Charles F. Hockett

View on linguistics

In his paper "A Note on Structure", he proposes that linguistics can be seen as "a game and as a science." A linguist as a player in the game of languages has the freedom to experiment on all utterances of a language, but must ensure that "all the utterances of the corpus must be taken into account." Late in his career, he was known for his stinging criticism of Chomskyan linguistics.

Key contributions

Criticisms of Noam Chomsky and the Generative Programme

Hockett was initially receptive to Generative grammar, hailing Chomsky's Syntactic Structures as "one of only four major breakthroughs in the history of modern linguistics" (1965). After carefully examining the generative school's proposed innovations in Linguistics, Hockett decided that this approach was of little value. His book The State of the Art outlined his criticisms of the generative approach. In his paraphrase a key principle of the Chomskyan paradigm is that there are an infinite number of grammatical sentences in any particular language.

The grammar of a language is a finite system that characterizes an infinite set of (well-formed) sentences. More specifically, the grammar of a language is a well-defined system by definition not more powerful than a universal Turing machine (and, in fact, surely a great deal weaker).

The crux of Hockett's rebuttal is that the set of grammatical sentences in a language is not infinite, but rather ill-defined. Hockett proposes that "no physical system is well-defined".

Later in "Where the tongue slips, there slip I" he writes as follows.

It is currently fashionable to assume that, underlying the actual more or less bumbling speech behavior of any human being, there is a subtle and complicated but determinate linguistic "competence": a sentence-generating device whose design can only be roughly guessed at by any techniques so far available to us. This point of view makes linguistics very hard and very erudite, so that anyone who actually does discover facts about underlying "competence" is entitled to considerable kudos.

Within this popular frame of reference, a theory of "performance" -- of the "generation of speech" -- must take more or less the following form. If a sentence is to be uttered aloud, or even thought silently to oneself, it must first be built by the internal "competence" of the speaker, the functioning of which is by definition such that the sentence will be legal ("grammatical") in every respect. But that is not enough; the sentence as thus constructed must then be performed, either overtly so that others may hear it, or covertly so that it is perceived only by the speaker himself. It is in this second step that blunders may appear. That which is generated by the speaker's internal "competence"is what the speaker "intends to say," and is the only real concern of linguistics: blunders in actually performed speech are instructions from elsewhere. Just if there are no such intrusions is what is performed an instance of "smooth speech".

I believe this view is unmitigated nonsense, unsupported by any empirical evidence of any sort. In its place, I propose the following.

All speech, smooth as well as blunderful, can be and must be accounted for essentially in terms of the three mechanisms we have listed: analogy, blending, and editing. An individual's language, at a given moment, is a set of habits--that is, of analogies, where different analogies are in conflict, one may appear as a constraint on the working of another. Speech actualizes habits--and changes the habits as it does so. Speech reflects awareness of norms; but norms are themselves entirely a matter of analogy (that is, of habit), not some different kind of thing.

Despite his criticisms, Hockett always expressed gratitude to the generative school for seeing real problems in the preexisting approaches.

There are many situations in which bracketing does not serve to disambiguate. As already noted, words that belong together cannot always be spoken together, and when they are not, bracketing is difficult or impossible. In the 1950s this drove some grammarians to drink and other to transformations, but both are only anodynes, not answers

Design features of language

One of Hockett's most important contributions was his development of the design-feature approach to comparative linguistics. He attempted to distinguish the similarities and differences among animal communication systems and human language.

Hockett initially developed seven features, which were published in the 1959 paper “Animal ‘Languages’ and Human Language.” However, after many revisions, he settled on 13 design-features in the Scientific American "The Origin of Speech."

Hockett argued that while every communication system has some of the 13 design features, only human, spoken language has all 13 features. In turn, that differentiates human spoken language from animal communication and other human communication systems such as written language.

Hockett's 13 design features of language

1. Vocal-Auditory Channel: Much of human language is performed using the vocal tract and auditory channel. Hockett viewed this as an advantage for human primates because it allowed for the ability to participate in other activities while simultaneously communicating through spoken language.
2. Broadcast transmission and directional reception: All human language can be heard if it is within range of another person's auditory channel. Additionally, a listener has the ability to determine the source of a sound by binaural direction finding.
3. Rapid Fading (transitoriness): Wave forms of human language dissipate over time and do not persist. A hearer can receive specific auditory information only at the time it is spoken.
4. Interchangeability: A person has the ability to speak and hear the same signal. Anything that a person is able to hear can be reproduced in spoken language.
5. Total Feedback: Speakers can hear themselves speak and monitor their speech production and internalize what they are producing by language.
6. Specialization: Human language sounds are specialized for communication. When dogs pant it is to cool themselves off. When humans speak, it is to transmit information.
7. Semanticity: Specific signals can be matched with a specific meaning.
8. Arbitrariness: There is no limitation to what can be communicated about and no specific or necessary connection between the sounds used and the message being sent.
9. Discreteness: Phonemes can be placed in distinct categories which differentiate them from one another, like the distinct sound of /p/ versus /b/.
10. Displacement: People can refer to things in space and time and communicate about things that are not present.
11. Productivity: People can create new and unique meanings of utterances from previously existing utterances and sounds.
12. Traditional Transmission: Human language is not completely innate, and acquisition depends in part on the learning of a language.
13. Duality of patterning: Meaningless phonic segments (phonemes) are combined to make meaningful words, which, in turn, are combined again to make sentences.

