Vous êtes sur la page 1sur 13

Journal of Neurolinguistics 43 (2017) 120e132

Contents lists available at ScienceDirect

Journal of Neurolinguistics
journal homepage: www.elsevier.com/locate/jneuroling

Perspective

The evolution of language: Sharing our mental lives


Michael C. Corballis
School of Psychology, University of Auckland, Private Bag 92011, Auckland 1142, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history: There are two broad views on the evolution of language. One is that language involved a
Received 12 March 2016 restructuring of thought, unique to humans, which occurred within the past 100,000 years.
Received in revised form 7 June 2016 The other view, sketched in this article, is that human thought evolved gradually and
Accepted 7 June 2016
largely independently of language, with properties of generativity and theory of mind.
Available online 24 June 2016
Generative thought includes mental time traveldthe capacity to mentally relive past ex-
periences, imagine future ones, and even create purely imaginary scenarios. Behavioral
and neurophysiological evidence suggests that this may go far back in evolution. Theory of
mind may be more recent, and allows us to imagine being someone else; it is also critical
to communicative language itself, which depends heavily on shared trains of thought.
Language emerged as a device for sharing our mental stories; it probably began in the
Pleistocene as pantomime, but was gradually conventionalized, depending increasingly on
culturally defined symbols and rules rather than on physical resemblance. The emergence
of speech itself was a late stage in the process of conventionalization, and maximized
communicative efficiency. Language allows us to share the knowledge and experiences of
others, vastly increasing our mental resources.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Language is widely considered to be uniquely human, and the result of some emergent process rather than the product of
evolution through natural selection. The Bible teaches that language was a gift from God, and this view was held by many of
the major philosophers, such as Rene  Descartes, Thomas Hobbes, and John Locke. Even Alfred Russel Wallace, jointly credited
with Darwin as the founder of the theory of evolution through natural selection, could not believe that natural selection could
account for the vast differences in mind between human and apesddifferences that included the faculty of language. Frie-
drich Max Müller, Professor of Philology at the University of Oxford, greeted Darwin’s theory with the statement that “There is
between the whole animal kingdom on one side, and man, even in his lowest state, on the other, a barrier which no animal
has ever crossed, and that barrier isdLanguage” (Müller, 1873, p. 666).
To Descartes, nonhuman animals were automata, operating according to mechanical principles, but humans were capable
of a freedom of thought that defied mechanics. This was most evident, he thought, in language itself; in a 1646 letter to the
Marquess of Newcastle, he wrote that:
… none of our external actions can show anyone who examines them that our body is not just a self-moving machine
but contains a soul with thoughts, with the exception of words …
(Descartes, 1646/1970, p. 206).

E-mail address: m.corballis@auckland.ac.nz.

http://dx.doi.org/10.1016/j.jneuroling.2016.06.003
0911-6044/© 2016 Elsevier Ltd. All rights reserved.
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 121

The freedom from mechanical constraint lent language its generativity, the capacity to create a potential infinity of possible
meanings. The Prussian philosopher and linguist Wilhelm von Humboldt (1836/1999, p. 91) described this as the “infinite use
of finite means.”
The idea has persisted into recent times. The psychologist David Premack (1985) writes that “Human language is an
embarrassment for evolutionary theory because it is vastly more powerful than one can account for in terms of selective
fitness” (p. 282). Noam Chomsky, echoing von Humboldt, regards generativitydor what he has also called “discrete infinity”
(e.g., Chomsky, 2007)das the critical feature that makes human language and thought distinctive, and like Premack suggests
that it could not have evolved through natural selection. Instead, he proposed that it emerged within the past 100,000 years,
at first in a single individual whom he whimsically names Prometheus. Unlike Premack (2007), Chomsky (2007) downplays
the magnitude of change that produced generative language: “Perhaps it was an automatic consequence of absolute brain size
… or perhaps some minor chance mutation” (p. 18). In Chomsky’s view, the essence of generativity is what he terms “un-
bounded Merge,” the capacity to merge elements recursively, with merged elements themselves merged, create hierarchical
structures of potentially unlimited complexity and variety (Chomsky, 2007).
Based on changes in the archeological record, archaeologists have similarly proposed a sudden cognitive change within
the past 100,000 years, or even more recently, that transformed the human mind. Richard Klein (2008) writes that it is “at
least plausible to tie the basic behavioral shift at 50 ka to a fortuitous mutation that created the fully modern brain” (p. 271).
John Hoffecker (2007) writes similarly:
Language is a plausible source for the sudden and dramatic change in the archaeological record [after 40 ka] because:
(a) it is difficult to conceive of how the system for generating sentences (i.e., syntax) could have evolved gradually, and
(b) it must have had far-reaching effects on all aspects of behavior by creating the collective brain (p. 379).
Ian Tattersall (2012) seems almost bemused that such a momentous event could have taken place so recently and so
rapidly:
Our ancestors made an almost unimaginable transition from a non-symbolic, nonlinguistic way of processing infor-
mation and communicating information about the world to the symbolic and linguistic condition we enjoy today. It is a
qualitative leap in cognitive state unparalleled in history. Indeed … the only reason we have for believing that such a
leap could ever have been made, is that it was made. And it seems to have been made well after the acquisition by our
species of its distinctive modern form (p. 199).
These various statements pose a problem for the theory of evolution itself. Darwin himself wrote:
If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous,
successive, slight modifications, my theory would absolutely break down. But I can find no such case (Darwin, 1859,
p. 158).
Could language be the case that Darwin feared?
But not all hold to the view that language emerged suddenly. In a classic article, Pinker and Bloom (1990) argue that “there
is every reason to believe that language has been shaped by natural selection as it is understood within the orthodox
‘synthetic’ or ‘neo-Darwinian’ theory of evolution (p. 708).” Given the extraordinary success of evolutionary theory to account
for the emergence of complex biological phenomena, we should, in my view, be reluctant to make an exception of the human
mind, or more specifically of language. The problem is daunting, and has even been called “the hardest problem in science”
(Christiansen & Kirby, 2003, p. 1).
In an attempt to meet the evolutionary challenge, in the remainder of this article I sketch out scenarios that are more
consistent with an evolutionary account than with the idea that language was the result of a cataclysmic event restricted to
our own species.

1.1. What is language?

In much of the discussion about the evolution of language, there is ambiguity about what is meant by language itself. Most
are clear that language cannot be simply defined as communication. If that were so, there could be no suggestion that human
language is unique, since communication is ubiquitous among species of all kindsdland animals, birds, fish, insects.
At an opposite extreme, language is often considered an aspect of thought itself, with the implication that humans evolved
a uniquely symbolic way of thinking, with communicative language a mere by-product. Thus Terrence Deacon (1997) referred
to humans as “the symbolic species.” To Chomsky, too, language is primarily a mode of thought, which he terms “internal
language,” or I-language, and the communicative aspect is relatively uninteresting, or even trivial. But it is still the generative
nature of thought that captures its uniqueness; Unbounded Merge, then, is fundamentally a property of symbolic thought,
and not of the manner in which our thoughts are communicated. Indeed Chomsky considers unbounded Merge to be the basis
of what he terms universal grammar, common to all humans and yet also uniquely human (Chomsky, 2010). External lan-
guages, or E-languages, are extraordinarily diverse, with some 6000 different language in the world, each more or less
impenetrable to every other (Evans, 2009).
An intermediate view, perhaps a more commonsense one and the one taken here, is that language is indeed primarily a
means of communicating our thoughts, and some of its distinctive properties derive from the nature of the thoughts we wish
122 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

to communicate. Thus Daniel Dor (2015) defines language as “the sharing of experience,” and also notes that the emphasis on
the communicative aspect “turns the Chomsky proposal on its head” (p. 2). Daniel Everett (2012) writes similarly that
“languages are tools. Tools to solve the twin problems of communication and social cohesion. Tools shaped by the distinctive
pressures of their cultural nichesdpressures that include cultural values and history and which account in many cases for the
similarities and differences between languages” (p. 6).
By distinguishing language from thought, we can then consider how thought itself evolved independently of how lan-
guage evolved as a device for communicating our thoughts. Language, then, does depend on thought insofar as it is shaped to
convey thoughts, but the structure of thoughts probably long precedes the emergence of language itself. As explained below,
this may well include generativity itself, the capacity to generate new thoughts. What makes human language arguably
differs from other forms of animal communication is that it is designed to share thoughts in intentional fashion, whereas most
if not all nonhuman communication is based on simple aspects of the external environment or on emotional states, and lacks
the generativity that enables humans to construct a potential infinity of different messages. But that generativity applies to
our thoughts, and was probably present long before our species invented ways to communicate them. Nonhuman animals
may well have generative thoughts too, but lack the capacity to share them.
Another capacity necessary to language is what has been termed theory of mind, the capacity to understand what others
are thinking. This may well have emerged late in hominin evolution, and may be one of the components that does make
human language distinct. Even so, there are probably recognizable precursors in other species, notably in great apes, that cast
a more continuous light on the manner in which language evolved.
I next discuss how these aspects of thought evolved, and then consider how humans developed appropriate ways of
communicating their thoughts.

