Ship of Fools
Page 18
Finite clause subordination is the prime example of a complex feature of language, and there is actually very good evidence that this has evolved in the course of history. Referring to Deutscher’s Syntactic Change in Akkadian (2000) Geoffrey Sampson observes:
Akkadian is one of the earliest languages to have been reduced to writing, and Deutscher claims that if one looks at the earliest recorded stages of Akkadian one finds a complete absence of finite complement clauses. What’s more, this is not just a matter of the surviving records happening not to include examples of recursive structures that did exist in speech; Deutscher shows that if we inspect the 2,000-year history of Akkadian, we see complement clauses gradually developing out of simpler, non-recursive structures which did exist in the early records. And Deutscher argues that this development was visibly a response to new communicative needs arising in Babylonian society. (Sampson 2009a: 11)
Karlsson rightly states that typical everyday spoken language tends to be very brief and simple, and:
Before the advent of writing in the third millennium BC, the major expository genre was oral narrative, which has been shown to be aggregative and paratactic rather than subordinating. Phrases are strung together into loosely conjoined shallow sequences (Lord 1960). This pattern of ancient additive structure is found across the world. As a case in point, Leino (1975) analysed a section of the Kalevala (402 lines of verse, some 1,300 words) and found only three subordinate clauses, all ‘when’-clauses embedded at depth immediately below the main clause. For comparison, 1,300 words of current written Finnish would typically contain some sixty finite subordinate clauses…. (Karllson 2009: 195)
While he agrees that some pre-literate languages do make limited use of finite subordinate clauses:
…there is ample evidence, for example from Semitic, Indo-European, and Finno-Ugric languages that the emergence of more elaborate grammaticalized patterns of finite clausal subordination is related to the advent of written language, especially to the conventionalization of various written registers. Proof of this development is provided for instance by Deutscher (2000) for Akkadian, by W.P. Lehmann (1974) for subordinate clauses in Vedic, by O’Neil (1976) for subordinate clauses in Old English, by M. Harris (1988) for concessive clauses in English and Romance, and by König and van der Auwera (1988) for subordinate clauses in Dutch and German. (ibid., 195-96)
Kalmar refers to studies showing lack of subordinate clauses in Australian languages, hunter-gatherers in Siberia, the Bushmen of South Africa, the Ojibway of North America, and Benveniste’s claim that proto-Indo-European did not have relative clauses. (Kalmar 1985: 158-59) He also describes in some detail the emergence of subordinate clauses in Inuktitut in response to the use of writing and familiarity with English (ibid., 159-64).
So, to sum up, the linguist F.N. Akinnaso gives the following characteristics of written as distinct from spoken language:
Preferential usage of elaborate syntactic and semantic structures, especially nominal constructions (noun groups, noun phrases, nominalizations, relative clauses, etc.) and complex verb structures.
Preference for subordinate rather than co-ordinate constructions
Preferential usage of subject-predicate constructions instead of reference-proposition
Preferential usage of declaratives and subjunctives rather than imperatives, interrogatives, and exclamations
Preferential usage of passive rather than active verb voice
Preferential usage of definite articles rather than demonstrative modifiers and deictic terms
Higher frequency of certain grammatical features, e.g., gerunds, participles, attributive adjectives, modal and perfective auxiliaries, etc.
The need to produce complete information or idea units and make all assumptions explicit
Reliance on a more deliberate method of organizing ideas, using such expository concepts as “thesis”, “topic sentence”, and “supporting evidence” (1982: 104, quoted in Goody 1987: 263-64)
According to Karlsson, the maximum number of levels for initial clausal embedding is two, for Central clauses is three, and for Final clauses is five. These maxima were reached by Akkadian, which then influenced Greek, which in turn influenced Latin (Karlsson 2009: 201-202) which then influenced the European languages. These in turn have been extremely powerful models for non-European cultures, especially through colonisation and more recently the United nations, for the official languages of other governments throughout the world.
And then, once this new language had been invented, generative linguists would come along and point to it as yet further corroboration of the idea that human beings share innate cognitive machinery which imposes a common structure on all natural languages…. I believe essentially that process has been happening a lot with Third World languages in modern times. (Sampson 2009a: 17)
It might seem rather obvious to most of us that writing would have an important effect on grammatical complexity, but “Earlier linguists had resisted the idea of the distinctiveness of spoken and written languages. Despite his new insights into orality, or perhaps because of them, Saussure takes the view that writing simply re-presents spoken language in visible form … as do Edward Sapir, C. Hockett and Leonard Bloomfield” (Ong 1982: 17). UG makes the whole question of writing and literacy irrelevant in principle, so it is not surprising that Pinker in The Language Instinct does not mention it at all, but it is significant that the topic is almost as generally ignored in most of the other works on linguistics that I have consulted.
