Making a Zapotec Dictionary

Making a Zapotee Dictionary Pamela Munro San Lucas Quiavini Zapotee (SLQZ) is a language with several thousand speakers ofall ages in the town ofSan Lucas Quiavini, Oaxaca, Mexico. With my collaborator Felipe Lopez, I have been involved for two years in studying this previously undescribed language of the Zapotecan family,1 through traditional linguistic fieldwork and the analysis of narratives about immigration experiences collected in Los Angeles (where about a quarter of the speakers now live) and in Oaxaca. In the course of this work, we have assembled a dictionary that currently contains over 4,000 entries in Zapotee with English and Spanish translations (Munro and Lopez, and others, in preparation), a draft of which will be published along with the edited narratives (Lopez and Munro, eds., in preparation). Making the dictionary has required a number of difficult aesthetic and analytical decisions with broad implications for the construction of other dictionaries, particularly those for other languages of the Americas. In this paper, I will briefly discuss four general questions that we have confronted in this project: • How should the language be written? • What form of an inflected word should be selected as the main entry in the dictionary? Zapotecan is a family ofa number ofmutually unintelligible languages, each of which is generally referred to as "Zapotee." There is no consensus on how many separate Zapotee languages exist; a conservative estimate is four (Terrence Kaufman, personal communication), but other varieties of the Valley group, which includes SLQZ, show considerable diversity in all areas of grammar. Apparentiy the most similar Zapotee language to SLQZ that has been described is San Juan Guelavía Zapotee (Jones and Knudson 1977), but this language appears to differ from SLQZ in terms of phoneme inventory and the lexicon, at least. 132Pamela Munro • How should main entries in the dictionary be translated? • How should syntactic information be encoded in dictionary entries? Each ofthese questions is important for many dictionaries, if not for most. In discussing how we have dealt with these questions for SLQZ, I will consider examples from several other languages, citing dictionaries of American Indian languages I have worked on or with (as well as other work) in addressing these questions. In addition, I will refer to standard orthographies (or spelling systems) of a number of languages without making reference to particular dictionaries. How should the language be written? One of the most important issues in preparing a dictionary is choosing the orthography or spelling system to be used. For languages with an established orthography, there usually seems to be only one option ,2 but few American Indian languages have standard orthographies, so almost every dictionary project must begin with decisions about spelling. There are two basic principles for developing an adequate phonetic orthography: ( 1 ) Each letter or letter combination in the system should represent the same sound each time it is used, according to the pronunciation rules of the language. (In other words, if a learner who knows the rules sees a new word, he should be able to pronounce it correctly without further assistance.) (2) Each occurrence of the same sound of the language should be spelled the same. (In other words, if a learner who knows the rules hears a new word, he should be able to spell it correcdy without further assistance.) These are non-trivial, and the caveat "according to the pronunciation rules ofthe language" is important. No one can guess how to implement an unfamiliar orthography without learning its pronunciation rules. 2Even for languages with well-established orthographies, like that ofEnglish, this is often less ofa fixed choice than it might at first seem. For instance, in my work compiling dictionaries ofUCLA student slang (undertaken with the members of several undergraduate classes: Munro and others 1991; Ali and others 1993), such issues as that of how to spell words in which final -ing is universally pronounced as if it is spelled -in have occasioned heated discussion. ____________________Making a Zapotee Dictionary_________________^33 (This is true of familiar European languages as well as unfamiliar American Indian ones, of course. For instance, the letter j represents completely different sounds in Spanish, French, German, and English. No one can be sure how to...