2. Mental precursors of language

2.1. Generative thinking

The theme to be developed below is that generativity derives, not from language, but from the ability to generate thoughts.
This derives in turn from experience in time and space, and the ability to travel mentally away from the present. The capacity
to mentally relive past happenings or imagine possible future ones has come to be called mental time travel (Tulving, 1985;
Suddendorf & Corballis, 1997, 2007). It is fundamentally generative because it involves combinations of disparate elements in
often novel combinations, and allows us to imagine events that have never occurred. It corresponds one of the “design
features” of language known as displacement, the capacity to refer to events that are not present (Hockett, 1960). Derek
Bickerton (2014) has suggested that displacement, rather than the use of arbitrary symbols, is “the road into language” (p. 93).
Hockett suggested that displacement distinguishes language from all other forms of animal communication, but the un-
derlying capacity to escape from the present may well be ubiquitous in animals that move in space, and need to remember
where they have been and what happened there, and imagine where they might go next and what they might do there.
Language is a device for communicating our mental travels, but is not part of the travels themselves.
Mental time travel of course lacks the precision of actual travel, but compensates in speed. I can travel physically to Paris
and enjoy the sights, sounds, and tastes, but it takes a long time to get there, even in the modern age. Mentally, though, I can
be there in an instant. Mental time travel is also open to generativity and embellishment. One aspect of mental time travel is
what Tulving termed episodic memory, and it has long been known that our memories are constructed rather than simply
replays of the past (Bartlett, 1932). Indeed, our memories are often distorted, or even false (Loftus & Ketcham, 1994; Roediger
& McDermott, 1995). As Ulric Neisser (2008) put it, “Remembering is not like playing back a tape or looking at a picture; it is
more like telling a story” (p. 88). Even more generative is the construction of future episodes, since these have not yet
occurred, and indeed may never do so. It is now established that the generation of past and future events activates closely
overlapping areas in the brain (Addis, Wong, & Schacter, 2007). At the hub of the common system is the hippocampus
(Martin, Schacter, Corballis, & Addis, 2011), and case studies show that destruction of the hippocampus results in a profound
inability to recall past events or imagine future ones (Corkin, 2013; Wearing, 2005).
It is often claimed that mental time travel is uniquely human (Suddendorf & Corballis, 1997, 2007), perhaps because
language itself seems ideally calibrated to allow us to share our time travels. However, lack of language capacity in nonhuman
species need not mean an inability to travel mentally in time, and evidence now suggests that some form of mental time travel
may go far back in evolution. Behavioral evidence suggests that scrub jays can refer to both past and future events (Clayton,
Bussey, & Dickinson, 2003). But the more compelling evidence for a precursor to mental time travel may come from studies of
hippocampal activity in nonhumans, and more particularly in the rat.
It has long been known that the firing of single cells in the rat hippocampus records the current location of the animal,
leading to the notion of the hippocampus as a cognitive mapdan internal GPS system (O’Keefe & Nadel, 1978). Since then,
experiments have shown that hippocampal activity may persist in short-wave ripples (SWRs) after the animal has been
removed from a spatial environment, such as a maze, and these ripples map out trajectories in the environment. These
trajectories are sometimes “replays” of trajectories previously taken, sometimes reverse of those trajectories (Foster & Wilson,
2006), sometimes trajectories the animal did not take (Gupta, van der Meer, Touretzky, & Redish, 2010), some of which may
be anticipations of future trajectories (Pfeiffer & Foster, 2013). Reviewing this evidence, Moser, Rowland, and Moser (2015)
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 123

write that “the replay phenomenon may support ‘mental time travel’ … through the spatial map, both forward and backward
in time (p. 6).”
Hippocampal firing is also modulated by cells in the adjacent entorhinal cortex. These cells are laid out in a grid pattern,
and different grid-cell modules are dedicated to different geometric aspects of the environment and its relation to the animal.
These include its contours, the orientation of the animal’s head in the environment, and the spatial scale (Moser et al., 2015).
The combinations of just a few of the entorhinal grid modules underlying aspects such as orientation and scale can generate a
vast number of patterns of activity in the hippocampus, since each grid module can assume different levels. Moser and her
colleagues write “The mechanism would be similar to that of a combination lock in which 10,000 combinations may be
generated with only four modules of 10 possible values, or that of an alphabet in which all words of a language can be
generated by combining only 30 letters or less (p.11).” In short, the mechanism is generative.
Within the hippocampus itself, different spatial scales are represented along the axis of the hippocampus, ranging from a
more detailed, close view at the rearward end toward a broader, more distant view at the front end. In the rat, the rearward
portion operates on a region of about 1 m in width, while at the frontward the ventral end the width is about 10 m (Strange,
Witter, Lein, & Moser, 2014). These various properties allow rapid adjustment of the internal map to accommodate changes in
viewing angle, relation to contours, and zooming.
Zooming is also a ubiquitous property of human spatial understanding. I can locate myself in my office as I type this
sentence, or I can imagine where I am located in my apartment, in my city, or even within the country as a whole. Zooming
may also apply to remembered locations, or imagined future ones. Variations in scale may apply also to our understanding of
events in the world as well as simply to spatial awareness. In one study, people shown sequences of four videos of different
events, along with narratives describing the events. At one level, narratives were linked to each video, encouraging attention
to individual details. At the next level narratives linked a pair of videos, and at the final level a narrative linked all four videos.
As the people processed these narratives, activation in the hippocampus progressed from the posterior to the anterior end as
the scales of the narrative shifted from small and detailed to larger and more global (Collin, Milivojevic & Doeller, 2015). This
shift probably occurs as you read a novel, with specific information registering as you read each page, but more global un-
derstanding as you proceed and later remember the novel.
Orientation can also be adjusted, in imagined as well as in perceived space. In a classic experiment, patients with left
hemingelect, a condition in which damage to the right hemisphere results in failure to attend to objects or events on the left
side of space, were asked to imagine themselves in the Piazza del Duomo in Milan. They were first asked to imagine
themselves at one end of the square facing the cathedral, and to list the landmarks they could identify. The systematically
neglected those on the left. When then asked to imagine themselves at the cathedral end, facing the other way, they sys-
tematically neglected those they had previously identified, now on the left, and identified those they had previously
neglected (Bisiach & Luzzatti, 1978). This is a striking demonstration not only of the flexibility of spatial orientation, but also of
the deficits hemineglect can induce in imagined as well as in perceived space.
Moser et al. also summarize evidence that place cells in the rat hippocampus respond not only to specific locations, but
also to features of environments they have explored, such as odors, touch sensations, and the timing of events. This suggests
that the hippocampus acts not only as a cognitive map, but as a template for mental time travel that includes nonspatial
elements. Similar properties apply to hippocampal action in humans. In one study, human patients about to undergo surgery
had electrodes implanted in place cells in the medial temporal lobe, in an attempt to locate the source of epileptic seizures.
They were given the task of navigating a virtual town on a computer screen, and delivering items to one of the stores in town.
They were then asked to recall only the items, and not the location to which they were delivered. The act of recall, though,
activated the place cells corresponding to that location, effectively mirroring the replay of place-cell activity in the rat brain
(Miller et al., 2013).
The hippocampus and neighboring entorhinal cortex, in rats as in humans, can therefore accommodate a vast number of
patterns of activity corresponding to episodic eventsdpast, present and future, associated with nonspatial elements, and
subjected to transformations of scale and orientation. It is of course likely that humans can accommodate a wider range of
possibilities in space, time, and associated elements than can the humble rat, but the fundamentals of the system seem to go
far back in evolution. The generativity of the system, moreover, may be the principal source of generativity in language. That
is, language is generative because it allows us to describe the products of generative experience and imagination, and not
because of some fundamental property restricted to the linguistic system itself. One might say that walking is generative,
because it allows us to move to a potentially infinite number of locations, but the generativity lies in the variety of locations
and not in the act of walking itself.

