of Specific Groups and Special Educational Needs
• Teaching Skills and Techniques
• Browse content in Environment
• Applied Ecology (Social Science)
• Climate Change
• Conservation of the Environment (Social Science)
• Environmentalist Thought and Ideology (Social Science)
• Natural Disasters (Environment)
• Social Impact of Environmental Issues (Social Science)
• Browse content in Human Geography
• Cultural Geography
• Economic Geography
• Political Geography
• Browse content in Interdisciplinary Studies
• Communication Studies
• Museums, Libraries, and Information Sciences
• Browse content in Politics
• African Politics
• Asian Politics
• Chinese Politics
• Comparative Politics
• Conflict Politics
• Elections and Electoral Studies
• Environmental Politics
• European Union
• Foreign Policy
• Gender and Politics
• Human Rights and Politics
• Indian Politics
• International Relations
• International Organization (Politics)
• International Political Economy
• Irish Politics
• Latin American Politics
• Middle Eastern Politics
• Political Methodology
• Political Communication
• Political Philosophy
• Political Sociology
• Political Behaviour
• Political Economy
• Political Institutions
• Political Theory
• Politics and Law
• Public Administration
• Public Policy
• Quantitative Political Methodology
• Regional Political Studies
• Russian Politics
• Security Studies
• State and Local Government
• UK Politics
• US Politics
• Browse content in Regional and Area Studies
• African Studies
• Asian Studies
• East Asian Studies
• Japanese Studies
• Latin American Studies
• Middle Eastern Studies
• Native American Studies
• Scottish Studies
• Browse content in Research and Information
• Research Methods
• Browse content in Social Work
• Addictions and Substance Misuse
• Adoption and Fostering
• Care of the Elderly
• Child and Adolescent Social Work
• Couple and Family Social Work
• Developmental and Physical Disabilities Social Work
• Direct Practice and Clinical Social Work
• Emergency Services
• Human Behaviour and the Social Environment
• International and Global Issues in Social Work
• Mental and Behavioural Health
• Social Justice and Human Rights
• Social Policy and Advocacy
• Social Work and Crime and Justice
• Social Work Macro Practice
• Social Work Practice Settings
• Social Work Research and Evidence-based Practice
• Welfare and Benefit Systems
• Browse content in Sociology
• Childhood Studies
• Community Development
• Comparative and Historical Sociology
• Economic Sociology
• Gender and Sexuality
• Gerontology and Ageing
• Health, Illness, and Medicine
• Marriage and the Family
• Migration Studies
• Occupations, Professions, and Work
• Organizations
• Population and Demography
• Race and Ethnicity
• Social Theory
• Social Movements and Social Change
• Social Research and Statistics
• Social Stratification, Inequality, and Mobility
• Sociology of Religion
• Sociology of Education
• Sport and Leisure
• Urban and Rural Studies
• Browse content in Warfare and Defence
• Defence Strategy, Planning, and Research
• Land Forces and Warfare
• Military Administration
• Military Life and Institutions
• Naval Forces and Warfare
• Other Warfare and Defence Issues
• Peace Studies and Conflict Resolution
• Weapons and Equipment

• < Previous chapter
• Next chapter >

1 The Origins and the Evolution of Language

Salikoko S. Mufwene is Frank J. McLoraine Distinguished Service Professor of Linguistics and the College at the University of Chicago, where he also serves as Professor on the Committee on Evolutionary Biology and on the Committee on the Conceptual and Historical Studies of Science. His current research is on language evolution, including the indigenization of English and other colonial European languages worldwide. He is the author of The Ecology of Language Evolution (Cambridge University Press 2001), Créoles , Écologie Sociale , Évolution Linguistique (l’Harmattan 2005), and Language Evolution: Contact, Competition and Change (Continuum Press 2008); and the (co-)editor of Africanisms in Afro-American Language Varieties (UGA Press 1993) and African American English: Structure, History and Use (Routledge 1998). He edits the book series Cambridge Approaches to Language Contact .

• Published: 01 July 2013
• Cite Icon Cite
• Permissions Icon Permissions

This chapter is a selective comparative and critical survey of speculations/hypotheses, since Antiquity, on the phylogenetic emergence of language in mankind. It highlights topics and explanations that have been recurring, how some of them have been refined and/or enriched by modern thinking about hominin evolution since Charles Darwin’s account of the process by natural selection under specific ecological pressures. It also shows how some questions have been shaped by the manifold evolution of linguistics itself since the nineteenth century, including variation on what counts as language, and by intellectual exchanges between linguistics and other disciplines such as primatology, neurology, and paleontology. It concludes with an itemization of accomplishments, after articulating a long list of question-begging accounts and still unanswered questions.

1.1 Introduction 1

Although ‘language evolution’ is perhaps more commonly used in linguistics than ‘evolution of language’, I stick in this chapter to the latter term, which focuses more specifically on the phylogenetic emergence of language. The former, which has prompted some linguists such as Croft ( 2008 ) to speak of ‘evolutionary linguistics’, applies also to changes undergone by individual languages over the past 6,000 years of documentary history, including structural changes, language speciation, and language birth and death. There are certainly advantages in using the broader term, especially to uniformitarians who argue that some of the same evolutionary mechanisms are involved in both the phylogenetic and the historical periods of evolution. For instance, natural selection driven by particular ecological pressures putatively applies in both periods, and social norms emerge by the same principle of the ‘invisible hand’ or ‘self-organization’ (e.g. Hurford 2006 , Mufwene 2008 ). However, I focus here only on phylogenetic evolution.

In this chapter I provide a selective history, since antiquity, of this complex but still largely speculative topic which, over the past two decades alone, has prompted numerous publications and has aroused much controversy among linguists and informative exchanges between them, primatologists, psycholinguists, anthropologists, neurolinguists, evolutionary biologists, paleontologists, and computational linguists. This intellectual engagement has been in sharp contrast with most of the twentieth century, during which linguists appear to have abided by the the Société de Linguistique de Paris' 1866 ban on discussing the subject at its meetings. It appears also to have resurrected several positions by—and controversies among—especially eighteenth- and nineteenth-century European philosophers and philologists. I show below that the differences between the two periods lie especially in the stronger empirical foundations of some recent hypotheses, and in the realization by their authors of the need for interdisciplinary scholarship.

My discussion is organized around the following questions (which do not necessarily determine the section structure of the chapter):

Was language given to humans by God or did it emerge by Darwinian evolution?

From a phylogenetic perspective, did language emerge abruptly or gradually? If the emergence of language was protracted, what plausible intermediate stages can be posited and what would count as evidence for positing them? Assuming that the structure of modern languages is modular, would gradual evolution apply to any of the modules, only to some of them, or only to the overall architecture? What is the probable time of the emergence of the first real ancestor of modern language?

Does possessing Language, the non-individuated construct associated exclusively with humans, presuppose monogenesis or does it allow for polygenesis? How consistent is either position with paleontological evidence about the evolution of the Homo genus? How did linguistic diversity start? Assuming Darwinian (variational rather than transformational) evolution, can monogenesis account for typological variation as plausibly as polygenesis?

What is the chronological relationship between communication and language? What light does this distinction shed on the relation between sign(ed) and spoken language? Did some of our hominin ancestors communicate by means of ape-like vocalizations and gestures? If so, how can we account for the transition from them to phonetic and signed languages? And how can we account for the fact that modern humans have favoured speaking over signing? Assuming that language is a communication technology, to what extent are some of the structural properties of languages consequences of the linearity imposed by the phonic and signing devices used in their architecture?

Is the evolution of language more biological than cultural? Or is it the other way around, or equally both? Are languages as cultural artifacts deliberate inventions or emergent phenomena? Who are the agents in the emergence of language: individuals or populations, or both?

What is the relationship between language and thought? Did these entities co-evolve or did one cause the other?

Is there such a thing as ‘language organ’ or ‘biological endowment for language’? How can it be characterized relative to modern humans' anatomical and/or mental makeups? What are the anatomical, mental, and social factors that facilitated the emergence of language?

Can we learn something about the evolution of language from historical language change, especially from the emergence of creoles and pidgins? Can we learn something from child language and/or from home sign language? And what can be learned from ‘linguistic apes’? Does it make sense to characterize these particular communicative ‘systems’ as fossils of the human protolanguage (cf. e.g. Bickerton 1990 )? In the same vein, what can modelling contribute to understanding the evolution of language. This is definitely the kind of thing that scholars could not do before the twentieth century; it is important to assess its heuristic significance.

As noted by Kirby ( 2007 ), the subject matter of the origins and evolution of language is very complex. It lies at the intersection of several academic disciplines and requires an interdisciplinary approach. I have listed all the above questions, which are still but a subset of the larger range of questions one can address in a book, so that the reader may empathize with the daunting task I have accepted in writing this synopsis, and appreciate the synthetic approach I adopt in focusing on noteworthy positions and issues, aiming at the big picture. Unfortunately, this strategy entails omitting many equally relevant references, aside from forcing me to be topically selective. The positions of the scholars I discuss may not even be presented in their entirety, due largely to space limitations. More interested readers are encouraged to read recent publications such as Fitch ( 2010 ) and Hombert and Lenclud (in press) for complementary and/or alternative accounts. I must also apologize for focusing exclusively on Western scholarship, which reflects my embarrassing ignorance of the other traditions. I will seek no excuse for the fact that European colonial expansion, which has shaped me intellectually, has generally downplayed what we could be learning from the other scholarly traditions.

1.2 A Historical Synopsis

Speculations about the origins of language and linguistic diversity date from far back in the history of mankind. Among the most cited cases is the book of Genesis, in the Judeo-Christian Bible. After God created Adam, He reportedly gave him authority to name every being that was in the Garden of Eden. Putatively, God and Adam spoke some language, the original language, which some scholars have claimed to be Hebrew, the original language of Bible. Adam named every entity God wanted him to know; and his wife and descendants accordingly learned the names he had invented.

Although the story suggests the origin of naming conventions, it says nothing about whether Adam named only entities or also actions and states. In any case, it suggests that it was necessary for Adam's wife and descendants to learn the same vocabulary to facilitate successful communication.

Up to the eighteenth century, reflecting the impact of Christianity, pre-modern Western philosophers and philologists typically maintained that language was given to mankind, or that humans were endowed with language upon their creation. Assuming that Eve, who was reportedly created from Adam's rib, was equally endowed with (a capacity for) language, the rest was a simple history of learning the original vocabulary or language. Changes needed historical accounts, grounded in natural disasters, in population dispersals, and in learning with modification, to which I return below.

The book of Genesis also deals with the origin of linguistic diversity, in the myth of the Tower of Babel (11: 5–8), in which the multitude of languages is treated as a form of punishment from God. According to the myth, the human population had already increased substantially, generations after the Great Deluge in the Noah's Ark story. To avoid being scattered around the world, they built a city with a tower tall enough to reach the heavens, the dwelling of God. The tower apparently violated the population structure set up at the creation of Adam and Eve. God brought them down (according to some versions, He also destroyed the tower), dispersed them around the world, and confounded them by making them speak in mutually unintelligible ways. Putatively, this is how linguistic diversity began. 2 The story suggests that sharing the same language fosters collaboration, contrary to some of the modern Darwinian thinking that joint attention and cooperation, rather than competition, facilitated the emergence of language (see e.g. Tomasello 2008 ).

Another story often reported in linguistics is the following:

According to Herodotus ( Histories 2.2) Pharaoh Psammetichus I [also known as Psamtik, of the 26th dynasty, seventh century bc ] wanted to determine the oldest nation and establish the world's original language. For this purpose, he ordered two children to be reared by a shepherd, forbidding him to let them hear a single word, and charging him to report the children's first utterance. After two years, the shepherd reported that on entering their chamber, the children came up to him, extending their hands, calling bekos . Upon enquiry, the pharaoh discovered that this was the Phrygian word for ‘wheat bread’, after which the Egyptians conceded that the Phrygian nation was older than theirs. ( Wikipedia , Jan. 2011)

The story may be interpreted to suggest monogenesis, according to which a single language was the ultimate ancestor of all modern languages. This would correspond to a protolanguage, such as proto-Bantu or proto-Indo-European, in genetic linguistics. However, this is not the theme we find in Plato's Cratylus , which focuses on how the first words emerged (in Greek). According to the dialogue with two disciples, Cratylus and Hermogenes, Socrates (the teacher and Plato's mouthpiece) claims that names originally captured the essence of the entities they denote; transmission from generation to generation has affected their transparency, making them (rather) opaque, reducing them to conventional, arbitrary signs. Opaqueness is accordingly more obvious in words borrowed from other languages, then considered ‘barbarous,’ especially since their roots are harder to trace. Socrates' comparison of the putative initial baptismal practice with the work of a painter makes his account a precursor of modern synesthetic approach, as he associates particular sounds with specific meanings. He thus anticipated some eighteenth- and nineteenth-century philologists who saw the origins of language in ‘natural sounds’ produced by animals and other entities in nature.

Anticipating Johann Gottfried Herder, Socrates rejects the hypothesis that names had divine origins because, according to him, they are so imperfect that they could not have been made by the gods. The Cratylus is also one of the earliest works that associate language change with imperfect learning and language contact. The latter phenomenon complicates the evolutionary trajectories of particular languages, which, in contemporary metalanguage, need not be considered as unilinear.

