Grover Hudson

Linguistic Analysis of the 1994 Ethiopian Census

The first national census of Ethiopia was conducted in 1984 and the second in 1994, with some recounting in Afar and Somali regions in 1996. In addition to information on age, gender, marital status, education, migration patterns, housing, and economic activity type, the 1994 census includes considerable information of interest concerning language (Office of Population and Housing Census Commission 1998). Numbers reported in the census are not actual counts of census takers, but projections based on statistical regularities observed in the actual counts. Information on language was collected in even fewer households than other sorts of information, so it is derived even more as statistical projections (Office of Population 1998, 1:1-5). In seeking linguistic data, census takers had to overcome politically significant ethnolinguistic sensitivities, as well as confusion presented by the varying names of Ethiopian ethnic groups and languages, and logistical difficulties of reaching and objectively sampling the diverse, multilingual Ethiopian population. In fact, the National Population Policy of Ethiopia (Office of the Prime Minister 1993) says nothing that assigns importance to Ethiopian linguistic or ethnolinguistic differences, which would necessitate the collection of linguistic information by the census. Thus, it is a pleasant surprise to find so much linguistic information in the 1994 census, and important also to observe that this information is generally quite consistent with our usually unquantified and often rough intuitive knowledge of the complex Ethiopian linguistic scene. Linguistically relevant information is tabulated in six tables in vol. 1 of the national census, Results at Country Level:

On a Defence of Autosegmentalism.

Haile and Mtenje (I988) raise counter-arguments to my criticisms of an autosegmental analysis of Arabic verb morphology and also offer criticisms of the alternative analysis which I presented in Hudson (I986). The Arabic facts and the issues are somewhat complex, and to review them thoroughly here would require repetition of much of the three earlier articles. Perhaps I should trust those who have read all three articles to judge accurately the second in relation to the first and third. A response seems necessary, however, because the issues are important as well as complex, and because Haile and Mtenje do not give a clear or even reasonable presentation of my argument.

The Principled Grammar of Amharic Verb Stems

The first purpose of this paper is to present a grammar of stem-formation of Amharic basic verbs. The second purpose is to present an argument for the principle of grammar known äs the true generalization condition (Hooper 1976:13). In section 2 the basic facts of Amharic verb-stem formation are presented and in section 3 the grammar of these facts governed by the true generalization condition. Since many linguists appear to think that the true generalization condition is overly restrictive and unreasonable, it seems important just to show how a grammar observing the true generalization condition is formulated and how it works, and how in it are expressed the empirically supportable generalizations of grammars, äs in the case of Amharic verb-stem formation. Perhaps most linguists are aware of the important need for constraints on grammar. In sections 4 and 5 I will briefly illustrate the problem which this need creates in the grammar of Amharic verb-stem formation: the existence, in the absence of a principle like the true generalization condition, of innumerable unprincipled analyses. This part of the argument must be brief since the possibilities ARE innumerable. The argument is reviewed in section 6 along with some concluding discussion about the differences between grammar governed by the true generalization condition and its alternatives. Amharic verb stems are an interesting case, since they show a number of alternations of form äs the apparent result of the loss of five back consonants: je, ft, ?, A, and ?, the latter pair in some environments only. The problem of accounting for these alternations has, naturally enough, raised in the minds of some linguists (Taddese 1972, Bender and Fulass 1978, Podolsky 1980) the possibility of expressing generalizations over the complexities by supplying the lost consonants in the lexical representations of verb roots, providing rules conditioned by the underlying presence of these consonants, and subsequently deleting the consonants. I hope to show here that this is an unnecessary and unreasonable procedure, for it simply leads to complexities of a different sort: theoretical and analytic complexities, versus the actual grammatical complexities which the data

Highland East Cushitic Languages

Highland East Cushitic (HEC) is a group of five languages with some nine named dialects in south-central Ethiopia: from north to south Hadiyya, Kambaata, Sidaama, Gedeo, and Burji. Neighbor to Semitic and Omotic languages, HEC speakers are some 5.5 million people. HEC languages are consistently head-final and inflectional, with glottalic ejective stops, implosive d, long consonants and vowels, and productive n-metathesis in verb formation. Verbs have regular causative, passive, and reflexive derivatives. Nominative case is marked, and the singular as well as plural of nouns. Sidaama marks gender in first person, and has gender exclusive vocabulary.

A Comparative Dictionary of the Agaw Languages by David Appleyard (review)

The dictionary has four parts: a 20-page introduction providing essential linguistic background; the dictionary of some 720 English-Agaw entries comparing words of the four principal Agaw varieties—Bilin, Kemant, Xamtanga (also known as Chamir), and Awngi (Awiya)—plus reconstructions, where possible, and discussion; as appendices, a list of reconstructed forms (about 400) and Agaw-English wordlists of the four Agaw varieties; and a four-page bibliography. Most entries of the dictionary include etymological notes and discussion, with word variants from the older literature, and possible cognates in other Afroasiatic languages. The Agaw-English wordlists are words of main entries but also those that arise in the discussion of these words. The introduction is a concise review of Agaw-language studies, to which Appleyard has, for some time, been the principal contributor. It has three parts: “the Linguistic Context,” concerning the linguistic and geographic setting of the languages; “The Dictionary,” which explains how entries of the dictionary are constructed; and “The Basis for

A Grammar of Konso

Orkaydo’s grammar of the Lowland Cushitic language Konso is a comprehensive and richly examplified description of an important language of interest in a number of ways including absence of a voice contrast, marked nominative case, and an unusually distributed set of subject-person agreement clitics. Highlighted in this review are the subject clitics, adjectival notions expressed through a class of stative verbs, and a rule of glottal stop insertion argued here to result from an analogical not phonological rule.

A Neglected Ethiopian Contribution to Semitic and Afroasiatic Reconstruction

Proceedings of the Twentieth Annual Meeting of the Berkeley Linguistics Society: Special Session on Historical Issues in African Linguistics (1994)