C H A P T E R 14

The evolution of human

communication and language
James R. Hurford

14.1 Introduction territorial songs, and by vervet monkey alarm

calls. We are not yet sure exactly what is conveyed
Human language stands out in a number of ways by whale songs, but a reasonable default hypoth-
from the topics of almost all of the other chapters esis would seem to be that they convey messages of
in this book. Although every communication sys- the same expressive power as complex birdsongs.
tem can claim in some way to be unique, human We may be wrong about this, but the current belief
language is spectacularly unique in its complexity is thus that any human language is capable of com-
and expressive power. municating the sum total of all that any animal spe-
Complexity is hard to measure, but a clue is cies can communicate, and more. More, because we
given by the fact that The Cambridge Grammar of alone, as far as we know, can tell each other about
the English Language (Huddleston & Pullum, 2002), fictional or abstract objects, and about events far
which is just a description of Modern Standard distant in time and space.
English, weighs in at over 1700 pages. The head- In the bulk of this chapter, I will list and discuss
ings of the first half-dozen descriptive chapters, some of the most important differences and simi-
out of eighteen, are: The verb, The clause: comple- larities between human languages and nonhuman
ments, Nouns and noun phrases, Adjectives and communication systems, with an evolutionary
adverbs, Prepositions and preposition phrases, perspective, in particular drawing on results from
and The clause: adjuncts. No non-human com- comparative psychology pertaining to our closest
munication system demands anything like this relatives, the non-human primates (see related dis-
degree of detail to describe it. And English is just cussion on language in Chapter 13).
one of over 6,000 human languages, all of compa-
rable complexity.
As for expressive power, this is also hard to meas- 14.2 Diversity
ure. We cant see far into the minds of nonhuman
animals to know what exactly they can commu- We must first make the vital distinction between
nicate with each other, but it seems a fair bet that Language, the biologically given universal human
any factual information, and any affective content1 capacity, and languages, such as English, Swahili,
that can be conveyed by an animal communication Cantonese, Dyirbal and Navajo, which are cultur-
system can also be conveyed, or at least satisfacto- ally developed systems enabled by the biological
rily paraphrased, in any human language. We can, capacity. Noone speaks Language; Language (with
we believe, concisely summarize the information a capital L) is not a language. This contrasts with
given by honeybee waggle dances, by chaffinch animal communication systems. True, different
chaffinch, and other songbird, dialects exist, but
1
A glossary of terms routinely used by linguists about their range is far less than that among human
language is given at the end of this chapter. languages.

249

14-Dettore-Chap14.indd 249 3/11/2008 10:55:06 AM

 250 S O CI O B I O LO GY O F CO M M U N I C AT I O N

By the usual count, there are over 6,000 different long as they put the words in the same order to
human languages. Defining the difference between convey who did what to whom, and use the same
a language and a dialect is ultimately not possible, conventional inflections to convey such details as
but a rough criterion is that different languages the timing of the event reported and the speakers
are mutually unintelligible, whereas there is some attitude to it, the language works. So languages are
degree of mutual intelligibility among different fairly free to evolve different grammars and sound
dialects of the same language. By this criterion, systems within the limits imposed by the commu-
there are in fact several different Chinese lan- nicative needs of the group.
guages, of which Mandarin and Cantonese are two, A contentious issue within linguistics is the
but Norwegian and Swedish actually count as the degree to which language learning, and therefore
same language. In the past, many other languages linguistic diversity, is further constrained by biol-
existed, but are now extinct. It seems likely that ogy. One can certainly imagine crazy languages
the peak number of languages spoken by humans that would be impossible to learn. A frequently
occurred some time in the last few millennia, when given example is a language which expressed ques-
the earth was as yet sparsely populated by small tions by completely inverting the word order of the
groups of humans living in relative isolation. Now, corresponding statements. In such a language, you
languages are being lost, and we are in an age of would ask the way to the station by saying Station
mass linguistic extinction, with predictions that the to way the me tell can you?. The strain on short-
about half the worlds languages will die out in term memory in computing how to express ques-
the next century. The great diversity of human lan- tions in such a language rules it out as a possible
guages is made possible by the fact that they are natural language. The contention within linguistics
learned, rather than biologically transmitted from arises because there may be some such constraints
generation to generation via the DNA. The fact that on language-learnability which are not attribut-
languages are learned is not, however, sufficient to able to nonlinguistic factors such as short-term
account for their great diversity. memory, but operate only in the specific domain of
The diversity of biological species arises through language. Here is a candidate for such a Language-
accumulated genetic copying errors, geographical specific constraint. Beware, like all such examples,
isolation, and selective adaptation to new niches. it involves thinking the impossible, something
Copying errors in learning and geographical iso- that linguists are skilled at. Consider the following
lation are also responsible for the great diversity pair of sentences.
of human languages. As early humans spread out
The man built the house. The house fell down.
over the planet, their group languages accumu-
lated changes which were not constrained by any We can make a single sentence, expressing the
need to communicate with the groups they had left same information, thus:
behind, and these languages struck out on their
The house that the man built fell down.
own. But adaptation to new niches is not a factor
affecting the diversity of languages, aside from the So far, so good. The compressed sentence was
relatively simple matter of vocabularylanguages formed by making a relative clause (underlined) out
of African pygmies have no unborrowed word of the first sentence, and attaching it to the shared
for snow. In matters of grammatical structure noun phrase, the house. The original right-hand
and structure of their sound systems, there is no sentence The house fell down is wrapped around the
correlation between languages and the physical outside of this underlined relative clause. Now lets
environments of their speakers. try it again, with this last complex sentence as one
A factor permitting the diversification of lan- of the inputs to the process:
guages is the arbitrariness of the sign1. A rose by
The house that the man built fell down. The man escaped.
any other name would smell as sweet. So long as
people tacitly agree to use the same words for the Here again, there is a shared noun phrase, the man.
same things, a language works. In grammar, as So in principle, it ought to be possible to use the

14-Dettore-Chap14.indd 250 3/11/2008 10:55:06 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 251

