Do submarines swim?

Joe McLaren


In light of such a question, some have concluded that there is no immediate relation between the external world and the expressions with which it is described. It is hopeless to study the movement of submarines in order to determine whether or not they swim. It is not a question of submarines, but of the connections between the way we think and the way we speak. The question of how to go about understanding such connections is contested, as with all attempts at knowledge of ‘human nature’. Such contests often devolve into claims for scientific or literary-humanistic monopoly. I don't think that, in the case of language, this is the most fruitful course to pursue.

One roughly-defined group of linguists, ‘relativists’, have tended towards the position that the grammatical structure of a language is connected to the thought-structure of a speaker of that language, both of which are in some way connected to the culture of the society within which the language is spoken. On the other hand, the linguistic ‘universalists’ contend that there is no good reason to believe in the existence of such connections. Even supposing that there were connections between grammatical structure and thought-structure, the universalist believes the grammatical structures of different societies’ ‘languages’ are only superficially different. Once the kaleidoscopic veil is pulled back, language is like the rest of nature: masses of ostensibly diverse phenomena generated by a few simple formulae.

A champion of the universalist position is Noam Chomsky. The scientific nature of Chomsky’s approach can be found in a classic statement on the application of abstraction to the study of language, in his Aspects of the Theory of Syntax (1965):

Linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogeneous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance…

Thus Chomsky is interested in postulating abstract mechanisms that explain the fundamental, universal features of language use, rather than studying the particular ways it is used in everyday life.

Colloquially, we often refer to ‘learning how to talk’. A more accurate locution, says Chomsky, is to say that language grows. That the acquisition of language is largely directed by a genetic endowment is the most controversial and most fundamental of Chomsky’s arguments. This must be so, says Chomsky, due to the incredible gap between the rate and depth of linguistic information a child is able to acquire, and the paucity of the data with which they are presented. A child acquires roughly a word an hour between the ages of two and eight, typically with a single exposure, in ‘highly ambiguous circumstances’. The words, or rather the concepts that words represent, are understood with a richness and complexity at which a dictionary can only hint. People have acquired normal language with no more input than putting their hands on another person’s face and throat. We take it for granted that the design of our visual system, for example, is due to a genetic endowment. Why not the same for the linguistic system?

The most prominent characteristic of language, for Chomsky, is what he terms its ‘creativity’. The speaker of language is capable of producing and understanding hierarchically structured expressions of indefinite length that they have never heard or said before. Take the English sentence, ‘The cat that sat on the mat… died.’ To this sentence it is possible to perpetually insert prepositional phrases in place of the ellipsis. For example, ‘The cat that sat on the mat that was wet before it rained on the hill in Australia but not the mat that was not on the hill where the brother of the cat that sat on the mat...and so on… died.’ A language user has most likely never heard this sentence before (it has probably never been uttered before), and yet is capable of producing and interpreting it. Hence, language’s ‘creativity’. To be able to explain a universal property such as this, thinks Chomsky, is to get at the heart of language.

Not all are convinced, however. Many see Chomsky’s approach as dismissive of the elementary feature of language, diversity. As George Steiner put it, such linguists assume that any study of language must begin and end with its ‘enigmatic largesse’. An example of this ‘largesse’ is the range of differences in the way the meaning of words is constructed out of smaller units of meaning. Vietnamese speakers do not add these smaller units together to modify meaning, but gain this information from other, isolated words. For example, ngôn ngữ means ‘language’ and ‘-học’ means ‘the study of’, so ngôn ngữ học means ‘linguistics’. One the other hand, in the North American Cayuga language, ‘Eskakhehona’táyethwahs’ is one word, and means ‘I will plant potatoes for them again.’ A further demonstration of diversity is the fact that there are languages in which word order is argued to be free, such as Warlpiri, in which Ngarrka- ngku ka wawirri panti-rni (‘The man is spearing the kangaroo’) can be spoken with the elements of the sentence in any order. Thus, (1), (2) and (3) are understood by speakers of Warlpiri as repetitions of each other:

(1) Ngarrka- ngku ka wawirri panti-rni

(2) Wawirri ka panti-rni ngarrka- ngku

(3) Panti-rni ka ngarrka- ngku wawirri

‘Linguistic universals’, argue the linguists of largesse, are often no more than tendencies. The claims of universalists are either false or unfalsifiable. Upon hearing the proposition of a Universal Grammar, the typological linguist raises an eyebrow peppered with the dust of the Real World, and murmurs, Seriously, Noam?

Whether or not a linguist thinks languages have a common, innate structure is going to influence what they think of the relationship between language and thought. Relativists attribute a much deeper role to a person’s experience in the formation and design of the language they speak. As a person’s social environment also dramatically influences the belief systems that person possesses, many have seen reason to believe in a relation between culture, thought and grammar. The linguist most famous for this belief is Benjamin Whorf. This network of relations constitutes what Whorf termed the language user’s ‘thought-world’. For Whorf, the world in which speakers of different languages live is a distinct world, not just the same world with different labels attached. Contrary to the popular notion of Whorfian determinism, or the ‘Sapir-Whorf hypothesis’, Whorf did not believe an individual’s thought to be determined and paralleled by their native language. Rather, he posited subconscious connections between some aspects of thought and language, connections implicit in grammatical forms. For Whorf, some of these grammatical forms can influence the way we classify and categorise the things around us and our reactions to them. These influences reflect, in a fragmentary way, the particular thought-world of the culture to which the language belongs.