Recently, the significance of population movements and language contacts in the evolution and diversification of languages has been underscored especially by Cavalli-Sforza ( 2000 ). Assuming that the exodus of Homo sapiens sapiens out of East Africa was protracted, he argues that some of the later migrant populations came in contact with earlier ones. Though he says nothing about monogenesis vs polygenesis, the idea appears to be that the original language changed as human populations migrated away from the homeland. Later contacts between the dispersing populations produced even more changes. No more reason other than population dispersal is given for the change, which is also problematic in typical accounts of speciation in language families such as Bantu and Indo-European.

The dominant trend in genetic linguistics, which inspired Cavalli-Sforza ( 2000 ) but had been disputed by Trubetzkoy ( 1939 ), has indeed been for monogenesis, positing a protolanguage from which all the members of a language family can be derived. This account of the evolution of language has also been adopted in particular by Ruhlen ( 1994 ), who attempted to reconstruct the ultimate phylogenetic protolanguage since Homo sapiens , on the model of proto-Indo-European or proto-Bantu. This ‘protolanguage,’ identified by some as ‘proto-world,’ should not be confused with the ‘protolanguage’ (without a hyphen) posited by Bickerton (e.g. 1990) and discussed below.

Writing in the first century bc , the Roman poet and philosopher Titus Lucretius Carus questioned one particular brand of monogenesis that is not necessarily Adamic:

[…] to think that one individual then distributed names to things and that humans learned the first words from him is absurd. For why would he be able to mark everything with utterances and emit different sounds of the tongue, and at the same time others not being capable of having done it? Besides, if others too had not used their voices with one another, from where was the notion of utility implanted, and from where was this power first granted to him, to know what he wanted to do and conceive of it in his mind? Similarly, one person could not have prevailed and forced so many to want to learn the names of things so thoroughly […] (Lucretius Carus 2003 [?54 bc ]: ll. 1041–51).

Lucretius thereby suggests that language emerged and evolved from the collective communicative acts of individuals interacting with each other. We may, in modern terms, think of different interactants innovating on different occasions and the successful innovations being copied by others. This is the position articulated by Michel Bréal in the late nineteenth century (see below), in contrast with the vast majority of scholars who have simply ignored the question.

There doesn't seem to have been much speculation on the origins of language since Lucretius until the eighteenth century, ‘the (Age of) Enlightenment.’ The contribution of the Renaissance period appears to be negligible, as the focus was on (the logic of) the structure of language, epitomized by the Port-Royal Grammar, published in 1660 by Claude Lancelot and Antoine Arnauld. It's not evident what the reason for this return to the subject matter of the origins of language was, except perhaps that the post-Renaissance social philosophers, so interested in defending the natural rights of people and freeing fellow citizens from superstition and the creationist dogma of Christianity, may have wanted also to have a better understanding of the origins of mankind. Convinced that rationality distinguishes mankind from other animals, they were interested in the apparent chicken-and-egg connection between humans' mental capacity and language.

A name that was particularly influential in the eighteenth century was Étienne Bonnot de Condillac, who, according to Aarsleff ( 1982 ), then launched debates on the origins of language with his Essai sur l'origine des connoissances humaines ( 1746 ). He argued that language is a consequence of humans' being rational and needing this tool to express their thoughts. Although he saw language as constrained by its phonetic architecture to linearize thought, he also claimed that language gives more structure to thought processes and is the foundation of (the growth of) human knowledge. This sounds similar to Bickerton's (e.g. 1990 ) claim that the emergence of language, especially syntax, enhanced human capacity for thought (see below). 3

Contrary to the received doctrine of the Catholic Church, the dominant one at his time, Condillac, an abbot, concluded that language was man-made, the product of humans' capacity for creative thought, and not God-given, a position adopted by other eighteenth century philosophers. He is also reported to have contributed to, if not started, the hypothesis that language emerged from natural cries. Although it would be derided by Friedrich Max Müller in the nineteenth century (see below), this position addresses the question of how humans evolved from the mere production of ‘natural cries,’ identified today as holistic vocalizations, to phonetic ones, which Condillac characterized as ‘vocal signs,’ at least according to Aarsleff ( 1982 ). This is a question that still awaits a conclusive answer (see esp. Wray 2002 , Tallerman 2007 , and Bickerton 2010 ) and on which MacNeilage ( 2008 ) contributes some significant insights (see below).

The hypothesis that the original ancestor of language lies in the natural cries and gestures was also developed by Jean-Jacques Rousseau in his 1755 essay on the origin of language, in Discours sur l'origine et les fondements de l'inégalité parmi les hommes . For him, cries and gestures are the language most expressive of humans' passions, which dominated in the earliest phylogenetic stages of mankind. The evidence can allegedly still be found in ‘savage’ or less advanced populations, particularly in southerly, warmer climates, where humans are, according to him, closer to nature. It is not that those populations are still in the primordial or less evolved stages of human evolution,

the order of their progress is different. In southern climates, where nature is bountiful, needs are born of passion. In cold countries, where she is miserly, passions are born of need, and the languages, sad daughters of necessities, reflect their austere origin. ( 1755 ; Moran and Gode's translation, 1966 : 46)

According to Rousseau, the passions are still best expressed through tones (and intonation) and gestures, and thus in tonal languages. However, ‘while visible signs can render a more exact imitation, sounds more effectively arouse interest’ (Moran and Gode 1966 : 9), which is why, as communication became less and less passionate and more and more referential/rational, speech prevailed as a means of communication. Like most philosophers and philologists of the eighteenth century, Rousseau did not realize that tones play a contrastive lexical and/or grammatical function in many languages, although this is not the case in most European languages. On the other hand, like some modern students of the origins of language (e.g. Tomasello 2008 , MacNeilage 2008 , Corballis 2010 , Dor and Jablonka 2010 , Mufwene 2010b ), Rousseau also assumed that modern language emerged under social ecological pressures, especially out of the need to help each other understand what they had to do in order to survive danger (pp. 47–8). On the other hand, unlike today's scholars, Rousseau interpreted evolution as progress towards a more explicit architecture meant to express reason more than emotion. According to him,

Anyone who studies the history and progress of tongues will see that the more words become monotonous, the more consonants multiply; that, as accents fall into disuse and quantities are neutralized, they are replaced by grammatical combinations and new articulations. […] To the degree that needs multiply […] language changes its character. It becomes more regular and less passionate. It substitutes ideas for feelings. It no longer speaks to the heart but to reason. (Moran and Gode 1966 : 16)

Thus, Rousseau interpreted the evolution of language as gradual, reflecting changes in the Homo genus's mental, social, and environmental structures. He also suggests that consonants emerged after vowels (at least some of them), out of necessity to keep ‘words’ less ‘monotonous.’ Consonants would putatively have made it easier to identify transitions from one syllable to another. He speaks of ‘break[ing] down the speaking voice into a given number of elementary parts, either vocal or articulate [i.e. consonantal?], with which one can form all the words and syllables imaginable’ (p. 17). This account appears to anticipate Peter MacNeilage's notion of ‘syllabic variegation’ (see below).

Like his contemporaries and predecessors, Rousseau did not (always) distinguish sounds from the letters, but he also had curious positions about the latter. He associated pictographic writing with ‘a savage people, signs of words and propositions [with] a barbaric people, and the alphabet [with] civilized peoples’ (p. 17).

This stratification of populations was a common belief until the early twentieth century (see below). However, it is not out of place to discuss, in the context of the evolution of language and of writing systems as technology designed to overcome some of the shortcomings of speech and signing. Writing does not just extend our capacity to remember and carry to longer distances what was or could have been spoken or signed. For instance, Chinese ideograms are additionally efficient in enabling speakers of mutually unintelligible Sinitic language varieties to understand each other. In this particular respect, they also illustrate why evolution should not be thought of in rectilinear and unilinear terms, as there is room for variation. While alphabetic writing systems, designed to capture speech, may be preferred (by the principles of economy and productivity) for their simplicity, they cannot accomplish the role Chinese ideograms play in bridging dialectal differences with regard to meaning. Scholars who think of language as technology (Smith and Szathmáry 1999, Lee et al. 2009 , Mufwene 2010a ) will hail Rousseau for bringing writing as derivative technology into the picture. It's undoubtedly also relevant to ask to what extent writing has influenced language evolution during the historical period (Wang 2011 ).

Rousseau questioned the Adamic hypothesis on the origins of modern language, arguing that the language that God had taught Adam and was learned by the children of Noah perished after the latter abandoned agriculture and scattered. Modern language is therefore a new invention (Moran and Gode 1966 : 36). Rousseau may have been concerned more about the diversification of the language that Noah's children had spoken before they dispersed than about the origins of language itself. He assumed the speciation to have happened before the Tower of Babel explanation in the Judeo-Christian tradition.

Language diversification is a topic that has not been sufficiently discussed in today's literature on the evolution of language(s). The focus has typically been language as a common endowment of all humans, thus obviating the question of whether the origins of modern languages were monogenetic or polygenetic. If they evolved ultimately from one language, was this original language internally variable or not? Accounts of how linguistic diversity emerged should vary, depending on whether one assumes monogenesis without internal variation or polygenesis with the possibility of variation from one hominin colony to another.

It is thus noteworthy that, unlike most of his contemporaries and somewhat anticipating variational evolutionary theory, Rousseau also addressed the question of the consequences of inter-idiolectal variation in the emergence of language as a communal phenomenon:

[E]ach individual is unique, possessed of, even in some ways identical with, his own nature or ‘essence’ while participating in the whole of nature, the whole of reality, so speak. In so far as there is plurality of individuals, and one individual (or group) practices any of the arts on others, there is a basis for contrasting nature (the nature of one) and art (the art of another). (Moran and Gode 1966 : 76)

In modern terms, every idiolect differs from others. This situation raises the interesting question of how they converge toward the same communal norm (Mufwene 2008 , 2010b ). Does normalization as emergence of a communal norm entail elimination, or just reduction, of variation? What does it really mean when two or more individuals are said to speak the same language? One should also ask: what role has inter-idiolectal variation played in the evolution of language?

A contemporary of Jean-Jacques Rousseau, the German philologist Johann Gottfried Herder contributed to the debates on some of the above issues, with his Über den ursprung der Sprache (1772), translated and published in Moran and Gode ( 1966 ) as Essay on the origin of language (which is cited here). Herder especially argued that human language was not God-given, and that it started in animal communication (p. 94). Like Lucretius, he thought that even Hebrew, assumed then to be the oldest language, was too imperfect to be God's creation (pp. 94, 96), though he could have made allowance for change, which normally disturbs the original design, over time. Likewise, he observed:

Now trace, if you can, divine order in the fact that a god, who saw the plan of language as a whole, invented seventy words for the stone and none for all the indispensable ideas, innermost feelings, and abstractions, that in one case he drowned us in unnecessary abundance while leaving us in the other in the direst need which obliged us to steal and usurp metaphors and talk half nonsense, etc. (p. 153)

The distribution of the vocabulary within and across languages appeared to Herder to be too inconsistent for the latter to be God's creation(s). Like Rousseau, he concluded that such varying reality could only reflect the work of mankind.

Herder was ambivalent about the origins of language. On the one hand, he argued against Rousseau's and Condillac's position that it evolved from emotional cries (p. 102). On the other, he admitted that it may have started as animal-like cries, with the difference that human utterances in the form of speech are volitional and driven by reason (p. 99). He concludes several pages later that early human language ‘was an expression of the language of all creatures within the natural scale of the human voice’ (p. 137).

Herder also argued that knowledge of particular languages is not instinctive; the child learns the language of its social environment. Anticipating modern linguists, he clarified that what is being discussed is the capacity for language, what Ferdinand de Saussure referred to as the faculté de langage and generativists as Universal Grammar or biological endowment for language. He observed that this capacity, which is also shared by the deaf (p. 118), enables humans to learn naturalistically, through interactions or by immersion, whatever language they have been exposed to. This of course leaves unanswered the question of how in the first place this particular capacity for language evolved in mankind and in what form. It also leaves open the question of how particular languages displaying both structural diversity and common/universal features evolved (see below).

Herder also speculated that language started with the practice of naming. He claimed that predicates, which denote activities and conditions, were the first names; nouns were derived from them (pp. 132, 160). He thus partly anticipated Heine and Kuteva ( 2007 ), who argue that grammar emerged gradually, through the grammaticization of nouns and verbs into grammatical markers, including complementizers, which make it possible to form complex sentences. An issue arising from Herder's position is whether nouns and verbs could not have emerged concurrently. Not quite in the same way, Allan ( 2010a : 230) comments that Herder was more concerned with proving that ‘God could not have invented human language because, as the Western Classical Tradition affirms, the logical order is to name entities first and then predicate acts and attributes of them.’

On the other hand, as hypothesized by William Dwight Whitney (discussed below), the original naming practice need not have entailed the distinction between nouns and verbs and the capacity to predicate. At that time, naming may have amounted to pointing with (pre-)linguistic signs; predication may have started only after hominins were capable of describing states of affairs compositionally, combining word-size units in this case, rather than holophrastically. This issue cannot be addressed independently of what Bickerton's ( 1990 ) ‘protolanguage’ is and when it may have emerged. The question of the order in which other grammatical categories emerged remains open, there being no conclusive evidence in support of the particular order proposed by Heine and Kuteva ( 2007 ). In any case, Herder also argued that language was ‘the child of reason and society’ (p. 91). He thought that ‘vowels are the first, the most vital things, the hinges of language’ (p. 95), which appears to suggest evolution from primate-like vocalizations.