2.2. Theory of mind

The generativity of mental travel is even further extended by our capacity to take the mental perspective of others, through
what has been termed theory of mind. This refers to the ability to attribute mental states, such as beliefs, attitudes, desires, or
intentions to others, and to oneself, and to understand that the mental states of others may differ from our own. To the extent
that we can do this, we can extend our mental time travels beyond our own experiences into the mental experiences of others.
This is well illustrated by fiction, TV soaps, or the general round of office gossip. Like mental wandering in time and space,
wandering into the minds of others is generative, as when we create the imaginary characters that inhabit our fictional lives,
or the “imaginary companions” that many young children invent (Taylor, 1999). Mind wandering, which occurs
124 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

spontaneously nearly 50 percent of the time during our waking hours (Killingsworth & Gilbert, 2010), depends on a broad
network dubbed the default-mode network (Raichle et al., 2001), which includes the hippocampus. It is active, according to
Buckner, Andrews-Hanna, and Schacter (2008) “when individuals are engaged in internally focused tasks including auto-
biographical, memory retrieval, envisioning the future, and conceiving the perspectives of others, p. 1).” These three in-
gredients form the basis of story-telling, which may well be a uniquely human pastime, although it may derive from play, an
activity widely observed in species other than humans (Boyd, 2009).
Premack and Woodruff (1978) once raised the question “Does the chimpanzee have a theory of mind,” and the more
general question of whether this applies to nonhuman animals has since had a checkered history. As late as 2008, Penn,
Holyoak and Povinelli argued that even chimpanzees, our closest nonhuman relatives, have no theory of mind, describing
such attributions as “Darwin’s mistake,” while Call and Tomasello (2008) conclude, more generously, that the 30 years of
research showed chimpanzees to have an understanding of the goals, intentions, perceptions, and knowledge of others, but
no understanding of others’ beliefs or desires. However this is issue is resolved, it seems reasonable to conclude humans have
a much more highly developed theory of mind than do other living species.
To summarize, then, some of the fundamental properties of language probably derive from the nature of thought itself,
and our manner of internally representing experience. These properties include the basic generativity of experiences in time
and space, and in our capacity to understand the minds of others. Theory of mind effectively adds a dimension to our mental
time travels, whether real or imaginary, and has enabled humans to become storytellers. The literary scholar John Niles (2010)
suggested that our species should be renamed Homo narransdthe story-tellers. Our stories, our mental travels, are based
strongly on imagination, in part because they generally refer to travels that have not yet occurred, or indeed never occur, but
even our stories of past experiences are often more imaginary than real. Generativity is often regarded as one of the defining
features of language, but this is derived more from the nature of the thoughts to be communicated rather than from the
means of communication itself.

3. Communicating experience

3.1. Underdeterminacy

Theory of mind also plays a critical role in language as an effective communication system, and not just as providing extra
content. For two people to have a meaningful conversation, each must have an understanding of what the other is thinking
prior to any words being uttered. The philosopher Paul Grice (1989) expresses this as follows:
He said that P; he could not have done this unless he thought that Q; he knows (and knows that I know that he knows)
that I will realize that it is necessary to suppose that Q; he has done nothing to stop me thinking that Q; so he intends
me to think, or is at least willing for me to think that Q (pp. 30e31).
That is, you should not only know what the other is thinking, but also know that the other knows this too. The manner in
which people extract information from often minimal utterances is explored in some detail by Sperber and Wilson (2002).
Because communicative language depends critically on theory of mind, it is characterized by what Scott-Phillips (2015)
calls underdeterminacy. Uttered words themselves are seldom if ever sufficient to specify precise meaning. This can be
partially illustrated through the phenomenon of polysemy, the fact that many words have multiple meanings, and speakers
and listeners must agree on the intended meaning. The English word set, for example, can serve as verb, adjective or noun,
and my dictionary identifies 105 different meanings. According to Scott-Phillips, it is this underdeterminacy that makes
language unique to humans. On this view, then, the uniqueness of language lies not so much in the symbolic nature of
underlying thought, as proposed by Chomsky, as in our ability to read each other’s minds, and so convey our thoughts to each
other.
Underdeterminacy need not be restricted to language. Scott-Phillips gives the example of a person in a coffee shop,
catching the attention of a waiter, and tilting his coffee cup in a stylized way. The waiter comes over and fills his cup. Here, the
gesture is underdetermined, but the thirsty customer and the waiter both know what’s in the mind of the other. Scott-Phillips
refers to this kind of communication as ostensive-inferentialdostensive because it involves an act of showing, and inferential
because it require an act of interpretation. We carry out many such acts, such as shrugging, pointing, raising eyebrows, but the
most complex form is language itself, whether spoken or signed.
It is questionable whether such acts are unique to humans. For instance, Hobaiter and Byrne (2011) recorded over 4000
instances of gestures among chimpanzees in the wild, most of them occurring during play. These appeared to be intentional
and directed to other chimps, and to convey meaning. Their meanings, though, were somewhat loose and ambiguous, and
dependent on context. In short, they appeared to meet Scott-Phillips’ definition of ostensive-inferential. Language, then, may
be a refinement and extension of such gestures. And as outlined earlier, great apes may well be capable of theory of mind,
albeit perhaps to a limited degree.

3.2. Gestural origins

The nonlinguistic examples of ostensive-inferential communication given by Scott-Phillips are gestural rather than vocal,
and there indeed seems good reason to suppose that the origins of language itself lie in gesture. In primates, at least, the
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 125

requirements for intentional communication are better met by the gestural system than by the vocal system. Vocal calls are
well adapted to relatively fixed situations, such as danger, kin identification, or discovery of food, but are for the most part
fixed and involuntary, perhaps in part to ensure that vocal signals are trustworthy. Some primates, especially great apes, do
appear capable of limited vocal learning, an essential requisite for language, but this more often seems to involve modulation
of existing calls (e.g., Slocombe & Zuberbühler, 2007) than the creation of new ones, or the production of sounds through
vibrations of the tongue and lips rather than activation of the larynx (e.g., Hopkins, Taglialatela, & Leavens, 2007). Reviewing
the evidence, Petkov and Jarvis (2012) write
… we would interpret the evidence for vocal plasticity and flexibility in some non-human primates as limited-vocal
learning, albeit with greater flexibility via non-laryngeal than laryngeal control. But they do not have the consider-
able levels of laryngeal (mammalian) or syringeal (avian) control as seen in complex vocal learners (p. 5).
Of course, as primates, we humans are something of an exception in that we are complex vocal learners with flexible
intentional control over output, perhaps due to late-forming connections between the motor cortex and the nucleus ambi-
guus, the nucleus that activates the muscles involved in speech and swallowing. These connections, according to Jürgens
(2002), “seem only to exist in humans (p. 248).”
Although speech is the dominant mode for linguistic communication in humans, there is a strong case for supposing that
the precursors of language, and perhaps the earliest forms, were gestural, at first involving pantomimic movements of the
body, and especially the hands, but eventually becoming conventionalized, losing much their iconic aspect. The arguments
have been elaborated elsewhere (e.g., Arbib, 2005, 2012; Armstrong, Stokoe, & Wilcox, 1995; Corballis, 2002; Hewes, 1973;
Rizzolatti & Sinigaglia, 2006; Tomasello, 2008) and need not be detailed here, but the main points are as follows. In
nonhuman primates, the hands and body are much better adapted to intentional action than the voice, and offer the flexibility
needed to generate varieties of signals. Attempts to teach great apes to talk have failed, but reasonable progress has been
made by teaching them simplified forms of sign language or having them point to abstract symbols on a keyboard, and as we
have seen chimpanzees in the wild use bodily gestures to communicate intentionally and flexibly. People gesture sponta-
neously as they speak, and the sign languages of the deaf are widely recognized as true languages.
Gesture precedes speech in ontogeny as well. Pointing seems to be especially important in early development. In a
metanalysis, Colonnesi, Stams, Koster, and Noomb (2010) note that young children use pointing gestures before they learn to
articulate words, and describe pointing as “the royal road to language (p. 352).” And of course pointing is often used as an
accompaniment to speech when we give directions or indicate a focus of discussion.
Speech itself can be conceived as a gestural system, comprising movements of the lips, the larynx, the velum, and the
blade, body and root of the tongue (Studdert-Kennedy, 2005). The shift from manual gesture to speech can be understood not
so much as a shift in sensory modality as a gradual shift in the locus of gestures from the hands to the face, and then to
movements located in the mouth, with sound added to render these gestures accessible. Although manual, facial, and speech
movements are all still evident in expressive language, speech is now dominant, at least in the hearing population, and the
progression from expansive outward gestures to gestures largely internal to the mouth might be described as an example of
miniaturization (Corballis, 2015), freeing the rest of the body for other activities.

