"In this book, Richard Kern explores how technology matters to language and the ways in which we use it. Kern reveals how material, social and individual resources interact in the design of textual meaning, and how that interaction plays out across contexts of communication, different situations of technological mediation, and different moments in time."
EDITOR: Katamba, Francis X. TITLE: Morphology SUBTITLE: Critical Concepts in Linguistics SERIES: Critical Concepts in Linguistics YEAR: 2003 PUBLISHER: Routledge (Taylor and Francis) ISBN: 0415270782 (for the six-volume set)
Kalyanamalini Sahoo, language consultant
[This review divided into two parts. This part contains the Overview and Synopsis of Volumes I-III. The second part in a subsequent issue contains the Synopsis of Volumes IV-VI, the Critical Evaluation and References. To read Part II: http://linguistlist.org/issues/16/16-1352.html -- Eds.]
This set of 6 volumes is a collection of 78 reprinted articles from various journals and books. The anthology of articles beginning from 1939 to the year 2000 has been carefully planned and nicely coordinated by the editor between the various contributors. Volume 1 ''Word Structure: a variety of views'' is constituted of 18 papers, a preface by the editor, and a general introduction to all the volumes in the set. Is it words or morphemes that are the basic units of morphological analysis? - has been discussed in many articles of volume I. Volume II ''Morphology: Primes, Phenomena and Processes'' is constituted of 13 papers. Articles in this Volume survey morphological phenomena in a huge number of genetically and typologically different languages. The interaction of morphology with other levels of linguistics (phonology, syntax and semantics) has been dealt with in volumes III, IV and V. Volume III ''Morphology: its relation to Phonology'', Volume IV ''Morphology: its relation to Syntax'', and Volume V ''Morphology: its relation to Semantics and the Lexicon'' are constituted of 10 papers each. Volume VI ''Morphology: its place in the wider context'' deals with articles which discuss morphology emerging in an interdisciplinary context or application-oriented field (e.g. psycholinguistics, computational linguistics and historical linguistics). This Volume is constituted of 17 papers, and an index of linguistic terms referred to in all the volumes of the set.
Three articles (one in each of the volumes II, III and VI) are in French, while all the other articles are in English. Each article has its own Notes and References. There is no special effort to put all the articles in the same format, they have just been reproduced from the original/source.
VOLUME I -- WORD STRUCTURE: A VARIETY OF VIEWS Volume I presents a variety of approaches to word- structure. 'Whether words or morphemes are the basic unit of morphological description' is the topic of discussion in many articles. Chapter 2, 3, 4, and 13 defend a morpheme-based approach, while ch. 5, 7, 8, 17, 18 argue for a word-based approach. Ch.16 takes a radical view and denies the existence of words. Overall, this volume presents a unique collection of early influential studies in word structure analysis by leading scholars in the field. Not only that the authors have highlighted what phenomena a morpheme-based approach or a word-based approach will be able to tackle, but also in many articles, the authors have preferred to take a rather critical stand to many key notions of this theoretical approach.
This volume starts with Roman Jakobson's article 'Quest for the essence of language' (source: Diogenes 51 (1966): 21-37). It surveys the development of theories of the linguistic sign from Ancient Greece to the twentieth century. Beginning with Bloomfield's 1933 manual regarding the study of language with respect to 'sound' and 'meaning', Jakobson discusses Saussure's interpretation of the sign - 'significant' and 'signifié'; St. Augustine's Latinized term 'signum' comprising both 'signans' and 'signatum'; Charles Sanders Peirce's interpretation of 'sign' as 'semiosis' in terms of 'icon', 'index' and 'symbol'; Greenberg's grammatical universals and near-universals; Bolinger's importance of cross influences between sound and meaning. He discusses the nature of the word and the place of morphology in the study of language.
In chapter 2, Zellig S. Harris's article 'Morpheme alternants in linguistic analysis' (Source: Language 18 (1942): 169-180) focuses on morpheme-based approach to morphology. It presents regular phonology, morphophonemics, sandhi, morphological processes like vowel change, morpholexical variation, suppletion etc. as cases of a single linguistic relation - morpheme alternants. It proposes techniques for determining the morphemes of a language, gives an account of key concepts and analytical techniques of morpheme analysis in American structural linguistics.
