JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Linguistic Society of America is collaborating with JSTOR to digitize, preserve and extend access to Language.The following paper is a descriptive analysis of the phonemes of the language represented by the Zographensis manuscript,' based on the distribution of the symbols used by the scribe. Wherever two alternative analyses are possible from the descriptive data, historical evidence is used in making a choice. Such evidence is also used in making an initial phonetic assumption. Otherwise it is not used, either to determine the phonemes or to supplement the distributional analysis.2The Codex Zographensis, a manuscript in the Glagolitic alphabet dating from ca. 1000 A.D., consists of a translation of the four gospels into Old Bulgarian. It is one of the oldest OB texts, and for that reason the phonological structure of its language is of significance for historical linguistics. The present analysis is based on Luke 5-13, that portion of the original which Leskien gives (Handbuch der altbulgarischen Sprache 179-205) in a faithful transliteration of the original Glagolitic into Cyrillic symbols. This amounts to 27 pages of printed material in the Handbuch and includes enough material for a phonemic analysis, as will be evident from the completeness of the evidence presented here.The Cyrillic text makes use of 21 consonantal symbols, for which I will use the following conventional transliteration: p, b, m, v, t, d, s, z, k, g, x, y, ) 6, ,it, id, c, 3, 1, n, r. The evidence that p, b, m, v, etc. represent (approximately) a voiceless labial stop, a voiced labial stop, a voiced labial nasal, a voiced labial fricative, and so on, is historical. There are 18 vocalic symbols, of which only 14 are of significance in this analysis. One symbol, w, occurs in a few foreign words, and in a few native words which also appear with the symbol o.0 It is assumed, at least for OB words, that c and o are two symbols for the same phone. The symbol y is used only in a few foreign words and has no significance for the general distribution. (Foreign words are not considered in this paper because, although they may have been adapted to the phonemic system of the language, they may not have been similarly adapted to the graphic system. Such words may be subject
