This book "asserts that the origin and spread of languages must be examined primarily through the time-tested techniques of linguistic analysis, rather than those of evolutionary biology" and "defends traditional practices in historical linguistics while remaining open to new techniques, including computational methods" and "will appeal to readers interested in world history and world geography."
Date: Fri, 01 Apr 2005 10:23:17 +0200 From: Edward Vajda Subject: Linguistic Theory and Complex Words: Nuuchahnulth word formation
AUTHOR: Stonham, John T. TITLE: Linguistic Theory and Complex Words SUBTITLE: Nuuchahnulth word formation PUBLISHER: Palgrave Macmillan YEAR: 2004
Edward J. Vajda Western Washington University, Bellingham, Washington, and Max Planck Institute for Evolutionary Anthropology, Leipzig
DESCRIPTION OF THE BOOK
Nuulchahnulth (formerly called Nootka) is an endangered Southern Wakashan language spoken in several dialectal forms along the western coast of Vancouver Island (British Columbia, Canada). Ever since the pioneering work of Franz Boas (1890) and Edward Sapir (Sapir 1915, Sapir and Swadesh 1939), the highly complex morphological structure of this language has attracted the attention of general linguists. Key features of typological interest include the extreme suffixing polysynthesis, which permits only one root per grammatical word form, as well as the use of tense affixes on nouns as well as verbs. There is also a highly intricate series of morphophonological rules that operate on different levels of word formation. This book builds on the author's earlier work on the language, notably Stonham's (1999) treatment of the phonology, which itself is quite complex. Much of the data analyzed derives from fieldwork performed by Edward Sapir in the early decades of the 20th century (cf. Sapir and Swadesh 1939), when the Nuulchahnulth dialects were in a much better sociolinguistic condition.
The book contains nine chapters, including an introduction and conclusion. There is also a brief appendix containing examples of Nuulchahnulth numeral classifiers (277-278). The bibliography (285-292) is extremely thorough and references the earliest ethnographic works as well as the increasing number of relevant publications on Southern Wakashan that have appeared during the past several years.
The introduction (1-12) pays particular attention to earlier treatments of Nuuchahnulth (henceforward "N") morphology and phonology, and to the position of N in the southern branch of the Wakashan language family. Although the book does not attempt to treat dialectal variation with any degree of thoroughness, the dialect map on p. 6 and a number of comments on local variation made throughout the book are very useful.
Chapter 2 (13-31) covers N segmental and prosodic phonology. Stonham provides charts (7, 13-14) illustrating the correspondence between IPA symbols and the actual phonetic symbols used in the book. The domain of primary stress is identified as the first two syllables of the phonological word. Also identified here is the phonological and morphophonological patterning of such processes as glottalization, lenition, coalescence, epenthetic rules involving the glottal stop, rules involving the realization of length in variable-length vowels, assimilations of various kinds, and processes of labialization and delabialization.
Chapter 3 (32-63) initiates the discussion of such key notions as "root", "stem", and "grammatical word". These definitions are an important matter, since N is generally regarded as a language with allows only one lexical root per word, and also displays significant morphophonemic differences between its derivational and inflectional suffixes. Stonham follows tradition by describing N lexical morphology as based on a single root per word, which can be followed by a number of suffixes chosen from a semantically rich array of functional groups. A chart (37) illustrates that the prosodic structure of root morphemes varies rather sharply according to part of speech, with 85% of verb roots being monosyllabic, but only 15% of noun roots being multisyllabic. Also discussed are the free forms of roots and their bound-form allomorphs, whose final coda or final syllable often differ significantly. This dichotomy functions in recently borrowed vocabulary, as well. Stonham posits the notion of "stem" vs. inflected grammatical word to account for allomophy triggered by whether the morpheme in questions occurs in combination with aspectual affixes or derivational affixes. The author also makes a proposal new to N linguistics, to distinguish between verbal suffixes and what he calls "bound verbs". The latter are verbal morphemes of rich semantic content that can be added to semantically bleached "roots". What this discussion seems to reveal is that, while N morphology is certainly characterized by an obligatory word-initial root in a structural sense, the semantic center in many words is located in the following "suffix", rather than in the word-initial morpheme. The word- initial morpheme in such cases simply acts as a sort of slot filler. It is important to take into account this occasional dichotomy between formal vs. semantic head when attempting to form a typological assessment of N as a language with word-initial roots followed by suffixes. The presence of semantically empty verb "roots" allows for a great deal more semantic richness in the "suffixes" that follow them. This chapter also lays out a series of means for identifying major form classes, such as noun and verb, in N.
Chapter 4 (64-120), which is entitled "Suffixation", discusses the difference between inflectional and derivational meaning in the language. Subsections treat each respective tense, mood, and aspectual category in verbs, as well as categories such as alienable vs. inalienable possession, causation, and plurality. A distinction in derivation is drawn (110) between so-called "governing" suffixes, which add a new central notion to the stem, and "restrictive" suffixes, which simply modify or clarify some property already present in the base to which they are added (such as dimension, quality, etc). A restrictive suffix, for example, might add a notion such as "perceive X", where X is the referential notion expressed in the base to which this suffix is added. Once again, the notion of semantic "headedness" is appealed to distinguish between these types of functions.
