“Monkeys
… very sensibly refrain from speech lest they should be set to earn their
livings.”
-Kenneth
Grahame
The Golden Age: Lusisti Satis (1895)
I am a primate. I share ninety-eight percent of my genetic information with
the other species of primates, such as chimpanzees, orangutans,
and bonobos. Yet, I feel more than just two percent different than
these other animals. For one, I have the capacity for language. I have never
heard of a nonhuman primate that can write a novel, tell a story, sing a song,
or read a newspaper. But just because these other animals fall short of
possessing my language expertise, does that mean that other primates do not have
any capacity for language?
Human
philosophers and linguists throughout the ages, and many still today, do not
believe animals other than Homo sapiens have the capacity for language.
In fact, many believe language to be the distinguishing feature that sets humans
apart from the beasts. René Descartes wrote in the 17th
century that animals are mechanical “automata” and do not possess
“feelings and reason.”] Three hundred years later, 20th century
linguist Noam Chomsky continued the tradition of language’s human-specificity.
Chomsky believes the quest to quantify language in apes is in vain; he asks why
apes do not use language if they have the ability. (Hart, 1) But perhaps the
apes are using their own language. Julien Offray de la Mettrie’s challenge
posed to Descartes’ view in 1748 still applies to Chomsky’s camp, over two
hundred years later. De la Mettrie cautions that
“proof of man’s uniqueness and rationality could be undermined by teaching
an ape to speak.” (Rumbaugh, 12) Obviously, apes do not have the full
capacity for human language, just as humans do not have the capacity to
communicate fluently with apes. And here we stumble upon the primary conundrum
of primate language research: What is language?
A
specific definition of language is necessary in order to test a primate’s
ability of language capability. Language is the ability to communicate with
members of an animal’s own species using symbolic output, as well as the
ability to perform somewhere along the continuum of language. Considering that
language is a function of an animal’s consciousness, researchers must tackle
the question of whether primates are conscious. This task proves difficult,
although necessary. “Consciousness cannot be understood unless accurately
described.” (Savage-Rumbaugh, 910) Sue Savage-Rumbaugh, a renowned
researcher of primate language, believes that consciousness “is general among
animal species.” (Savage-Rumbaugh, 911) Additionally, Rumbaugh stresses that
we can come to a more clear understanding of a primate’s ability for language
by shaping an individual ape’s worldview using human culture. This
way, we can rate the performance of an ape based on criteria familiar to all
humans. “The culture in which an ape is raised will significantly affect the
form of consciousness it develops. If reared in a human culture, ape
consciousness will be molded according to a form that human beings can
recognize.” (Savage-Rumbaugh, 911) Regardless of how well an ape lives up to
human linguistic and cultural standards is irrelevant if we assume that primates
perform perfectly when communicating with each other. Still, in order for
researchers to develop a clear understanding of an ape’s linguistic
performance, she must test the ape in accordance to rules she can understand as
a human, and as an investigator. Resulting from this trans-species outlook of
language capacity is a theoretical “continuum” of language understanding and
capability. (Savage-Rumbaugh, 920) Along this continuum reside the capacity to
understanding a language and the capacity to produce a language. Considering
human language employed on a nonhuman primate, apes might perform better at the
former. “Humans first learn to understand a language, then later learn to
produce it. It follows, they claim, that animals might learn to understand
language better than they learn to produce it.” (Snowdown, 225) The continuum
of language ends at human capability and degrades as a gradient. If an
animal’s abilities reside somewhere along this line, then that animal is said
to have some capacity for language as we know
it.
For
years, the preferred method for testing a primate’s ability to use symbolic
language was to submit apes to experiments in which they interacted with humans.
In 1971, Duane Rumbaugh began the “Lana Project” in which researchers
attempted to teach a chimpanzee named Lana to use the American Sign Language.
