Animals and language learning

Animal languages are forms of non-human animal communication that show similarities to
human language. Animals communicate by using a variety of signs such as sounds or movements.
Such signing may be considered complex enough to be called a form of language if the inventory of
signs is large, the signs are relatively arbitrary, and the animals seem to produce them with a degree
of volition (as opposed to relatively automatic conditioned behaviors or unconditioned instincts,
usually including facial expressions).

A. Teaching spoken English to apes

1. The first scientific attempt: with an orangutan
The earliest-known scientific attempt at teaching language to an ape was that of
Furness (1916) in the USA, who attempted to teach an orangutan to speak. The brief four-
month project ended in tragedy, when the animal died with a high fever while repeating
the two words it had learned to say ‘papa’ and ‘cup’. Later more comprehensive scientific
attempts to teach speech to animals were made with chimpanzees.

2. Gua: the chimp raised with a human ‘sibling’

Winthrop and Luella Kellogg (Kellogg and Kellogg, 1933; Kellogg, 1968) raised a
female chimp named Gua along with their own son, Donald. Their idea was that by giving
the chimpanzee the same input and social interaction as a human child, the chimpanzee
would learn language in the same way that the human child learns its first language.
Gua, a female, and Donald, the boy, were born less than three months apart, with
Gua being the younger. For nine months they were raised in the home as siblings. Initially
on problem-solving tests and tests of mental ability the two scored the same, but over time
the boy surpassed the chimp. The Kelloggs noted that Gua demonstrated an exceptional
ability to learn by imitation but that the boy was more versatile and continuous in his
learning. Gua was raised in an ordinary speech environment with no specific language
training being given. The Kelloggs wanted to see if Gua could learn language.