same relative clause-forming process. If we do, we grammar, the result of the Norman Conquest of
get: England in 1066 and centuries of contact ever since.
Many languages are of such hybrid types, owing
*The man that the house that built fell down escaped.
their structure to multiple sources. The possibility
This is an impossible sentence, as indicated by the of such hybridization adds to the overall diversity
linguists conventional asterisk. And generally, of languages.
across languages, we find that sentences like this, Summarizing the factors contributing to linguis-
and their analogues, adjusting for the differences tic diversity, (1) the fact that languages are learned,
between languages, such as word order, are also not rather than coded into the genes, (2) the arbitrari-
well-formed. The interesting question is Why? Is it ness of the sign, and (3) the prevalence of hori-
due to a constraint specific to Language, a putative zontal transmission allow for great diversity, but
Law of Language, that sentences such as this do this is significantly constrained by (4) biological
not occur in languages? Or is this fact due to a more factors such as memory and processing limitations,
general constraint on processing any kind of serial which may or may not be specific to the Language
information, linguistic or otherwise? Both opinions domain.
are held in the field, probably with a swing under
way to the general non-domain-specific explana-
tion. The original discoverer of this family of con-
14.3 Learning
straints, known as Syntactic Island Constraints Human languages are learned. Many animals are
was J.R.Ross (Ross, 1967, 1986), in the early heyday also adept at learning. But they cant learn human
of the generative linguistics1 movement, whose goal languages. Kanzi has learned about 500 words, but
was, in part, to discover facts peculiar to the human this seems to be close to his limit. And Kanzi is
Language faculty. The alternative view that such a highly human enculturated bonobo1. If animals
constraints arise from general constraints on learn- can learn, why cant they learn languages? What
ing any sequential behaviour has been argued by is different about languages? A major factor is the
Morten Christiansen, among others (Christiansen arbitrariness of the Sign. A chimpanzee can learn
et al., 2002, Christiansen & Ellefson, 2002). Note to use a tool to reach a banana. In this case the
that there are dozens of similar examples, a fact function of the tool is transparently mechanical.
which underlines the great complexity of human Physical laws govern the interaction of the tool and
languages, as compared with animal communica- the banana. A word, it can be argued, is also a kind
tion systems, where such considerations of such of tool. If I want a banana, just saying banana may
abstractness and complexity do not arise. be enough, if my hearer is cooperative, to get me
Linguistic inheritance is both vertical, as the banana. But the word banana has no physically
when children more or less faithfully acquire causal relation to the outcome. In this sense, words
the languages of their parents, and horizontal, are magic; just using words, if the hearers are coop-
as when languages mix and borrow each others erative, makes physical things happen. Apes have
words and constructions. Branching family tree a good understanding of physical cause and effect
diagrams are still popular in historical linguistics. with everyday objects, and can learn practical
But such genealogies are misleading. According tasks. But the arbitrary nature of human symbols
to a common classification, English is a Germanic is a far greater challenge to learning, because its
language (along with German, Dutch, Icelandic, not obvious how they work.
Danish and others), while French is a Romance To begin to understand, as human infants do,
language (along with Romanian, Portuguese, that the noises made by conspecifics carry some
Italian, Spanish and others). But such always- informative message, there needs to be, in the
diverging, never-converging tree diagrams distort child, a presumption of their relevance to its life
the extent to which languages have influenced each (Sperber & Wilson,1986). This is the idea that utter-
other across language family boundaries. English ing a sentence such as Its late not only conveys
and French share a lot of similar vocabulary and the information that something is late, but also

14-Dettore-Chap14.indd 251 3/11/2008 10:55:06 AM

 252 S O CI O B I O LO GY O F CO M M U N I C AT I O N

that the speaker intends the hearer to know that stage in babies is like play among animals, in that it
something is late. So an infant on hearing an utter- seems to have no immediate purpose. Play-fighting
ance in a human language knows that the speaker in young animals is plausibly accounted for in evo-
intends the hearer to know something. Count the lutionary terms in that it refines motor skills which
instances of intentional verbs1 in this last sentence will be useful later in life. Likewise, both babbling
(knows, intends, know), and we see three embedded and vocal imitation are practice for use of language
levels of intentionality1. (Intentionality is about- later in life. Human babies, unlike other apes, have
ness, some mental attitude to an external state of a natural disposition to engage in these activities,
affairs.) Hallidays (1975) provocative title for his whose payoff only materializes long afterwards.
book was Learning how to mean. Taken literally, this Childrens natural dispositions to imitate and to
might suggest a tabula rasa1 in the child, in which participate in group activities are signs of a kind of
even an understanding that utterances mean some- sociability special to humans, a topic to which we
thing has to be learnt. What we see in humans, as will return later in this essay.
opposed to non-humans, is a developmental proc-
ess whereby this understanding emerges well
14.4 Complexity
within the first year of life. It is developmentally
programmed into normal early human matura- We have already touched on the great complex-
tion, in normal circumstances, as opposed to being ity of languages, compared to any non-human
strictly learned. The human child is pre-disposed communication system. How does this complex-
to understand that utterances mean something. ity relate to biology? In linguistic theorizing, two
Play, babbling and imitation are aids to achieving opposing tendencies are felt. On the one hand, peo-
this understanding. ple are struck by the universality of such complex
A useful distinction has been made between facts as the Island Constraints mentioned above,
learning by emulation2 and learning by imita- and a host of other universal tendencies involving
tion. Emulation involves achieving the same goal intricate facts about the reference of pronouns, the
as was observed, but not necessarily by the same scope of quantificational words such as all and each,
means. For example, if a chimpanzee sees me push the varying semantic effects of verbs such as prom-
a door open with my foot, it may learn to push the ise and persuade and adjectives such as easy and
door open with its hand; this is emulation. But if eager3 and so forth. Children learn such abstract
the chimp slavishly follows my actual method of facts with no overt tuition and from examples
opening the door, using its foot, that would be which by no means logically determine the con-
learning by imitation. Whiten et al. (2004), in a sur- clusions the children come to. The childrens more
vey of ape learning, conclude that there is more or less faithful learning is evidenced by their later
emulation than pure imitation in apes, and both usage in general conformity with other members
kinds of copying occur much more readily when of the community. This led linguists in the 1970s
the demonstrator is a human trainer than spon- and 1980s to hypothesize a rich innate structure
taneously among apes themselves. None of the guiding the acquisition process, consisting of a set
work surveyed, however, involves the copying of of distinct innate modules such as the X-Bar mod-
clearly communicative behaviours. In some sense, ule, the Binding module, the Theta module, and
emulation is more intelligent; it gets the job done. the Case module (Chomsky, 1986). It is not neces-
Human children are natural imitators. They imi- sary here to explain the purported content of these
tate for no apparent reason, as shown by Meltzoffs modules; note that each is a set of propositions
(1988) well-known experiment, in which very determining some independent, but interlocking,
young babies imitated the facial gestures of adults. aspect of the linguistic knowledge that the child
Non-human apes are very poor at vocal imitation,
3
but human children are expert at it. The babbling Compare I promised John to go with I persuaded John to go.
Who, in each case, is to do the going? Also compare John is
easy to please with John is eager to please. Who, in each case, is
2
The term emulation is due to Wood (1988). understood as doing the pleasing?

14-Dettore-Chap14.indd 252 3/11/2008 10:55:06 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 253