With writing comes schooling, which has further potential for grammatical development if it goes beyond mere rote learning. Schooling and formal education that involve taking the pupils out of the context of their normal daily lives and their active participation and discussion with their teachers is of particular importance. It is closely involved with the ability to explain verbally one’s reasons for making particular choices in test situations, and it also seems to develop the search for rules for the solution of problems, and the awareness of one’s own mental operations. But we should also remember that medieval European universities conducted philosophical and theological debates in Latin, and that disputation was a fundamental basis of ancient philosophy in Greece, India, and China. One is not saying, therefore, that verbal discourse cannot engage a high level of syntactic complexity, only that it requires special circumstances, including people who are already literate, to do so. Bernstein found that there were marked differences in schooling between users of the “restricted” and the “elaborated code”:
[T]he group whom Bernstein found using this [restricted] code were messenger boys with no grammar school education. Their expression has a formula-like quality and strings thoughts together not in careful subordination but “like beads on a frame” (Bernstein 1974 p.134)—recognizably the formulaic and aggregative mode of oral culture. The elaborated code is one which is formed with the necessary aid of writing, and, for full elaboration, of print. The group Bernstein found using this code were from the six major public schools that provide the most intensive education in reading and writing in Britain. (Ong 1982: 83)
It is an axiom of UG that all native speakers have equal competence in their language, so:
The fact that native speakers vary in grammatical competence has deep implications for linguistics and related disciplines.… I suggest that many linguists have an ideological objection to native speaker variations in grammatical competence, which they regard as socially dangerous, in that they can be used to justify social discrimination based on class and race. I also suggest that generative grammarians have a theoretical objection to individual differences. They are committed to the notion of an innate universal grammar, and individual differences are fundamentally incompatible with this notion. (Chipere 2009: 190-91)
So once a written literature has developed in the context of high civilization it has a number of important consequences for thinking and consequently for the use of language. Michael Barnes lists a more critic
al attitude to tradition because literacy allows the documentation of the past; a greater awareness of language as a tool of thought and the expression of ideas; the promotion of more complex systems of classification and their organization into coherent systems; and the greater privatization of thought and the objectification of personal experience (Barnes 2000: 82-3). Halverson refers in general to “the preservative potential of writing” and its cognitive consequences:
[T]he amount of available information can increase far beyond the carrying capacity of human memory, individual or collective; it means that each generation of thinkers can therefore build on the work of its forebears without starting all over again, thus making possible a much more rapid advancement of knowledge than is possible under oral conditions; it means that thought can be communicated more easily and accurately over space as well as time; that it can provide intellectual stimulation beyond the possibilities of isolated oral societies; that it can, in short, expand the mind and sharpen intelligence. These are the kinds of possibilities—only possibilities—opened up by writing, all of them probable and rather obvious. (1992: 315-16)
The complex societies in which the written word becomes normal therefore develop a whole range of cultural subjects that require increasingly complex forms of thought: administrative documents, legislation, legal disputes and arguments, technical manuals for a variety of tasks, and more abstract interests such as the theory of government and philosophy, the natural sciences, and of course literature.
More generally, in modern literate languages such as English, often read from the newspapers or in novels, it requires far more elaborate syntactical structures than are needed in primitive societies in order to provide enough context of time, place, and social circumstances to make statements comprehensible. And in our society most people we meet are strangers with very different life experiences, especially due to the enormous division of labour in advanced industrialized societies.
We can also expect the expression of more abstract thought to require greater precision of language than is necessary for the communication of concrete ideas. For example, “skid– crash– hospital” conveys a perfectly comprehensible sequence of concrete events with no syntactical assistance at all, because we can easily visualise the circumstances referred to. But when we are trying to convey abstract ideas we have to supply the necessary conceptual context as we go along, and this requires very precise syntactical tools to make each sentence comprehensible. Such sentences often become extremely long and complex in order to integrate a number of ideas into an intelligible whole. For example:
Thus in spite of the important respects in which Aristotle’s use of opposites resembles, and indeed is influenced by, earlier notions, his physical doctrine of hot, cold, dry and wet may and should be distinguished both from the hypotheses of modern scientific method and also from the vague accounts common in pre-philosophical myths and early philosophical cosmologies: for if this doctrine cannot be said to give rise to predictions which can be tested experimentally, it is, on the other hand, far removed from the myth that derives all things from Sky and Earth or from symbolic classifications of phenomena which deal globally with the entire spectrum of reality. (G.E.R. Lloyd Polarity and Analogy 1966, p.85)
Finally, it should be noted in relation to the notion of a genetically based language instinct that writing involves different brain functions and areas from those of speech:
[D]ifferent brain areas are involved in hearing speech and reading it, and different comprehension centers in hearing words and reading them.… This finding refutes the conventional theory of comprehension, which argues that a single center in the brain understands words, and it doesn’t really matter how (by what sense or medium) information enters the brain, because it will be processed in the same way and place.(Doidge 2008: 308)
Writing, however, and especially literacy , is a very recent historical development, and the differences in brain functioning that it produces scarcely seem consistent with the idea of a hard-wired cognitive organ of language.