Unfortunately, “her signs were often inarticulate and difficult to
decipher.” (Savage-Rumbaugh, 911) Others criticized the work “for its lack
of control, for the iconicity of the signs not being truly symbolic, and for
displaying little true evidence of grammar.” (Snowdown, 220) For example, the
controversial symbol for “flower” on the keyboard somewhat resembled a
flower. Most of the symbols, however, were entirely abstract. Lana only learned
about twenty-five different signs, and exhibited no mastery of the symbols.
“It was also not clear why she sometimes made what seemed to be
incomprehensible errors and formed nonsensical strings.” (Savage-Rumbaugh,
912) Clearly, Lana could not master the higher level meaning the sign language
represents. “Lana’s use of symbols was somehow not humanlike.” (Savage-Rumbaugh,
912) In attempt to salvage the research, Rumbaugh allowed Lana more leeway by
allowing her to try various sentence combinations while attempting to
communicate a desire. Yet the results show that Lana was not learning from her
errors, and “it could take five to fifteen attempts” to arrive at the
correct answer. This “absence of systematic elimination” shows a lack of
understanding human syntax. D. Rumbaugh went on to develop a “lexical
keyboard” containing various pictograph symbols representing real world
objects. “The lexical keyboard system … provided a potential means of
propelling apes beyond the limitations posed by these other methodologies,”
such as sign language. (Savage-Rumbaugh, 911) Although chimpanzees mastered the
keyboard easier than the sign language system, research using the keyboard
cultivated little evidence for an ape’s understanding of language. In all of
these experiments, Lana communicated with a human being, not a member of her own
species. Attempts to gauge the language capacity of a chimpanzee using
human-primate interaction failed. Researchers have proposed a reason for the
disaster.
One
objection to the research involving primate interaction with humans is the
potential for reward expectation on the part of the primate. Were Lana and the
other chimpanzees involved in the primate-human experiments simply responding to
Pavlovian stimuli? Perhaps these experiments clouded any evidence
for actual language capacity by setting up a positive-feedback
loop. Researchers questioned whether the primates were simply using symbols and
sign language in order to achieve the goal of getting a food reward or if they
grasped the concept of symbolic language in the same manner as a human’s
understanding. Michael Beran confirmed these suspicions in 2001 with a study
designed to test this exact hypothesis. In fact, Beran actually used Lana, the
chimpanzee from D. Rumbaugh’s original study. He found that the chimpanzees
did expect a reward. The chimpanzees would become agitated if the human
experimenters withheld the reward after a correct response. “What is clear is
that both Lana and Mercury (another chimpanzee) had a clear expectation of
receiving a food reward when a trial was completed correctly that they did not
have when a trial was incorrect.” (Beran, 181) These primates expected a
reward, yet all of these experiments were based on primate-human interaction.
Could a change in research philosophy solve the problem of reward expectation?
In
accordance with the above definition of language, a paradigm shift in research
could allow for a greater understanding of primate language. In fact, this
revolution would come with the development of primate-primate research, as
opposed to the awkwardness of primate-human experiments. In 1983, D.