will ultimately acquire. Thus syntactic theory, spectacularly unique among animal communica-
at this stage, responded to the complexity of lan- tions systems, its biological foundations rest in a
guages by postulating a complex biological endow- combination of many factors, many of which are
ment specifically devoted to Language. not specific to Language. Enhanced memory for
On the other hand, of course, there was always large numbers of arbitrary associations is one
Occams Razor, the normal scientific pressure to such factor. Humans typically have vocabularies
adopt theories which are as simple as possible. The of about 50,000 items. Kanzi, the best-performing
postulated richness of the innate language acquisi- non-human primate in this regard, has mastered
tion mechanism was a biological embarrassment, as about one hundredth of that number5, and is not
each of these modules presumably had to be coded expected to learn significantly more. It was always
somehow into the genome. Further, their interact- recognized that the lexical component of a lan-
ing4 nature in modern languages made it necessary, guage necessitated rote-learning, and this made
but difficult, to imagine stages in the evolution of it the least theoretically interesting component
the modern Language faculty when some of these of a language. A recent movement in the theory
modules were present and others had not yet of grammar, known as Construction Grammar1,6,
emerged. Certainly it is possible to imagine such suggests that there is in fact no principled distinc-
undeveloped versions of the modern Language tion between lexical items and grammatical con-
faculty, but it adds to the strain on credibility of the structions. Grammatical constructions are like
whole story, in an already speculative field. This lexical items with variables in them, permitting the
kind of gradual evolution of the Language faculty insertion of a more or less wide range of permit-
was proposed by Pinker & Bloom (1990), in a land- ted other constituent items. An example of such a
mark article arguing the proposition that the most construction in English is The + COMPARATIVE +
obvious explanation for the complexity of natural CLAUSE + the + COMPARATIVE + CLAUSE, as for
language is that it evolved by Darwinian natural instance in The more you eat, the fatter you get or The
selection. To many, this had seemed obvious, but bigger they are, the harder they fall. This theory places
it is a sign of the intellectual climate within the less emphasis on economy of statement in the
dominant paradigm in linguistics in the late 20th grammar, and recognizes that the representation of
century that it needed arguing at all. language in the brain may be somewhat redundant,
The simpler a theory of the innate Language and uneconomical, taking advantage of humans
faculty could be made to look, the more it appealed undoubtedly great powers of memorization.
to biologists. Quite recently, a theoretical move has The human Language faculty, rather than being
been made toward an extremely simple account a richly structured independent module of the
of the human language faculty; this is known as mind, is a mosaic of many factors which have come
the Minimalist Program1 (Chomsky, 1995), propos- together in a unique combination in humans. Many
ing that the language faculty is nothing more than of these factors can be observed, often in a less pow-
a facility to recursively merge lexical structures erful form, in other animals. The recognition that
(precisely specified dictionary entries) to form this is so is seen in a distinction made by Hauser
larger structures such as phrases and sentences. et al. (2002) between the faculty of language in the
It is stressed that this is a program rather than a broad sense (FLB1) and the faculty of language
theory, and its empirical and predictive delivery in the narrow sense (FLN1). FLN is whatever is
remains also minimal; one is reminded of String unique to human language; Hauser et al. (2002)
Theory in physics. suggest that this may, at most, be limited to the
Linguists have become persuaded in the last human capacity for recursion1, the execution of a
decade or so that, while human language is clearly computation of a certain type during the execution

4 5
These hypothesized modules of the Language faculty are Kanzi uses a lexigram board, with abstract symbols that he
interacting in roughly the same sense as subsystems of points to, as a substitute for uttering spoken words.
6
physical organisms, such as the respiratory system and the Goldberg (1995), Croft (2001), Fillmore (2003), Culicover
circulation system, work together. (2005).

14-Dettore-Chap14.indd 253 3/11/2008 10:55:07 AM

 254 S O CI O B I O LO GY O F CO M M U N I C AT I O N

of a similar computation at a higher level. For no sense in which the meaning of the whole gib-
example, to grasp what Johns fathers brothers neigh- bon song is understood as any combination of the
bours cat refers to, you have to identify the referent meanings of these subunits. In this sense, the term
of the subpart Johns fathers brothers neighbour, and song is appropriate, as such complex animal calls
to understand that, you have to identify the refer- are more like music than human language, which
ent of Johns fathers brother, and so on. Hauser et al. expresses semantic content through the application
leave it open whether such a capacity for recursion of compositionality.
can be found in any non-human animals. If it can The odd exception mentioned above is the wag-
be, then the human faculty for language in the nar- gle dance of honeybees, which has two meaningful
row sense is, in their view, actually empty, leaving components, the speed of the dance and its orien-
us with a picture of FLB as a mosaic of factors all of tation on the honeycomb. The speed conveys the
which can be found in some form or other outside approximate distance of food from the hive, and
of the domain of human Language. One candidate the orientation conveys the angle, relative to the
for recursion in animals is navigation; figuring out suns position, at which the food is to be found.
how to get from A to B might involve recursively Both distance and angle are necessary to specify
embedded processes. The technical definition of the foods location, and the meanings of the two
recursion, and how to recognize whether it is in aspects of the dance combine to compose this
play in a specific animal activity, is not, however information. The dance behaviour of honeybees is,
satisfactorily pinned down, and there is room for however, completely specified in their genes. The
argument about the use of recursion in animal example shows how utterly different mechanisms
activities. The radical view that human Language may achieve a communicative effect. The human
may have no unique individual properties is con- mechanism of learning languages is clearly more
troversial, and I will review below other candi- versatile in allowing the possibility of conveying a
dates for a categorical difference between human wide range of different messages, in adaptation to
Language and animal communication systems. a complex and changing world.
The nearest to an example of compositionality
in the communication of an animal closely related
14.5 Compositionality
to humans is described by Klaus Zuberbhler
The example of how we parse a recursive struc- (Zuberbhler, 2002, 2005). As any kind of precur-
ture such as Johns fathers brothers neighbours cat sor to the compositionality in human language, the
highlights another feature of human language example is problematic, as it involves the responses
that is not found in any animal communication of one species, Diana monkeys, to the alarm calls of
system that we know of (with one odd exception). another species, Campbells monkeys. Campbells
The Principle of Compositionality1 states that The monkeys have different unitary alarm calls for
meaning of the whole is a function of the mean- leopards and eagles. Diana monkeys interpret these
ings of the parts, and the way they are structured appropriately. Occasionally a Campbells monkey
together. You understand the meaning of a whole utters a boom about thirty seconds before such an
sentence because you know the meanings of the alarm call, and the Diana monkeys then react with
individual words, and you know the contribu- less panic than to boom-less alarm calls. . . . adding
tion the grammar makes to this understanding. booms before the alarm call series of a Campbells
This is how you know that Mary kissed Bill means monkey created a structurally more complex utter-
something different from Bill kissed Hillary. While ance with a different meaning than that of alarm
many complex animal calls are combinatorial, that calls alone. (Zuberbhler, 2005:279). Zuberbhler
is, they are made up of several reusable subunits himself is frank about the limitations of non-human
strung together, none is compositional in this primates: there is no evidence that they are able
sense. The songs of gibbons are sequences of units to invent and incorporate new call types into their
which occur in other contexts, and can therefore repertoires or to combine calls creatively to produce
be identified as independent subunits, but there is novel meanings (Zuberbhler, 2005:281)

14-Dettore-Chap14.indd 254 3/11/2008 10:55:07 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 255

