Merleau Ponty would have loved this, I think. CJ
http://www.phonetik.uni-muenchen.de/~hoole/kurse/hs_evolution/studdertkennedygoldstein_launchingLanguage_2003.pdf introductory excerpt: 0.1 Introduction ‘Discrete infinity’ refers to the property by which language constructs from a few dozen discrete elements an infinite variety of expressions of thought, imagination and feeling. The property ‘seems to be biologically isolated’, because it is unique among systems of animal communication. From another point of view, however, it is not isolated at all, but rather an instance of a general principle common to all natural systems that ‘make infinite use of finite means’ (Humboldt 1836/1999: 91), including physics, chemistry, genetics and language, namely, ‘the particulate principle of self-diversifying systems’ (Abler 1989). 0.2 The Particulate Principle According to the particulate principle, the only route to unbounded diversity of form and function is through a combinatorial hierarchy in which discrete elements, drawn from a finite set, are repeatedly permuted and combined to yield larger units higher in the hierarchy and more diverse in structure and function than their constituents. The particulate units in physical chemistry include atoms, ions, and molecules, in biological inheritance, chemical radicals, genes and proteins, in language, gestures (as will be argued below), segments, syllables, words and phrases. A parallel between languages and genetic systems has repeatedly been remarked by physicists (e.g. Schrödinger 1944), linguists (e.g. Jakobson 1970), and biologists (e.g. Jacob 1977, Pollack 1994). Jacob, for example, wrote: ‘Linguistics has furnished genetics with an excellent model. The image which best describes heredity is that of a chemical message…written…with the combination…of just four chemical radicals. The four units…are combined and permuted infinitely, just as are the letters of the alphabet throughout the length of a text. As a phrase corresponds to a segment of text so does a gene correspond to a segment of the nucleic acid fiber’ (Jacob 1977:187). Like Jakobson (1970), Jacob emphasized that for such a system to work its basic units must themselves be devoid of meaning or function. In language, only if phonetic units have no meaning can they be commuted across contexts to form new words with new meanings. Jacob went on to observe that the principle of combining discrete units to form successive levels of a hierarchy ‘…is not limited to language and heredity …[but]…appears to operate in nature each time there is a question of generating a large diversity of structures using a restricted number of building blocks’ (1977:188). But Jacob did not try to explain why systems as apparently diverse as language, physics, and genetics converge on a common structural principle. That was left to Abler (1989) who first recognized correspondences among Fisher’s (1930) genetical theory of natural selection, the atomic theory of physical chemistry, and Humboldt’s (1836/1999) description of language. 3 Fisher (1930) reasoned that if parents’ characteristics were to blend in their offspring, they would vanish in an average; variation, critical to the process of natural selection, would then decrease from one generation to the next. In fact, of course, variation is conserved, or even increased, across generations and parental characters lost in one generation may reappear unmodified in the next. From such facts Fisher (like Mendel before him) inferred that biological inheritance was necessarily effected by a particulate mechanism: unbounded biological diversity can only be maintained by permutation and combination of discrete genetic entities. Abler (1989) saw that Fisher’s logic of particulate combination applied to physics, chemistry and language no less than to genetics. Moreover, Humboldt’s characterization of the language hierarchy could be extended to these other domains: all four achieve unbounded diversity by‘…a synthetic [i.e. combinatorial] process…[that]…creates something…not present per se in any of the associated constituents’ (1836/1999:67). Novel structures and functions arise at each level of a hierarchy because units do not blend and disappear, but combine as integral units to form new integral units, whose properties are not limited by, and cannot be predicted from, the properties of their constituents. We cannot derive the fire-extinguishing properties of water from the combination of hydrogen (which burns) and oxygen (which sustains burning), nor the properties of proteins from the genes that control their formation. In language we cannot derive the meaning of a word from the phonetic elements that compose it, nor the meaning of a phrase from the lexical meanings of its words without regard to their syntax. Indeed, it is precisely because the properties of units at each new level cannot be derived from the properties of their constituents that successive levels in the language hierarchy (phonology, morphology, syntax) are independent and subject to their own characteristic rules of combination. Thus, the particulate principle rationalizes and generalizes across diverse domains the combinatorial mechanisms and the independence of successive levels in a hierarchy that standard linguistic theory adopts as axioms of linguistic analysis. The principle is a mathematical constraint to which any system that has the property of discrete infinity necessarily conforms. That is why, despite their different modalities, signed and spoken languages arrive at analogous hierarchies of phonology (or sign formation) and syntax (Klima & Bellugi 1979). By assimilating language to other particulate domains, we do not ignore the unique properties of syntax and phonology essential to its function. We do, however, emphasize the roots of language in biophysics, and the critical importance for both lexicon and syntax of the prior evolution of phonetic capacity. 0.3 Discrete Phonetic Units as Conditions of a Lexicon and Syntax Discrete phonetic units of some kind must have emerged relatively early in the evolution of language. For, as Bickerton remarks, ‘…syntax could not have come into existence until there was a sizable vocabulary whose units could be organized into complex structures’ (1995:51). And a sizable vocabulary could not have come into existence until holistic vocalizations had been differentiated into categories of discrete phonetic units that could be organized into words. A critical early step into language therefore was (as it still is) the breakthrough into words, or symbolic verbal reference, by means of a particulate phonetics (Studdert-Kennedy 1998, 2000). 4 Less often remarked, though no less important, syntax could also not have come into existence until there was a code, a phonetic form, for short-term storage of words, independently of their meaning and syntactic function, during preparation of an utterance by a speaker and comprehension of an utterance by a listener. Independent phonetic segments, devoid of meaning, are indeed taken for granted by virtually every approach to the evolution of syntax. Berwick (1998), for example, in his account of the development of hierarchical syntactic concatenation of words by the operator ‘Merge’ in the Minimalist framework starts his derivation with a ‘bag’ of unordered words, each marked by independent phonetic, formal and semantic features. Kirby (2000), for another example, models the emergence of syntax from ‘holistic’ utterances associated with decomposable meanings. His initial utterances are semantically holistic, but consist ‘physically’ of discrete symbols, randomly concatenated into strings of ‘phonetic gestures’. Thus, a necessary condition of compositional syntax (discrete phonetic units) is included in the initial conditions: compositionality can only emerge, because ‘holistic’ utterances readily fractionate along the fault lines of their discrete components. Similarly, Wray (2000), deriving words from holistic utterances as a first step into syntax, assumes that ‘…arbitrary phonetic representation developed not in the service of words, but of complete, [semantically] holistic utterances…long before words or grammar appeared’ (293). Thus, phonetic breakpoints between portions of a semantically holistic utterance (portions that, in Kirby’s and Wray’s models, eventually become words, if they happen to correlate with presupposed breakpoints in the field of reference) are built into the utterance. Where, then, do these phonetic breakpoints come from? What is the physical basis for phonetic segments? The standard units, consonants and vowels, will not do, because they and their descriptive features are purely linguistic and therefore precisely what an evolutionary account must explain. What we require is a prelinguistic unit of motor action that takes on linguistic form and function as it is put to communicative use. _______________________________________________ Marxism-Thaxis mailing list [email protected] To change your options or unsubscribe go to: http://lists.econ.utah.edu/mailman/listinfo/marxism-thaxis
