PART ONE
Literacy and the Constructs of Language
A
Introduction
Chapter One
Graphic Relativity and Linguistic Constructs
Ranko Bugarski
University of Belgrade
Twenty years ago, while spending a year in the United States as a Ful-bright lecturer in linguistics, I presented a paper to the Chicago Linguistic Society entitled âWriting Systems and Phonological Insights,â dealing with the impact of writing on native perception of linguistic units and on the direction of linguistic research and speculation (Bugarski, 1970). In retrospect, this paper appears to have foreshadowed, from a linguistic angle, some of the major notions and preoccupations associated with the interdisciplinary focus on writing and literacy that has developed over the last 2 decades. Robert Scholes recently singled it out as an extremely rare specimen of its kind (Scholes, 1989; Scholes & Willis, 1990).
Twelve years later, this text was substantially expanded as a chapter (entitled âWriting Systems and Language Awarenessâ) in the second, revised edition of a book of mine in Serbo-Croatian (Bugarski, 1984). I have since not returned to these considerations in any systematic way, and the only other article I have published specifically dealing with writing and literacy is a recent introductory review for interested nonspecialists (Bugarski, 1988). I have, however, occasionally had some further thoughts on these issues. In this chapter, I summarize very briefly the main points of my earlier treatments, develop them somewhat in line with my more recent thinking and reading, and finally propose the notion of graphic relativity as emerging from such concerns.
Speech and Writing in Linguistics
Broadly speaking, and in rough chronological terms, three main approaches can be distinguished in linguistic studies regarding the relationships between language, speech, and writing. The first approach, associated with philology and traditional grammar, as disciplines largely concerned with written texts, gave writing primacy over speech. Reflecting ancient beliefs in the magic of the written sign and concerned with the establishment and preservation of canonical versions of highly valued texts, both sacred and profane, this approach identified writing as ârealâ language. Following the invention of printing, it derived further support from the impact of books and other printed matter on human life and civilization. Echoes of this view are, of course, still present in the popular mind.
The ground for a gradual reversal of this approach was prepared in the late 19th century with the rise of phonetics and dialectology and with Neogrammarian interest in immutable sound laws. Language increasingly came to be equated with speech, whereas writing was practically excluded from linguistic science. This complete shift of emphasis continued into early structuralism, culminating in American descriptive linguistics. Although similar pronouncements had already been made by leading European theorists like Hermann Paul and Ferdinand de Saussure, it was Leonard Bloomfieldâs influential and often-quoted dictum that âwriting is not language, but merely a way of recording language by means of visible marksâ (Bloomfield, 1933, p. 21) that most effectively blocked the study of writing by linguists for some time to come. The phylogenetic and ontogenetic primacy of speech was insisted on, which relegated writing to a secondary and incidental graphic reproduction of the sounds of speech as ârealâ language.
The scene began to change again by the middle of the present century, as the stage was set for a third approach, balancing the previous ones by restating the opposition between language and writing as one between spoken language and written language. This resulted from a convergence of theoretical insights originating in different quarters including glossematics, Prague School linguistics, and the work of prominent individual scholars more recently (see, erg., Halliday, 1986; Haas, 1970; Uldall, 1944; Vachek, 1988, an updated version of his earlier work). Essentially, on this view, speech and writing are two partially autonomous, complementary and synchronically equally valid media of language, not simply or fully convertible into each other, with interacting and overlapping but distinct sets of norms. Thus, writing is not a mere reproduction of spoken language in another medium, but a code or formal system in its own right, at least partly independent of the system manifested in speech and amenable to study as such.
It is safe to say that this last paradigm is now dominant among students of writing, although harmful remnants of the preceding approach can still be found, especially, but not exclusively, in the United States. In linguistics itself, most researchers are currently working within it, though naturally with variations (see, e.g., Akinnaso, 1982, 1985; Biber, 1988; Chafe, 1985; Coulmas & Ehlich, 1983; Tannen, 1982, 1984; on the relevance of written language to linguistics, see Linell, 1982; Coulmas, 1989, especially chapter 14). Written language, in the context of the implications and consequences of literacy, is now the subject of a large body of multidisciplinary research in various fields (reported or reviewed in such publications as Olson, 1977; Olson et al., 1985; Goody, 1987; and Scholes, 1989). Taken as a whole, none of these aspects can be my concern here (some discussion and more references can also be found in Bugarski, 1988).
