Writing

Introduction

Language is a ubiquitous feature of our lives. As such it may not seem to stand out as a particularly remarkable characteristic of being human, unless one considers what the human animal would be like without the ability to comprehend and use language. In fact, it could be argued that human language is one of the few traits which completely distinguishes human behavior from that of other animals, as so much of our success depends upon our manipulation of complex, language-based, communication. Language involves combining and recombining concepts. You use language in an internal language of thought. It can be used for communication through either speech, signing, or writing. However, in the context of linguistics, writing is considered an externalization which builds upon language, but is not language itself. The capacity for language, rather than the way it is externalized is the issue which has inspired the deepest curiosity. Although over the past seventy years many scientists have studied other animal’s communication systems in the search for insight into the origins of human language, it is cognition which now seems to hold the key to understanding its evolution.

What is the Definition of Language?

When people developed an interest in how human language evolved, they began to study animal communication systems. But to compare animal communication systems to human language they first had to have a clear understanding of the features that defined language. Although people have been analyzing language and describing its grammar for centuries, the modern field of linguistics developed only in the last century. Researchers looked to this emerging field, and to the work of Noam Chomsky in particular, for help in defining the essentials of human communication (Jackendoff & Pinker, 2005).

Human language is comprised of certain core elements which are present whether you’re using speech or sign language; phonology (sound/spatial-temporal quality), semantics (symbols with meaning, like words), grammar (the particular rules of a given language), and syntax (a subset of grammar; more general rules referring to sentence structure) (Suzuki, et al. 2019; Jackendoff & Pinker, 2005). When considering syntax, Suzuki et al. suggest three criteria for compositional syntax: 1. That the meaning of individual signals, and combinations thereof reflect a context; 2. That the meaning of a combination of words is only understood because the component parts are meaningful; 3. And further, that the precise order of the signals (i.e. words in a sentence) can determine the meaning (2019). With this kind of analysis, researchers could determine if animal communication systems had these important characteristics of human language.

What is Unique About Human Language?

Some researchers came to the opinion that there is another essential aspect of human language which is not only characteristic but unique. Bolhuis et al. (2014) locate this uniqueness in a cognitive ability to combine mental representations, then recursively re-combine these to create mental representations of almost infinitely complex ideas—they call this ability “merge.” They claim that merging concepts recursively is the core of human grammar, and that this is innate and uniquely human. The merge concept suggests that these combinations are pairwise: taking n elements, placing them into pairs, and recombining those in nearly infinite compositions.

A competing if similar theory is termed “Unification” (Jackendoff & Pinker, 2005). Based partly on considerations of the nature of idioms, Jackendoff and Pinker propose that a binary rearrangement of similar elements is inadequate to explain the richness of human language. Bolhuis et al. present no explanation for idioms where Jackendoff and Pinker insist that no explanation of human language can ignore sentences which are consistent with proper grammatical usage, but have an actual meaning which is different from their literal meaning (e.g. ‘Bite the bullet’). The merge theory assumes that all sentences are comprised of smaller elements for which we have mental representations. The Unification theory proposes that mental representations are much more flexible, even for an entire sentence, which would be the case for an idiom like ‘Stick ‘em up.’ Humans can think non-linguistically, but a large portion of our stream of consciousness and mental life is in the form of language.

The similarity in both the Unification and merge theories is that we compose sentences by recombining elements. It is these processing characteristics which researchers generally agree most fully typify the core of human language. Therefore these features are what were looked for as a basis for comparison in animal communication systems as well as in the cognitive capabilities of animals.

Animal Communication as an Insight into the Evolution of Human Language

Clearly animals communicate too. When researchers first began investigating the question of the evolution of human language they looked mainly to our closest genetic relatives, the great apes, and to birdsong, the most complex observed example of animal communication. Everyone who has heard birdsong can recognize that it consists of different notes, combined in different orders, and one might imagine that it is used for communication rather than expending energy for no purpose. Writing of chick-a-dee birdsong, Hailman observed that “the staggering variety of call-types created from combinations of note-types and their repetitions is not likely to be haphazard variation” (1985, p. 1). As the title The “chick-a-dee” calls of Parus atricapillus: A recombinant system of animal communication compared with written English suggests, in 1985 there were very high hopes for finding close analogies between birdsong and human language.

