http://www.newscientist.com/channel/being-human/mg19826571.700-so-you-think-humans...
So you think humans are unique?
21 May 2008
From New Scientist Print Edition. Subscribe and get 4 free issues.
Christine Kenneally

THERE was a time when we thought humans were special in so many ways. Now we know better. We are not the only species that feels
emotions, empathises with others or abides by a moral code. Neither are we the only ones with personalities, cultures and the
ability to design and use tools. Yet we have steadfastly clung to the notion that one attribute, at least, makes us unique: we alone
have the capacity for language.

Alas, it turns out we are not so special in this respect either. Key to the revolutionary reassessment of our talent for
communication is the way we think about language itself. Where once it was seen as a monolith, a discrete and singular entity, today
scientists find it is more productive to think of language as a suite of abilities. Viewed this way, it becomes apparent that the
component parts of language - everything from gesticulation and babbling to meaning and syntax - are not as unique as the whole. In
fact, a boom in research into animal cognition and communication has gradually picked off most items on the list one by one.

Take gesture, arguably the starting point for language. Until recently it was considered uniquely human - but not any more. Mike
Tomasello of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, and others have compiled a list of gestures
observed in monkeys, gibbons, gorillas, chimpanzees, bonobos and orang-utans, which reveals that gesticulation plays a large role in
their communication (Gesture, vol 5, p 39). Ape gestures can involve touch, vocalising or eye movement, and individuals wait until
they have another ape's attention before making visual or auditory gestures. If their gestures go unacknowledged, they will often
repeat them or touch the recipient.

An experiment carried out in 2006 by Erica Cartmill and Richard Byrne from the University of St Andrews in the UK underscores the
similarity between the way humans and apes use gesture (Current Biology, vol 17, p 1345). They got a person to sit on a chair with
some highly desirable food, such as banana, to one side of them, and some bland food, such as celery, to the other. The orang-utans,
who could see the person and the food from their enclosures, gestured at their human partners to encourage them to push the
desirable food their way. If the person feigned incomprehension and offered the bland food, the animals would change their gestures
- just as humans would in a similar situation. If the human seemed to understand while being somewhat confused, giving only half the
preferred food, the apes would repeat and exaggerate their gestures - again in exactly the same way a human would.

Such findings highlight the fact that the gestures of non-human primates are not merely innate reflexes but are learned, flexible
and under voluntary control - all characteristics that are considered prerequisites for human-like communication. The fact that we
can interpret ape gestures also suggests that there is a shared evolutionary basis for gesticulation in humans and other primates.
The innate similarities were demonstrated by Joanna Blake from York University in Toronto, Canada, who examined the literature on
the gestures of human infants aged between 9 and 15 months and that on gestures by apes of various ages. She found that both human
babies and apes use similar gestures to make requests, such as extending a hand to beg for food and raising both arms to be picked
up and carried. Both use their whole hand to point. Infants and apes alike make the same gestures of protest, pushing someone away
or turning away themselves while shaking their heads. They also emote by stamping their feet, flapping their arms and rocking. When
they want someone to do something, both take another individual's hand and place it on the object they want to manipulate.

As well as gesturing, pre-linguistic infants babble, and it turns out they are not alone in this either - dolphins, and even
songbirds, do it too. At about five months babies start to make their first speech sounds, which some researchers believe contain a
random selection of all the phonemes humans can produce. But as children learn the language of their parents, they narrow their
sound repertoire to fit the model to which they are exposed, producing just the sounds of their native language, as well as its
classic intonation patterns. Indeed, they lose their polymath talents so effectively that they are ultimately unable to produce some
sounds - think about the difficulty Japanese speakers have pronouncing the English "l" and "r".

Dolphin calves also pass through a babbling phase. Laurance Doyle from the SETI Institute in Mountain View, California, Brenda
McCowan from the University of California at Davis and their colleagues analysed the complexity of baby dolphin sounds and found it
looked remarkably like that of babbling infants, in that the young dolphins had a much wider repertoire of sound than adults. This
suggests that they practise the sounds of their species, much as human babies do, before they begin to put them together in the way
characteristic of mature dolphins of their species (Journal of Comparative Psychology, vol 116, p 116).