14.6 Double articulation modelling, shows a process of self-organization

at work. They start with utterances which are not
The contrast between songs consisting of identi-
recognizably made up of discrete elements, but
fiable subunits and truly compositional signals
are merely random walks (trajectories) through
such as human sentences brings out another
articulatory space. To imagine such a trajectory, try
unique characteristic of human languages, their
to make some inarticulate vocal sound, moving
so-called double articulation1,7; at the phonologi-
your jaw, lips and tongue around while sporadi-
cal level, the expressions of human languages are
cally vibrating your vocal cords, and altering the
combinatorial but not compositional. That is to say
airflow through nose and mouth; avoid visiting
that the signals consist of systematically reusable
known phonemes of your language. By a process
subunits which themselves carry no meaning. The
of imitative transmission through a population of
separate phonemes1 making up a word have no
agents, with copying errors, and a requirement that
meaning. The word cat consists of three phonemes
the separate trajectories should not collapse with
{/k/ + /a/ + /t/}, but the meaning of the word is
each other, the set of such trajectories through
not derived from the meanings of the phonemes,
articulatory space gradually settled down to a
because they have no meanings. At the morpho-
more systematically organized set. In this evolved
syntactic level, the word cat does have a mean-
system, the same starting and end points were
ing, which contributes, for example, to the overall
used by many different trajectories.
meaning of a sentence such as The cat sat on the mat.
This can be envisaged in visual terms. Imagine
So languages are structured in two layers, a seman-
a square with random scribbles on it. The only
tically compositional morphosyntactic layer, and a
constraint is that each scribble is a single continu-
phonological layer which is merely combinatorial.
ous line. Agents are required to copy these lines
All human languages have this property.
and pass their copies on to a successive generation,
Double articulation clearly contributes to the
keeping the same overall number of lines. What
massive expressive power of human languages.
happens, over time, is that a more systematically
New meaningful words can be invented by simply
organized set of lines emerges, which start and end
combining phonemes from a given set. The pho-
at the various corners of the square. At the begin-
neme inventories of languages vary in size from
ning of the simulation, the corners of the square
a mere dozen to over a hundred. Obviously, lan-
had no special status in the formation of the lines/
guages with fewer phonemes at their disposal tend
scribbles. At the end of the simulation, there is con-
to have longer words. The combinatorial power
vergence on a system of lines which re-use a small
afforded by a phonological layer of structure pro-
set of starting and end points, and move economi-
vides languages with their vocabularies of tens of
cally between them. See Figure 14.1. This suggests
thousands of meaningful words.
that, even though humans are biologically capa-
The evolution of communication systems with
ble of making inarticulate vocal gestures, and of
this feature of double articulation is thought-pro-
attaching some meaning to them, what happens
voking. It clearly has a biological aspect. Humans
over time in the continuous trade of such gestures
must have the facility for combining elementary
is a self-organizing process by which articulate,
sounds from a small fixed set in highly flexible and
jointed speech, re-using a small set of focal points,
productive ways. But the fixed sets of phonemes
emerges.
vary widely from language to language, so these
are not biologically fixed, although the articula-
tory apparatus within which they are defined is 14.7 Self-organization
a matter of biological endowment. Recent work
The example of the emergence of combinatorial
by Zuidema & de Boer (in press), using computer
phonology introduces what may be a potent and
pervasive force in the evolution of languages, in
7
Also sometimes called duality of patterningthe terms their grammars as well as in their sound systems.
are equivalent. The investigation of such self-organizing processes

14-Dettore-Chap14.indd 255 3/11/2008 10:55:07 AM

 256 S O CI O B I O LO GY O F CO M M U N I C AT I O N

In other words, each individual involved in the

process could have behaved in a variety of ways,
as far as any direct pressure from the genes is con-
cerned. Bu the accumulation of hundreds of tiny
unconscious facultative acts led to the language
concerned being the way it is.
Here is an example. A salient feature of very
many languages is a correlation between fre-
quency and irregularity. For example, in English,
Figure 14.1 Self-organization in articulatory space. The the most common verbs (be, have, do, make, go, etc.)
left-hand box contains ve randomly scribbled lines, schematically
are all irregular. This correlation between fre-
representing random gestural trajectories in articulatory space.
The right-hand box shows ve trajectories approximately optimized quency and irregularity in languages comes about
for simplicity and distinctiveness from each other. After Zuidema through the repeated action of several processes.
and de Boer (in press). One process is the tendency to slur or phonetically
erode high frequency words. This erosion creates
irregularities. It is well known that children are
in the context of language evolution is relatively somewhat resistant to irregularities, tending natu-
new. It is a distinct process from natural selec- rally to regularize even irregular verbs. Thus chil-
tion, but entirely compatible with it. In an early dren learning English go through a stage where
pioneering work on self-organization, Thompson they use *goed instead of went and *comed instead
(1961) tended to depict self-organization (laws of of came. In the case of the most frequent irregular
growth and form) and natural selection as mutu- verbs, however, the irregular usage in the environ-
ally exclusive alternatives. More recently, Oudeyer ment overwhelms the childs natural disposition
(2006) gives a clear discussion of the relationship to regularize, with the result that irregular forms
between natural selection and self-organization. persist in the language, just in the more frequent
Self-organization can affect both organic phenom- forms. For less frequent forms, the child is not pre-
ena (e.g. snail shells) and non-organic phenomena sented with enough evidence to overrule its natu-
(e.g. snow crystals). Self-organization narrows the ral tendency to regularize, and less frequent forms
search space of possibilities from which natural are mostly regular. This process has been modelled
selection selects. computationally by Kirby (2001).
In the evolution of language, the most promis- Such self-organizational processes have been
ing cases of self-organization arise through the dubbed phenomena of the third kind by Keller
interaction of users of a language over historical (1994). In his taxonomy, phenomena of the first
time. The self-organized object which emerges is kind are natural phenomena, like oceans and vol-
not a physical object like a snail shell or a snow canoes; phenomena of the second kind are human
crystal, but the language itself, an abstraction over artefacts, deliberately made, like telephones and
the common behaviours of the speakers of the lan- the written constitutions of nations; phenomena of
guage. However, a physical, non-linguistic exam- the third kind are the outcome of human action,
ple may help. Consider an informal well-worn but not deliberately made by any single, or even
footpath diagonally crossing a field. The path was collective, conscious decision. Keller also invokes
not deliberately designed by any one person, but Adam Smiths Invisible Hand (Smith, 1786),
is simply the end product of hundreds of people paraphrased in modern terminology as market
taking the shortest route across the field. In the forces. Keller argues that much of the evolution
case of language, repeated usage over generations, of language should be seen as an Invisible Hand,
with idealized copying of the observed patterns i.e. self-organizing, process. The self-organization
by new learners, results in features of language of a communication system along the lines illus-
which were not the invention of any one person, trated here can only happen in a relatively complex
and further, were not closely dictated by the genes. learned system, such as humans have. With such

14-Dettore-Chap14.indd 256 3/11/2008 10:55:07 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 257