In chapter 3, Eugene A. Nida's article 'The identification of morphemes' (source: Language 24 (1948): 414-441) also focuses on morpheme-based approach to morphology. This is in response to Hockett's (cf. next chapter), and Bloch's (Language 23: 399-418) morphemic analysis. He treats some of the specific problems raised by Bloch's and Hockett's papers, which constitutes a background for the development of certain principles for the analysis and classification of morphemes. Adopting Bloomfield's definition, he provides sub- morphemic distinctions of phonetic form as well as of semantic value. According to it, a definition which contrasts one phonetic-semantic entity with all other entities in the language still permits the sub- morphemic distinctions of phonetic form and semantic areas within the basic distinctiveness which sets off such a form from other possibly related forms. Providing examples from different languages, he discusses 13 principles which govern the identification of morphemes. He classifies morphemes based on (i) the types of phonemes which comprise the morphemes, (ii) the positional relationship of the parts of the morphemes, and (iii) the positional relationship of the morphemes to other morphemes.
In chapter 4 'Problems of morphemic analysis' (Source: Language 23 (1947): 321-343), Charles F. Hockett provides a critical evaluation of the theory of morphemic analysis as propounded by Harris (cf.chapter 2).
Charles E. Bazell in chapter 5 'On the problem of the morpheme' (Source: Archivum Linguisticum 1 (1949): 1-15.) stands his ground and gives a good explanation of the morphemic analysis. He tries to remove any skepticism about this approach by posing a more critical questioning of the fundamental assumptions and analytical techniques of morpheme theory.
In chapter 6, Hockett's article (source: Word 10 (1954): 210-234) discusses two models of grammatical description: Item and Process (IP) and Item and Arrangement (IA). Besides, he makes passing mention of the traditional Word and Paradigm (WP) approach employed since classical antiquity to describe the morphology of Sanskrit, Greek and Latin as well as many modern languages but which had not found favour with modern linguists in the first part of the twentieth century. According to IP model, morphological description involves identifying morphological items and the processes which they undergo, while IA model essentially talks of things and the arrangements in which those things occur. In both the types, the minimal grammatical item is the morpheme, and this is the fundamental unit for all subsequent grammatical analysis. However, neither approach turns out to be entirely satisfactory. The paper concludes with a proposal for a set of five criteria for evaluating models of grammatical description namely, generality, specificity, inclusiveness, productivity and efficiency. Judged using these criteria, none of the models is found to be completely satisfactory, although Hockett leans towards IA. However, he is concerned about its inability to deal effectively with a number of phenomena (e.g. internal vowel change as in take - took).
Chapter 7 'In defence of WP' (Source: Transactions of the Philological Society (1959): 116-144) is written by R. H. Robins as a response to the apologetic passing mention of WP by Hockett (ch.6). Robins's paper presents a defence of a word-and-paradigm approach and shows how revamped WP can be reinstated as a model suitable for aspects of grammatical analysis, in particular the description of languages with inflecting morphology. The distinctive characteristics of a WP model is: the word is taken as the basic unit of both syntax and morphology, and variable words are grouped into paradigms for the statement of their morphological forms and the listing of their various syntactic functions. WP has advantages over Hockett's IA or IP in the sense that (i) the word as a unity is more easily susceptible to grammatical statement than is the individual bound morpheme; (ii) WP avoids some of the difficulties in morphophonology (morphophonemics), in the relating of grammatical structuring, which beset IA and to a lesser extent IP. However, WP has its disadvantages also: (i) with its use of Process terminology, WP, like IP, implies a historic perspective and confuses synchronic linguistics with diachronic linguistics. (ii) WP appears to be less tidy and economical in requiring both Process and Arrangement as separate terms (in morphological and syntactic description respectively) than either IA or IP with their exclusive use of one or the other. (iii) a descriptive model is intrinsically less desirable if it makes a non-minimal element, the word, basic in the hierarchy of structures at the same general level of analysis, as against both the other models that make the minimal grammatical element, the morpheme, also the basic element of structure. However, Robins admits that because of the immense complexity of language, none of the three models, IA, IP and WP, has worked out to be equally suitable for every part of a grammatical system in every language.
In chapter 8 'Some concepts in word-and-paradigm morphology' (Source: Foundations of Language 1 (1965): 268-289.), Peter H. Matthews takes up the baton and develops an explicit account of WP articulating, in particular, the notion of ''paradigmatic structure''.