Rumbaugh’s wife, Sue Savage-Rumbaugh, developed what Charles Snowdown of the
University of Wisconsin Madison calls a “major breakthrough in the study of
language analogues.” (Snowdown, 223) Savage-Rumbaugh endeavored on one such
study in 1979. She and other researchers analyzed communication between two
chimpanzees named Sherman and Austin. Unlike previous attempts at measuring
language competency in chimpanzees, this study did not involve any primate-human
interaction during the course of the experiments. “It meant for the first time
in the field of animal language, the experimenter was removed as half of every
subject-experimenter interaction.” (Savage-Rumbaugh, 913) By eliminating the
role of the human, Savage-Rumbaugh undermined the challenges involved with reward
procurement on the part of the primate. No longer could critics challenge the
experiment’s results on charges of reward expectation. “The issue is no
longer one of data, of potential cueing or experimenter effects, or of
conditioning.” (Savage-Rumbaugh, 919) In these experiments with Sherman and
Austin, experimenters created situations that required the two chimpanzees to
communicate to each other using a symbolic language. A keyboard provided the
means for communication. The keyboard featured 256 unique symbols. Each symbol
represented a common object or concept. One example of an experiment involved a
circumstance where one of the chimpanzees knew where food was hidden and the
other did not. In order to receive the food, one chimpanzee would have to
communicate the location of the food to the other so that the other chimpanzee
could locate the food. The chimpanzees nearly always completed the task
successfully. After analyzing the data obtained by the experiments, researchers
concluded, “symbolic communication of a high level, with the use of an
abstract code and with mutual understanding and cooperation, was possible
between nonhuman creatures.” Without the primate-primate setup of the
experiment, researchers would not have been able to conclude any significant
facts about primate communication because human interaction would have tainted
the results by creating unrealistic situations. “Linguistic communication
necessarily takes place between individuals in a multiplicity of exchanged that
cannot be controlled from the outside either by intentionally setting the stage
of the preceding stimulus or effecting a particular event.” (Savage-Rumbaugh)
Primates
not only successfully use symbolic visual language, such as with the
aforementioned keyboard, but also vocally. Although nonhuman primates lack the
proper vocal chords to make the broad range of human sounds, research has shown
bonobos use audible messages to communicate with members of their own species.
In 2000, investigators led by Jared Taglialatela painstakingly analyzed over
four hundred hours of bonobo vocalization. They found that the sounds bonobos
use in nature alter substantially during unique situations. Statistical analysis
verified the linguistic implications of the sounds, or phrases. “Structural
differences do exist between those vocalizations produced in different
communicative contexts. These results suggest that the vocalizations possess
lexical significance.” (Taglialatela, 95) Taglialatela used computer
technology to decipher the hidden meaning in what sounds like indiscriminate
shrieks to the human ear. What sounds like a casual grunt to a human primate
actually means “banana” to a bonobo primate. “A vocalization produced in
conjunction with a gesture to a banana was assigned to the communicative
context, ‘banana.’” (Taglialatela, 95) Primates, human and nonhuman alike,
have the capacity to understand symbolic visual language as well as their own
vocal language.
In
addition to communicating among themselves with their natural vocal
communication system, researchers have evidence of primates teaching sign
language to their peers of the same species. In 1974, researchers at the Institute of Primate Studies in Norman, Oklahoma, witnessed a sign language
competent chimpanzee teach another chimpanzee signs. “Manny, a young
chimpanzee in the colony, has acquired from Washoe the ‘come hug’ sign,
which is used correctly when the chimpanzees greet one another or are engaged in
mutual comforting.” (Sebeok, 279) This drifting of contagious symbolism
between human primates and chimpanzees is the most stunning verification of
language in primates. Humans using signs to communicate with each other is
language. And when the technique is transmitted to an ape, it remains
language.
Whether
the reason for the human’s adamant rejection of primate language is due to taxonomic arrogance or
cross-species misunderstanding, scientific research shows
that primates utilize the power of language for the same reasons as humans: for
survival, culture, and play.
Work Cited
Hart, Stephen. “Apes may hold clues to language’s origins.” BioScience, June 1998 v48 i6 p437(2)
Rumbaugh, D.M. Language Learning by a Chimpanzee. New York: Academic Press. 1997.
Savage-Rumbah, S., Fields, W. M., Taglialatela, J.P. (2000). Ape consciousness-human consciousness: A perspective informed by language and
culture. American Zoologist, 40 (6): 910-921.
Sebeok, T.A., Umiker-Sebeok, J.U. Speaking of Apes. New York: Plenum Press. 1980.
Snowdown, C.T. (1990). “Language capacity of
Nonhuman Animas.” Yearbook of Physical Anthropoloby 33: 215-243
“Vocalization production and
usage in language-competent, captive bonobos (Pan paniscus).” American Journal
of Primatology, 51 (Supplement 1): 95 Abstract.