limited systems as the mostly innate 3-way alarm Animals can plan future actions, to some
calls of vervet monkeys, there is far less scope, if degree. Mulcahy & Call (2006) report on experi-
any, for the accumulation of tiny facultative actions ments in which bonobos and orangutans col-
determining the historical course of the system. lected and hoarded appropriate tools for tasks as
far ahead as 14 hours before the task was carried
out. They comment that These findings suggest
AQ: We 14.8 Stimulus-freedom
have that the precursor skills for planning for the future
renumberd There is a considerable difference of degree between evolved in great apes before 14 million years ago,
this A
head to
humans and non-humans in the extent to which when all extant great ape species shared a common
ensure their mental processes are immediate reactions to ancestor. (p. 1038)
continuity. their environment. Humans can recall, and muse There is a symmetrical relationship between
Kindly
confirm.
about, specific events from long in the past, and planning and memory, to the extent that planning
can plan complex series of actions far in the future. has sometimes been classified as prospective mem-
One can find the tiny seeds of stimulus-freedom ory (Meacham & Singer, 1977). In one experiment
in animals closely related to humans. In object- (Cook et al. 1983), rats searching a 12-arm radial
displacement experiments, for example, a desirable maze for food were taken out while still searching
object is hidden from an animals view, but the ani- and replaced in the maze later. They showed simi-
mal still seems to know it is there, and searches for lar accuracy of recall in relation to (1) number of
it. Thus the animal has a mental representation of arms already previously searched and (2) number
an object not currently perceived. Dogs are good of arms not yet searched and therefore remaining
at this. Panzee, a chimpanzee, could remember to search. This indicates an overlap of the mecha-
after a nights sleep where food had been hidden nisms of retrospective and prospective memory.
the day before (Menzel, 2005). It is often claimed Such memories are stored in the animals brain
(Tulving, 1972, 2005; Suddendorf & Corballis, 1997, and are not dependent on its current perceptions.
in press) that only humans have episodic memory1, Humans, however, have much longer retrospec-
a recall of specific events, as opposed to non-time- tive and prospective memories than non-humans.
indexed knowledge of some state of affairs (which Suddendorf & Corballis (1997) write of the uncon-
may result from having observed some event in strained mental time travel of humans.
the past). Experimentally sorting out the difference Humans can think about absent things. This
between recall of events and knowledge of result- stimulus-freedom of human mental processes is
ing facts is problematic. In recognition of this diffi- reflected, naturally, in our communication sys-
culty, experimenters have attributed episodic-like tems. We can talk about absent things, and in fact
memory to animals such as scrub jays, which show this is the norm for human communication. We
evidence of remembering what food they hid, and constantly bring to mind distant events or possible
where, and how long ago (Clayton & Dickinson, future events, and talk about them. The structure
1998; Clayton et al, 2001). of modern languages makes this possible, but this
The evidence from animals who show some is probably a case where language structure has
slight signs of episodic memory means that such evolved to meet the need to express such time-
memory is not absolutely dependent on the prior travelling thought, rather than the structure of
evolution of language. Certainly in humans, epi- language actually enabling such time-travelling
sodic memories are aided by public language. There thoughts in the first place. A simple story, probably
was presumably some co-evolution of the faculty of partly correct is: first the private thought capacity,
Language and a capacity for episodic memory. In then a communication system adapted to make the
humans, the earliest memories of specific lifetime private thoughts public.
events are typically from roughly around 2 years The relation between language and thought is
of age, when syntactic language begins to develop. a hot philosophical issue. Most comparative psy-
This suggests some interdependence between chologists, and a growing number of philosophers,
episodic memory and language. are willing to concede some thoughts and concepts

14-Dettore-Chap14.indd 257 3/11/2008 10:55:07 AM

 258 S O CI O B I O LO GY O F CO M M U N I C AT I O N

to non-humans. But clearly there are thoughts that thought goes through the mind of the threatener or
can only be attained with the help of language. the threatened animal. A tiny number of human
Examples are the concepts expressed by words utterances have only this bare illocutionary force.
and phrases such as Tuesday, unicorn, ninety-three, For example, Hello just does greeting; it has no
zero, generosity and legal. Examples such as these declarative content, and doesnt describe any state
rely on verbal definitions made possible by the of affairs. The vast majority of human utterances
productive generative capacity of languages. Given have some social purpose, an intended impact on a
compositionality (as discussed above), it is possible hearer8, in addition to whatever descriptive content
to arrive deductively at meanings not previously they may have. For example, Its raining, in addition
entertained by the mind. For example, given the to describing the current weather, will always in
concepts expressed by white, horse, single, horn and normal circumstances be said with some intended
forehead, compositionality allows one to deduce effect on a hearer, such as to warn them to put on a
what the expression white horse with a single horn in raincoat, or to prove that your prediction was right,
its forehead should mean, even though we are never or to jokingly complain about the local climate. As
likely to experience such a beast. Presumably such this dyadic doing things to each other function is
thoughts are permanently denied to non-humans. basic to both animal and human communication,
Once fictions can be expressed, and thus shared it seems likely that this is a remote evolutionary
between people, they can become potent cultural foundation of human language, and that the vast
forces, defining group identity. A commitment to referential, descriptive, triadic power of language
the proposition that Jesus Christ is the son of God came later. The set of social acts which can be car-
is what centrally divides devout Christians from ried out using language exceeds the range of things
devout Muslims. Thus, beside the obvious prac- that non-human animals can do to each other with
ticality of generative language, for transmitting symbolic signals, such as threat or submission ges-
real-world information, enabling us to build space- tures. All of the acts that can be carried out by non-
ships that reach the Moon, generative language humans can also be carried out using words (see
provides for the construction and sharing of rich related discussion in Chapter 1). Thus a threat can
structures not corresponding to any perceptible be made by purely nonverbal means, e.g. by shak-
reality, defining complex cultures. ing ones fist in a persons face; and it can also be
carried out in the calmest of ways with words, with
little of emotion, by saying If you move, Ill shoot.
14.9 Interpersonal function
Social acts between animals are mostly dyadic, only
The vast potential of languages for describing the involving the sender and the receiver of the signal
real world, and fictitious worlds, in detail, should (see detailed discussion in Chapter 3). Humans can
not lead us to ignore the fact that making descrip- overlay their social acts with descriptive content,
tive statements about a world must have a social as in the previous example, which refers to actions
purpose. Austin (1962) wrote of the descriptive such as moving and shooting. This vastly increases
fallacy, the idea that the point of language is to the subtlety and fine-grained detail of things that
describe a world. He famously stressed that when humans can do to each other, using language.
we use any language at all, we are thereby doing Some things that humans can do to each other,
something, carrying out some social act. Much using language, can only be done with language, or
animal communication carries this purely social
force, and it is inappropriate to paraphrase such 8
An exception may be private soliloquizing, praying, or
signals in declarative terms. For example, a threat talking to oneself. It seems likely that these are uniquely
human activities which evolved on top of a prior purely
signal is just doing the threat, or it just is the threat. social communicative form of language. Chomsky (1980) is
Translating an animal threat signal into a human in a minority in holding that such talking to oneself may be
declarative sentence, such as If you dont back off, the main function of human language.
9
Another clear difference between the vervets code and a
I will attack you may be useful for our purposes, but human language is along the innate/learned dimension, of
there is no reason to suppose that any such complex course.

AQ: Please provide the

corresponding note cue
for note 9.

14-Dettore-Chap14.indd 258 3/11/2008 10:55:07 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 259