3.3. The mirror neuron system

The gestural theory was strengthened by the discovery of neurons in the prefrontal cortex of macaques that respond both
to actual execution of intentional manual actions, such as grasping, and observation of the same actions carried out by another
individuals. These neurons have become known as mirror neurons, and were initially discovered in area F5, considered ho-
mologous to Broca’s area in humans. This led to the view that mirror neurons provided a platform for the evolution of speech,
while at the same time implying an origin in gesture rather than in vocalization (Rizzolatti & Arbib, 1998). Subsequent
research has revealed a more extensive mirror neuron system (MNS), involving parietal and temporal regions besides the
prefrontal areas (Rizzolatti & Sinigaglia, 2010), and this system overlaps quite extensively with the homologs of the language
circuits in the human brain (Corballis, 2015).
Neurons with mirror properties have also been recorded directly in the human brain (Mukamel, Ekstrom, Kaplan,
Iacoboni, & Fried, 2010), and a meta-analysis of human brain-imaging studies shows that mirror activity occurs in regions
that include those homologous to the mirror network identified in macaques (Molenberghs, Cunnington, & Mattingley, 2012).
Another meta-analysis of areas activated during the observation or imitation of actions points to a network that goes “far
beyond” the MNS as identified in the macaque (Caspers, Zilles, Laird, & Eickhoff, 2010), and some have referred to the system
inferred from human imaging studies as an “expanded MNS” (Fabbri-Destro & Rizzolatti, 2008; Iacoboni, 2009). Brain im-
aging also suggests a confluence of language and gesture within this system. Xu, Gannon, Emmorey, Smith and Braun (2009)
used fMRI to record activation during production of language and perception of pantomimes and emblems, and found
common activation in the left hemisphere in Broca’s area and areas in the posterior temporal lobe along the Sylvian fis-
suredareas long regarded since the mid-19th century as the core of the language system in the human brain. They suggest
that this network can be regarded more generally as a system for the processing of symbols, be they “words, gestures, images,
sounds, or objects” (p. 20,664). Although these results might be taken to support the idea that language evolved from manual
actions, and more specifically from the primate MNS, Xu et al. suggested instead that the circuit was primarily one of
associating action with meaning. Yet Hamzei et al. (2003) showed, in 10 right-handers, that Broca’s area in left inferior frontal
126 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

cortex was activated not only during speech but also during tasks as simple as grasping an object or seeing a picture of an
object being grasped.
Rather than supposing that the primate MNS evolved as a circuit for the representation of meaning, it may be more
realistic to suppose that it was an ancestral system that later gave rise to multiple daughter systems. During the Pleistocene,
the habitat shifted from forest to savanna, perhaps also passing through an early aquatic phase when our forebears may have
foraged at the edges of lakes or sea (Verhaegen, 2013). These changes in habitat posed new challenges to survival, leading to a
tripling of brain size and the emergence of such distinctively human qualities such as the manufacture and use of tools,
enhanced social understanding and differentiation of roles, and of course language. Brain circuits grew more specialized,
resulting in what Barrett (2012) calls “hierarchical specialization,” with specialized networks arising from common ancestral
ones. Oakley and Rivera (2008) write similarly of how new circuits are formed, whether through fissioning of ancestral
systems, or copying and differential modification, or sometimes through modified circuits fusing to create new functions.
Such processes are of course more compatible with evolution through natural selection than through the creation of new
networks.

3.4. Lateralization

Given the space limitations imposed the size of the skull, itself limited by the size of the birth canal, the establishment of
new circuits was probably accompanied by increasing lateralization, reducing duplication of function. Of course lateralization
of structure and function is not unique to humans, nor to the brain (Rogers, Vallortigara, & Andrew, 2013), but the demands of
human survival probably led to enhanced lateralization, and an increase in the number of lateralized circuits. For instance,
Behrmann and Plaut (2015) document evidence that the fusiform gyrus in the primate brain is specialized for face recog-
nition, but in the course of human evolution fissioned into a right-hemispheric system for face recognition and a left-
hemispheric one for the recognition of printed words. This hemispheric separation is graded rather than absolute, and
lateralization is weighed against the advantages of bilateral symmetry (see also Corballis, 2006).
One approach to understanding the nature and number of lateralized circuits is to use fMRI to derive indices of lateral-
ization in different brain regions, and then apply factor analysis to discover the underlying dimensions. One useful index is the
laterality index (LI) developed by Wilke and Lidzba (2007), which ranges between þ1 for extreme left-hemispheric domi-
nance to 1 for extreme right-hemispheric dominance. Liu, Stufflebeam, Sepulcre, Hedden, and Buckner (2009) measured
laterality indices in lateralized areas of the brain while subjects simply looked at crosshairs on an otherwise blank screen.
Factor analysis of the intercorrelations among the indices produced four factors, representing independent lateralized net-
works. Two were biased to the left hemisphere, one corresponding to the language network and the other to the DMN itself,
and two were biased to the right, one related to visual processing and the other to an attentional network. This implies that
lateralization is not a unitary dimension of brain activity.
We have applied this approach to the analysis of brain asymmetries to assess the possible relations between language and
gesture. We used two language tasks, one requiring covert production of words beginning with designated letters, and one
requiring judgments of whether printed word pairs were synonyms or not. We also used two tasks involving observation of
gestures, one involving meaningful actions (pantomimes of common actions) and the other involving sign language, which
was largely meaningless to the subjects. Half of the subjects were left-handed. Here, factor analysis of LIs recorded in different
areas is response to the tasks yielded three uncorrelated factors, implying three independent networks (see Haberling,
Corballis and Corballis, in press; for details). These networks are shown schematically in Fig. 1. All are left-lateralized.
This analysis provides some support for the idea of separate networks derived from the primate MNS. The first factor is a
language factor, involving Broca’s area and temporal regions, activated both by word generation and synonym judgments. The
second involves activity primarily in the parietal lobe in response to the observation of gesture, and is the only factor
correlated significantly with handedness. The third factor also represents activity induced by gesture observation, and ac-
tivates Broca’s area and a region of the middle temporal gyrus, and is uncorrelated with handedness. This third factor clearly
overlaps the language factor, but is functionally distinct from it.
This analysis must be considered somewhat preliminary, but does offer general support for the idea the primate MNI
fissioned into several networks, including the three left-hemisphere-biased networks described here. The networks show
different degrees of lateralization, with the language factor the most left-lateralized and the third, handedness-independent
gestural network the least, perhaps corresponding to the order in which the networks separated off. A possible scenario is
that the third factor is the residual of the original DMN, with the first factor separating off to mediate language and the second
to mediate tool use and handedness itself. Lateralization is seldom if ever absolute, and seems to represent directional de-
partures from bilateral symmetry. Mean laterality indices do not dip below zero even among left-handers, and are generally
positive, implying an overall but variable left-hemispheric bias. Of course activities other than gesture or language, such as
face recognition or spatial judgments, induce right-hemisphere biases (Badzakova-Trajkov, H€ aberling, Roberts, & Corballis,
2010; Behrmann & Plaut, 2015).
Asymmetry itself is not unique to humans. A general left-hemispheric bias for action dynamics exists in other species,
including marine animals and some primates (MacNeilage, 2013), and probably preceded the fissioning into separate
specialized networks for gesture, language, and tool use in hominin evolution. Conversely, a right-hemisphere dominance for
emotion seems to be present in all primates so far investigated, suggesting an evolutionary continuity going back at least
30e40 million years (Lindell, 2013).
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 127

Fig. 1. Lateralized neural networks based on factor analysis of laterality indices (LIs) derived from fMRI activation of selected regions in response to language
tasks (word generation and synonym judgments) and observation of gestures (pantomime and sign language). Red areas represent a language factor, activated by
both language tasks. Blue areas represent a gesture factor uncorrelated with handedness, while green areas represent a gesture factor correlated with hand-
edness. (BA ¼ Brodmann’s area; MTG ¼ middle temporal gyrus, A ¼ anterior, P ¼ posterior, TO ¼ temporal/occipital; PFt ¼ rostral inferior parietal lobule; 7a/
hiP3 ¼ part of the intraparietal area). Figure simplified from Fig. 4 of Ha€berling, Corballis and Corballis (in press).