Another important philosopher of the eighteenth century was Pierre Louis Moreau de Maupertuis, author of Réflexions sur l'origine des langues et la vie des mots (1748). Among other things, he sought to answer the question of whether modern languages can ultimately be traced back to one single common ancestor or whether current diversity reflects polygenesis, with different populations developing their own languages. Associating monogenesis with the Tower of Babel myth, which needs a deus ex machina , God, to account for the diversification of languages, he rejected it in favour of polygenesis. Note, however, that his position needs Cartesianism, which assumes that all humans are endowed with the same mental capacity and suggests that our hominin ancestors could have invented similar communicative technologies at the same or similar stages of our phylogenetic evolution. This position makes it natural to project the existence of language as the common essence of languages beyond their differences. Saussure ( 1962 [1916] ) may be credited with similar thinking when he observed that le langage ‘language’ is heteroclitic, anterior to languages and more natural than them, and yet deriving its unity from the latter (pp. 25–26). These considerations provide the background for speaking of universals in the architecture of language and of (constraints on) parametric typological variation.

In the nineteenth century, scholarship on the origins of language was enriched with an alternative perspective. Charles Darwin commented in The Descent of Man (1871) that the evolution of language was in several ways reminiscent of that of mankind itself. He hypothesized that it had emerged gradually, had not been given by God or invented by design by humans, and could also be explained by natural selection. He was among the first to correlate the evolution of language with that of the human mind (see also Müller 1880 [1861] ), thus accounting for why parrots cannot produce original spoken messages intentionally, although they can imitate human speech fairly accurately. Showing what an important driver role the human mind has played in the evolution of language, he argued that it was for the same reason that other primates do not use their buccopharyngeal structure to speak.

We now know that Charles Darwin was only partly right. The other primates' buccopharyngeal structure is not shaped in exactly the same way as that of humans, although, based on the parrot's phonetic accomplishments, we must wonder how critical this particular structure was for the emergence of language (not speech!) in the first place. After all, humans who cannot speak produce signed language, which is just as adequate for communication. This argument may be claimed to support the position that the emergence of the capacity for language must be distinguished from the emergence of languages. However, one must also wonder whether the two questions can be considered independently of each other (see below).

On the other hand, like eighteenth-century philosophers, Charles Darwin also claimed that complex thought could not ‘be carried on without the aid of words.’ Many modern linguists doubt that the language of thought is structured just like spoken or signed language. It does not appear to be constrained by linearity (see below). In fact, in its most fundamental form it does not appear to depend on these communication media and is ontologically anterior to them. Just because the language of fundamental thought is probably structured differently, it does not follow that it is less complex than spoken or signed language. The evidence appears to be lacking regarding the role that speech and signing allegedly play in structuring human thinking. It seems so natural to claim that complex language evolved in response to the communicative needs of social minds that were becoming more and more complex.

Charles Darwin should be credited for subsuming the topic of language vitality, as it should be under the umbrella of language evolution (Mufwene 2001 , 2008 ). He paid attention to the spread of some languages at the expense of others, a topic that linguistics has dealt with recently under the heading of ‘language endangerment.’ However, he also thought of some populations and their languages as less evolved than others, although he did not establish any obvious correlation between the alleged less evolved populations and less evolved languages. This is a recurrent claim throughout the eighteenth and nineteenth centuries, whereby non-Europeans are often described as ‘savages’ and the position of their languages on a putative evolutionary trajectory as ‘primitive,’ simply because their morphologies are too complex (the case of agglutinating and polysynthetic languages), or they have no morphophonology (the case of isolating languages), and/or they are tonal. Though Charles Darwin also concluded that races are probably the counterparts of subspecies in biology, he was still a prisoner of the social prejudices of his time (Mufwene 2008 : ch. 6). His hypotheses on the evolution of language were thus tainted by them.

It is worth mentioning in this context the contribution that George Howard Darwin, Charles Darwin's son and an accomplished astronomer and mathematician, made to the subject of the evolution of language. He defended his father and Dwight Whitney against Friedrich Max Müller, both of whose views are discussed below. In his essay titled ‘Professor Whitney on the Origin of Language’ (1874), George Darwin especially supported the idea that human language may have started from ‘the imitational and interjectional sources of [Aryan] roots,’ that the number of initial roots must have been very small at the early stages of true language and everything else developed later. He elaborates:

It is surely probable that many generations of quasi-men passed away, who used a small vocabulary of conventionalised cries, that these cries became more and more conventionalised, by departing more and more from the sounds of exclamations, from which they took their origin. Many roots would probably propagate themselves by fission, and give rise to new roots, gradually to become entirely separate from their onomatopoeic originals. (Harris and Pyle 1996 : 288)

Max Müller had ridiculed as ‘bow-wow theory’ the hypothesis that human language had started from imitations of animal sounds, interjections, etc. In his essay titled ‘The theoretical stage and the origin of language’ (1861), Müller argues that what distinguishes humans from other animals is not so much speech but the ‘inward faculty which is called the faculty of abstraction, […] which is better known to us by the homely name of Reason.’ Against Charles Darwin's unjustified assertion that there are languages without abstract terms (of course spoken by ‘savages’), he observes that every (denoting?) word ‘contains a predicative root’ which ‘expresses a general concept’ (Harris and Pyle 1996 : 197). Against the ‘bow-wow theory,’ Müller argues that although there are interjections and onomatopoeic terms in every language, ‘as yet no language has been discovered that was so formed.’ According to him, ‘interjections are only the outskirts of real language,’ which begins where they end (Harris and Pyle 1996 : 23). Although it is conceivable that ‘some kind of language might have been formed’ based on onomatopoeias and interjection, it could not have been ‘a language like that which we find in numerous varieties among all races of men’ (p. 24).

In his 1873 ‘Lectures on Mr. Darwin's Philosophy of Language,’ Müller is undecided about whether the roots emerged in a protracted fashion or all at the same time. His overall position raises the question of when grammar emerged in the phylogeny of human language and whether, in the first place, our hominin ancestors were capable of producing phonetic sounds at the time they developed the initial vocabulary. Nonetheless, the original roots evolved gradually into the vocabulary of modern spoken languages, some of them becoming grammatical terms, as argued today by Heine and Kuteva ( 2007 ). 4

On the other hand, Müller also thought that some languages are primitive and simpler, especially those with an isolating morphosyntax. Within the context of complexity/simplicity in language, this is fundamentally the thesis defended recently by McWhorter ( 1998 , 2001 ), according to whom creoles are not only young languages but also the world's simplest. According to the latter, creole ‘prototypes’ lack derivations, inflections, and tones, all being features that older languages have putatively acquired through much longer histories of evolution and accretion. Independent of the forceful and extensive rebuttal provided by DeGraff ( 2001 ), how ironical it is that, for reasons that are no sounder, much of the eighteenth- and nineteenth-century literature on the evolution of language considered inflections and tones to be primitive features! As we will see soon in the discussion of Otto Jespersen's views, creoles could thus be considered more evolved than their European lexifiers and other languages.

Objecting to Charles Darwin's hypothesis that human languages, like different races of man, have evolved from a common ancestor, Müller ( 1873 ) states:

[B]ecause the merest tyro in anatomy knew that the different races of man constituted so many species, that species were the result of independent creative acts, and that the black, brown, red, yellow, and white races could not possibly be conceived as descended from one source. (Harris and Pyle 1996 : 175)

This remark is reminiscent of objections made by some scholars such as Maine ( 1875 ) and Freeman ( 1881 , 1886 ) to Sir William Jones's 1786 hypothesis that Sanskrit, Greek, Latin, and other Indo-European languages had all evolved ultimately from the same protolanguage, Proto-Indo-European. They thought that the Indians were too ‘barbaric’ to share genetic ancestry, racially and linguistically, with Europeans. Otherwise, Müller's objection conjures up the question of whether monogenesis and transformational evolution, as typically suggested in linguistics, can account adequately for the emergence of linguistic diversity, especially if no allowance is made for internal variation in the protolanguage à la Bickerton ( 1990 ). In this respect, modern linguists would be remiss to overlook the fact that Charles Darwin invoked natural selection as applying to variation which he assumed to obtain in any population (see below).

Müller thought that ‘collateral development’ (polygenesis) was more likely to account for some of the differences between dialects and languages. According to him, there is no reason why different individuals at different places and/or different times would have solved the same communicative challenges in identical ways, even when they are endowed with the same ‘instinct, gift, talent, faculty, proprium ’ for language (1873: 228–9). He was clearly not Cartesian! Nonetheless, he maintained that language was a means ‘for the formation of thought’ (pp. 231–2), oddly in agreement with Darwin in this case.

Müller was also strongly opposed to the hypothesis that humans are phylogenetically related to the great apes and monkeys. He concluded that Darwin must have been confused, ignoring the fact that human language is unattainable by other animals (p. 183). The question is whether this state of affairs is a consequence of Müller's suggestion that the great apes are not phylogenetically related to humans. One wonders what he would think of today's attempts to get some great apes to communicate with humans in approximations of sign language or with lexigrams, or even of claims that they understand speech.

Like Jean-Jacques Rousseau, Müller stipulated a distinction between ‘emotional language’ and ‘rational language.’ Accordingly, the former is something that humans share with animals and in which imitations of ‘natural cries’ fit, whereas the latter is the outer side of the mind and is unique to mankind. Müller was curious how one may account for the evolution from ‘emotional’ to the ‘rational language’ (1873: 225). This question has remained hard to answer, though one may suggest that our hominin ancestors may have started with modulating their vocalizations into sequences of contrasting syllabic peaks, thus producing different vowels. However, as discussed below, more was involved in the process; we need to learn from paleontology and other relevant disciplines about how we evolved mentally, anatomically, and socially from Homo habilis to Homo sapiens sapiens to be able to account adequately for the transition.

William Dwight Whitney responded to Müller in his article titled ‘Nature and Origin of Language’ (1875), by first articulating a distinction between the ‘capacity for language,’ with which every normal human is endowed, and ‘speech.’ The critical point is that the ‘capacity’ has made it possible for humans to develop language or learn whatever is spoken and/or signed in their social environment. This ‘capacity’ distinguishes mankind from animals, although, as recent findings about bird songs have made clear (e.g. Margoliash 2010 ), the observation should be mitigated (see below). Whitney argues that ‘the only conscious motive’ for developing language was communication, which is certainly at odds with Bickerton's ( 1990 ) claim that it was made to enhance human capacity for thought. Then he reformulates the ‘bow-wow theory’ as follows:

Spoken language began […] when a cry of pain, formerly wrung out by real suffering, and seen to be understood and sympathized with, was repeated in imitation, no longer as a mere instinctive utterance, for the purpose of intimating to another, ‘I am (was, shall be) suffering.’ (Harris and Pyle 1996 : 298)

Whitney thus saw the foundations of language in the intentional use of the cries and other sounds. Then he proceeded to address the question of how spoken language has emerged as the dominant mode of explicit communication in mankind:

[I]t is simply by a kind of process of natural selection and survival of the fittest that the voice has gained the upper hand, and come to be so much the most prominent that we give the name of language (‘tonguiness’) to all expression. There is no mysterious connection between the thinking apparatus and the articulating apparatus, whereby the action that forms a thought sets the tongue swinging to utter it. (Harris and Pyle 1996 : 300)

As we shall see below, ‘natural selection’ is not much of an explanation if one does not mention the factors that influenced the resolution of the competition in this particular direction. On the other hand, like Charles Darwin, Whitney seems also influenced by the social prejudice of his time, as in the following passage that should not resonate well to speakers of tone languages:

[T]one, and still more gesture, has assumed the subordinate office of aiding the effectiveness of what is uttered. And the lower the intellectual condition of the speaker and the spoken-to, the more indispensable is the addition of tone and gesture. (Harris and Pyle 1996 : 302)

The bias against non-Indo-Europeans is equally strong in the following passage:

An infinity of things can be said in English which cannot be said in Fijian or Hottentot; a vast deal, doubtless, can be said in Fijian or Hottentot which could not be said in the first human language. (Harris and Pyle 1996 : 307)

A great deal can be said in Fijian, Hottentot, and other non-European languages that cannot be said in European languages either, just as there are things that can be said in English but cannot be readily expressed in French, for instance, and vice versa. Whitney also claimed that the earliest form of linguistic communication must have been holographic, consisting of one-word utterances, without a formal distinction between entities and actions; parts of speech and predication emerged later, and even later the combinations of words belonging in different lexical categories into complex utterances (Harris and Pyle 1966 : 306, 308). As noted above, this comes as an apt rejoinder to Herder's speculations, although, as with everything else, this must be verified by future research. Like his contemporaries, Whitney thought that inflectional or fusional languages represent a high level of ‘cultivation.’ However, he also thought of the evolution of language as the ‘accidental […] product of forces and circumstances so numerous and so indeterminable that we cannot estimate them and could not have predicted their result’ (pp. 312–13). In this respect, he is like today's emergentists, for whom evolution is largely driven by self-organization.