at least in a language-defined context. Thus, prom- could see two food items, and was also in a posi-
ising, for example, requires some understanding tion to see that a dominant chimpanzee could only
of what is promised, which can only be expressed see one of these food items. When both animals
with words. True, I can effectively promise some- were released from their positions, the subordi-
thing merely by nodding, but in such a case what nate chimpanzee reliably went for the food item
I am thereby committed to has previously been that had been invisible to the dominant.
spelled out in language. Another class of uniquely All work of this kind is centered around the com-
human communicative acts is those where a social plex question of the extent to which non-humans
fact or convention is made to exist solely by using have a Theory of Mind1, the ability to know that
an appropriate verbal formula, as in examples like another organism is just like, and therefore thinks
I name this ship the Mary Rose or I hereby declare like, oneself. Note that there are two main com-
you man and wife or Ego te absolvo. Such acts are, of ponents here (1) the obvious one, just stated, and
course, impossible in the non-human world. (2) what oneself is like. Informal character attri-
butions among people show a tendency to project
ones own vices and virtues onto other people.
14.10 Mindreading, manipulation
Thus a generous person will tend to assume that
and cooperation
other people are generous; and a miserly person
Encounters between animals can be either adver- will tend to assume that other people are also
sarial or mutually beneficial. In both cases, it is miserly. Crucially, a naturally uncooperative ani-
advantageous to an animal to be able to predict mal will not be able to read cooperative intentions
and influence the actions of the other. Predicting in another animal, although it may well be able to
events can involve various degrees of intentional- read competitive intentions in another.
ity. Predicting that a falling rock will land near you There is experimental evidence that chimpan-
involves no understanding of the mental processes zees can read competitive intentions in human
of another organism. Predicting that a lion skulk- experimenters but not generous cooperative inten-
ing nearby will chase you may, or may not, involve tions, when the stimuli presented to the animal are
attributing some attitude to the lion. A zebra may very similar. In one condition, a human made a
simply have a built-in avoidance response to nearby reaching gesture, with hand spread for grasping,
skulking lions, just as some people may possibly toward a container; in this condition, the observ-
have built-in, or epigenetically easily-triggered, ing chimpanzee anticipated the humans reach and
arachnophobia. But some ability to mind-read got to the container first. In another condition, the
accurately the intentions of competitors, predators gesture was very similar, but with fingers together
and prey would clearly be advantageous to any in a whole-hand pointing gesture, indicating the
animal (Krebs & Dawkins, 1984). container. The chimpanzee subject did not get
Experiments with chimpanzees show that they the point of this cooperative pointing gesture.
can tell whether a human experimenter is teasing A natural interpretation is that chimpanzees can
them or merely being clumsy (Call et al. 2004), thus read the intentions of others, but they do not expect
demonstrating a degree of mind-reading. There those intentions to be cooperative. Thus, a certain
are also many reports of tactical deception among category of others intentions (the cooperative
primates, and Byrne & Corp (2004) also found a intentions) remains obscure to them.
correlation between neocortex size and rate of tacti-
cal deception. Thus one thing bigger brains is good
14.11 Reference
for is deception, which involves both prediction of
the likely actions of another organism and delib- Communicative acts in the animal world are mostly
erate manipulation to influence them. Hare et al. dyadic, not involving any third entity beside the
(2000) showed experimentally that Chimpanzees sender and receiver of the signal. A widespread
know what conspecifics do and do not see (their exception is alarm calls. The alarm calls of vervet
title). In this experiment, a subordinate chimpanzee monkeys (Cheney & Seyfarth, 1990) are especially

14-Dettore-Chap14.indd 259 3/11/2008 10:55:08 AM

 260 S O CI O B I O LO GY O F CO M M U N I C AT I O N

well known, but many other species of birds and after hearing, for example, an eagle alarm call fol-
mammals also have ritualized alarm calls for spe- lowed by the growl of a leopard. The logic is this.
cific classes of predators, typically with separate If, on hearing an eagle alarm call you are pre-
signals for aerial and terrestrial predators. Alarm pared to be wary of an eagle in the area, you are
calls are triadic because they involve the sender, the less disturbed by hearing the actual eagle shriek.
receiver, and the referent of the call. Triadic com- The eagle shriek merely confirms what the earlier
munication is about something, whereas dyadic alarm call told you. But if you hear an eagle alarm
communication is not. Animal alarm calls are and then hear a leopard growl, the growl is new
largely genetically determined, in both production information, telling you about a kind of predator
and reception, with very little room for voluntary that you hadnt been made aware of by the previ-
control. In young vervets there is some learning ous call. The researchers did, of course, try out all
of the specific class of aerial objects for whom it is the necessary control conditions to consolidate this
appropriate to make the eagle alarm call. And there conclusion. The conclusion is that the alarm calls
is also an audience effect, with mothers being more do not merely trigger the relevant evasive action,
likely to make an alarm call when their own off- with no representation of the specific source of
spring are nearby. Since both the stimulus-to-call danger being kept in the head; the Diana monkeys,
behaviour and the call-to-response evasive behav- on hearing a leopard alarm call, keep the idea of a
iour (e.g. climbing a tree when hearing the leopard leopard in their minds for at least five minutes; and
call) are strongly specified in the animals genes, likewise with the eagle alarm call. This behaviour
the question arises whether the animals are refer- meets the criteria set by Marler et al. (1992) for calls
ring, in anything like a human sense, to the preda- being functionally referential. It seems likely that
tor. It could be the case, for instance, that natural similar results would be obtained with all species
selection has acted in parallel to favour two inde- with small inventories of predator-specific alarm
pendent but mutually adaptive behaviours: (1) Bark calls.
when seeing a leopard, and (2) Climb a tree when The term reference is used by animal researchers
hearing a bark. If this were the case, there would be with less care than by most linguists. There are two
no human-like sense in which the animals alarm senses that need to be distinguished. In the discus-
call means, or brings to mind, the appropriate class sion above of alarm calls, the question is whether
of predators. some class of calls, such as a vervets bark, has a
Klaus Zuberbhler has described experiments referential meaning in roughly the same way that
which can be naturally interpreted to suggest that the English word leopard has. Of course transla-
animal alarm calls do in fact bring the concept tion even from one human language to another is
of the appropriate predator to mind, at least for a seldom, if ever, perfect, so we should not expect
short period. Zuberbhler et al. (1999) worked with to have a perfect English translation of what the
Diana monkeys of the African forest who have dis- vervets bark means. But the idea is that the vervets
tinct calls for leopards and eagles. Female monkeys have a (very limited) code, shared by the whole
both give spontaneous alarm calls on sensing a community, according to which bark means what
predator and respond to alarm calls from males by we humans roughly translate as leopard. When a
repeating the call. Beside recording the alarm calls, vervet hears the bark, it brings a certain concept to
the researchers also recorded characteristic noises mind. Certainly, the vervets concepts are only pro-
associated with the two predators, such as the toversions of ours, because they cannot expound on
growl of a leopard and an eagles shriek. Next, they the nature of leopards, and presumably do not ever
played back three different kinds of pairs of stim- mutely reflect dispassionately on the nature of leop-
uli, where the stimuli in each pair were separated ards8. But nevertheless, we may see in such alarm
by an interval of five minutes silence. On hearing calls a skeletal version of our own shared codes
first an eagle alarm call, then (after five minutes) (vocabularies), by which reference to classes of
the shriek of an eagle, female monkeys showed objects and actions is conventionalized to arbitrar-
less sign of alarm (giving fewer repeat calls) than ily related signals. Putting it anthropomorphically,

14-Dettore-Chap14.indd 260 3/11/2008 10:55:09 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 261