Lateralization is often regarded as a critical feature of language, but recent research suggests that it has more to do with the
manner in which brains evolve more varied and complex functions. In this respect, language may be no more “special” than
other complex skills, such as the manufacture and use of tools, or the capacity for theory of mind, that have also led to the
emergence of specialized circuits.

4. Grammar

Grammar is generally regarded as the critical feature that sets language apart. In modern times, at least, this dates from
Chomsky’s (1957) book Syntactic Structures, which established an approach to language in direct contrast to Skinner’s (1957)
Verbal Behavior, published in the same year, in which language was treated simply as a form of complex behavior, explainable
through the principles of operant conditioning. Two years later, Chomsky (1959) wrote a devastating review of Verbal
Behavior. To Chomsky, language could not be reduced to behavioral principles, and grammar could only be understood in
terms of universal computational rulesdsubsequently called universal grammar. Although Chomsky’s arguments won the
day, linguists appear never to have agreed as to precisely what those rules are, and McCawley (1982) was once moved to
publish a book entitled Thirty Million Theories of Grammar. He was joking, of course, but the point was well made. In more
recent times, and based on the remarkable differences in structure between languages, Evans and Levinson (2009) concluded
that “the emperor of Universal Grammar has no clothes (p. 438),” and in the same year Tomasello (2009) remarked that
“Universal grammar is dead (p. 740).”
Chomsky may well have been right, though, in suggesting that the fundamentals of grammar are to be found in the in-
ternal language rather than external languagedthat is, in the processes of thought rather than in language as actually spoken
or signed. In this article, though, the point of departure from Chomsky lies in the nature of internal thought. To Chomsky, I-
language is innately endowed through some event that took place within the past 100,000 years which effectively
restructured the human mind. I-language is based on the manipulation of arbitrary symbols, combined into grammatical
structures through the recursive principle of unbounded Merge. Chomsky argues that language cannot have evolved through
natural selection because the internal symbols of I-language have no reference to the external world, and so could not have
been “selected.” In one recent chapter, he writes: “Crucially, even the simplest words and concepts of human language lack
the relation to mind-independent entities that appears to be characteristic of animal communication (Chomsky, 2010, p. 87).”
These symbols, he supposes, are arbitrary, with nothing in their shape or sound to link them to anything in the natural world.
They therefore cannot have been shaped by natural selection.
The argument seems spurious, since there is nothing about natural selection that demands a link between internal
structures and the external world. All that is required is that a given property, be it internal or external, should influence
biological fitness, or the chances of reproduction and survival.
In this article I argue, contrary to Chomsky, that the properties of thought that give rise to the basic structure of language
are not unique to humans, and evolved through natural selection. The capacity to mentally relive past episodes or imagine
128 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

future ones, or even purely fabricated ones, is fundamentally generative, and perhaps goes far back in evolution. It is probably
based on an ancient adaptation to the capacity to move through spatial environments, leading to the ability to remember and
plan spatial and temporal excursions. It is this that is primarily responsible for the generativity of language, as it maps onto
episodes that occur in time and space. At a simpler level, walking or flying may be considered generative, allowing animals
and birds to locate themselves in different environments, record past excursions, and plan new ones, and perhaps even
generate untraveled trajectories so that knowledge of environments can be extended beyond experience itself. Nonspatial
elements can be attached, so that individual trajectories can be understood as episodes, feeding episodic memory and
episodic forethoughtdand storytelling.
Our episodic thoughts can also be considered to involve merge operations, as individual components are combined. In one
experiment, for example, subjects are asked to recall 100 episodes they had experienced in the past 10 years, and identify a
person, and object and a location from each. These were then presented in new combinations, and the subjects were asked to
imagine future episodes involving each new combination. Brain imaging showed the hippocampus to be activated when they
imagined these episodes, and also when they later recalled the imagined episodes (Martin et al., 2011). This experiment
revealed the involvement of the hippocampus, but also illustrates people’s capacity to remember and construct episodes by
combining elements. Of course the elements in this study were named, but it is likely that the episodes themselves were
experienced as images rather than as linguistic entities.
The capacity to create the combinatorial structures underlying memory and imagination is no doubt common to all people,
and the universality of the language faculty may have as much to do with the universal structure of experience as with the
nature of language itself. We all move and walk in much the same way, creating experiences that are at least broadly similar, at
least in extent and scale. Again, language mimics experience. Of course not all of our experiences and thoughts are based on
spatiotemporal life; we also have thoughts that are abstract or emotional rather than derived from experience of the physical
world. My conjecture, though, is that language emerged from the benefits to be derived from sharing information about
spatial experiences and plans, perhaps in the context of early hunter-gathering.

4.1. Pantomime

Episodic communication may have originated in pantomime. Although there are some suggestions of pantomime in great
apes (e.g., Boesch, 1993; Russon & Andrews, 2001; Tanner, Patterson, & Byrne, 2006), the capacity to mime sequences may
have emerged in the Pleistocene, as our forebears relayed information about actual or planned foraging expeditions (McBride,
2014). Donald (1991) suggested that what he called “mimetic culture” evolved with the emergence of our ancestor Homo
ergaster from around 1.9 million years ago. In ergaster and the later members of the genus Homo, moreover, bipedalism shifted
from facultative to obligate, and assumed a more free-striding gait. This would have freed the hands for carrying the spoils of
foraging, and for more elaborate gesturing. Brain size increased dramatically with the emergence and subsequent devel-
opment of the genus Homo, which might be taken as evidence of selection for more complex communication.
Pantomime may have set the stage for grammar, the construction of sequences involving combinations of elements, and
perhaps even some recursion as sub-sequences were inserted into the main flow of the story. The elements of pantomime are
pictorial, or iconic, rather than abstract, and in that sense, at least, pantomime does not resemble language. Over the course of
time, though, mimed elements would have become conventionalized (Burling, 1999) in the interests of greater accuracy,
efficiency and speed. One example is Nicaraguan Sign Language (NSL), which first emerged in the 1970s when a school was
established for deaf children, and the children themselves invented the language, which gradually assumed more gram-
matical structure over succeeding cohorts. A similar transformation occurred in the newly-emerged Al-Sayyid Bedouin Sign
Language in the Negev Desert in Israel (Aronoff, Meir, Padden, & Sandler, 2008). The autonomy and rapid rise of these sign
languages suggests that language itself might well have developed and persisted in a predominantly manuofacial form until
quite late in hominin evolution, and persists as an accessible option to this day. Tomasello (2008, p. 55) writes that “it is
possible that the human capacity evolved quite a long way in the service of gestural communication alone, and the vocal
capacity is actually a very recent overlay.”
Conventionalization typically results in the loss of iconicity. For example, in American Sign language the sign for home was
once a combination of the sign for eat, which is a bunched hand touching the mouth, and the sign for sleep, which is a flat hand
on the cheek. Now it consists of two quick touches on the cheek, both with a bunched handshape, so the original iconic
components are effectively lost. With typical perspicacity, Charles Darwin (1872, p. 62) himself remarked “on the practice of
the deaf and dumb and of savages to contract their signs as much as possible for the sake of rapidity. Hence their natural
source or origin often becomes doubtful or is completely lost; as is likewise the case with articulate speech.” Yet sign lan-
guages retain a degree of iconicity. It has been estimated, for example, that in Italian Sign Language some 50 percent of the
hand signs and 67 percent of the bodily locations of signs stem from iconic representations (Pietrandrea, 2002). In American
Sign Language, too, some signs are arbitrary, but many more are iconic; for example, the sign for “erase” resembles the action
of erasing a blackboard, and the sign for “play piano” mimics the action of actually playing a piano (Emmorey, 2002).
Speech itself might be regarded as an endpoint in the process of conventionalization, in which most iconicity is lost. But
not all; even spoken words are not entirely arbitrary. Words sometimes do reflect the shapes of things they name. The German
psychologist Wolfgang Ko € hler (1929) showed people drawings of two meaningless objects, one smooth and rounded and the
other sharply inflected and spiky, and asked people to choose which one was named “baluma” and which “takete.” Ninety-
five percent of respondents chose “baluma” for the rounded one and “takete” for the spiky one. This has been repeated many
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 129