Several other scholars, many of them anonymous, published on the origins of language in the nineteenth century. One of the non-anonymous was the social anthropologist Edward Burnett Tylor. In a 1866 paper titled ‘On the Origin of Language’, he attempted to support the ‘bow-wow theory’ by invoking the ways in which ‘savages’ in the colonies named the goods the Europeans brought, using words based on sounds associated with the goods. For instance, the Sea Islanders in the Pacific allegedly used pu for musket, puhi for ‘to blow’ (as they thought the European blew in the gun), puff for the smoke coming out of the musket, and pupuhi for the barrel of a gun. He concluded:

If several languages have independently chosen like words to express like sounds, then we may reasonably suppose we are not deluding ourselves in thinking that such words are highly appropriate to their purpose. Thus we have such forms as pu, puf, bu, buf recurring in the most remote and different languages with the meaning of blowing or puffing. (Harris and Pyle 1996 : 91)

In a note, he illustrates his claim with the following list: ‘Tongan buhi , Mahjori pupui , Zulu pu , Hebrew puach &c.’ He likewise finds evidence for the common origin of language in the cross-linguistic similarities among words used for ‘father’ and ‘mother’, words which, according to him, vary more in their consonants than in their vowels (p. 95). It did not matter at all to him that some terms that are phonetically similar sometimes denote opposite entities. It is striking how nineteenth-century scholars really thought that the colonized populations were apparently less evolved anatomically and/or mentally and therefore may provide evidence for how language evolved. Nowadays, we have to deal with Bickerton's ( 1990 ) controversial claim that pidgins (typically those based on European languages) represent fossils of his ‘protolanguage.’

Nobody articulates the above thesis as explicitly as the Revd Frederic William Farrar, who, in his 1865 book Language and Languages , asserts:

Savage languages are […] the best to show us what must have been the primitive procedure; but we can trace the same necessary elements of words in languages far more advanced. (Harris and Pyle 1996 : 59)

Arguing that language is too imperfect to be God's creation, he also interpreted the multiplicity of languages as evidence that language is an invention of mankind, ‘developed by intelligence and thought. […] It may be unable to keep pace with the advancing power of abstraction, but it can never by any possibility anticipate or outstrip it’ (p. 45). He adduced evidence for humans' ability to invent languages from what is now known as ‘home sign language’ and from the ability of abandoned children living in groups to develop a language of their own (pp. 54–5). This evidence should actually be used to highlight the fact that, from an evolutionary perspective, the language phenomenon under discussion is a communal one, which does not emerge unless there is population of individuals, at least two, who interact with each other. (See also Lieberman 2006 : 354ff.) Unless a situation such as the Nicaraguan boarding school for the deaf arises, no particular communal sign language emerges from the practices of isolated home signers interacting only with their speaking relatives.

Not unlike Bickerton ( 1990 ) with pidgins and creoles, Farrar thought that the modifications of European languages in the colonies might shed light on how language evolved, just like the invention of ‘Argots’ by ‘the dangerous classes throughout Europe’ (Harris and Pyle 1966 : 66). According to him, because they are not intelligible to speakers of the languages from which they have evolved or been developed, they ‘must, from their very nature, remain uncultivated’ (p. 66). Although he assumed that language emerged gradually, he discussed the complexity of ‘savage languages’ in a way that reveals again strong prejudice against non-Europeans. This was indeed the century of ‘la mission civilisatrice’ or ‘the white man's burden’ ideologies developed by the French and the British respectively to justify exploitation colonization. Being non-European, isolating languages and, according to Farrar, also agglutinating and polysynthetic languages were deemed primitive. Putatively, the ‘apparent wealth of synonyms and grammatical forms is chiefly due to the hopeless poverty of the power of abstraction ’ (Farrar's italics, p. 78). This would allegedly be obvious in languages that lack the copula. All such remarks that are undoubtedly offensive today, at least to some of us, underscore how cautious we must be in how we use our findings about some modern linguistic systems to make inferences about the evolution of language.

We should, of course, not ignore Friedrich Wilhelm Christian Karl Ferdinand Freiherr von Humboldt, who conceived of language dynamically in terms of the ‘energeia’ that translates the ‘inner linguistic sense’ into the outer expression, in which the universe of experience is categorized differently from one community to another. He may be considered a forerunner of the Sapir–Whorf hypothesis. What is especially relevant to the study of the evolution of language (a topic on which Humboldt did not say much) is the individuality of the inner sense, which makes every idiolect different but also every dialect and every language different, as the dynamics leading to social norms vary from one community to another. Humboldt also claimed that different populations have not evolved identically in developing their linguistic individuality. He characterized the evolution of language as what Harris and Taylor ( 1989 : 177) paraphrase as ‘the continuous outcome of [the] dialectic between the inner linguistic sense and sound-form; that is, between energeia and ergon .’ Every individual speaker contributes to this process, as they reshape in not reproducing perfectly, the language of their social environment.

Then we must now ask how the different individuals, innovators and copiers, ultimately converge toward shared communal norms (Mufwene 2008 ). Note that invoking either the ‘invisible hand’ or ‘self-organization’ is simply admitting that we cannot yet articulate explicitly how the mutual accommodations that speakers/signers make to each other, in their ever-changing dyadic and triadic interactions, evolve to these ‘conventions.’ It is like saying that languages take on lives of their own when in reality the agents and hosts are the speakers or signers (Mufwene 2001 ). The conclusion does not take us farther than Saussure's 1916 correct observation that ‘la parole fait évoluer la langue’ (‘speech makes language evolve’), without explaining how it does it.

The foregoing gives us a representative canvas of the state of the art in the eighteenth and nineteenth centuries concerning the evolution of language. It also gives us a sense of the kinds of controversial speculation that led the Société de Linguistique de Paris in 1866 to ban any linguistic discussions on the subject matter at its meetings. Only one more scholar is worth noting from the period, the semanticist Michel Bréal, who argued against the French ban on the ground that it impoverished the subject matter of linguistics. Bréal saw languages as being reshaped constantly by their speakers, and rejected his contemporaries' organic approach to them. He thought the approach was inaccurate in casting some languages not only as less evolved than others but also as decaying or dying. He would undoubtedly have opposed the present discourse about language birth, vitality, and endangerment, as well as about moribund languages, though it can be argued that languages conceived of as species (Paul 1880 , Mufwene 2001 ) are born and may die in the same protracted ways biological species do, unlike individual organisms (Mufwene 2008 : 208–9).

As noted above, the French ban appears to have been respected even outside France. It became almost taboo to discuss the evolution of language throughout most of twentieth century, until the 1990s, which I discuss in the next section . 5 Among the exceptions to the rule are the Dane Otto Jespersen, in his book Language: Its Nature, Development and Origin (1922a) and the American Morris Swadesh, whose book Origin and Diversification of Language , written in 1967 but published posthumously in 1971, also changed the nature of the discourse.

Otto Jespersen's contributions to the study of the origins of language include his argument that the ‘bow-wow’ theory (claiming the origins of language in the imitation of sounds in nature), the ‘pooh-pooh’ theory (based on human interjections), and the ‘yo-he-yo’ theory (based on human sounds during collective physical work) need not dismissed offhand. ‘Each of the three chief theories enables one to explain parts of language but still only parts, and not even the most important parts—the main body of language seems hardly to be touched by any of them’ (1922a: 416).

A more important and relatively uncontroversial contribution of Jespersen's is his position that we can learn indirectly about the origins of language by focusing on infant language during the first year of what is still nonlinguistic interaction with the caretakers, focusing on its cooing, babbling, and gestures. Later scholars such as Tomasello ( 2008 ) have suggested the development of joint attention, observable in human infants but not in great apes, as an important determinative factor in the evolution of language. Babies' ability to take turns in vocalization games also appears to be evidence of joint attention.

Jespersen also advocated paying attention to trends in how human languages have evolved in documented history, though the conclusions he suggests are controversial. He points out that European languages such as English and French have evolved from more complex morphosyntax to simpler, analytic ones and from structures putatively harder to learn and full of irregularities to more regular and systematic ones. ‘The direction of the movement is toward flexionless languages (such as Chinese, or to a certain extent Modern English) with freely combinable elements’ (1922a: 425). If, like Jespersen, one adopted from the misguided nineteenth century the view that some languages and related populations are less evolved than others, this would not rank German (which Jespersen does not discuss in this context) very high on the scale, nor Basque, which he finds excuses for not lumping into the category of ‘primitive languages.’ His conclusion is that the initial language must have had forms that were more complex and non-analytic; modern languages reflect evolution toward perfection which must presumably be found in languages without inflections and tones. It is not clear what Jespersen's position on derivational morphology is. In any case, his views are at odds with Bickerton's ( 1990 ) hypothesis that the protolanguage, which allegedly emerged by the late Homo erectus , was much simpler and had minimal syntax, if any. While Bickerton sees in pidgins fossils of that protolanguage and in creoles the earliest forms of complex grammar that could putatively evolve from them, Jespersen would perhaps see in them the ultimate stage of the evolution of language to date. Many of us today find it difficult to side with one or the other position.

Rather outrageous is Jespersen's claim that languages of ‘savages’ in Africa and the Americas could inform us about the origins of language, not only because they have longer words (with complex morphology, 1922a: 421), but also because they use difficult sounds such as clicks and rely on tones (p. 419), which, according to him, suggests that their speakers are ‘passionate’ (p. 420). ‘Primitive languages’ were accordingly sung, poetic, and figurative (p. 432). Being tonal and using numeral classifiers (pp. 429–30), Mandarin would be low on Jespersen's scale of evolved languages, though it might be better off than languages that are both tonal and have complex morphological structures. It is of course worse for languages that have no terms such as ‘colour’ for abstract concepts or general categories. Jespersen concludes, among other things: ‘Primitive units must have been more complicated in point of meaning, as well as much longer in point of sound, than those with which we are more familiar’ (p. 425). As pointed out in Mufwene ( 2008 : ch. 6), it is noteworthy how late race lingered as a factor in accounts of language evolution in linguistics.

In contrast, Morris Swadesh's arguments are grounded in the then state of the art concerning phonetic and morphological properties of several languages around the world, as well as in paleontological and archaeological evidence. The examination of these led him to draw (among others) the following conclusion, which anticipated Mufwene's 2010b comparison of the pace of the evolution of language with that of computers, in shorter and shorter intervals of time as we near the present: ‘It seems probable that language developed in the same general lines as other aspects of human culture: very slowly at first and gradually faster and faster’ (Swadesh 2006 : 45).

However, like many others before him, Swadesh hypothesized that language started with naming. The words may originally have been imitative of sounds heard in nature; then they were allegedly replaced by ‘exclamative’ ones, and later by ‘a purely expressive paradigm and an attention-calling or demonstrative one’ (p. 182). He believed that numerals ‘were among the last to take on their present character’ (p. 183). His world-wide comparison of demonstrative forms led him to the conclusion that ‘before the neoglottic period, perhaps in the paleoglottic, fewer phonemes were differentiated than in contemporary languages’ (p. 199), suggesting that even the phonetic inventories of modern languages must have evolved gradually, not becoming fully modern until as late at the emergence of agriculture.

Philip Lieberman ( 2002 ) believed phonetic language to have emerged earlier with the late Homo erectus or archaic Homo sapiens . Although this position has been revised (see below), the most relevant point here is that different parts of language appear to have evolved incrementally and no particular module seems to have emerged abruptly. It does not appear likely that Homo erectus or archaic Homo sapiens waited until a complete phonetic inventory was in place before producing their first words, or waited until there was a complete vocabulary with identifiable morphemes before producing phrases and sentences. Although ontogeny does not recapitulate phylogeny, child language acquisition discourages us from speculating about the phylogenetic emergence of language in strictly linear terms. Then, as now, early lexical and phonetic developments must have proceeded concurrently. One may also speculate that the expansion of the lexicon drove the elaboration of a wider phonetic inventory, as this enables more lexical distinctions.

In the style of evolutionary biology, Swadesh proposes a monogenesis account which assumes inter-individual variation in the ‘vocal behavior’ of the relevant hominins: ‘in addition to individual differences, there could have been variations by sub-species and by locality, but all within essentially ‘one language’’ (1971 [2006 edn: 213]). Putatively, hominin populations equipped with similar anatomical and mental structures, living in different localities, and having developed comparable communities in which they experienced similar pressures to interact explicitly, would have developed comparable but non-identical means of communication. This sounds quite plausible, as East Africa, where most of the hominin fossils have been found, is a vast geographical area; to date no paleontological evidence suggests that an early Homo habilis or Homo erectus population dispersed out of one single locality to the rest of the world.

As argued in Mufwene ( 2008 , 2010b ), different individuals endowed with the same capacity for language need not have innovated exactly the same strategies for the same communicative needs. 6 Locally and regionally, there must have been plenty of variation, as argued by Johann Gottfried Herder, which set the innovators' productions up for competition among their imitators. This would have set things up for variational evolution, through competition and selection among available alternatives even within the same language, as members of the relevant populations converged toward their respective norms. Dor and Jablonka ( 2010 : 138) call this normalization process ‘canalization.’