a vervets bark denotes something like the class of The absence of pointing in primate communica-
leopards. Such denotation1 is one sense of the term tion in the wild highlights the absence of a human
reference, and in this sense it is the signals them- level of cooperation.
selves, the proto-words, that refer.
The other sense in which reference is used does
not involve a shared code, conventionally mapping 14.12 Conclusion
a set of signals onto corresponding classes of objects There is indeed a wide gap between human lan-
and actions. In this sense, it is the individual users guage and non-human communication, in the
themselves who do the referring. Here is an exam- various ways I have surveyed here. The difference
ple. Third person personal pronouns, such as he, she, cannot be attributed to any single factor. Apes are
it, this and that can be used with variable reference; different from us in many qualitative ways. It seems
what they point out on any occasion depends on most likely that at some time a critical combination
the circumstances at the time. If I write here Thats of factors arose in our ancestors, which gave rise to
good, you, as a reader remote in time and space, the rapid expansion of the Language faculty, in its
dont know what I am referring to, and nothing in many facets, and a concomitant diversification and
the word that gives you any clue about what class of enrichment of individual languages and cultures.
objects the referent of that might belong to. Words Exactly what the components of that critical combi-
such as these are known to linguists as deictic1, or nation were is still to be discerned, and it is not clear
pointing, words. The word is used in a particular what further evidence we may be able to call upon.
context to draw attention to some particular thing,
and on other occasions to draw attention to totally
different things. All that is conventionalized about Summary
such words is that they are used for pointing to Human languages are far more complex than
things in the context of the current discourse. What any animal communication system. Furthermore,
exactly they point to is left to the pragmatic infer- they are learned, rather than innate, a fact which
ence of the observer. Pointing with the index finger partially accounts for their great diversity. Human
(or in some cultures the lower lip) is a non-linguistic languages are semantically compositional, gener-
analogue of the use of deictic words. What precur- ating new meaningful combinations as functions
sors can be found in non-human behaviour for this of the meanings of their elementary parts (words).
kind of pointing behaviour, whereby an animal This is unlike any known animal communica-
draws the attention of another animal to some spe- tion system (except the limited waggle dance of
cific object in their shared context? In the wild, none; honeybees). Humans can use language to describe
in captivity, some, but only with their human keep- and refer to objects and events in the far distant
ers. In thirty years of observing chimpanzees in the past and the far distant future, another feature
wild, Jane Goodall (Goodall, 1986) never observed a which distinguishes language from animal com-
chimpanzee point to an object with a view to draw- munication systems. The complexity of languages
ing the attention of some other chimpanzee to it. arises partly from self-organization through
Primates in the wild just dont point to things. In cultural transmission over many generations of
this sense, they dont refer to specific things. This users. The human willingness altruistically to
kind of reference is totally absent from non-human impart information is also unique.
communication in the wild. In captivity, chimpan-
zees and other primates learn to communicate their
needs to human keepers by pointing, but the cir- References
cumstances are limited to the fulfilment of current Austin, J.L. (1962). How to Do Things with Words. Harvard
desires, as when a chimpanzee points to some food University Press, Cambridge, MA.
item that he wants to be given. Even in captivity, Byrne, R.W. and Corp, N. (2004). Neocortex size predicts
apes do not point to things just to share information deception rate in primates. Proceedings of the Royal
about some interesting property they may have. Society of London, Series B, published online.

14-Dettore-Chap14.indd 261 3/11/2008 10:55:09 AM

 262 S O CI O B I O LO GY O F CO M M U N I C AT I O N

Call, J., B. Hare, M. Carpenter, and Tomasello, M. (2004). Hauser, M.D., Chomsky, N. and Fitch, W.T. (2002). The
Unwilling versus unable: chimpanzees understand- faculty of language: What is it, who has it, and how
ing of human intentional action. Developmental Science, did it evolve? Science, 298, 15691579.
7, 488498. Huddleston, R. and G.K. Pullum (2002). The Cambridge
Cheney, D. and Seyfarth, R. (1990). How Monkeys See the Grammar of the English Language. Cambridge: Cambridge
World: Inside the Mind of Another Species. University of University Press.
Chicago Press, Chicago, IL. Keller, R. (1994). On Language Change: the Invisible Hand in
Chomsky, N. (1980). Rules and Representations. Basil Language. London: Routledge. (Translation and expan-
Blackwell, London sion of Sprachwandel: Von der unsichtbaren Hand in der
Chomsky, N. (1986). Knowledge of Language. Praeger, Sprache. Tbingen: Francke.).
New York. Kirby, S. (2001). Spontaneous evolution of linguistic struc-
Chomsky, N. (1995). The Minimalist Program. Current ture: an iterated learning model of the emergence of
Studies in Linguistics 28, pp.420. MIT Press, Cambridge, regularity and irregularity. IEEE Journal of Evolutionary
MA. Computation, 5, 102110.
Christiansen, M.H., Dale, R., Ellefson, M.R. and Conway, Krebs, J. R. and R. Dawkins (1984). Animal signals: mind-
C. (2002). The role of sequential learning in language reading and manipulation. In JR Krebs and N B Davies,
evolution: Computational and experimental studies, pp. eds. Behavioural Ecology: an Evolutionary Approach.
165187. Springer Verlag, London. Blackwell Scientific Publications, Oxford.
Christiansen, M.H. and Ellefson, M.R. (2002). Linguistic Marler, P., Evans, C.S. and Hauser, M.D. (1992). Animal
adaptation without linguistic constraints: The role of signals? reference, motivation or both? In H Papoucek,
sequential learning in language evolution. In A Wray, U Jrgens and M Papoucek, eds. Nonverbal Vocal
ed. The Transition to Language, pp. 335358. Oxford Communication: Comparative and Developmental
University Press, Oxford Approaches, pp. 6686. Cambridge University Press,
Clayton, N.S. and Dickinson A. (1998). Episodic-like Cambridge.
memory during cache recovery by scrub jays. Nature, Meacham, J.A. and Singer, J. (1977). Incentive effects in
395, 272274. prospective remembering. Journal of Psychology, 97,
Clayton, N.S., Griffiths, D., Emery, N. and Dickinson A. 191197.
(2001). Elements of episodic-like memory in animals. Meltzoff, A. (1988). Infant imitation after a one-week
Philosophical Transactions of the Royal Society of London delay: long-term memory for novel acts and multiple
B, 356, 14831491. stimuli. Developmental Psychology, 24, 470476.
Cook, R., Brown, M., and Riley, D. (1983). Flexible memory Menzel, C. (2005). Progress in the study of chimpan-
processing by rats: Use of prospective and retrospec- zee recall and episodic memory. In HS Terrace and
tive information. Journal of Experimental Psychology. J Metcalfe, eds. The Missing Link in Cognition: Origins
Animal Behavior Processes, 11, 453469. of Self-Reflective Consciousness, pp. 188224. Oxford
Croft, W. (2001). Radical Construction Grammar: Syntactic University Press, Oxford.
Theory in Typological Perspective. Oxford University Mulcahy, N.J. and Call, J. (2006). Apes save tools for future
Press, Oxford. use. Science, 312, 10381040.
Culicover, P. and R. Jackendoff (2005). Simpler Syntax. Oudeyer, P.-Y. (2006). Self-Organization in the Evolution of
Oxford University Press, Oxford. Speech. Oxford University Press, Oxford. (Translated
Fillmore, C., Kay, P., Michaelis, L.A. and Sag, I. Construction from French Lauto-organisation de la parole by James R.
Grammar. University of Chicago Press, Chicago. Hurford).
Goldberg, A. (1995). Constructions: A Construction Grammar Pinker, S. and Bloom, P. (1990). Natural language and
Approach to Argument Structure. University of Chicago natural selection. Behavioral and Brain Sciences, 13,
Press, Chicago. 707784.
Goodall, J. (1986). The Chimpanzees of Gombe: Patterns of Ross, J. R. (1967). Constraints on Variables in Syntax. Ph.D.
Behavior. Harvard University Press, Cambridge, MA. thesis, MIT. (Published as Ross, 1986.).
Halliday, M.A.K. (1975). Learning how to mean: Explorations in Ross, J. R. (1986). Infinite Syntax! Ablex Publishing
the Development of Language. Edward Arnold, London. Company, Norwood, New Jersey.
Hare, B., Call, J., Agnetta, B. and Tomasello, M. (2000). Smith, A. (1786). An Inquiry into the Nature and Causes of
Chimpanzees know what conspecifics do and do not the Wealth of Nations: in three volumes. Fifth edition, A
see. Animal Behaviour, 59, 771785. Strahan and T Cadell, London.