Of course language is more than mere sound - it also has meaning. While the traditional, cartoonish version of animal communication
renders it inchoate, unpredictable and involuntary, it has become clear that various species are able to give meaning to particular
sounds by connecting them with specific ideas. Dolphins use "signature whistles", so called because it appears that they name
themselves. Each develops a unique moniker within the first year of life and uses it whenever it meets another dolphin. Elephants
also use sounds in a word-like way according to Katy Payne, who, before she retired, led the Elephant Listening Project at Cornell
University's Bioacoustics Research Program. Working with Joyce Poole of the Amboseli Elephant Research Project in Kenya, Payne began
to compile a dictionary of sounds produced by individual elephants for various purposes, such as greeting a fellow member of the
clan. Whales have a similarly diverse vocal repertoire. This year, Rebecca Dunlop from the University of Queensland in Australia and
colleagues announced that they had put together a catalogue of 34 different humpback social sounds that remained stable over several
years and were distinct from the whales' song (The Journal of the Acoustical Society of America, vol 112, p 2893).

One of the clearest examples of animals making connections between specific sounds and meanings was demonstrated by Klaus
Zuberbühler and Katie Slocombe of the University of St Andrews in the UK. They noticed that chimps at Edinburgh Zoo appeared to make
rudimentary references to objects by using distinct cries when they came across different kinds of food. Highly valued foods such as
bread would elicit high-pitched grunts, less appealing ones, such as an apple, got low-pitched grunts. Zuberbühler and Slocombe
showed not only that chimps could make distinctions in the way they vocalised about food, but that other chimps understood what they
meant. When played recordings of grunts that were produced for a specific food, the chimps looked in the place where that food was
usually found. They also searched longer if the cry had signalled a prized type of food (Animal Behaviour, vol 72, p 989).

While working in the Budongo Forest in Uganda, Slocombe and Zuberbühler discovered that wild chimpanzees make distinctive noises
during fights which other individuals can interpret. Victims produce screams with a very consistent pitch, while the screams of
aggressors have a variable pitch that falls at the end. Slocombe's recordings reveal that in high-risk situations victims' screams
tend to be long and high-pitched, whereas in low-risk situations they are shorter and lower in pitch. What's more, she found that if
a high-ranking individual was nearby the victim's screams were higher-pitched, suggesting that they were exaggerating the severity
of the threat to get more help (DOI: 10.1073/pnas.0706741104).

What a hoot
Wild chimps also use loud cries known as pant hoots to communicate at a distance, and careful observation has revealed that these
too are neither meaningless nor involuntary. Pant hoots can last between 3 and 23 seconds and have a complex internal structure,
including an introduction phase of low-pitched tones, a build-up phase of panting sounds, a climax that might include long wails or
roars and a let-down phase where pitch and volume gradually decline. They are uttered in specific situations such as resting,
feeding and during travel and display, and are often used to rally support and keep individuals in a group together.

Pant hoots are learned, and differ between individuals and groups (Animal Behaviour, vol 58, p 825). While nobody has demonstrated
that the sounds refer to specific meanings as human words do, there does seem to be a connection between the kind of call and the
situation in which it is made. For example, pant hoots made while travelling or in response to finding a lot of food are more likely
to have a let-down phase than other pant hoots (Animal Behaviour, vol 70, p 177). Given that chimpanzees are so closely related to
us, some ape experts believe that the connection between context and call-type could be an evolutionary precursor to the human
ability to make specific words refer to specific things.

The most hotly contested territory in the language evolution debate is syntax, the grammatical rules we use to combine words in a
meaningful way. It has long been believed that only we are capable of understanding and deploying any of the structural devices
found in human syntax, but in recent years Zuberbühler has shown that wild monkeys use a rudimentary syntax when they communicate.
Campbell's monkeys in the Tai Forest of Ivory Coast have an alarm call that warns other monkeys of crowned hawk eagles and a
different call for leopards (see Primal scream). They also use a combination cry, in which either of the alarm calls is preceded by
a boom sound. This indicates a lesser threat such as the detection of a far-off predator or breaking branches (Animal Behaviour, vol
63, p 293). Zuberbühler likens the boom to qualifiers we use, such as "maybe" or "kind of".

The meaning of pyow
Male putty-nosed monkeys also combine two basic calls to add meaning to a message. Typically, they produce a "pyow" sound in various
situations - most often as an alarm in response to the sighting of a leopard - and they make a "hack" sound when they see an eagle.
Zuberbühler and colleague Kate Arnold discovered that these monkeys also make a "pyow-hack", a combination call that means something
like "let's go" (Animal Behaviour, vol 72, p 6430). Recently they have shown that the various hacks and pyows contain at least three
types of information: the caller's identity, the threat it has witnessed and whether or not it intends to move to avoid the danger.
Other monkeys are able to extract all this meaning from the calls (Current Biology, vol 18, p R189). The findings, says Zuberbühler,
suggest that primates have some naturally occurring syntactic abilities, challenging the widespread belief that the transition from
non-combinatorial to combinatorial communication was an essential step in the evolution of human language.