Whorf contends that the grammatical structures of the languages of Europe, those he terms ‘Standard Average European’ (SAE), produce a particular ‘thought-world’. The structure of this thought-world is responsible for some of the cultural practices and understandings of European societies. The SAE thought-world, Whorf argues, interprets reality in terms of forms and contents. This carries over into conceptions of non-spatial phenomena as well, particularly time. The SAE tense-system of past, present and future allows SAE speakers to conceive of time as individual units standing in a row. In this way, English speakers conceive of the duration and intensity of temporal phenomena in spatial terms. I can say that it was a ‘long’ morning that ‘dragged along’. When it ‘hit’ five I had a beer, however, and the next hour ‘rushed past’. According to Whorf, the spatialisation of time is responsible for European historicity and record-keeping. The commercial structure of Western societies, including the use of rent, time-wages, credit, and interest, would function under any linguistic treatment of time once it had been built and was got going. But that it got going at all, for Whorf, is a ‘fact decidedly in consonance with the patterns of the SAE languages.’

Whorf contends that neither Hopi grammar nor cosmology evince a spatial conception of time. In his account, the structure of Hopi gives its speakers a conception of reality in terms of ‘events’. These events are processes of becoming, growing, metamorphosing or vanishing. The essences of the ‘eventings’ of Hopi consists in their mode of their unfolding. The form of the manifestation of one state of an event is essential to the quality of the following event-state, and it is in this unfolding process that a link may be detected between the Hopi thought-world and the Hopi culture. In Hopi culture, Whorf says, there is an emphasis on preparation. Whorf cites what he sees as elaborate formality in introductions, rituals intended to prepare the better development of such events as crop-growth, and an emphasis on the power of desire to alter events. These conceptions of reality are influenced by the tense system of the Hopi language, which is ‘subtle, complex, and ever-developing’.

For claims like Whorf’s to be true, it would be necessary for the grammatical structures of different ‘languages’ to more than superficially differ: something Chomsky has always denied. In the last couple of decades he has posited that Universal Grammar consists only of ‘recursion’, or Operation Merge. Ostensibly at odds with Chomsky’s earlier insistence on a rich and highly articulated innate structure, full of baroque rules and operations, Merge is the result of the ‘Minimalist Program’: the search for the simplest possible solution to what was referred to above as ‘creativity’. The basic idea is that you take one item, X, and another thing, Y, perform Merge on them and get {X,Y}. The product can likewise be operated upon by Merge, and you can get {{X,Y}Z}. These items are concepts such as ‘the boy’, ‘eats’ and ‘the apple’. Thus an infinite range of ordered expressions may be produced. Diversity between languages occurs only when Merge-constructions are mapped to the system that controls our ears, eyes, hands, and mouths, at which point conditions of this articulatory-perceptual system are imposed. Different languages are then just different solutions to the problem of translating internal linguistic constructions into sound or gestural patterns.

If the differences between the sentence-structures of French and Hindi exist at the level of sound and gesture, not at the level of thought, then Chomsky can hardly believe it is sentence-structure that provides deep differences in perceptions of the world. Then why does the way we talk apparently imitate the way we think? For Chomsky, the answer lies in the lexicon: the place from which our ideas are plucked in order to be combined into complexes of ideas. Unlike grammatical structure, the universality and simplicity of which allow for scientific study, the range of concepts made available by the lexicon is affected by experience. Yet for this reason, among others, Chomsky thinks the workings of even such deceptively simple concepts as ‘person’ and ‘house’ lay far from current scientific understanding of language and mind, and might always do so.

The study of language is a domain rather jealously guarded by many involved in humanistic study. In the opinion of George Steiner, the ‘relativism’ of Whorf is more appropriate a conception for someone who reads novels and focuses on the particularity of certain phrases and such particularities’ propensity to build the author’s ‘thought-world’. For Steiner, there is an affinity between Whorf’s relativism regarding language and culture and a literature student’s regarding authorial styles. Yet I would argue that this is not necessarily the case. Although Chomsky believes in the superiority of science’s capacity to explain certain aspects of language, his scepticism regarding its ability to articulate the links between how we live, think and speak is actually rather encouraging. It radically circumscribes the scope of possible scientific understanding of human mind and culture, and leaves these open to humanistic and literary understandings that need not mould themselves into quasi-scientific projects. Thus Chomsky, who in the domains of history and politics holds a Platonic aversion to the arts as a source of understanding, proposes that literature makes possible far deeper insight into ‘the full human person’ than scientific inquiry may hope to do.


About the author: Joe McLaren, co-editor of sugarcane, is a student living in Melbourne.

Recent Posts

See All