Swadesh assumed that in the earliest, longest stages of the emergence of language, communication among hominins remained instinctive and did not vary significantly from one locality to another; therefore it is normal to assume that our hominin ancestors spoke the same language. According to him, significant diversity started to emerge about ‘half a million or so years ago,’ when the earliest forms of phonetic and symbolic communication, which he calls ‘formal language,’ started to emerge ( 2006 : 214–15). The estimated period is consistent with that proposed by Corballis ( 2002 ) and Lieberman ( 2002 ), though they now think otherwise (see below). This is a stage when Swadesh believes it was possible for different individuals to innovate different linguistic forms for the same denotata and presumably different structures for the same propositions. (Which is reminiscent of Herder's account of the origin of synonyms in various languages.)

Swadesh's hypothesis raises the question of whether his monogenesis position is not really polygenesis; it leaves open the possibility that two late Homo erectus or archaic Homo sapiens populations developed languages that were not structurally identical and/or mutually intelligible. As is obvious from Bickerton's ( 1990 ) hypothesis of the protolanguage from which ‘true language’ putatively evolved, all may depend on what particular stage in the evolution of the Homo genus and what particular phase of its vocal communication one decides to identify as the beginnings of modern language. This entails particular assumptions about the size of the phonetic inventory and the nature of grammar, which are captured eloquently by Ray Jackendoff's 2010 title ‘Your Theory of Language Evolution Depends on Your Theory of Language.’

Swadesh is also one of the very few scholars who have considered the implications of population movements for language evolution. As the migrants' languages come into contact, often coexisting in competition with each other for the same communicative functions within the same larger population, some may drive others to extinction. Typically, the prevailing language undergoes structural changes and can even speciate into separate languages. Seldom have linguists who are concerned with language endangerment and loss today cast the subject matter from this perspective, which Mufwene ( 2001 , 2008 ) articulates in his ecological approach. The contact-based approach to language birth, endangerment, and death makes language evolution more similar to biological evolution, especially regarding the consequences of language practice under differing ecological pressures. The relevant ecology includes not only the mental and anatomical structures of hominins and humans but also the socioeconomic conditions that determine their population structures and their particular interactional dynamics. Indeed, the latter also trigger migrations, which history has shown to affect both the vitality and structures of languages.

1.3 Recent Developments

As a research topic, the evolution of language has expanded into a productive and stimulating, though diverse, area of scholarship since the 1990s. The scholarship has also expanded beyond the origins of language to include language birth and death, as well as language speciation. While philosophers and philologists no longer appear to deal with it, linguists can hardly claim it as a private domain. No insightful or informative linguistics publication on the subject matter is based exclusively on linguistic data. Interestingly, this is also an area where generative syntax, which has claimed centre stage since the late 1950s, has probably been unable to prevail over other areas, especially since the notion of Universal Grammar (UG), or ‘biological endowment for language,’ or ‘language organ’ (Chomsky 1986 , Anderson and Lightfoot 2002 ), or ‘bioprogram’ (Bickerton 1981 ) is still a black box whose contents have not been articulated in sufficient detail and whose capacity to account for how language works and/or is learned has increasingly been disputed (see below).

Noam Chomsky's occasional contributions to the discourse (e.g. Hauser et al. 2002 , Chomsky 2010 ) have aroused controversy, primarily for not considering much of the non-linguistic evidence and for ignoring objections to his claim that recursion is the most important characteristic of the capacity for language that is not shared by other animals. Others have objected that recursion distinguishes human languages from other animals' means of communication only to a degree. For instance, Margoliash and Nusbaum ( 2009 ) argue that some form of it occurs in some bird songs. Moreover, it may be a general cognitive, problem-solving strategy, as it is attested outside language, such as in mathematics and musical scores, unless the latter domains are claimed to be consequences of language. According to Lieberman ( 2006 : 4–5; 2010 : 164), it can be identified in dancing too. In addition, some scholars argue that there is little, perhaps nothing, in the structure of the human brain that exists only for language and is not part of the general learning adaptation. Language has also increasingly been interpreted as the gradual cumulation of exaptations of particular mental capacities and anatomical organs for communication (Hurford 2006 , Oudeyer 2006 ).

Chomsky's ( 2010 : 51) stipulation ‘The study of the evolution of language is specifically concerned about UG and its origins’ is questionable. An important reason why several scholars have raised issues with it has to do with whether language boils down to UG only, to the exclusion of the physical architecture of language(s). Chomsky's usual equivocations with the disjunctive phrase ‘mind/brain’ has not been informative about the nature of UG. Neurolinguistics has revealed that there is no particular part of the brain that can be identified as the ‘language organ.’ The fact that the parts of the brain implicated in language are not only situated in different regions but also associated with domains other than human communication precludes the possibility of a discontinuous modular language organ. The fact that UG appears to be mental, a property of the mind rather than of the brain as physical matter, clearly leaves open the possibility that it is a (by)product of something else in the many brain activities, including its capacity to produce language. Anderson and Lightfoot ( 2002 ) do not address these issues, although the book is specifically on this topic. Taking the notion for granted, they decide to define it ‘in functional rather than in anatomical terms,’ as it is ‘not localized in the manner of the kidney’ (p. xiii). As a matter of fact, they sometimes identify language itself, like the ‘knowledge of language,’ as the language organ (e.g. p. 8). 7

One must also note an important difference between, on the one hand, how ‘modularity’ is invoked here in reference to concurrent engagements of different parts of the brain during the production of utterances and, on the other, the way the concept is used in technology to characterize the way different parts of a complex machine just complement each other. While complementarity is also true in the case of language, it is not evident that the brain parts are specialized for language only. For instance, Broca's area plays a central part in coordinating sensorimotor activities that have nothing to do with language. Mirror neurons, which have been invoked recently as playing a role in the reproduction of sounds, also play an important part in the reproduction of other physical activities and have been identified in other primates. The lateralization of the brain is not exclusively associated with language either. According to Lieberman ( 2010 : 171), the FOXP2 gene, which was initially too hurriedly associated with language alone, also appears to facilitate ‘learning and precise motor control in human and other species.’

If UG contains no properties that are unique to language, then we are perhaps back to the interest of Condillac and other eighteenth-century philosophers in the evolution of language as a way of learning about the evolution of mankind and their mind. Thus some linguists such as Jackendoff ( 2010 ) justifiably object to focusing on a questionable notion of faculty of language , especially ‘in the narrow sense’ (Hauser et al. 2002 ). It impedes investigating the evolution of language in relation to that of, say, human cognition in general and animal communication.

Several scholars appear to align themselves with Pinker and Bloom's ( 1990 ) position that an all-purpose mental capacity, or various phases of its development, at a particular stage (or stages) of the Homo phylogeny, would have sufficed to produce language. 8 Assuming that what emerged are individual languages but not Language per se (a position consistent with Saussure's ‘la parole […] est nécessaire pour qu'une langue s'établisse’, 1962 [1916]: 37), an alternative interpretation of UG is that it is the common denominator of the properties and architectures of the different languages. Thus, UG may not be a particular mental infrastructure that emerged at some particular phylogenetic stage of the Homo genus and enabled or facilitated the emergence of language, but simply a consequence of this evolution (MacNeilage 2008 : 298).

Subscribing to the distinction between I-language and E-language, Chomsky correctly dismisses the hypothesis that language emerged in the form of ‘language of thought’ (LOT), citing lack of linguistic evidence and the fact that ‘we have almost no idea what LOT would be’ (2010: 226, n. 24). However, he associates language diversification with the externalization of language. According to him, the reason why there are so many languages ‘might be that the problem of externalization can be solved in many different and independent ways, either before or after the dispersal of the original population [out of Africa]’ (p. 61).

Consistent with some remarks in § 1.2 , one may want to justify this position by invoking the Cartesian view that the mind is the same in all members of Homo sapiens sapiens and would work the same way (allowing a limited number of alternatives) in speaking or signing. However, this position does not entail that they must of necessity be endowed with a language-specific UG in order to accomplish this. We just do not know yet. A general-purpose problem-solving cognitive capacity can lead to the same results, if interactants develop similar technologies for communication. UG could amount to common properties of these technologies, i.e. languages of particular communities, properties that are tantamount to universals of language and typological variation on particular parameters. Alternatively conceived of as a body of constraints on the architecture of language, UG can boil down to specifications of what the general-purpose problem-solving cognitive capacity permits and does not permit, bearing in mind that some of the constraints may simply be consequences of the materials used in the technology.

Chomsky too speculates that the externalization ‘might have been a process of problem-solving using existing cognitive capacities’ (2010: 61). This appealing position need not be wedded to his assumption of UG. Those who believe that modern language emerged to facilitate communication among humans can ask why I-language, associated with UG, need be considered anterior to E-language; it may also be conceived of as patterns emerging from successful utterances, as suggested in Construction Grammar or by Complexity Theory. In other words, ‘knowledge of language’ may be considered as internalization of what the communicator can(not) do vocally and/or with manual signs in his/her attempts to express meaning, i.e. a mental representation of the technology developed by a particular population for communication. As hypothesized by Saussure (cited above), the internalization may be considered as a consequence of practice.

Chomsky also argues that only I-language should be in the domain of investigations on the evolution of language. In his own words, ‘any approach to the evolution of language that focuses on communication, the SM [sensory–motor interface] system, or statistical properties of spoken language, and the like may well be seriously misguided’ (2010: 61). This position raises the issue of whether in some cases students of the evolution of language should not start by agreeing on the particular conception of language they are assuming. This is especially important because Chomsky's reaction to the question of ‘why languages appear to vary so widely’ is that this phenomenon ‘is an illusion, much like the apparent limitless variety of organisms’ (p. 62). He is of course driven to this remark by his strong minimalist theory, which appears to treat typological variation as a linguistic epiphenomenon less important than the core of language putatively determined by UG.

Could language really have originated as an abstract and uniform UG, thanks to the brain-rewiring event Chomsky hypothesizes? Or, as surmised above, is UG only the consequence of similarities among the ways members of the Homo genus have gradually solved their communicative problems? As remarked above, this evolution would have been enabled by the same general-purpose mental capacity that evolved gradually in them, and would have led them to coopt their anatomical structures to produce the relevant technology for communication, but not necessarily in identical ways.

It is certainly necessary to agree on a particular definition of language, so that we may determine whether or not we seek to explain the same subject matter. As pointed out in Mufwene ( 2001 , 2008 ), the Saussurean conception of language as ‘system,’ which still prevails in linguistics, is at odds with the folk notion of language as the particular way a population speaks. In fact, lay people speak of languages, not Language (which is a philosophical concept); for them a language is just a way of speaking. It is not evident that the earliest speculations about the origins of human communication were not about languages but about Language, hence the long-held belief among some that Hebrew was the original language.

A problem in linguistics about what is language also arises from the status of phonetics. It is not obvious that linguists agree on whether it is part of language proper or is just a modality, as suggested, for instance, by Hombert and Lenclud (in press). This is a legitimate question, as some like to focus on rules and constraints seemingly ignoring the fact that these apply to physical items called words, which couple meanings (abstract entities) with forms. The architecture of language is built on them. It is hard to imagine that any grammar at the UG level or at the specific-language level, say I-language, could exist without physical entities that it applies to.

The above considerations make it natural to investigate how typological diversity emerged between languages and sometimes within individual languages. The diversity regards, among other things, the specific phonetic inventories that different populations of speakers have chosen and whether or not they made tones phonemic. It also has to do with whether they chose agglutination, polysynthesis, inflections, or isolating morphosyntax to code information around the main verb, whether the verb comes second or in another position in the sentence, whether they use Nominative/Accusative or Ergative/Absolutive syntax to code agency, what strategies they use to specify and track reference (for instance, do they use noun classifiers or genders?), how they articulate tense distinctions, etc. (See Hurford 2008 for a complementary discussion.)

Although syntax has long been privileged in formal linguistics, it has by no means claimed centre stage in the scholarship on the evolution of language, despite all of Bickerton's ( 1990 ) claims about the nature of his phylogenetic protolanguage. Very little has been written, for instance, about the evolution of combinations of words, constraints on the positions of particular constituents within larger units, and movements of constituents to particular positions in sentences. If Chomsky is correct in claiming that typological variation is an illusion, then something should be said about how the common aspects of these syntactic phenomena evolved.

The above question may be more difficult to answer than that of why delimiters such as tense , aspect , and mood for the verb as well as number and class for the noun evolved in language. One can surmise that for communication to be more precise, or less vague, events and conditions must be situated in time and reported differently according to whether they are facts or not, and whether the referents of nouns must be specified according to cognitive requirements that interest particular populations. It must be equally informative to find out why, for instance, the verbal complements of volitional verbs in inflectional languages are more likely to be used in the subjunctive or infinitive. Are the constraints purely linguistic or cognitive?

An answer to the question of why predication emerged, one that Herder considered to be central to the study of the evolution of language, can also be attempted here. We can resort to the way the distinction between Topic / Subject and Predicate has been traditionally explained in grammars, viz. what the utterance is about (the topic ) and what state of affairs ( activity or state ) is associated with the topic. However, much more is involved in predication than just having a head of the predicate phrase. The evolution of the organization of an utterance into Topic / Subject + Predicate   Phrase for most languages needs some explanation, as much as the ways in which materials are structured into the predicate phrase. Would a UG-based account be satisfactory? Or would it be more informative to invoke general-purpose problem-solving cognitive capacity to explain how different populations developed their communicative technologies which nonetheless share similar principles? We probably need considerations not exclusively grounded in linguistic theory to answer this question.