3. Viki: another chimp raised in a human household

As far as speech understanding is concerned, Cathy Hayes (1951, p. 227) reports,
‘Are there any words which Viki comprehends without need of supplementary cues? There
are a few . . . She obeys the commands: “Go to your room”, “Go outside” and “Go upstairs”
without error.’ This result is surprising compared to the much longer list of items that Gua
could understand. Yet the Hayes did seem to be aware of the need to neutralize the effect
of environmental cues when testing for speech comprehension. Even so, because chimps
have a great ability to mimic and to read facial expressions and body language (Fouts and
Mills, 1997; Savage-Rumbaugh et al., 1998), special measures have to be taken to control
such variables. Perhaps it is an oversight on the part of the Hayes, but they did not include
the four words that Viki could say, ‘mama’, ‘papa’, ‘up’, and ‘cup’ (the first author viewed a
videotape of their home movie). Perhaps Viki could say but did not know the meaning of
these words, but this does not seem likely.
B. Teaching sign language to the chimpanzee, gorilla and orang-utan
1. Washoe: the first signing chimp
In 1966, another husband-and-wife team of psychologists, Allen and Beatrice
Gardner (1969, 1975), began to teach sign language to a baby chimp, a female they
called Washoe (rhymes with ‘show’). They reasoned that any attempt to teach chimps
to speak was doomed to failure because of the simple fact that chimps do not possess
the necessary vocal apparatus for human speech. Viki’s failure to learn to speak could
plausibly be said to be a simple physiological failure and not a mental one. Since chimps
are very adept at using their hands, the Gardners conceived of the idea of teaching
them a simplified form of American Sign Language (ASL).
The two- or three-word length of utterance seems to be similar to that
produced by human children around the age of 1 to 2 years. The Gardners,
unfortunately, did not focus on comprehension. Many of the gestures that the Gardners
recorded as signs seem to have already been part of the chimpanzees’ natural gesture
system, such as shaking the hand at the wrist for ‘hurry’. These gestures were among
the most frequent of the signs used by Washoe and were often included in her two- and
three-word utterances. Concerning such word combinations, it is not clear whether they
were actual reliable syntactic combinations rather than single signs formed in close
proximity to each other.
2. Loulis, son of Washoe, and a community of signing chimps
The Fouts regard their chimp subjects fondly and have established a primate
reserve where chimps are taught some of the rudiments of human culture, such as
using tools, for example, with the aim that they start their own community.
However, Washoe’s signs seem not to have been picked up by chimps other
than Loulis, who unfortunately died when still very young. As yet, this line of research
has not provided strong evidence for the creation and learning of signs from
chimpanzee to chimpanzee.
3. Nim Chimpsky and the Chimpskyan revolution
a psychologist who was an early enthusiast of chimps being able to learn sign
language, worked with a chimp that he named Nim Chimpsky. By giving the chimp this
name, Terrace evidently set out to make a monkey out of Noam Chomsky by proving
that animals can learn language.
Like the Gardners, Terrace used a modified form of American Sign Language for
teaching language to Nim. Examples of Nim’s two-, three-, and four-sign sequences are
‘more drink’, ‘tickle Nim’, ‘banana Nim eat’, ‘banana eat Nim’, ‘eat drink eat drink’, and
‘banana me eat banana’.
When Nim made longer utterances, Terrace says that all he was doing was
mainly imitating what the teacher signed and adding words almost at random until he
got what he wanted. (Terrace’s examples in the previous paragraph seem to bear this
out.) Terrace thus came to the conclusion that chimpanzees were capable of learning
only a few of the most elementary aspects of language.
s. The confinement of Nim within a small room for the language learning
sessions was meant to remove the effects of extraneous variables, variables which were
not controlled for in other studies. Such an approach, the Fouts say, fails to get at the
spontaneity of a good relationship between a researcher and an animal. According to
 Roger Fouts (1983b), ‘We talk to people we like – and people we like don’t ask us the
same dumb questions 50 times in a row. We converse about things.’
However, here we may note that even in the cases of human children who are
badly treated, but, spoken to, by their parents, such children, if they are not born with
physical and mental defects, generally learn language.
4. Teaching sign language to Koko, the gorilla
Francine Patterson (1978a, 1978b, 1980) reports remarkable results with the
gorilla Koko, born in 1971, whom she has trained in American Sign Language and speech
since 1972. Unlike Washoe, Koko received speech input from her trainers as well as
sign. Teaching has consisted of ten hours per day of sign and speech exposure, with
help in moulding the hands for sign making.
One interesting fact Patterson discovered was that Koko was productive in her
sign language. Koko could make new words to describe new objects by combining
previously known ones. Koko, for example, created ‘eye-hat’ for mask, ‘white-tiger’ for
a toy zebra, ‘finger-bracelet’ for ring, and ‘bottlematch’ for a lighter. Human languages,
of course, include similar devices for word-making, e.g. ‘blackbird’ in English and ‘white
bird’ for swan in Japanese.
Koko is friendly and apparently tries to start up sign-language conversations
with strangers. Sometimes she signs to herself when she is alone. For example, on a
videotape of Koko that we have seen, Koko spontaneously made the sign for smoking
while browsing through a magazine and coming across an advertisement for cigarettes
On the other hand, we are sure Terrace would point out that while the extent
of Koko’s vocabulary is substantial, her syntax, like that of the chimpanzees, is quite
rudimentary. Further testing, however, could show that Koko’s syntax for
comprehension might be more advanced.
5. Teaching sign language to Chantek, the orang-utan
Miles’s project differed from others that have focused on language; she was
more concerned with the cognitive and communicative processes that might underlie
language development. Chantek was not just trained to use signs but was immersed in a
human cultural environment to learn the appropriate ways for behaviour and
interaction. No attempt, though, was made to raise Chantek like a human child. Rather,
his natural arboreal habitat was maintained to the extent possible at the research
centre.
An interesting observation that she noted concerned Chantek’s use of the verb
‘give’ (and only this verb): Chantek was more likely to name a physical object first
(‘Object + give’) if the object was present.

