Characteristics of Human Language

A Comparison to Animal Communication Systems

Is human language anything like what animals do? If so, how is it similar? More importantly, how is it different? How could the differences have arisen?

Continuous vs Discontinuous Evolution

One reason to compare human and animal communication systems is that similarities and differences may reveal whether language evolved continuously or discontinuously; i.e. whether our ancestor species' had language systems that were increasingly sophisticated, evolving slowly over a period of time, or whether language appeared suddenly, fully formed.

There are different systems that are involved with language:

Auditory perception

The systems involved with the perception of language sounds has continuously evolved. In fact, some other animals (rhesus monkeys, chinchillas) are generally capable of the same categorical perception of sound as humans (i.e. the tendency to categorise sounds into one category or another, even in borderlinne cases)

Brain systems

Continuity seems to be moderate. There are some centers in the human brain which have abilities that are not found in other primates.

Articulatory production

The greatest apparent discontinuity with other primates is the vocal tract. Although there are some continuities (chimp 'pant-hoots' are related to syllables), humans possess a lowered larynx, which allows more vowel sounds (and also choking on food).

Chomsky has claimed (1988) that a half-evolved language would not be adaptive, so it must have been discontinuous.

Others (Pinker & Bloom 1990) claim that language, being highly complex, must have evolved by gradually, using exaptations of features orignally used for some other purpose.

Characteristics of Human language

Hockett, a linguist and anthropologist identified thirteen characteristics of human language.

These characteristics can be used as the basis for comparison with other animal communication systems:

Bee dancing

Bees returning to the hive perform a 'dance' followed by other bees, which indicates the location of a food-source they're located:

• Angle of figure-of-eight pattern = direction with respect to the sun
• Rapidity of repetitions = distance
• Vigour of waggle = quality/quantity of food source

Bird song

Many bird species sing for various social functions:

• Mating display
• Contact with neighbors - territorial control
• ...

Whale song

The song of some species of whale have been found to identify individuals and can take years to 'learn'

Primate calls in the wild

Vervet monkeys, for example, make calls in wild which seem to 'refer' to various predators

• Eagle
• Snake
• Leopard

Apes in Language experiments

Chimpanzees, Bonobos, etc. raised in captivity have been involved in experiments where human experimentors have attempted to teach them human-like languages

• American Sign Language
• Languages involving gesture and symbolic plastic tokens

The idea is that the characteristics of language which are shared with animal communication systems probably need less explaining than those which no animal communication system has.

Characteristic		Bees	Birds	Whales	Wild Primates	Apes in Expts	Human
Vocal Auditory Channel
Broadcast Transmission / Directional reception
Transitoriness	(rapid fade)
Interchangability	(both emit or receive)
Total feedback	(you hear what you say)
Specialisation	(specially for communication, not merely a symptom of an inner state, like panting)
Semanticity	(referential rather than affective)
Arbitrariness	(symbol, not index or icon)
Discreteness	(vs. Continuity)
Displacement	(ebility to refer to absent objects)
Productivity	(ability to say new things)
Traditional transmission	(learn it from parents)
Duality of Patterning	(meaningless components combined into meaningful components)

Hockett & Ascher claim (in The Human Revolution) that the essential aspects in which human language differs from animal communications systems are:

• productivity (primate call systems are closed)
• displacement (primates only make a food call when food is present)
• duality of pattering (the difference between two primate calls is global)
• traditional transmission (primate calls are more or less innate)

What about specialisation?

They go on to tell a story about how these features maay have developed:

1. early hominids were forced out of their forest environment as the forest became savanna
2. carrying of tools and food developed foresight, and freed the mouth for 'chattering'
3. development of hunting ties up the hands more, requiring the call system to be more elaborate
4. blending two calls together creates hitherto nonexistent meanings:
   if "ABCD" = food
   and "EFGH" = danger
   and then by accidental blend, "ABGH" = food + danger then "AB" = food, "CD" = no danger
   and "EF" = danger, "GH" = no food
   This starts the process of opening a closed call system.
5. The habit of building (putting together meaningful components two or more units long) allows this to continue.
6. This more complex system is (can only be) propagated down generations by traditional transmission
7. This teaching of the open call system would be supported by the facility of "verbal play", which creates displacement, as utterances are practised outside the situations in which they are truly valid.
8. Meanwhile, the upright walking is becoming more biologically supported, changing the shape of the vocal tract, thus allowing more vowel sounds.
9. As the pre-language call system became more densely packed with meanings, 'pre-morphemes' (smallest meaningful call) became more and more similar to each other. This led to the possibility of ambiguity, which was solved by allowing pre-morphemes to be not necessarily wholly distinct from each other, but rather distinct by different meaningless internal sound features. Thus pre-morphemes became true morphemes, and introducing duality of patterning.

Studdert-Kennedy argues that 'duality of patterning' (as Hockett calls it) is an inevitable consequence of the fact that human language 'makes infinite use of finite means'; Given the task of getting infinite possible meaning from finite signals:

1. Assigning a single meaning to each signal means that the signals run out
2. each part of a meaning could be represented by a signal, combining them together creating compound meanings. Thus instead of:
   • a = "X"
   • b = "X verbs"
   • c = "X verbs Y"
   • d = "Y"
   • e = "Y verbs"
   • f = "Y verbs X"
   (6 signals) you use:
   • a = "X"
   • b = "Y"
   • c = "verbs"
   • ac = "X verbs"
   • acb = "X verbs Y"
   • bc = "Y verbs"
   • bca = "Y verbs X"
   (3 signals combined in different ways), breaking holistic signals with complex referents into basic components of meaning. Thus was born subject-predicate syntax.
3. This means that any language like this inevitably have a formal (syntactic) system standing between the signals and the world.
4. This could allow the beginnings of displacement (in Hockett's terms), by placing cognitive distance between signal and meaning.

Studdert-Kennedy says that this combination of parts to form meanings (which he calls the 'particulate principle') is a property which any system that makes infinite use of finite means ('physically and mathematically') necessarily conforms.