However, the study of birdsong as well as other animal communication systems continues. Suzuki et al. give a brief overview of the current state of the field (2019). They describe observations which suggest that various species may combine predator warning calls with contact calls. The significant point is that different calls with consistently observed responses (i.e. meanings) may when combined with other calls provoke entirely different responses. Nevertheless, the evidence is not overpowering. For instance, paired meanings in putty-nosed monkeys are given as a possible example of idiomatic usage (Suzuki et al., 2019, p. 5). However, this may simply be a recombination of two sounds, each with a meaning, where the combined sound has a third meaning such as in the German word Kühlschrank—kühl meaning cool, and Schrank meaning cupboard, which combined means refrigerator.

A different research path was to try to teach language to our closest genetic relatives (chimps, gorillas, and bonobos). They were taught sign languages or language through a symbolic interface, as it was clear that they could not mimic sounds. While people enthused over early results, which exhibited how many concepts apes could learn to sign, combinations were another matter. In fact, one of the more impressive outcomes was announced in 1977, when a chimpanzee named Washoe saw a swan in the park and signed ‘Water + Bird’ (Suzuki et al., 2019). However, given that there were never repeated examples of this kind of spontaneous generation of a combination of words/signs to indicate a concept it is not clear that Washoe’s communication even rose to the level of the putty-nosed monkey’s Kühlschrank.

Cognition as an Insight into the Evolution of Human Language

The research program in the latter half of the previous century attempting to teach great apes human language highlighted the simplicity of great ape signaling behavior, but this stands in great contrast to ongoing research showing the sophistication of great ape cognitive capacities. Tecumseh Fitch points out that “animal signals do not equal animal concepts” (2019, p. 4). In fact, there is now clear evidence that many species have considerable cognitive sophistication. Dolphins for example can demonstrate in their behavioral responses that they are able to interpret complex sentences with grammatical order, as well as concepts like ‘same’ and ‘different’ (Tecumseh Fitch, 2019). Yet they have never been successfully trained to communicate in a way which could be viewed as either resembling or even roughly translating to the system of grammar, and semantics that underlies human language.

Many animals display evidence of highly complex mental representations, intricate concepts, planning for the future, social relationships, and mental maps of their environment (Tecumseh Fitch, 2019). Tecumseh Fitch (2019) describes evidence that some animals even have a theory of mind, which does not begin to develop in humans until about age 3-5. This could be deduced from the fact that they will hide something desirable only once their competitor is not looking, or that they can recognize themselves as an individual in a mirror.

There is always skepticism about recognizing whether animals have these cognitive capacities, because they cannot directly tell us. However, more and more researchers are conducting subtle experiments that can demonstrate that the animal subject is not simply being rewarded for giving the ‘right’ answer without truly having a cognitive representation of the concept under examination. For example, to demonstrate the capacity for the perception of sequencing inherent to syntactic structure, researchers designed an experiment to show that it was not simply incidental rewards that lead to the conclusion that monkeys could make cognitive inferences like, if a is to the left of b, and b is to the left of c, then it follows that a is to the left of c (Jensen et al. 2019). Illustrative of trends in the field, Jensen’s senior coauthor Herbert Terrace famously lead a project in the 1970s attempting to teach language to a chimpanzee subject whom they playfully named, Nim Chimpsky (Sukuki et al., 2019). Clearly in the intervening decades Terrace shifted his focus from animal communication to animal cognition as a central means of insight to the evolution of human language.

Conclusion

The most unique attribute of human language, the process of recombining mental representations, whether best described as merge or Unification, is a capacity which has yet to be conclusively demonstrated in animals. Tecumseh Fitch (2019) emphasizes how important it would be to discover whether the merge process is present in animal cognition, saying that it is crucial to “…explore in detail animals’ abilities to combine concepts. To the extent that they can do so in a flexible, hierarchical manner, I think we can see the germs of the recursive symbolic system that underlies human linguistic concepts” (p. 6). Semantics (i.e. meaningful concepts) are clearly present in animals. The question of whether these concepts can be recombined in a way which resembles recursive syntax is less clear. If the recursive ability which underlies syntax could be found in the cognitive systems of animals, then animals may have something closer to a language of thought than we are currently able to demonstrate. Given the direction that research has previously gone and its current projections, it seems unlikely that the study of animal communication systems alone will provide the breakthrough insights into the origins of human language. The study of animal cognition currently looks more promising.

References