times with slightly different names. Ramachandran and Hubbard (2001) attribute it to synesthesiada natural (though widely
varying) tendency to associate stimuli of different modalities, in this case sound and vision. They also suggest more direct
mappings. For instance, the words for referring in the second person (“you” in English, “tu” or “vous” in French, “thoo” in
Tamil) involve pushing the lips forward to the listener, while words referring to the self (“me” in English, “moi” in French,
“naan” in Tamil) seem to point inwards with tongue and lips to the speaker herself.
Even so, the majority of spoken words give no physical clue as to their meaning, and the fact that there are some 6000
languages in the world (Evans, 2009) is testimony to this. The important point, though, is that arbitrariness is not a necessary
or defining property of language, as traditionally argued (Saussure, 1916). Rather, it is the outcome of the progressive
refinement of communication so that elements referring to real-world entities are standardized and calibrated to increase
efficiency. Conventionalized forms are maintained in culture, stored through association, and passed on through generations,
albeit with alterations that eventually lead to the multiplicity of different languages. Changes in the forms of words also result
in language becoming something of a secret code, available only to speakers of the particular languages they know. In World
War II, bilingual American Marines in the Pacific theater used the Navajo language to transmit coded messages, in the sure
expectation that the enemy would not be able to understand them. This is known as code talking, and was pioneered in the
World War I by Cherokee and Choctaw Indians.
The origins of grammatical language can therefore be largely understood in terms of the communication of episodic in-
formation, at first through pantomime and later through the conventionalization, creating standardized versions of elements
to apply to the ingredients of episodes. Of course, modern language goes well beyond the transmission of mental travels in
time and space, but even when talking of nonspatial events or ideas we use metaphors based on the spatiotemporal world.
Lakoff (2014) suggests that the metaphors we use to describe thought itself fall into four categories, each involving reference
to the physical world. First is the idea that thinking is moving, as in reach a conclusion, or go off on a tangent. Second is the idea
of understanding as seeing, as in see what I mean, or shed light on. Third is the metaphor of thinking as object manipulation, as
in toss an idea around, or gave him the idea. Fourth is the notion of thinking as eating, as in food for thought, or raw facts, or half-
baked ideas. The ubiquity of such metaphors may be taken as evidence that language originated in the description of bodily
events in a spatiotemporal world, where we move about and manipulate things. Even processes as abstract as deductive logic
may depend on “mental models” that often involve visual images (Johnson-Laird & Byrne, 1991), which in turn may derive
from the ability to imagine events.
There is of course much more to grammar. We need to specify time, conditionality, degrees of certainty, number, quantity,
and the like. In large measure the conventions by which we do this are idiosyncratic, and specific to individual cultures. In
many languages, for example, such specifications are accomplished through inflections, modifying nouns or verbs to carry
information about number, time, and many other elements of meaning. English makes relatively little use of inflections, but
uses auxiliaries to indicate tense, aspect, mood and voice. One highly inflected language is Turkish, in which inflections of the
verb may result in millions of possible forms; one amusing example is the single verb “Çekoslovakyalilastiramadiklar-
imizdanmisinizor?” which can be translated into English as “Are you one of those we couldn’t make into a Czechoslovakian?”
At this level of analysis, grammar probably depends more on the vagaries of culture and human ingenuity, as evident in the
vast complexities of the manufactured world as in language. It was the sheer variety of languages that prompted Evans and
Levinson (2009) to conclude that the concept of universal grammar has no substance. As suggested above, though, there are
universal aspects of language, but these probably derives from the common features of thought, rather than from the various
forms that language takes as a communication device.

5. Conclusions

In some respects, the scenario sketched in this article is in accord with that proposed by Chomsky, in that the essential
features of language lie in the structure of human thought. The main difference has to do with the nature of thought itself.
Chomsky proposed that thought itself was restructured at a singular point in recent human evolution, and so endowed with
the unique attributes of symbolic representation and combinatorial rules. Here, I have proposed that the critical properties of
human thought arose through natural selection, and in some respects go far back in evolution.
In this view, the critical feature of thought that underlies language is the capacity to engage mentally with the nonpresent.
This is the feature that gives language the property of displacement, one of the design features of language proposed by
Hockett (1960) and arguably the most critical. Engaging with the nonpresent includes reliving past episodes, imagining future
ones, taking the emotional and mental perspectives of others, and combining all three of these in the construction of stories.
The capacity to travel mentally in space and time probably has a long evolutionary history. The study of short-wave ripples in
the hippocampus suggests that even rats have some capacity to escape the present and imagine trajectories in spatial en-
vironments, whether past, future, or purely imaginary. Mental travels in space and time are probably critical to survival in
birds, as well, especially to those that cache items of food in multiple locations and must later find them.
The ability to take the mental perspective of others probably evolved much more recently, and appears to be critical to
language as a communicative system. Apes may have some capacity for theory of mind, but it seems to be much more
developed in humans. Indeed the expansion of interpersonal understanding was perhaps that aspect of cognitive evolution
that most clearly sets humans apart from other primates, creating the so-called “cognitive niche” and making language itself
possible. Even so, it was not a sudden development; it appears to be present to a limited degree in great apes, and probably
gained its added complexity during the Pleistocene from nearly three million years ago. It is evident not only in language, but
130 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

in thought itself. Its recursive nature is well illustrated by David Premack’s (2007) mischievous comment on sexual politics:
“Women think that men think that they think that men think that women’s orgasm is different” (p. 13, 864).
The ability to communicate our thoughts probably emerged much later and more slowly than the ability to actually
entertain complex thoughts. The humble rat may well imagine trajectories in a maze, but appears to lack any mechanism to
relay these imaginingsdexcept through hippocampal activity impervious to all but the invasive neuroscientist. Communi-
cation of complex thoughts requires an output system that is intentional and flexible enough to convey complex messages
that are often generated anew. Primates have something of a head start in that their limbs are adapted to intricate movement,
whether in negotiating the branches of trees, plucking leaves or fruit, catching insects, or rough-and-tumble play. My
conjecture is that this manual and bodily capacity provided the source for the later emergence of language as a communi-
cative tool.
There can be no question that language had a huge impact on the human mind, distinguishing us from other living species.
That influence, though, was not so much a restructuring of thought, as proposed by Chomsky and others, but arose from the
sharing of knowledge and experience. Much of our knowledge is derived from other people, whether through formal edu-
cation or the instruction of parents and specialists. Indeed specialization itself is a consequence of language. The Hungarian
biologist Eo€rs Szathm ary (2015) writes:
Language allows for something unprecedented: negotiated division of labor. Just as the evolution of powerful genetic
and epigenetic inheritance systems allowed the evolution of complex multicellularity, natural language allowed the
emergence of complex human societies. Social life itself would have imposed demands on language, as individual
assumed different roles and social structure grew more complex. People would have assumed different roles in their
communities, which would itself have required more explicit communication (p. 10,109).
But it’s not just knowledge. We also share our experiences. We like to tell each other what we have done, what we plan,
what we like and don’t like. We also like to tell stories, or immerse ourselves in fiction or TV soaps. Of course other species also
share simply by living together in communities and witnessing the same events, but language allows us to vastly expand our
sharing in space and time, and mentally live the lives of others, whether real or imaginary. It is the sharing of experience that
comes most easily and naturally, and probably goes back further in our evolutionary history. The sharing of knowledge, in
contrast, seems more effortful, as witnessed by Shakespeare’s lines from As You Like It:
Then, the whining school-boy with his satchel
And shining morning face, creeping like snail
Unwillingly to school (Act II, Scene VII).
By the same token, this chapter, alas, is probably much more unpalatable than the novel you are currently engaged in.
What of the future? In trying to put together the various system that underlie our capacity to share our thoughts, I have
drawn on a number of neural systems, including the mirror system, the default mode network, and the hippocampal-
entorhinal system. We do not know how these systems interact with the language system itself to create meaningful lan-
guage. This is perhaps the next challenge for neurolinguistics.