Writing Systems and Linguistic Intuitions
The psycholinguistic, neurolinguistic, and cognitive aspects of the role of reading and writing in developing linguistic consciousness have been ably dealt with in the literature represented by, and surveyed in, such sources as Olson et al. (1985). and Bertelson (1987), as well as in the other contributions to this volume. What I propose to do is look, from a linguistic point of view, at what still seems to be a large grey area related to such studies. This is the role that writing systems appear to play in shaping native intuitions about linguistic units and patterns and, by extension, in the rise and development of various traditions of linguistic thoughtâup to and including theoretical constructs devised by modern linguists to account for language structure.
Although this chapter can lay no claim to being a systematic treatment of this intriguing subject or even a review of what literature there is, it at least attempts to stimulate further explorations by suggesting possible lines of enquiry.
One major and still unresolved dilemma is the precise relationship between native intuitions on the âbuilding blocksâ of language and the emergence of different writing systemsâlogographic, syllabic, and alphabeticâeach of which appears to imply a kind of linguistic analysis. In his classic paper of 1933, Edward Sapir stated that many years of fieldwork on unwritten languages, American Indian and African, had led him to realize that âin the physical world the naive speaker and hearer actualize and are sensitive to sounds, but what they feel themselves to be pronouncing and hearing are âphonemesââ (Sapir, 1933/1949, p. 47). Coming from such an authority, and based on extensive experience, statements like these cannot be dismissed lightly. On the other hand, there is much evidence in work previously cited that, even if Sapir was right, intuitions on linguistic units are further shaped and channeled in a crucial way by the acquisition of literacy. A concrete example of this relationship is the question: âWhich came first, the phoneme or the alphabet?â On the face of it, this looks very much like a chicken-and-egg question, but such a way of phrasing it is hardly helpful. Methodologically, it is more productive to see here an intricate dialectic of mutual conditioning and reinforcement. Just how the mechanism works is, however, still an open issue.
Another factor in this interplay is the structural properties of the given language that are reflected in writing it. We can say briefly that there exists a complex interrelationship, with much feedback, among the structure of a language, the writing system devised or adapted for it, and the intuitions underlying and directing further perceptions, thus preparing the ground for the emergence and development of different linguistic theories. Hence, the nature of the available writing systems and graphic conventions leads from practical matters determining their creation (preserving texts, compiling glossaries, reforming orthographies, teaching languages, etc.) and influencing language development to techniques of linguistic analysis and theories on the structure of language. (See Bugarski, 1980, on âfolk linguisticsâ as antedating and accompanying linguistics proper; for historical reviews of the phoneme/grapheme relationship, see Amirova, 1977, and Kohrt, 1985).
As pointed out in my 1970 paper (Bugarski, 1970) containing more discussion and some illustration from the major ancient and medieval traditions in the history of linguistics, a unit of a language is the element that the prevailing graphic practice recognizes as such. Thus, in Greece, Rome, and India, the smallest structural element of language was conceived as a âletterââclearly a precursor of the latter-day phoneme. In Latin terminology, this was known as litera and identified under the three aspects of nomen (name), figura (written character), and potestas (sound value). In various ways, such a conception seems to underlie and link together efforts at identifying the basic linguistic units, from the creators of the Greek alphabet, through the First Grammarian of 12th-century Iceland and the Korean King Se Yong in 1446, to modern phonemicists like Jan Baudouin de Courtenay, Daniel Jones, and Kenneth Pike, whose contributions to phonological theory largely grew out of their search for a method of broad transcription. (In fact, the last-named scholarâs book on phonemicsâPike, 1947âbears the characteristic subtitle, A technique for reducing languages to writing).
A minuteâs digression at this point might be of interest with regard to...