C. Teaching artificial languages to chimpanzees

1. Lana: the computer chimp
The Rumbaughs (Rumbaugh, 1977; Savage-Rumbaugh and Rumbaugh, 1978)
(another husband and wife team!) taught the chimp Lana a simple artificial language
called Yerkish (after the Yerkes’ Primate Center). Lana was named after the research
programme, which was called the LANguage Analogue project. Lana was just over 2
years old when the project began.
 2. Sarah: the magnetic plastic token chimp
An ingenious piece of research was one conceived by David Premack of the
University of Pennsylvania involving a chimp by the name of Sarah.
Rather than using sign language or electronic keyboards, Premack (1970, 1971,
1976) gave Sarah 130 plastic tokens with magnets so that they could be manipulated
easily by her and others. These included tokens for the names of colours such as ‘red’
and ‘blue’, for different fruits such as ‘banana’ and ‘peach’, and for actions such as
‘wash’, ‘cut’, and ‘take’, and some functions such as ‘QUESTION’. A typical question that
might be posed to Sarah was ‘QUESTION banana red’ (Is the banana red?), to which she
would correctly answer ‘no’ by means of a token. The tokens would adhere to a metallic
board.
Premack’s research with Sarah makes it very clear that chimps are intelligent
creatures. abstract word ‘colour’, which is, in itself, a notable achievement.) Other apes
in Premack’s research were also able to distinguish between strings of words differing
only in word order, such as ‘red on green’ and ‘green on red’.
3. Kanzi: a pygmy chimp produces synthesized speech
A pygmy chimpanzee, or bonobo, is said by the principal researcher, Sue Savage-
Rumbaugh (of Lana fame – see above), to be more similar to humans than to other apes
in terms of evolution and with respect to communicative behaviours such as eye
contact, gestures, and vocalizations AIT_C05.pm5 112 17/10/2005, 16:57 113 5 •
Animals and language learning (Savage-Rumbaugh et al., 1986). Such being the case, she
and others believed the bonobo to be a better candidate for language research than the
other apes that researchers had used. Thus it was that the bonobo male chimp, Kanzi,
and his younger sister, Mulika, were selected for study.
The lexigrams (visual word symbols) on the keyboard were made up of arbitrary
geometrical symbols, each matching an object, action, or location. When touched, the
lexigrams on the computer keyboard would produce synthesized English speech sounds
for a particular word. Gestures and some sign-language signs were not taught but
allowed to develop spontaneously. No attempt was made to teach language.
The researchers state that ‘Kanzi showed an incipient ability to use difference in
symbol order to signal difference in meaning’ (Greenfield and SavageRumbaugh, 1990,
p. 567). The word ‘incipient’ properly weakens the claim because symbol order may not
have been acquired; strict scientific controls, particularly regarding knowledge of the
world, were not employed.