8. Conclusions
The purpose of this paper has been to assemble some evidence and arguments that language does indeed become more complex in relation to social and cultural complexity, especially as a result of writing and literacy, and cannot therefore be an instinct, organ or module as Universal Grammar maintains. But from the beginning we have encountered not just reason and evidence but the dominating influence of a number of ideologies. By “ideology” here I mean opinions that are held with complete devotion by True Believers and a determination to reject or evade what non-believers would consider to be important contrary evidence. Skinnerian Behaviourism, for example, had immense academic influence and claimed that the whole idea of the mind was an illusion and that language did not convey thoughts but simply consisted of stimuli that changed the hearers’ behaviour. On the other hand Chomsky and his school firmly believe that the brain is essentially like a machine, of which language is an inherited component and whose operation is specified by the genome, and has essentially nothing to do with culture or social relations.
Daniel Everett, in an interview with Geoffrey Sampson (2009b: 215) is obviously deeply committed to cultural relativism: “The Piraha’s culturally constrained epistemology can only be evaluated in terms of the results that it gives the Pirahas relative to their own values. Since it serves them very well, there is no sense in the idea that it is inferior.” We might choose to avoid terms like “inferior” as vague and tendentious, but it is entirely valid to compare Piraha culture with that of other societies, and in this more general scheme of things it appears to be unusually primitive, and the fact that the Piraha themselves appear quite happy has nothing to do with the matter. (Remarkably, in his book (2008: 272) Everett also says that he no longer believes in truth, a strange position for one who has spent a great deal of time and effort trying to prove that Universal Grammar is false, or who wishes his work to be taken seriously at all.)
We have also noted that linguistics has been significantly distorted not only by a dogmatic refusal for many years even to consider the subject of linguistic evolution, but by the obsession of secular liberal academics with intellectual equality, displayed in particular by linguists and cultural anthropologists. The Christian view is that all human beings have moral dignity because we are all children of God, a relationship which renders intellectual equality fairly irrelevant. (God, indeed, has generally been regarded as favouring the simple over the learned.) Secular liberals, however, dismiss the idea of God as superstition and believe that we are just another animal species in a Darwinian world, distinguished from other animals in the struggle for survival only by our intelligence. Nevertheless they still want to cling on to the traditional Western belief in the brotherhood of Man, and they can only do this by a fanatical conviction about human intellectual equality, despite all the evidence to the contrary. Liberal academics, especially during the Cold War, were also accustomed to look down on their colleagues in the Soviet Union, whom they saw as slaves to Marxist ideology. Yet at the same time, of their own free will and without the excuse of secret police, gulags, and firing-squads they eagerly enchained themselves in liberal political dogma about human intellectual equality that is just as devoid of evidence as Lysenko’s dismissal of genetics as bourgeois science.
Notes
I should like to record my thanks to Professor John Colarusso, Professor Norman Doidge, and Dr Iain McGilchrist for their very helpful comments on this paper.
1 . Indeed, the unfortunate Professor Everett, celebrated for claiming that the Piraha have no grammatical recursion, has been denounced as a racist for implying that they are therefore subhuman, and denied permission to return to them (Bartlett 2012: 5).
2 . It is interesting that, since descriptive linguistics in America was fundamentally concerned with field studies of the languages of the Native Peoples, the doctrine of equal complexity was not questioned at that time. But “When relative complexity was previously
a live issue in linguistics early in the twentieth century, syntax was not specially central to the discussions” (Sampson et al. 2009: 270) which were more concerned with comparative phonology.
3 . I am obliged to Prof. Dr. Georg Oesterdiekhoff for drawing my attention to this in an unpublished paper he was kind enough to send me.
4 . Rather similar “parameters” go back at least to the universals of Greenberg (1966).
5 . Tomasello claims that “…the ontogenetic process that Piaget hypothesized as crucial for infants” understanding of objects in space—namely, the manual manipulation of objects—cannot be a crucial ingredient since infants understand objects in space before they have manipulated them manually.… This ruling out of one potential developmental process is a significant scientific discovery” (Tomasello 2001: 50). It is a pity that Tomasello had not read Lourenço & Machado 1996:144, which conclusively refutes the infant studies cited by Tomasello.