14-Dettore-Chap14.indd 262 3/11/2008 10:55:09 AM

 T H E E VO LU T I O N O F H U M A N CO M M U N I C AT I O N A N D L A N GUAG E 263

Sperber, D. and D. Wilson (1986). Relevance : Communication Reflective Consciousness, pp. 356. Oxford University
and Cognition. Blackwell, Oxford. Press, Oxford.
Suddendorf, T. and Corballis M.C. (1997). Mental Whiten, A., Horner, V, Litchfield, C. and Marshall-Pescini,
time travel and the evolution of the human mind. S. (2004). How do apes ape? Learning and Behaviour, 32,
Genetic, Social, and General Psychology Monograph, 123, 3652.
133167. Wood, D. (1988). How Children Think and Learn. Basil
Suddendorf, T. and Corballis, M.C. The evolution of Blackwell, London.
foresight: What is mental time travel and is it unique Zuberbhler, K. (2002). A syntactic rule in forest monkey
to humans? Behavioral and Brain Sciences, in press. communication. Animal Behaviour, 63, 293299.
Thompson, D. (1961). On Growth and Form. JT Bonner, ed. Zuberbhler, K. (2005). Linguistic prerequisites in
Cambridge University Press, Cambridge. Abridged the primate lineage. In M Tallerman, ed. Language
edition, Origins: Perspectives on Evolution, pp. 263282. Oxford
Tulving, E. (1972). Episodic and semantic memory. In University Press, Oxford.
E Tulving and W Donaldson, eds. Organization of Zuberbhler, K., Cheney, D.L. and Seyfarth R.M. (1999).
Memory, pp. 381403. Academic Press, New York. Conceptual semantics in a non-human primate. Journal
Tulving, E. (2005). Episodic memory and autonoesis: of Comparative Psychology, 113, 3342.
uniquely human? In H. S. Terrace and J. Metcalfe Zuidema, W. and de Boer, B. The evolution of combinato-
(Eds.), The Missing Link in Cognition: Origins of Self- rial phonology. Journal of Phonetics, in press.

Glossary

Affective content: Part of the overall meaning spread of possible expressions in a language,
of an utterance which conveys the emotions including irregular and idiosyncratic ones.
or attitudes of the speaker, such as anger or Representative works are Goldberg (1995),
sarcasm. Croft (2001), Fillmore (2003) and Culicover
Arbitrariness of the sign: The non-iconic and (2005). Construction Grammar is sometimes
non-causal relationship between a word and its said to be opposed to generative linguistics, but
meaning. Onomatopeic words such as cuckoo both are concerned with the same goal, a clear
and miaow are rare exceptions to the general characterization of humans impressive syntactic
arbitrariness of the sign. abilities.
Bonobo: Pan paniscus, or pygmy chimpanzee, a Deictic: A deictic, or indexical, word has no
separate species from the common chimpanzee, constant reference in the external world, being
Pan troglodytes. applied to whatever ts in the particular context
of a conversation. For example, the English deictic
Compositionality: The principle whereby the
pronouns I and me do not refer constantly to any
meaning of a linguistic expression (such as a
particular person, such as the Queen of England
phrase or sentence) is a function of the meanings
or the Dalai Lama; if the Dalai Lama happens to
of its parts (e.g. the words), and the way they
say I am happy, then on that instance of use,
are structured together by the grammar of the
the pronoun I refers to the Dalai Lama.
language concerned.
Denotation: The constant relationship between
Construction Grammar: A cluster of theories
a word and what it picks out. Thus cat, for
in syntax, emphasizing the similarity between
instance, denotes the (fuzzy) set of cats in the
simple lexical items (words) and more complex
world, or alternatively, the concept of what is a
constructions. Such theories place more
cat shared by the language community.
reliance on human memory than other theories,
more driven by parsimony of representation. Double articulation: Alternatively known as
Construction Grammar focusses on the whole Duality of Patterning. This is the organization
continues

14-Dettore-Chap14.indd 263 3/11/2008 10:55:09 AM

 264 S O CI O B I O LO GY O F CO M M U N I C AT I O N

Glossary continued
of language structure into two distinct layers, a Intentional verb: A verb such as believe or
phonological layer assembling meaningless units, desire, which relates to some mental state, where
such as phonemes, into larger sequences such as the mental state often involves a representation
syllables, and a morphosyntactic layer assembling of some state of affairs in the external world.
meaningful units, such as stems and afxes, into Minimalist Program: The most recent
larger sequences such as phrases and sentences. incarnation of the generative linguistics research
All known human languages, including manually programme, summarized by Chomsky (1995).
signed languages of the deaf, have double Here the emphasis is on seeing how little
articulation. theoretical apparatus needs to be postulated
Episodic memory: An ability to recall specic to account for the complexity of languages.
events which have taken place earlier, as opposed The Minimalist Program attempts to subsume
to a memory for constant facts. Amnesics suffer organizational principles of language previously
from loss of episodic memory. thought to be independent under a single
FLB: The faculty of Language, in the broad abstract operation on the basic lexical elements
sense, incorporating all that is necessary to the of languages.
working of language, such as control over the Phoneme: A meaningfully distinctive phonetic
output mechanism (e.g. the vocal tract), memory, segment of a language. A phoneme does not
understanding of the intentions of other people, have a meaning in itself, but it is capable of
and so on. The concept is due to Hauser et al. signalling a difference in meaning. Thus English
(2002). /b/ and /p/ are distinct English phonemes,
FLN: The faculty of Language, in the narrow because bat and pat mean different things
sense, including only that which is distinctive of in English. In Arabic, by contrast, there is no
human language. If some property can be found corresponding difference between these two
in animal communication, or in human cognition sounds, and a [p] sound, if it is uttered, is not
outside of language, then by denition it does taken to distinguish a word from a word with a
not belong in FLN. The concept is due to Hauser [b] sound; thus in Arabic there is no phoneme
et al. (2002). /p/. (In linguistics, raw sounds are transcribed
between square brackets, and the phonemes
Generative linguistics: An approach to the
of a language are transcribed between oblique
study of language, pioneered by Chomsky and
slashes.)
largely associated with his ideas. In the early
days (roughly 19551970) it was characterized by Recursion: A property of computational
emphasis on explicit statement of formal rules, operations which call themselves. A classical
and generative was taken to mean explicit and formulation of the mathematical notion of a
rigorously formulated. More recently, generative factorial is recursive in this way. The factorial of
linguistics has become associated with a cluster of 1 is 1, and the factorial of any other number is
claims about the innateness of the human faculty the product of that number and the factorial of
of language and its distinctness from other the number immediately below it.
cognitive domains. Tabula rasa: A blank slate, the expression used
Illocutionary force: The socially conventional act to describe the most extreme form of empiricism,
embodied in the performance of an utterance. claiming that literally all knowledge comes from
Common illocutions are thanking, apologizing, experience. But any slate, even a blank one, has
congratulating, and greeting. Every utterance in a some properties determining what can, and
natural context has some illocutionary force, i.e. cannot, be written on it by experience.
is intended in some way to do something to the Theory of Mind: Put simply, the ability to
hearer. understand that another organism is an agent
Intentionality: Aboutness, some mental just like oneself. Severe autistics lack a Theory
attitude to an external state of affairs. of mind.

14-Dettore-Chap14.indd 264 3/11/2008 10:55:09 AM