References

Addis, D. R., Wong, A. T., & Schacter, D. L. (2007). Remembering the past and imagining the future: Common and distinct neural substrates during event
construction and elaboration. Neuropsychologia, 45, 1363e1377.
Arbib, M. A. (2005). From monkey-like action recognition to human language: An evolutionary framework for neurolinguistics. Behavioral and Brain Sci-
ences, 28, 105e168.
Arbib, M. A. (2012). How the brain got language: The mirror system hypothesis. Oxford: Oxford University Press.
Armstrong, D. F., Stokoe, W. C., & Wilcox, S. E. (1995). Gesture and the nature of language. Cambridge, UK: Cambridge University Press.
Aronoff, M., Meir, I., Padden, C. A., & Sandler, W. (2008). The roots of linguistic organization in a new language. Interaction Studies, 9, 133e153.
Badzakova-Trajkov, G., Ha €berling, I. S., Roberts, R. P., & Corballis, M. C. (2010). Cerebral asymmetries: Complementary and independent processes. PLoS One,
5(3), e9682.
Barrett, H. C. (2012). A hierarchical model of the evolution of human brain specializations. Proceedings of the National Academy of Sciences (USA.), 109(Suppl.
1), 10733e10740.
Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology. Cambridge: Cambridge University Press.
Behrmann, M., & Plaut, D. C. (2015). A vision of graded hemispheric specialization. Annals of the New York Academy of Sciences, 1359, 30e46.
Bickerton, D. (2014). More than nature needs: Language, mind, and evolution. Cambridge, MA: Harvard University Press.
Bisiach, E., & Luzzatti, C. (1978). Unilateral neglect of representational space. Cortex, 14, 129e133.
Boesch, C. (1993). Aspects of transmission of tool-use in wild chimpanzees. In K. R. Gibson, & T. Ingold (Eds.), Tools, language, and cognition in human
evolution (pp. 171e183). Cambridge, UK: Cambridge University Press.
Boyd, B. (2009). The origin of stories: Evolution, cognition, and fiction. Cambridge, MA: Belnap Press of Harvard University Press.
Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain’s default network e Anatomy, function, and relevance to disease. Annals of the New
York Academy of Sciences, 1124, 1e38.
Burling, R. (1999). Motivation, conventionalization, and arbitrariness in the origin of language. In B. J. King (Ed.), The origins of language: What human
primates can tell us. Santa Fe, NM: School of American Research Press.
Call, J., & Tomasello, M. (2008). Does the chimpanzee have a theory of mind? 30 years later. Trends in Cognitive Science, 12, 187e192.
Caspers, S., Zilles, K., Laird, A. R., & Eickhoff, S. B. (2010). A meta-analysis of action observation and imitation in the human brain. NeuroImage, 50, 1148e1167.
Chomsky, N. (1957). Syntactic structures. The Hague.
Chomsky, N. (1959). A review of Skinner’s “Verbal Behavior”. Language, 35, 26e58.
Chomsky, N. (2007). Biolinguistic explorations: Design, development, evolution. International Journal of Philosophical Studies, 15, 1e21.
M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132 131

Chomsky, N. (2010). Some simple evo devo theses: How true might they be for language? In R. K. Larson, V. De prez, & H. Yamakido (Eds.), The evolution of
human language (pp. 45e62). Cambridge: Cambridge University Press.
Christiansen, M. H., & Kirby, S. (2003). Language evolution: The hardest problem in science? In M. H. Christiansen, & S. Kirby (Eds.), Language evolution (pp.
1e15). Oxford: Oxford University Press.
Clayton, N. S., Bussey, T. J., & Dickinson, A. (2003). Can animals recall the past and plan for the future? Trends in Cognitive Sciences, 4, 685e691.
Collin, S. H. P., Milivojevic, B., & Doeller, C. F. (2015). Memory hierarchies map onto the hippocampal long axis in humans. Nature Neuroscience. http://dx.doi.
org/10.1038/nn.4138. advance online publication.
Colonnesi, C., Stams, G. J. J. M., Koster, I., & Noomb, M. J. (2010). The relation between pointing and language development: A meta-analysis. Developmental
Review, 30, 352e366.
Corballis, M. C. (2002). From hand to mouth: The origins of language. Princeton, NJ: Princeton University Press.
Corballis, M. C. (2006). Cerebral asymmetry: A question of balance. Cortex, 42, 117e118.
Corballis, M. C. (2015). What’s left in language: Beyond the classical model. Annals of the New York Academy of Sciences, 1359, 14e29.
Corkin, S. (2013). Permanent present tense: The man with no memory, and what he taught the world. London: Allen Lane.
Darwin, C. (1859). On the origin of species by means of natural selection. London: John Murray.
Darwin, C. (1872). The expression of the emotions in man and animals. London: John Murray.
Deacon, T. (1997). The symbolic species. New York: Norton.
Descartes, R. (1646/1970). Letter to Marquess of Newcastle. In A. Kenny (Ed.), Descartes’ philosophical letters (p. 206). Oxford: Clarendon Press (A. Kenny,
Trans.).
Dor, D. (2015). The structure of imagination: Language as a social communication technology. New York: Oxford University Press.
Emmorey, K. (2002). Language, cognition, and brain: Insights from sign language research. Hillsdale, NJ: Erlbaum.
Evans, N. (2009). Dying words: Endangered languages and what they have to tell us. Oxford: Wiley-Blackwell.
Evans, N., & Levinson, S. C. (2009). The myth of language universals: Language diversity and its importance for cognitive science. The Behavioral and Brain
Sciences, 32, 429e492.
Everett, D. L. (2012). Language: The cultural tool. New York: Pantheon.
Fabbri-Destro, M., & Rizzolatti, G. (2008). Mirror neurons and mirror systems in monkeys and humans. Physiology, 23, 171e179.
Foster, D. J., & Wilson, M. A. (2006). Reverse replay of behavioural sequences in hippocampal place cells during the awake state. Nature, 440, 680e683.
Grice, H. P. (1989). Studies in the ways of words. Cambridge, MA: Cambridge University Press.
Gupta, A. S., van der Meer, M. A. A., Touretzky, D. S., & Redish, A. D. (2010). Hippocampal replay is not a simple function of experience. Neuron, 65, 695e705.
Ha€berling, I. S., Corballis, P. M., & Corballis, M. C. (2016). Language, gesture, and handedness: Evidence for independent neural networks. Cortex (in press).
Hamzei, F., Rijntjes, M., Dettmers, C., Glauche, V., Weiller, C., & Buchel, C. (2003). The human action recognition system and its relationship to Broca’s area:
An fMRI study. NeuroImage, 19, 637e644.
Hewes, G. W. (1973). Primate communication and the gestural origins of language. Current Anthropology, 14, 5e24.
Hobaiter, C., & Byrne, R. W. (2011). Serial gesturing by wild chimpanzees: Its nature and function for communication. Animal Cognition, 14, 827e838.
Hockett, C. F. (1960). The origins of speech. Scientific American, 203(3), 88e96.
Hoffecker, J. F. (2007). Representation and recursion in the archaeological record. Journal of Archaeological Method and Theory, 14, 359e387.
Hopkins, W. D., Taglialatela, J. P., & Leavens, D. A. (2007). Chimpanzees differentially produce novel vocalizations to capture the attention of a human. Animal
Behaviour, 73, 281e286.
Humboldt, W. V. (1836/1999). On language. On the diversity of human language construction and its influence on the mental development of the human species.
In M. Losonsky (Ed.). Cambridge: Cambridge University Press. Translated from the 1836 German manuscript by P. Heath.
Iacoboni, M. (2009). Imitation, empathy, and mirror neurons. Annual Review of Psychology, 60, 653e670.
Johnson-Laird, P. N., & Byrne, R. M. J. (1991). Deduction. Hillsdale, NJ: Erlbaum.
Jürgens, U. (2002). Neural pathways underlying vocal control. Neuroscience and Biobehavioral Reviews, 26, 235e258.
Killingsworth, M. A., & Gilbert, D. T. (2010). A wandering mind is an unhappy mind. Science, 330, 932.
Klein, R. G. (2008). Out of Africa and the evolution of human behavior. Evolutionary Anthropology, 17, 267e281.
Ko€ hler, W. (1929). Gestalt psychology. New York: Liveright.
Lakoff, G. (2014). Mapping the brain’s metaphor circuitry: Metaphorical thought in everyday reason. Frontiers in Human Neuroscience, 9. article 958.
Lindell, A. K. (2013). Continuities in emotion lateralization in human and nonhuman primates. Frontiers in Human Neuroscience, 7. article 464.
Liu, H. S., Stufflebeam, S. M., Sepulcre, J., Hedden, T., & Buckner, R. L. (2009). Evidence from intrinsic activity that asymmetry of the human brain is controlled
by multiple factors. Proceedings of the National Academy of Sciences (USA), 106, 20499e20503.
Loftus, E., & Ketcham, K. (1994). The myth of repressed memory. New York: St. Martin’s Press.
MacNeilage, P. F. (2013). Vertebrate whole-bodyaction asymmetries and the evolution of right handedness: A comparison between humans and marine
mammals. Developmental Psychobiology, 56, 577e587.
Martin, V. C., Schacter, D. L., Corballis, M. C., & Addis, D. R. (2011). A role for the hippocampus in encoding simulations of future events. Proceedings of the
National Academy of Sciences (USA), 108, 13858e13863.
McBride, G. (2014). Story telling, behavior planning, and language evolution in context. Frontiers in Psychology, 5. article 1131.
McCawley, J. D. (1982). Thirty million theories of grammar. Chicago, IL: University of Chicago Press.
Miller, J. F., Neufang, M., Solway, A., Brandt, A., Trippel, M., … Schulze-Bonhage, A. (2013). Neural activity in human hippocampal formation reveals the
spatial context of retrieved memories. Science, 342, 1111e1114.
Molenberghs, P., Cunnington, R., & Mattingley, J. B. (2012). Brain regions with mirror properties: A meta-analysis of 125 human fMRI studies. Neuroscience
and Biobehavioral Reviews, 36, 341e349.
Moser, M. B., Rowland, D. C., & Moser, E. I. (2015). Place cells, grid cells, and memory. Cold Spring Harbor Perspectives in Biology, 7, a021808.
Mukamel, R., Ekstrom, A. D., Kaplan, J., Iacoboni, M., & Fried, I. (2010). Single-neuron responses in humans during execution and observation of actions.
Current Biology, 20, 750e756.
Müller, F. M. (1873). Lectures on Mr Darwin’s philosophy of language: Second lecture. Frazer’s Magazine, 7, 659e678.
Neisser, U. (2008). Memory with a grain of salt. In H. H. Wood, & A. S. Byatt (Eds.), Memory: An anthology (pp. 80e88). London: Chatto and Windus.
Niles, J. D. (2010). Homo narrans: The poetics and anthropology of oral literature. Philadelphia: University of Pennsylvania Press.
O’Keefe, J., & Nadel, L. (1978). The hippocampus as a cognitive map. Oxford: Clarendon Press.
Oakley, T. H., & Rivera, A. S. (2008). Genomics and the evolutionary origins of nervous system complexity. Current Opinion in Genetics and Development, 18,
479e492.
Petkov, C. I., & Jarvis, E. D. (2012). Birds, primates, and spoken language origins: Behavioral phenotypes and neurobiological substrates. Frontiers in
Evolutionary Neuroscience, 4. article 12.
Pfeiffer, B. E., & Foster, D. J. (2013). Hippocampal place-cell sequences depict future paths to remembered goals. Nature, 497, 74e79.
Pietrandrea, P. (2002). Iconicity and arbitrariness in Italian sign language. Sign Language Studies, 2, 296e321.
Pinker, S., & Bloom, P. (1990). Natural language and natural selection. Behavioral and Brain Sciences, 13, 704e784.
Premack, D. (1985). Gavagai! or the future history of the animal language controversy. Cognition, 19, 207e296.
Premack, D. (2007). Human and animal cognition: Continuity and discontinuity. Proceedings of the National Academy of Sciences (USA), 104, 13861e13867.
Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences, 4, 515e526.
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A., & Shulman, G. L. (2001). A default mode of brain function. Proceedings of the
National Academy of Sciences (USA), 98(2), 676e682.
132 M.C. Corballis / Journal of Neurolinguistics 43 (2017) 120e132