D. Teaching language to dolphins.

1. Elvar: the whistling dolphin.
There is much anecdotal lore about the intelligence of dolphins and whales. Until
the 1960s, though, no scientific attempt had been made to determine their
communicative abilities. In one of the first studies, Lilly (1962, 1965) tried to teach a
dolphin to force air through its blow-hole in such a way that it would allow the dolphin
to imitate human speech sounds. A young male dolphin named Elvar produced
approximations of the word ‘squirt’, which Lilly had been trying to teach him to
pronounce.
 Notwithstanding Lilly’s extravagant claims to the contrary (including claims that
dolphins have an intelligence and a religion(!) that is superior to those of humans),
research has yet to show that these animals use anything as complex as what we could
call language.
2. Akeakamai and Phoenix: learning artificial languages through sight and sound.
A radically different and more scientific approach to the teaching of language to
dolphins was later initiated by Louis Herman at the Dolphin Institute, University of
Hawaii. In an early study, rather than have dolphins mimic human sounds, Herman and
his associates (Herman and Wolz, 1984) trained a bottlenosed dolphin to mimic
computer-generated sounds.
In 1979, a teaching programme was begun with two dolphins, Phoenix and
Akeakamai (the latter’s name meaning ‘lover of wisdom’ in the Hawaiian language). Each
dolphin learned one of the two artificial languages. Akea was taught the gesture-based
language, while Phoenix was taught the soundbased language. Each was taught a
vocabulary of about 30 words, mainly names of objects, agents, actions, and modifiers.
Of special interest are Herman’s results, which show that generally the dolphins
correctly responded to what are often called ‘semantically reversible sentences’, i.e.
sentences for which the subjects and objects cannot be interpreted by meaning alone
but where the use of syntactic knowledge is required. For example, the English
sentences ‘Jack pushed Tom’ and ‘Tom pushed Jack’ describe two different events, one
in which Jack is doing the pushing and another in which Tom is doing the pushing. Given
our lack of knowledge about Jack and Tom, we can only judge that either event is
equally likely to occur.
The dolphins, Herman emphasizes, can also respond to novel sentences on the basis
of understanding words and their relations in a command structure. Once the structure
and relations are learned, then all new sentences with those characteristics should be
understood, providing, of course, that the meaning of component words is already
known. Thus, after acquiring the notions of direct and indirect object, Akea responded
correctly on her first exposure to the sentence ‘person left frisbee fetch’ (Take the left
frisbee to the person).
Herman’s research is one of the most scientific and methodologically reliable on the
learning of language by animals. It remains for further research to demonstrate whether
dolphins would be able to express in production what they have already learned in
terms of language comprehension. Devising such an appropriate means of production,
however, is not easy. Perhaps Herman should return to his earlier work where he
trained a dolphin to mimic computer-generated sounds.

E. Teaching spoken English to an African Grey parrot

Pepperberg (Pepperberg, 1987, 1993; Pepperberg and Kozak, 1986) has worked with
a male African Grey parrot she calls Alex. She used the speech mode because of the parrot’s
excellent vocal and hearing abilities. Alex is now able to understand and answer questions
on the colour, shape, and material of more than 100 objects. He can correctly name a host
of items such as key, chain, tray, toy truck, block, cup, and box. Furthermore, he can identify
them on the basis of seven colours (green, red, blue, yellow, grey, purple, and orange;
although grey and green are sometimes indistinguishable because of Alex’s pronunciation of
 green as ‘gree’), and of a number of shapes, up to those with six corners, including triangles,
squares, pentagons, and hexagons.
Alex is not only adept at identifying items but he can request them, refuse them,
and answer questions pertaining to the abstract categories of shape, colour, material, and
quantity. Alex’s accomplishments are admirable and in some ways, especially with his
recognition of abstract categories, he has surpassed aspects of language knowledge that the
apes and even the dolphins have demonstrated. He has not yet, though, reached the level of
syntax that the dolphins have mastered.
Why Pepperberg has been successful with her parrot while others over the centuries
have not is a puzzling question. It has been over 15 years since Pepperberg’s first
investigation was published (Pepperberg, 1987), yet no other parrot study has appeared to
our knowledge.
F. Teaching Rico the dog to understand spoken English words
The recent appearance of a language study with a dog has led us to wonder why
household pets have not been brought to the university for language instruction more often.
Rico is a 91 /2-year-old border collie who has learned to understand more than 200 words
for different objects, such as toys and balls, that have been said to him (Kaminski et al.,
2004). He can learn a new word after being shown an unfamiliar object just once (an
achievement that is the envy of these authors). No syntax, but still, Rico’s large vocabulary is
greater than a lot of the animals who have been given language instruction. It is not likely,
however, that any syntax would ever be learned. After all, household pets have been living
together with humans for many thousands of years under the close scrutiny of their owners
and nothing has ever been reported of dogs or cats learning more than single words, mainly
commands for action.