In Chapter 5 (121-143), Stonham appeals to a templatic model of analysis to account for reduplication and infixation in N. While infixes are rather restricted in their function, reduplication makes a variety of semantic contributions to both derivation and inflection. This chapter discusses inventories these processes from a semantic as well as formal standpoint.
Chapter 6 (144-177) discusses what the author calls "stratal segregation", following the contributions made by Lexical Phonology to the study of different morphological domains. Using formalisms provided by Optimality Theory, Stonham posits two layers in N morphology: a stem level (Stratum I) and a word level (Stratum II). Additionally, certain phenomena - such as clitic attachment in the morphology and aspiration in the phonology - are shown to occur post-lexically. His analysis succeeds in providing an elegant account of the specific behavior of a variety of phonological and morphological processes.
Chapter 7 (178-209) returns again to the issue of reduplication. Stonham introduces a foot-based template as a device for describing the specific rules that govern this process. He also touches on style of "abnormal speech", defined as involving "various permutations of conventional phonemes in order to convey certain extra-linguistic information to the listener." Here again the template is useful. This chapter only briefly mentions the interesting topic of special speech styles in N - a subject first discussed at length by Edward Sapir (1915) and which is treated in some detail in Stonham (1999:111-125).
Chapter 8 (210-271), entitled 'morphosyntax', begins with a discussion of tests for clitic-hood. Incorporation is also discussed here. Stonham argues that incorporation does not violate the "one root per word rule" in N since only bound verbal roots - defined as morphemes that have a free form as well as an bound form - can be incorporated. Bound forms that retain a morphophonemic relationship with a free form and thus resemble roots are attached to a semantically bleached root that acts as a base. Other incorporees display a suppletive relationship with respect to their free-form synonyms and can attach to fully semantic roots. This property of concatenation of richly meaningful suffixes somewhat resembles the type of polysynthesis found in Eskimoan languages. Stonham distinguishes between lexical incorporation - a process that results in new stems - and syntactic incorporation, where a sentence element is incorporated for pragmatic purposes. Unlike lexical incorporation, which involves an idiomatic semantic relationship between the morphemes brought together in a single word form, syntactic incorporation does not create new lexical items. In the latter case, the verb root plus incorporee do not yield a novel semantic unit, but retain all of the functional qualities of a free syntactic phrase. Given this morphological property, N even allows for the incorporation of non-head elements such as quantifiers. The difference between lexical and syntactic incorporation posited here by the author for N is well reasoned and is clearly demonstrated by the data presented. The chapter next examines the syntactic and semantic properties of six different types of noun + noun collocations, some of which provide close analogs to compounds in other languages. There is also a thorough treatment of the language's system of numeral classifiers and a balanced discussion of pros and cons regarding whether inflectional morphemes should be treated as suffixes or clitics. The functional aspects of inflectional morphology are not treated exhaustively here and the topic is left open for further investigation.
A brief final chapter (272-276) reiterates the various conclusions and findings put forward in the preceding chapters.
This thorough treatment of Nuuchahnulth morphology succeeds very nicely on a number of levels. It presents the most complete and detailed picture of the morphology available anywhere, and thus offers much to general typologists. In addition to the wealth of language data covered, its explanatory approach is informed by a number of current linguistic theories, making it highly relevant to morphologists of all backgrounds. The discussion should be of special interest to linguists working in the framework of Optimality Theory as well as to anyone interested in templatic morphological structures. Despite the complexity of the subject matter, the prose is consistently clear and the argumentation well reasoned. This book will prove essential to anyone attempting to unravel the many intricacies of word formation in this language. Anyone interested in the topic of polysynthesis cross-linguistically should certainly take stock of the data so expertly accounted for here. Finally, it could also serve to stimulate further synchronic as well as diachronic research into Southern Wakashan language structure.
Boas, Franz (1890) 'The Nootka.' Report of the 60th meeting of the British Association for the Advancement of Science, pp. 582-604.
Sapir, Edward (1915) 'Abnormal speech styles in Nootka.' Geological Survey, Memoir 62, Anthropological Series No. 5. Ottawa, Canada.
Sapir, Edward and Morris Swadesh (1939) Nootka texts, tales and ethnological narratives, with grammatical notes and lexical materials. Philadelphia & Baltimore, MD: Linguistics Society of America.
Stonham, John T. (1999) Aspects of Tsishaath Nootka phonetics and phonology. Lincom: Munich.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Edward J. Vajda is professor of linguistics, Russian language, and Eurasian studies at Western Washington University. He is an editor of the journal Word. His research interests include morphological typology, as well as the study of minority languages of the Russian Federation. For the past several years he has been intensively involved in descriptive research on the structure of Ket, a language isolate spoken by a few hundred people in Central Siberia near the Yenisei River.