Other researchers remain sceptical, and recent experiments show how difficult it can be to determine whether or not animals have
some grasp of syntax.

A simple grammar may account for some structures found in human language, for example, a so-called finite-state grammar that follows
the general rule (AB)n, where one syllable A is always followed by another syllable B, n times: as in AB, ABAB or ABABAB. However,
if you want to account for the most complicated human syntax, you need a more expressive construction, called phrase-structure
grammar. This can be captured by the basic rule An Bn, where a given number of A syllables is followed by the same number of B
syllables: as in AB, AABB, AAABBB.

In 2004, Tecumseh Fitch from the University of St Andrew's and Marc Hauser from Harvard University tested the ability of tamarins -
monkeys with whom we last shared a common ancestor 45 million years ago - to understand these two different types of grammar, by
playing them recordings of sequences of sounds. When the recordings mainly followed the (AB)n rule, monkeys would react with
surprise to sequences that violated it, suggesting that they had an expectation about how the sounds would be arranged. However,
they showed no sign that they could detect violations of the An Bn structural rule. Adult humans, in contrast, noticed
irregularities in sequences of sounds whether they represented a finite-state grammar or a phrase-structure grammar (Science, vol
303, p 377).

In 2006, Timothy Gentner from the University of California, San Diego, and colleagues used the same experimental approach with
starlings. Using natural starling sounds and exposing their subjects to many more examples than Fitch and Hauser had done, the team
found that the birds could detect violations of both types of grammar (Nature, vol 440, p 1204). Gentner's paper received a lot of
public attention, and many researchers were surprised by the results. Some welcomed the findings as proof that the syntax underlying
human language is not a monolithic ability - that the differences in our syntactic capabilities and those of other animals are
quantitative rather than qualitative. The experiment was not universally accepted, however. Some scientists questioned whether the
birds were really grasping syntax as opposed to just counting the strings of As and Bs. Others pointed out that the humans in the
original experiment might also have been counting the experimental stimuli - a possibility that Fitch acknowledges.

One reason why Gentner's results have been so hotly contested is that he claimed they indicate that starlings are capable of
understanding recursion - linguistic structures that are embedded inside other structures of the same type, for example, sentences
within sentences. That is a bold claim, given that Noam Chomsky, Hauser and Fitch have argued recursion may not only be the central
process of syntax, but may also be the only component of language that is unique to humans. Nevertheless, whether or not Gentner's
interpretation is correct, other linguists are increasingly questioning that special status for recursion. Daniel Everett from
Illinois State University in Normal has claimed that there is no recursion in the language of an Amazonian people called the Piraha.
Recently the outspoken linguist Derek Bickerton from the University of Hawaii in Honolulu even suggested that recursion does not
exist, it is simply an artefact of analysis.

What nobody disputes, though, is that human language is a spectacular phenomenon. It is hard, for instance, to overestimate the
intricacy and power of all the syntactic strategies that human languages deploy, and there remain many complexities of linguistic
structure that have no apparent analogue in the animal world. Similarly, animals may well show the ability to understand and even
use some words, but no other species can deploy the sheer number of words that humans do.

Speech also remains remarkable. Philip Lieberman from Brown University in Rhode Island argues that the athletic precision with which
we manipulate our mouths and tongues in speech is the singularly human part of the language suite.

Nevertheless, other animals clearly do have greater talents for communication than we realised. In particular, the finding that
primates in the wild use simple structural rules, not to mention the fact that many different kinds of animals can make use of human
words (see Who's a clever boy?), contradicts the idea that creating meaning with sound and structure is unique to humans. Sure, we
are still special, but it is a far more graded, qualified kind of special than it used to be.

Human Evolution - Follow the incredible story in our comprehensive special report.

The Human Brain - With one hundred billion nerve cells, the complexity is mind-boggling. Learn more in our cutting edge special
report.
--
Frederick Martin McNeill
Poway, California, United States of America
mmcneill@fuzzysys.com
w00t *********************************
“Enlightened statesmen will not always
be at the helm.” —James Madison
******************************************