As pointed out by Jackendoff ( 2010 : 69), an important problem with ‘syntactocentrism’ is that is does not account for ‘the evolutionary source of the lexicon.’ Questioning the centrality of syntax in generative grammar, Bolinger ( 1973 ) had argued, along with generative semanticists, that syntax was a consequence of the lexicon, being a body of generalizations from the ways that individual lexical items behave in utterances. It captures morphosyntactic similarities that lexical items display among themselves. Jackendoff ( 2010 : 70) is also right on the mark in pointing out that Chomsky's approach makes it hard to explain how lexical categories (and presumably the ensuing syntactic categories) emerged. Were they arbitrarily predetermined? Why do they not all occur in all languages or in identical ways? Is it also an illusion that some languages have articles while others do not, or that inflectionless languages may not have a finite/nonfinite distinction for the verb, or that the infinitive may not have an identical syntactic status from one language to another?

Ideologically germane to Chomsky's reliance on UG but drawing very different conclusions is Derek Bickerton's work since his book Language and Species (1990). Bickerton started with the claim that modern human language evolved almost abruptly from a ‘protolanguage’ used by our hominin ancestors up to Homo erectus . The protolanguage putatively consisted of a (limited) vocabulary without much grammar, and may have combined both words and gestures. 9 The protolanguage ‘is not a true language, but it's made up of languagelike elements’ (2010: 40). Its users produced ‘short and shapeless and disconnected utterances,’ as one may encounter in especially child language and incipient pidgins, which he considers to be its modern fossils (p. 40). They lack the kinds of syntactic rules and constraints one finds in a ‘true language.’

Like Slobin ( 2002 ), Mufwene ( 2008 , 2010b ) argues against this characterization of particularly pidgins and child language, products of humans endowed with Homo sapiens sapiens 's mind. Moreover, one must be cautious; the human child is not creating a language but learning the language of its social environment. The producers of a pidgin did not start from the absence of a language. Nor did their minds regress to the state of Homo erectus 's mind when faced with the challenge of communicating with another population in a language other than their own and without sufficient exposure to the target language. If anything, pidgins tell the extent to which a modern language can be reduced without losing the status of a language, therefore what are the most central/essential architectural materials a language cannot do without. Assuming that language has evolved gradually, they also tell us what in the architecture of language is so deeply entrenched that it cannot be dispensed with (Wimsatt 2000 ). Gradual emergence assumes a lot of scaffolding (Wimsatt and Griesemer 2007 ), a position quite implicit in grammaticization hypotheses, in which later developments are built on earlier ones. That order of evolution would more or less determine what can be dispensed with, in a less costly manner, if the system must be reduced to an earlier functional modern stage. We also learn that the architectural complexity of a language can be correlated with the communicative needs of its creators/users, not necessarily with the complexity or sophistication of their mental structure. Pidgins are by-products of contact settings where communication was minimal and sporadic (Mufwene 2008 ).

Bickerton also hypothesizes that language must have started with labels that were iconic. Symbolic communication would evolve later, making human language more different from animal means of communication. It's not clear whether symbolic items were already present in the putative protolanguage or whether they emerged in ‘true language.’ I am not sure that his quoting Terrence Deacon's assumption that ‘symbolism’ emerged ‘probably not until Homo erectus ’ (Bickerton 2010 : 50) answers the question, though he concedes to Deacon ( 1997 ) that symbolism, rather than syntax, is what distinguishes humans from animals (Bickerton 2010 : 49). Symbolism enabled what Hockett ( 1959 ) identified as ‘displacement,’ the ability to talk about entities and states of affairs that are not in the hic et nunc of interactions, and thus the ability to talk also about the past and the future, as much as about fictional scenarios. All human populations have developed the capacity to narrate stories and even construct myths of all kinds thanks to the world-creating power of language. This is not possible in animal communication, even after they have been taught to communicate with humans. The reason appears to lie not so much in our invention of symbolic language as in our being endowed with the mental capacity that enabled us not only to produce it but also to do more with it.

On the other hand, Bickerton appears to contradict himself in some ways, when he elaborates on the architecture of his ‘protolanguage’:

[T]he words of protolanguage, even if vocal, could not have been divided into component parts [i.e. sounds], and would likely sound to us like meaningless grunts or squawks. But, like today's words, each would have a fairly well-defined range of meaning, and that meaning, rather than relating directly to the current situation, would refer to some relatively stable class of objects or events, regardless of whether or not these were present at the scene. (2010: 66)

This sounds very much like symbolic communication minus phonetics and syntax. Except for symbolism, protolanguage would be a more elaborate version of primates' calls and gestures, raising the question of why Bickerton compared it to child language and incipient pidgins, which have human linguistic properties. These varieties have basic syntax, variable as it may be in the case of pidgins. In addition, it is not clear how consistent he is with the concession he makes to Deacon. If the latter version is right, reference would have started before ‘true language’ emerged, though ‘true language’ would refine reference by the addition of specifiers such as demonstratives and articles, as well as possessive constructions. The question of when such strategies developed is as worth investigating as that of when parts of speech emerged, and what the emergence entailed regarding the complexification of the architecture of grammar.

One of Bickerton's most problematic positions is his claim, like Condillac's, that language emerged to enhance human capacity for thought. In addition to Chomsky's ( 2010 ) observation that ‘we have almost no idea what LOT would be’ (p. 226, n. 24), we must ask why anybody would need a language of thought that would slow down their thinking process with the constraints of linearity? What is so more efficient about conceptual categories that are labelled linguistically when they can be identified nonlinguistically, as is often obvious when speakers do not have words for ideas they want to express? Granted, human languages have a world-creating capacity; but isn't language more for sharing conceptualizations across speakers rather than for conceiving the scenarios that are shared?

In a different vein, some linguists such as Croft ( 2000 ), Wang and Minett ( 2005 ), Mufwene ( 2008 , 2010b ), Beckner et al. ( 2009 ), and Lee et al. ( 2009 ) also now conceive of languages as complex adaptive systems, which presuppose no permanent sets of rules that guide linguistic behaviour. Instead, linguistic rules are interpreted as emergent patterns produced by self-organization, in a way similar to other natural phenomena involving complexity. This position does not remove from mankind its agency in the emergence of language; it simply means that, throughout the Homo genus phylogeny, the individual acts of solving communicative problems did not include anticipation or a plan to develop what Antoine Meillet identified as a ‘système où tout se tient’. The interactants never had/have any foresight of what their communicative ‘system’ will be like in the future or once it is presumably completed. The focus is always on the hic et nunc ecological pressures for adequate or successful communication.

Patterns, which linguists have identified as ‘rules,’ are therefore consequences of habits that the interactants have developed, based largely on analogies that obtained among items (Mufwene 2008 ), as when, in English, verbs of intention combine with verbal complements in the subjunctive or the infinitive but verbs of prohibition (such as prevent and discourage ) combine with verbal complements in the gerund, sometimes preceded by the preposition from . Because there are cross-linguistic similarities across languages, though the patterns are not identical, it is interesting in terms of evolution to understand why such variation is the case. 10 Thus, are there any particular cognitive pressures that impose on speakers only the typological options that have been attested in human languages but not others? Why would such a mood as the subjunctive , as opposed to the indicative , have emerged, even if it is not universal? Why didn't some other kinds of strategies develop for complements of verbs of intention and prohibition?

Would such constraints provide evidence for Charles Darwin's hypothesis that mental evolution drove the evolution of language rather than the other way around? This kind of question has generally not been addressed, though it arises as an issue from Bickerton ( 1995 ). He could not address it, because he assumes that language emerged to enhance human capacity for thought; therefore the conceptual infrastructure could not possibly influence how language would evolve. Is there any hope that cognitive grammar, functional grammar, construction grammar, or any other approach to syntax that does not rely overly on what Lieberman ( 2006 : 61) calls ‘theories of data’ may help us address the question adequately? Or are the approaches that assume that language is primarily a means of communication misguided? In any case, emergence is antithetic to design. If the claim that language emerged out of hominins' attempts to communicate at various stages of their evolution is correct, then it may be misguided to continue using Hockett's ( 1959 ) term/concept ‘design features.’

Much of the current scholarship on the evolution of language has been more global, focusing on the correlation between, on the one hand, the different stages of the evolution of the mental and anatomical structures of the Homo genus and, on the other, the apparently gradual emergence of language, especially since Homo habilis . These include but are not limited to Bickerton ( 1990 , 1995 , 2007 , 2010 ), Lieberman ( 1984 , 2002 , 2006 , 2010 ), Corballis ( 2002 , 2010 ), MacWhinney ( 2002 ), Fitch ( 2002 , 2010 ), Tomasello ( 2008 ), Tomasello et al. ( 2005 ), McNeill ( 2005 ), McNeill et al. ( 2008 ), MacNeilage ( 2008 ), Mufwene ( 2008 , 2010b ), and Hombert and Lenclud (in press). All but Bickerton argue for gradual, protracted evolution. Tomasello stresses the significance of ecological pressures exerted on hominins by their increasingly more complex social lives, which required management by means of efficient and explicit communication. Modern language would provide this, driven by the same mind that was ready to handle the corresponding complex social interactions. He argues that cooperation and joint attention played as important a role in the emergence of language as in social organization. He shares with Sperber and Wilson ( 2002 ) (see also Sperber and Origi 2010 ) the ‘theory of mind,’ which enables interactants to second-guess each other and thus to infer the intended meaning. All these factors enabled the emergence of symbolic language, the characteristic that indeed led Deacon ( 1997 ) to identify mankind as the ‘symbolic species.’ As noted above, symbolic communication is, according to the latter, the characteristic that clearly distinguishes human communication from animal communication. Sperber and Origi ( 2010 : 131) conclude:

From a pragmatic perspective, it is quite clear that the language faculty and human languages, with their richness and flaws, are only adaptive in a species that is already capable of naïve psychology [i.e. mind-reading ability] and inferential communication.

Corballis, MacWhinney, and McNeill also argue that the earliest ancestors of human language could not have been vocal. Whereas Corballis and MacWhinney originally estimated that the embryonic forms of speech may have started as early as 500,000 years ago, Corballis ( 2010 : 115–16, 119, 123) argues that only language, using gestures, may have started that early, with some complex grammar for that matter, and that the contribution of Homo sapiens since about 100,000 years ago was the introduction of speech. This may not have evolved to its modern forms until about 30,000 years ago. To be sure, Corballis does not claim that the switch was abrupt or that no phonetic vocalizations occurred before Homo sapiens . What he means is that gestural communication was dominant and verbal communication did not prevail as the dominant means of communication until Homo sapiens . It still took tens of thousands of years to evolve to modern phonetic norms.

Corballis' new position is echoed by Lieberman ( 2010 : 175):

McCarthy, Strait, Yates and Lieberman (forthcoming) found that the necks of the Middle Paleolithic fossils who lived about 100,000 years ago were too short to have a pharyngeal SVTv [vertical supralaryngeal vocal tract] that was equal in length to SVTh [horizontal SVT]. A similar constraint rules out Neanderthals having a human SVT. Surprisingly, neck lengths that would support a fully human SVT are not apparent in the fossil record until the Upper Paleolithic, some 50,000 years ago, when a blossoming of complex tools and art appears in the archeological record […] the sudden appearance of an array of advanced artifacts has been taken to be a sign of cognitive advance. […] The presence of a human SVT in a fossil hominid can be regarded as an index for the reiterative neural substrate that makes voluntary speech possible. And that neural substrate also plays a critical role in making syntax, cognitive flexibility, and, yes, dancing possible. Speech, language, and some degree of cognitive flexibility surely were present earlier, but the presence of a SVT specialized for speech at the cost of choking places a date stamp on when brains like ours definitely existed [and presumably on when, or after which, modern languages did too]. 11

McNeill's work certainly indicates that speech has not become the exclusive means of communication to date, as it is usually complemented or supplemented by gestures. Kegl et al. ( 1999 , on Nicaraguan Sign Language), Goldin-Meadow ( 2003b , on home sign language), and the rest of the literature on sign language (see Woll, Ch. 4 below) suggest also that mankind could have evolved to become predominantly signers rather than speakers. 12 It appears to me that biology-style natural selection did drive the evolution of language conceived of as the cumulative manufacture of particular communicative technology under specific ecological pressures that favoured speech as its medium. Givón ( 1998 , 2002 ) cites advantages such as the ability to work and communicate at the same time and the ability to communicate in the dark or in spite of barriers to vision. MacNeilage ( 2008 ) and Allan ( 2010a : 233) also invoke the broadcast capacity of speech, a factor that, according to Dunbar ( 1996 ), fostered the emergence of speech, as it enables the speaker to ‘groom’ (interpreted here charitably in the sense of ‘socialize with’) several rather than one other person at a time. 13 Broadcasting certainly widens the radius of message transmission. Corballis ( 2010 : 122) and Mufwene ( 2010b : 305) invoke, in addition, the fact that speaking uses less energy, as it depends on compact articulators that move in a much smaller space and proceeds faster. To be sure, signing compensates for this in not being absolutely linear, though the signer's hands probably cannot keep up with the speed of a normal speaker's speech organs.

These considerations are nonetheless not the full story. Signing has its advantages too. As John W. Wenzel (p.c., 24 Jan. 2009) pointed out to me, signing is useful when silence is required, such as during group hunting, or in situations where speaking would place the speaker in danger (such as before a carnivorous predator), or when one is diving. It looks as though our hominin ancestors would have weighed the pros and cons of speech vs signing as the primary technology for communication. All these dangerous situations are not part of humans' default mode of existence, in safe environments and interacting in dyads or triads rather than in large groups. If Tomasello ( 2008 ) is right about the significance of social life as an ecological pressure on the emergence of language (see also Corballis 2010 : 116), then interactions in situations of no danger must have favoured the advantages that speech offers over signing, though we now know that one can express in signed language anything that can be expressed in spoken language. Interactions in situations of danger might explain why gestures have not been completely eliminated, especially if one factors in their tendency to be iconic.

However, Fitch ( 2010 : 442–5) articulates more explicitly some of the counterarguments developed since Hewes 1996 about this evolution, highlighting more advantages of signing over speech. Auditory attention is freed while signing, and gestures can be more efficient while teaching a partner to make tools (aside from the fact that actions are more often learned by observation and imitation than from somebody else's verbal teaching). Speech may be more energy-efficient, as it depends on articulators that are smaller than those involved in signing. However, as MacNeilage ( 2008 ) points out, the latter is not structured in exactly the same way. So, according to Fitch, there is still no convincing explanation for why speech has prevailed as the demographically dominant medium of human language.

It appears that the study of the evolution of language will be enriched by a better understanding of changing ecologies of the Homo genus, within and outside the species, during its protracted evolution. It will be informative to learn more about the role played by obvious major ecological factors such as its neural, mental, and anatomical structures, the evolving social structure, and all the pressures they exerted on the emergence and evolution of language. It is crucial to identify individuals as the most direct ecology that filters the external ecological pressures, because the structures and vitality of languages are determined not by concerted behaviours of populations but rather by accumulations of individual behaviours, which occur without foresight of consequences but just happen to converge toward certain outcomes. (See also Dor and Jablonka 2010 for a related discussion.) Each communicative act is determined by particular ecological pressures to which the communicator responds in the hic et nunc of the interaction.

Much of the recent scholarship has focused just on the emergence of speech, especially regarding the transition from ape-like holistic vocalizations to phonetic communication, and the relation of this aspect of the evolution of language to that of the relevant neural circuitry and anatomical structure. This is probably also an area that is less abstract than syntax and semantics and easier to speculate on with more paleontological evidence. Space and time constraints force me to focus here on Philip Lieberman, Peter MacNeilage, and Alison Wray, though many others deserve attention.

MacNeilage ( 2008 ) presents perhaps the most extensive discussion to date, which, as noted above, also questions, like Lieberman ( 2006 ), the empirical justification for the notion of UG and its relevance to accounting for the emergence of language. According to him, speech evolved in several steps, starting with the cooption for phonation of organs that had evolved for ingestion. The rhythmic pattern of the relevant organs was subsequently exapted for vocalization in CV syllables, which could be reduplicated as in child language; but reduplication was abandoned for ‘syllabic variegation and (the related) restrictions on VC co-occurrences’ in the production of words, as ‘pressures on speech systems to expand the size of their message sets’ increased (MacNeilage 2008 : 320). Eventually, longer utterances corresponding to sentences would evolve, but MacNeilage does not discuss this particular aspect of the evolution of language. However, he leaves ‘some latitude for different dialects and for individual differences’ to have been part of the emergence process. (He does not specifically tackle the monogenesis/polygenesis issue.) Against the role that UG, he writes:

For language in particular, mirror neurons provide the foundation for a more encompassing embodiment-based neuro-cognitive alternative to UG, one that goes beyond the mechanisms that lie between meaning and sound, considered separately, by including meaning and sound in the same picture, and giving us a better basis for the relationship. The embodiment perspective was primary in my attempt to say how the first words were made. I suggested that the phonetic structure of the first words resulted from the cognitive pairing of an observed action […] with a concept. (p. 326)

To be sure, MacNeilage brings us closer to articulating Wray's ( 2002 ) hypothesis that the Homo genus evolved from holistic vocalizations to phonetic communication. However, it is difficult to link both scholars here, largely because they do not start from the same working assumptions. MacNeilage does not subscribe to Bickerton's protolanguage any more than to UG. A natural bridge between them is Carstairs-McCarthy ( 1999 ), who, not unlike Jean-Jacques Rousseau, argues that Wray-style vocalizations would have been articulated into syllables first and later into the segments that these consist of. This evolution would have resulted in phonetic communication, though, as noted in § 1.2 , it raises the question of whether vowels and consonants arose at the same time, and whether ‘syllabic variegation’ started in the way hypothesized by Rousseau and by MacNeilage (with CV syllables) or initially just with variation in the quality of vowels—which would raise the question of how long the initial polysyllabic words consisting only of vowels could be. In the relevant passage quoted in § 1.2 , Rousseau suggests that the initial vocalizations consisted of vowels only, and consonants were innovated to mark syllabic boundaries. The fact that in all languages around the world the vast majority of syllabic peaks consist of vowels makes these considerations an interesting question for students of the evolution of language.

An informative complement of the above discussion on speech comes from Fitch's ( 2010 : § 8.3 ) summary of the state of the art about the evolution and functions of what linguists call ‘speech organs,’ which, based on the foregoing, are but exaptations of anatomical structure that evolved primarily for breathing and ingestion of food and liquids. Their use for speech is a perfect illustration of exaptation as defined by Gould and Vrba ( 1982 ):

A character, previously shaped by natural selection for a particular function (an adaptation), is coopted for a new use—cooptation. (2) A character whose origin cannot be ascribed to the direct action of natural selection (a nonadaptation), is coopted for a current use—cooptation. (Gould and Vrba, copied from Wikipedia , 1 Mar. 2011)

Fitch starts by noting, ‘Many animals open and close their jaw in the course of a call […] and changes in lip position are almost as common’ (2010: 311). The role of the descent of the larynx in the emergence of speech has been exaggerated, especially also in the interpretation of the feature as uniquely human. It is attested in other animals too, though in many of them the descent is not permanent. Its role in non-humans is to exaggerate size, and humans too exploit this feature: ‘it is really the descent of the tongue root […] that is the critical factor in speech production, rather than the descent of the larynx per se ’ (p. 312). Fitch agrees with Lieberman et al. ( 1972 ) that ‘hominids must have had some form of speech [intended as ‘language’] before the descent of the larynx’ (p. 313)—which does not mean that they had modern language.

Based on Lieberman ( 2006 , 2010 ), discussed above, one must ask when (i.e. at what stage of hominin evolution) the larynx descended. According to Lieberman ( 1984 , 2006 ), this otherwise maladaptive phenomenon (which puts humans at the risk of choking while ingesting) was probably a consequence of the reconfiguration of the basicranial structure after the hominins became bipedal. This says nothing about the phylogenetic time of the emergence of the feature. However, it is informative to know more specifically that the descent of the larynx was a consequence of the descent of the tongue root down the pharynx, pushing the larynx down, as happens now in human infants (Lieberman 2007 : 46). This anatomical feature must have been selected because of the advantages it conferred to the further evolution of speech into its modern form. Fitch concludes:

Not only does the descent of the larynx enlarge our phonetic repertoire, but it does so in a way that enhances speech encoding and decoding […] and it give[s] us the point vowels [/i/, /a/, /u/] that are found in all human languages, particularly the ‘supervowel’ /i/, which plays a central role in the vocal tract normalization. (2010: 315) [ T ] here must be functions of a descended larynx other than increased phonetic versatility […] leaving size exaggeration as the most plausible explanation. (p. 321; Fitch's italics) [T]he primary evolutionary changes required for [modern] spoken language were neural, not changes in vocal anatomy. (p. 362)

This conclusion confirms Darwin's ( 1871 ) position that the mind drove the emergence and evolution of human language, as it enabled hominins at successive stages of their phylogeny (mental and physical) to coopt parts of their anatomy to develop various stages of the language technology. In other words, with the increasing power of their minds, hominins and humans gradually domesticated their anatomies to produce the communicative technologies called languages. I submit again that the mind is really the most important feature that distinguishes mankind from other primates, and certainly other animals, although it does not function identically in all individuals, not any more than their physiologies are identical. Language is after all a collective gradual invention (by emergence).

The implications of this position are worth exploring further, since no two speakers have identical competences in any language they speak and/or sign. This interpretation is consistent with the notion of idiolect , whose features, as noted above, are determined as much by the variation in the interaction histories of speakers/signers (Mufwene 2008 : 120, 126) as by their individual learning capacities as determined by their mental and anatomical singularities (Dor and Jablonka 2010 : 139).

It should be obvious by now that students of the evolution of language do not share identical working assumptions. Nor have they focused on the same research questions. Some have been more interested in the particular interactive dynamics that made it possible for language as a communal phenomenon to emerge. This is especially the case for Croft ( 2000 , 2003b , 2008 ), Tomasello ( 2008 ), Tomasello et al. ( 2005 ), and Mufwene ( 2001 , 2005 ). Croft and Mufwene have patterned their approaches on biological evolution. Assuming an emergentist construction grammar, Croft has assumed that utterances are replicators, which vary across individuals and are in competition , which is explained by Mufwene 2008 as a situation in which the variants are not equally rated by users. The competition is resolved by selection , which can be interpreted as in biology, when a variant prevails over another or others, for any number of reasons in the relevant ecology.

Mufwene has gone as far as to argue that individual languages are the counterparts of viral species, with their organisms being the idiolects of particular speakers/signers. He posits a feature pool in which the variants produced by different speakers/signers are in competition and the machine that runs selection lies in the ecologies in which languages are used. The challenge is to define ecology , which has usually been understood as the social environment, with all the pressures emanating from population structure. I now think that, regarding the evolution of language, the ecology that matters the most lies in the different evolutionary stages of the mental and anatomical structures of the Homo genus. They determine what forms the relevant means of communication could assume.

Where both Croft and Mufwene hope to inspire those focusing on strictly phylogenetic topics and issues is especially the way they invoke innovators and spreaders/copiers (concepts also used by Tomasello 2008 and Tomasello et al. 2005 ) to account for the emergence of new linguistic features, which can, for convenience, be explained roughly here as applying to forms and constructions. As different innovators need not introduce the same features, competition arises, and various ecological factors determine which variants will prevail for which specific functions, there being room for free variation too. Selection is not made consciously, but is the cumulative outcome of choices made at different times by speakers/signers in their utterances. Since most interactions are dyadic or triadic, and since speakers do not normally hold meetings to state which particular variants they prefer, the question arises of how norms emerge. Both linguists have at times invoked the ‘invisible hand’ but have been invoking ‘self-organization’ in their recent works, after familiarizing themselves with complexity theory.

Exaptation has been a recurrent concept in the literature, underscoring the (self-)scaffolding aspect of language evolution. This has been implicit in many of the discussions above, but the term has increased in currency especially regarding the emergence of speech (see Oudeyer 2006 for an extensive discussion.) It is also applicable to the emergence of grammar, especially in the process called ‘grammaticization’ or ‘grammaticalization,’ whereby some verbs or nouns are exapted to be used as function words, such as complementizers or prepositions. Regarding the emergence of grammar itself, the boldest attempt is to be found in Heine and Kuteva ( 2007 ), who, in the footsteps of Herder and Max Müller, claim that the initial language consisted just of nouns and verbs; all the other categories are derivatives from these. They do not explain how, among other aspects of grammar, predication and different strategies for specifying reference and time evolved, or under what particular ecological pressures, though they explain, on the basis of synchronic linguistic evidence, how particular markers may have acquired grammatical meanings.

I will conclude this selective survey of topics addressed in the past two decades on the evolution of language with a brief discussion of the emergence of linguistic diversity. It is particularly significant because universals and typological variation have been central in linguistics since Greenberg's ( 1966 ) landmark publication on the subject. Even the generative linguists' preoccupation with principles and parameters as they are constrained by UG is a consequence of the pioneering work of Greenberg, though UG is not synonymous with language universals. The question is critical especially because most of the literature has assumed or suggested monogenesis; it has typically not mentioned variation in the protolanguage or the earliest ancestor of modern language. As a matter of fact, as noted above, Swadesh (2006 [1971] ) assumed that because the original ancestor of modern language was instinctive, there could not be significant variation in it. Let's thus focus on when speech started to emerge. Here is what Jim Hurford, one of the veteran students of the evolution of language, has to tell us:

Summarizing the factors contributing to linguistic diversity, (1) the fact that languages are learned, rather than coded into the genes, (2) the arbitrariness of the sign, and (3) the prevalence of horizontal transmission allow for great diversity, but this is significantly constrained by (4) biological factors such as memory and processing limitations, which may or may not be specific to the Language domain. (Hurford 2008 : 251).

These factors account more for idiolectal variation, as there is no faithful replication in language learning (Lass 1997 ), than for the emergence of typological variation across languages. If populations can choose to build their languages on different words and only on overlapping phonetic inventories, what should keep them from developing different combination patterns of these units into larger utterances and therefore different grammars? If we interpret phonology as the grammar of sounds and assume that grammars are consequences of the ways units are combined together and structured into larger and larger (hierarchical) units, why should we expect the relevant hominin/human populations at the different stages of the evolution of language to have done exactly the same thing? 14 After all, the paleontological evidence does not suggest that Homo sapiens sapiens dispersed to the world out of one village in Africa; hominin fossils have been found in a vast area of East (and South) Africa. Shouldn't it be normal to assume that, having reached the same stage of mental and anatomical evolution, hominin populations developed languages that were comparable but not identical in their architectures? They did not have to package information in identical ways, no more than they developed identical cultures.