While Hockett believed that all communication systems, animal and human alike, share many of these features, only human language contains all 13 design features. Additionally, traditional transmission, and duality of patterning are key to human language.

Design feature representation in other communication systems

Honeybees

Foraging honey bees communicate with other members of their hive when they have discovered a relevant source of pollen, nectar, or water. In an effort to convey information about the location and the distance of such resources, honeybees participate in a particular figure-eight dance known as the waggle dance.

In Hockett's "The Origin of Speech", he determined that the honeybee communication system of the waggle dance holds the following design features:

1. Broadcast Transmission and Directional Reception: By the use of this dance, honeybees are able to send out a signal that informs other members of the hive as to what direction the source of food, or water can be located.
2. Semanticity: Evidence that the specific signals of a communication system can be matched with specific meanings is apparent because other members of the hive are able to locate the food source after a performance of the waggle dance.
3. Displacement: Foraging honeybees can communicate about a resource that is not currently present within the hive.
4. Productivity: Waggle dances change based on the direction, amount, and type of resource.

Gibbons are small apes in the family Hylobatidae. While they share the same kingdom, phylum, class, and order of humans and are relatively close to man, Hockett distinguishes between the gibbon communication system and human language by noting that gibbons are devoid of the last four design features.

Gibbons possess the first nine design features, but do not possess the last four (displacement, productivity, traditional transmission, and duality of patterning).

1. Displacement, according to Hockett, appears to be lacking in the vocal signaling of apes.
2. Productivity does not exist among gibbons because if any vocal sound is produced, it is one of a finite set of repetitive and familiar calls.
3. Hockett supports the idea that humans learn language extra genetically through the process of traditional transmission. Hockett distinguishes gibbons from humans by stating that despite any similarities in communication among a species of apes, one cannot attribute these similarities to acquisition through the teaching and learning (traditional transmission) of signals; the only explanation must be a genetic basis.
4. Finally, duality of patterning explains a human's ability to create multiple meanings from somewhat meaningless sounds. For example, the phonemes /t/, /a/, /c/ can be used to create the words "cat," "tack," and "act." Hockett states that no other Hominoid communication system besides human language maintains this ability.

Later additions to the features

In a report published in 1968 with anthropologist and scientist Stuart A. Altmann, Hockett derived three more Design Features, bringing the total to 16. These are the additional three:

1. Prevarication: A speaker can say falsehoods, lies, and meaningless statements.
2. Reflexiveness: Language can be used communicate about the very system it is, and language can discuss language
3. Learnability: A speaker of a language can learn another language

Cognitive scientist and linguist at the University of Sussex Larry Trask offered an alternative term and definition for number 14, Prevarication:

14. (a) Stimulus Freedom: One can choose to say anything nothing in any given situation

Relationship between design features and animal communication

Chomsky theorized that humans are unique in the animal world because of their ability to utilize Design Feature 5: Total Feedback, or recursive grammar. This includes being able to correct oneself and insert explanatory or even non sequitur statements into a sentence, without breaking stride, and keeping proper grammar throughout.

While there have been studies attempting to disprove Chomsky, Marcus states that, "An intriguing possibility is that the capacity to recognize recursion might be found only in species that can acquire new patterns of vocalization, for example, songbirds, humans and perhaps some cetaceans." This is in response to a study performed by psychologist Timothy Gentner of the University of California at San Diego. Gentner's study found that starling songbirds use recursive grammar to identify “odd” statements within a given “song.” However, the study does not necessarily debunk Chomsky's observation because it has not yet been proven that songbirds have the semantic ability to generalize from patterns.

There is also thought that symbolic thought is necessary for grammar-based speech, and thus Homo Erectus and all preceding “humans” would have been unable to comprehend modern speech. Rather, their utterances would have been halting and even quite confusing to us, today.

The : Phonetics Laboratory Faculty of Linguistics, Philology and Phonetics published the following chart, detailing how Hockett's (and Altmann's) Design Features fit into other forms of communication, in animals: