Subject: Re: [s]concept of representation in the speech sciences (trying again) From: Branka ZEI <Branka.Zei(at)HCUGE.CH> Date: Wed, 21 Sep 2005 20:42:01 +0200Hello List, Daniel Silva wrote: >..... concepts involving representations are too obscure in the >speech sciences. ...... authors are often telling >about different things by employing the same words. >In speech neurosciences, ......... we never know >exactly what does one means by the term "representation", and it gets even >worse when it is associated with the words "phonological" or "phonetic". The word « representation » refers to different concepts depending on the field where it is used. In cognitive psychology it refers to the mental constructs we acquire in interaction with the material world (external and internal). It includes for example mental images, or simply concepts. In a way all our knowledge of the world is but a RE-presentation of the world. Each Re-presentation is a reduced model of reality. It includes only some aspects of it to the exclusion of others. Its strength resides precisely in the reduction of data. Imagine a map of the world that would be as detailed as the world itself. It would be useless. Mental imagery (be it visual or acoustic, figurative or kinetic), concepts, etc - all are RE-presentations of the world constructed by our cognitive system. Obviously our mental constructs are not accessible to observation. Even in neuroscience, you can never observe a cognitive construct such as a concept or a mental image. All you can observe is the localization and sequencing of brain activities i.e. you can only observe the hardware and not the global qualitative result of the functioning of this hardware. The qualitative aspect is only indirectly « observable » . Linguistics is an ideal ground for demonstrating how systems of representations function: A linguistic sign itself is a "double-faced " mental construct (a semiotic structure). Double-faced because it is composed of an acoustic image (of a word for example) and its meaning (a concept). As all representations are reduced models of material reality (that served to their construction), they retain only the characteristics that are relevant for the task at hand. In the world of psycholinguistics a phonological representation of a word would contain only those features (of the sounds) that are relevant for the task at hand i.e. for communication in a particular language. The voiced-voiceless distinction for the consonants S/Z is relevant for the French phonological system but it is not relevant for Spanish. This is why in French, they constitute two phonemes. In Spanish, if you pay attention, you can hear S and Z as two acoustically different sounds, but they do not constitute two phonemes, as their distinction is useless for communicating in Spanish. For structural linguistics, phonology is concerned with phonemes as bundles of acoustic features relevant for differentiation of meanings. Phonetics is concerned with sounds as acoustic events produced by human organs of sound production. >At the moment I am struggling with questions concerning the demarcation >line (or grey zone) between phonetic and phonological representations. These concepts are discussed by Chomsky and Halle in their book The Sound Pattern of English (1968, p.5). They define phonetic representation as « a two-dimensional matrix in which rows stand for particular phonetic features (universal phonetic categories such as voicing, nasality, etc) ; the columns stand for the consecutive segments of the utterance generated. » . For them « the phonetic symbols are informal abbreviations for certain feature complexes; each such symbol stands for a column of a matrix of the sort just described » Chomsky and Halle have strong doubts about the existence of « phonemic representations » or a phonemic level (Ibid p. 11 and footnote 9). If I understand correctly, for Chomsky and Halle the term « representation » refers to the actually acoustically (materially) realized set of simultaneous and consecutive phonetic features (of an utterance) drawn from the « universal phonetics ». The latter would be something like a universal set of possible phonetic features. Chomsky and Halle define ' phonological representation » as « the representation given by the application of all readjustment rules » (p. 11). The latter actually shape feature matrices by replacing lexical and grammatical formatives (which can be seen as variables) by language specific sounds (constants). Personally, I think that Chomsky and Halle's usage of the term « representation » is very confusing. > It seems that there is no >great concern in clearing up these concepts in spite of the overspread use of >the terms I agree with you. A certain vocabulary is taken for granted in some schools of linguistics, and there is not much concern to disambiguate the usage of such terms. >(or maybe I am looking for answers in the wrong places) You may be right. Indeed, you will not find answers to epistemological questions in the sciences which study only material « facts ». A « fact » in linguistics is not a material fact. A linguistic fact is mental « fact » it is of cognitive nature. >I would appreciate some commentaries and indications of interesting papers. Take the greatest classics in European linguistics such as General Linguistics by Ferdinand de Saussure, and Troubetzkoy's Principles of Phonology. Best regards, Branka Zei Pollermann ________________________________________ Dr. Branka Zei Pollermann Psychologue Psychiatrie de Liaison Hôpitaux Universitaires de Genève 51 Bvd. De la Cluse, 1205 Genève tél. : 0041 22 382 48 81 Portable : 0041 79 203 92 17 ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. **********************************************************************