Another dimension of the scholarship on the evolution of language today lies in computer modelling, which I will not discuss here, due to lack of space. The rewards depend largely on the assumptions that underlie the models. When they are empirically grounded, they become important research tools, such as when used by Philip Lieberman and his associates to determine whether the Neanderthal was capable of speaking. When accurately informed and well designed, modelling can help empirical research reformulate some of its questions about a distant past that cannot be recreated. (See e.g. Oudeyer 2006 on self-organization in the emergence of language and Steels 2011 on the emergence of communal norms.)

Last but not least, there is all the research on animal communication, especially intraspecifically among non-human primates and between humans and some great apes. It is expected to inform research on the evolution of language insofar as scholars can identify both behaviours that may have been inherited from our common ancestors millions of years ago and later homologous evolutions from features shared earlier in our common phylogenetic ancestry. Unfortunately, I can do even less justice to this topic here than to those discussed above. Comparisons by Tomasello ( 2008 ) regarding joint attention and cooperation highlight the significant role which these social factors that we do not share with the other primates played in the phylogenetic emergence of human language. Fitch's and Lieberman's comparisons regarding primates' supralaryngeal vocal structures also reveal important differences that rule out the possibility that they would have developed human-like speech even if they were endowed with the same kind of mind as we have. On the other hand, discoveries that non-human primates share with us mirror neurons, the FOXP2 gene, and some of the specialized functions associated with Broca's area suggest that the human mind had a greater role to play in the emergence of language than may have been assumed before, which is precisely why our phylogenetic cousins have not even developed some symbolic-iconic system similar to sign language. Language may be a more cultural phenomenon than some of us have assumed. I submit that language is indeed one of the facets of human culture, and that both linguists and anthropologists may have been misguided in speaking of language and/in culture as if they were opposed to each other on the same plane.

On the other hand, there is a growing literature suggesting that differences between animals and humans are more a matter of degree than dichotomy. Some of the capacities having to do with mirror neurons and mind-reading are very similar, which raises the question of whether human intelligence is not a consequence of the particular ways various parts of the brains and modules of the mind interact.

It has long been assumed that animal means of communication are innate but that of humans are not. However, it has also become evident that a certain amount of learning is involved in, for instance, bird songs (Margoliash 2010 ). Past the critical period, the bird does not develop the right song for its con-specifics! Besides, some birds exposed to alter-specifics' songs acquire it rather than that of their con-specifics. This and other factors raise the question of whether there is such a thing as language or cultural ‘transmission,’ analogous to gene transmission in biology, especially among humans. Unlike transmission, which, in the absence of mutations, guarantees faithful maintenance of inherited traits, learning by inference almost ensures modification of the target features, which is more consistent with language ‘acquisition,’ interpreted as system reconstruction (Mufwene 2001 , 2008 ). Students of cultural evolution, such as Richerson and Boyd ( 2005 ), Mithen ( 2005 ), and Mesoudi et al. ( 2004 ), have kept up impressively with the scholarship on language evolution. We have everything to learn in reading them too.

Some of the more popular studies of animal communication have focused on what can be learned from teaching human language or an artificial system made by humans to primates (e.g. Segerdahl et al. 2005 ). It appears that lexigrams constitute a seriously impoverished system that does not go beyond the telegraphic stage in child language. Although great apes such as Kanzi have been credited with the ability to understand human speech, it is not obvious that they can follow a narrative the way a human child can. This highlights mental differences between non-human primates and us, though differences in mental capacities are also a matter of degree. Nonetheless, it appears that the less than 2 per cent genetic differences between chimpanzees and humans have entailed exponential cultural, and more specifically language-related, differences.

The overall approach has assumed that humans are more evolved than non-human primates, rather than just being different from them. We have not yet accounted for why we cannot learn to communicate the way they do! Answers to this question may equally well inform us about how different our minds really are from theirs or, more accurately, about how communication in all species is jointly constrained by their respective mental and anatomical ecologies. We have discussed culture as if it were peculiar to humans, whereas it can be interpreted as customary ways in which members of a particular population behave and do things. Cultural differences can also inform us about how different social structures have influenced what needs to be communicated and what kinds of systems are needed to convey the relevant pieces of information.

1.4 Conclusions: Older vs Current Approaches to the Evolution of Language

As aptly expressed by Fitch ( 2010 : 389),

regarding language evolution, there are very few new hypotheses under the sun, and current debates can and should pick up where our scholarly predecessors left off. […T]here are real insights in the older literature which remain unappreciated.

Hombert and Lenclud (in press) note likewise that a number of the positions assumed today were already defended by philosophers of the eighteenth century. For instance, the claim that language is what distinguishes mankind the most clearly from the animal kingdom is already evident in Condillac. It is also hard to sharply distinguish eighteenth-century arguments for the emergence of human language out of instinctive cries and gestures from Bickerton's position that the predecessor of his ‘protolanguage’ consisted of holistic vocalizations and gestures. The idea of gradualism in the evolution of language is not new either; and Rousseau had already articulated the significance of social interactions as a prerequisite to the emergence of language. And one can keep on identifying a number of current hypotheses which are hardly different from earlier speculations on the subject.

An important difference between us and those philosophers and philologists before the nineteenth century, and in some cases up to then, is that we no longer assume that our hominin ancestors up to 200,000–100,000 years ago were just like us, except that they were either created by God or just happened to inhabit our planet long before we did, or just were mentally inferior to us. We now approach the subject taking into account what communicative architecture would have been possible at various stages of hominin evolution. We ask: since Homo habilis was anatomically different from Homo erectus , what kind of language would those remote ancestors of ours have been capable of developing even if they were equipped with the same kind of mental capacity as us? The same applies to Homo erectus and archaic Homo sapiens . A similar question arises regarding the complexity of utterances relative to the complexity of the hominin mind and/or social organization. What kinds of ecological pressures did they exercise on the evolution of language? Lieberman ( 1984 ), Bickerton ( 1990 ), Tomasello ( 2008 ), Corballis ( 2002 , 2010 ), MacNeilage ( 2008 ), and Fitch ( 2010 ) are good illustrations of this ecological approach, although they do not draw identical conclusions. 15

Another important difference between us and philosophers and philologists before the nineteenth century is that, better than Socrates in Plato's Cratylus , we are more aware of the speculative nature of our hypotheses in this research area. With few exceptions, scholars have generally been more critical and more cautious, revealing more awareness of the limitations of the state of the art.

Whether or not we acknowledge it, Charles Darwin has also exercised a long-lasting impact on us: most scholars today do not assume that language was God-given (presuming creationists to be in the minority). Even Chomsky's account that UG emerged by some rewiring of the brain is a Darwinian explanation, because Darwin made allowance for mutations, and UG could have emerged only at a particular stage of hominin evolution, quite late. Besides, mutations are probably also the best explanations from all the changes in hominin evolution, with the mutants prevailing and the rest evolving as consequences of those mutations.

We also now think of the architecture of languages as modular. This is an idea that does not appear in the earlier literature. It also frees scholars from having to assume that every component of modern language must have evolved at the same time as the others. Nor do we have to assume that the anatomical and mental structures that were coopted in the apparently gradual emergence of language all evolved at the same time. Even in assuming that the mind domesticated hominin and human anatomy for the production of language, it need not have coopted the different organs concurrently. This is the kind of evolution suggested by the paleontological evidence that experts have adduced, leading both Michael Corballis and Philip Lieberman to now conclude that speech-dominated communication must have emerged more recently, 50,000–30,000 years ago, not 500,000 years ago. This thinking is consistent with Hombert and Lenclud's (in press) conclusion that the capacity for language is a derivative and consequence of hominins'/humans' evolving cognitive capacity.

It is more and more evident that the subject matter of the evolution of language is multifaceted, having to do with the mechanical/architectural aspects of language, with the particular anatomical organs coopted for its production and perception, with the mental aspects of the technology (including the formation of concepts and their combinations into larger chunks), and with the apparently social motivation for producing the technology. It would be difficult, if not impossible, to explain how modern humans' linguistic communication got where it is now without answering various questions that pertain to these different facets of the subject matter. It is part of understanding how the Homo genus has evolved over the past two or three million years biologically, anatomically, mentally, and socially.

I wrote this chapter while I was a fellow at the Collegium de Lyon, from 15 Sept. 2010 to 15 July 2011. I am very grateful to the Institute and its administrative staff for the financial and logistic support that enabled me to pursue my research on the phylogenetic emergence of language. I am also indebted to Keith Allan, Barbara Davis, Paul Keyser, and Ioana Chitoran for constructive comments on my first draft. I am alone responsible for all the remaining shortcomings.

Hombert and Lenclud (in press) identify another, less well-recalled account also from the book of Genesis. God reportedly told Noah and his children to be fecund and populate the world. Subsequently, the descendants of Sem, Cham, and Japhet spread all over the world and built nations where they spoke different languages. Here one also finds an early, if not the earliest, version of the assumption that every nation must be identified through the language spoken by its population.

Yet one can argue that syntax, also phonology and morphology, are just a consequence of linearity, constrained though it is by other, cognitive or pragmatic factors.

Note that, although the book is titled The Genesis of Grammar , Heine and Kuteva offer no plausible hypothesis of how the overall grammar evolved, beyond the emergence of free grammatical morphemes and a few inflections.

According to Hombert and Lenclud (in press), much of this practice has to do with what the linguists thought was the subject matter of their discipline. Ferdinand de Saussure was allegedly more interested in languages ( les langues ), which consist of systems, are unified, but are not organic. He was less interested in language ( le langage ), which he putatively considered ‘multiform and heteroclitic’ (as noted above), straddling domains that are ‘physical, physiological, and psychic’ [i.e. mental?].

According to Dor and Jablonka ( 2010 : 139), this variation ‘is inevitable given genetic differences, anatomical differences between brains, differences among ontogenies, and differences of processes of socialization,’ which amount to ‘different developmental trajectories.’ Mufwene ( 2008 ) underestimated the consequences of biological variation across individuals when he invoked ‘different interactional histories’ (pp. 120, 126) in his account of inter-idiolectal variation.

MacNeilage ( 2008 ) doubts that the notion of UG is worth positing at all. He suggests that it is a consequence of language emergence rather than its cause. According to him, ‘ there is currently no validity to the claim that UG has a specific genetic basis ’ (p. 298, MacNeilage's italics).

Hombert and Lenclud (in press) state more specifically: ‘The capacity for language is considered as a derivative capacity and its emergence as the secondary or induced effect of the emergence of a general cognitive competence. It may have followed from the aptitude that only humans would have been endowed with to read and share the other's intentions’ (my translation).

Note that some scholars, including Corballis ( 2010 ) and Lieberman ( 2010 ), now think that modern language may not have originated before 50,000 years ago or so, thus much later than Homo erectus , apparently during Homo sapiens sapiens , and this event may have coincided with the last exodus out of East Africa. (I return to this below.)

Dor and Jablonka ( 2010 : 140) comment on this as follows: ‘as more and more elements came to be canalized, and the language came to assume a certain architectural logic, the logic gradually imposed system constraints on what the next viable innovation would be.’ This underscores Wimsatt and Griesemer's ( 2007 ) idea that current forms and/or structures provide the scaffold for innovations. From the point of view of the evolving system, they refer to this extension of the notion as ‘self-scaffolding.’

[McCarthy, Strait, Yates, and Lieberman (forthcoming) is still being revised as we go to press, and its title is not yet determined.– Editor ] The shift from Corballis' ( 2002 ) and Lieberman's ( 2002 ) early conclusion about when phonetic language emerged underscores the stronger empirical foundations of today's speculations on the evolution of language. New paleontological discoveries and a better understanding to modern humans' neural circuitry will shed more light on the subject matter.

MacNeilage argues against this perspective, citing not only the assumption that the ability to vocalize started before Homo habilis but also ‘the greater organizational similarity between speech and birdsong than between speech and sign language’ (2008: 309).

Bickerton ( 2010 : 28) disputes this account, on the grounds that ‘it fails the ten-word test, what you might call the test of immediate utility.’ To be sure, grooming falls in the category of ecological explanation; it provides actuation for the emergence of language but says nothing about how the emergence occurred. It is undoubtedly one of the many social reasons and is not mutually exclusive with any particular account of how things proceeded, including Bickerton's own account.

A convenient nonlinguistic illustration of this may be found in how engineers using similar algorithms constrained by the same principles produce technologies (such as computers and derivative products) that are not identical in their architectures and functionalities.

As a matter of fact, Bickerton ( 2010 ) now discusses the evolution of language from the point of view of ‘niche construction,’ which Laland ( 2007 : 35) characterizes as ‘the process whereby organisms, through their metabolism, their activities, and their choices, modify [their] niches.’ (See also Odling-Smee et al. 2003 for a more elaborate discussion.) The subtitle of Bickerton ( 2010 ) captures the idea adequately: How Humans Made Language, How Language Made Humans .

• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.