Ramachandran, V. S., & Hubbard, E. M. (2001). SynaesthesiadA window into perception, thought and language. Journal of Consciousness Studies, 8, 3e34.
Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trends in Neurosciences, 21, 188e194.
Rizzolatti, G., & Sinigaglia, C. (2006). Mirrors in the brain: How our minds share actions and emotions. Oxford: Oxford University Press.
Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the parieto-frontal mirror circuit: Interpretations and misinterpretations. Nature Neuroscience, 11,
264e274.
Roediger, H. L., & McDermott, K. B. (1995). Creating false memories e Remembering words not presented in lists. Journal of Experimental Psychology:
Learning Memory and Cognition, 21, 803e814.
Rogers, L. J., Vallortigara, G., & Andrew, R. J. (2013). Divided brains: The biology and behaviour of brain asymmetries. Cambridge, UK: Cambridge University
Press.
Russon, A., & Andrews, K. (2001). Orangutan pantomime: Elaborating the message. Biology Letters, 7, 627e630.
Saussure, F. de (1916). Cours de linguistique ge nerale, C. Bally and A. Sechehaye (Eds.), with the collaboration of A. Riedlinger. Lausanne and Paris: Payot.
Translated as W. Baskin (1977), Course in general linguistics. Glasgow: Fontana/Collins.
Scott-Phillips, T. (2015). Speaking our minds: Why human communication is different, and how language evolved to make it special. Basingstoke, UK: Palgrave
Macmillan.
Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts.
Slocombe, K. E., & Zuberbühler, K. (2007). Chimpanzees modify recruitment screams as a function of audience composition. Proceedings of the National
Academy of Sciences (USA), 104, 17228e17233.
Sperber, D., & Wilson, D. (2002). Pragmatics, modularity and mind-reading. Mind and Language, 17, 3e23.
Strange, B. A., Witter, M. P., Lein, E. S., & Moser, E. I. (2014). Functional organization of the hippocampal longitudinal axis. Nature Reviews Neuroscience, 15,
655e669.
Studdert-Kennedy, M. (2005). How did language go discrete? In M. Tallerman (Ed.), Language origins: Perspectives on evolution (pp. 48e67). Oxford, UK:
Oxford University Press.
Suddendorf, T., & Corballis, M. C. (1997). Mental time travel and the evolution of the human mind. Genetic, Social, and General Psychology Monographs, 123,
133e167.
Suddendorf, T., & Corballis, M. C. (2007). The evolution of foresight: What is mental time travel, and is it unique to humans? Behavioral and Brain Sciences,
30, 299e351.
Szathm ary, E. (2015). Toward major evolutionary transitions 2.0. Proceedings of the National Academy of Sciences (USA), 112, 10104e10111.
Tanner, J. E., Patterson, F. G., & Byrne, R. W. (2006). The development of spontaneous gestures in zoo-living gorillas and sign-taught gorillas: From action and
location to object representation. Journal of Developmental Processes, 1, 69e103.
Tattersall, I. (2012). Masters of the planet: The search for human origins. New York: Palgrave Macmillan.
Taylor, M. (1999). Imaginary companions and the children who create them. New York: Oxford University Press.
Tomasello, M. (2008). The origins of human communication. Cambridge, MA: MIT Press.
Tomasello, M. (2009). Universal grammar is dead. Behavioral and Brain Sciences, 32, 470.
Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26, 1e12.
Verhaegen, M. (2013). The aquatic ape evolves: Common misconceptions and unproven assumptions about the so-called Aquatic Ape Hypothesis. Human
Evolution, 28, 237e266.
Wearing, D. (2005). Forever today. New York: Doubleday.
Wilke, M., & Lidzba, K. (2007). LI-tool: A new toolbox to assess lateralization in functional MR-data. Journal of Neuroscience Methods, 163(1), 128e136.
Xu, J., Gannon, P. J., Emmorey, K., Smith, J. F., & Braun, A. R. (2009). Symbolic gestures and spoken language are processed by a common neural system.
Proceedings of the National Academy of Sciences (USA), 106, 20664e20669.