In chapter 9 (Source: Journal of Linguistics 21 (1985): 321 -337.), Wolfgang U. Dressler elaborates a claim for the predictive (explanatory) power of Natural Morphology (NM). The first part of NM is a theory of universals; the second is of morphological typology; the third is of language specific system adequacy. Although it does not have a sociolinguistic or psycholinguistic theory of its own to account for norms and performances, however, it consistently refers to such theories.
In chapter 10 (Source: Foundations of Language 16 (1982): 181- 198.), Dieter Kastovsky outlines a functional approach to word- formation. He concedes that it is impossible to formulate a set of general rules that account for all word-formation patterns in a language. The dual functions of naming and syntactic recategorization affect word-formation in different ways, resulting in different outcomes depending on which of these functions is dominant.
Chapter 11 'Prolegomena to a theory of word formation' (Source: Linguistic Inquiry 4 (1973): 3-16.) is a study in morphology in early generative grammar. In this, Morris Halle proposes that morphology consists of three distinct components: a list of morphemes, rules of word formation, and a filter containing the idiosyncratic properties of words. The list of morphemes and the rules of word formation together define the potential words of the language. The set of actual words is obtained from that of the potential words by applying to the latter the modifications indicated in the filter. Morphology, thus, producing a long list of words designated as dictionary. Lexical insertion transformations select items from the dictionary and enter these in appropriate slots in structures representing the underlying constituent structure of particular sentences. It is to these underlying representations that the syntactic transformations apply and generate the surface structure. Contra Bresnan (1971, 1972), he does not accept any phonological rule applying as part of the transformational cycle of syntax.
Stephen R. Anderson's article 'Where's morphology?' (Source: Linguistic Inquiry 13 (4) (1982): 571-612.) is also a study in morphology in early generative grammar. In chapter 12, Anderson argues that in the domain ''inflection'', there is a nontrivial intersection between the theories of syntax and morphology. In order to capture both the relation of inflectional morphology to the syntax, and the exclusion of derivational morphology from ''syntactic accessibility'', he proposes a model of morphological operations, and provides the outlines of a theory of ''Extended Word and Paradigm'' morphology as an account of the mechanism of inflectional specification. He proposes principles governing the interaction of morphological rules, and also notes their interaction with ''phonological'' rules.
In chapter 13, 'A general theory of word structure' (Source: E. O.Selkirk, The Syntax of Words, Cambridge, Mass.: MIT Press, 1982, pp. 1-12.) Elisabeth O. Selkirk argues that word structure has the same general formal properties as syntactic structure. In generative grammar, as word-formation was considered as a part of syntactic transformations (Lees 1960), she argues that words, like sentences, have internal constituent structure, and they should be treated syntactically using phrase structure grammars.
In chapter 14, Stephen R. Anderson asks, ''How much structure do words have?'' (Source: S. R.Anderson, A-Morphous Morphology, Cambridge: Cambridge University Press, 1992, pp.256-291.). He denies any internal morphological structure for words and argues for a-morphous morphology in which word-internal structure is eliminated and replaced with conditions on the structure of the derivation of a word.
In chapter 15 'Autolexical syntax: a proposal for the treatment of noun incorporation and similar phenomena' (Source: Natural Language and Linguistic Theory 3 (1985): 379-439.), Jerrold M. Sadock puts forward a proposal for autolexical syntax, a theory in which morphology and syntax operate in a completely autonomous fashion and are held together by universal principles relating possible pairings of analyses sanctioned by each. One distinct advantage of an autolexical treatment of a polysynthetic language is that it allows a cross- componential analysis of phenomena such as noun incorporation while not requiring the mingling of syntax and morphology where there is no evidence for it.
In chapter 16, 'Distributed morphology and the pieces of inflection' (Source: Kenneth Hale and S. Jay Keyser (eds), The View from Building 20, Cambridge, Mass.: MIT Press, 1993, pp.111-176.), Morris Halle and Alec Marantz take a radical view and deny the existence of words, the lexical component and of lexical insertion. They propose a theory ''Distributed morphology'' (DM) (which adopts the basic organization of a principles and parameters grammar) and assume that it is only after all syntactic operations are completed that phonological expressions get inserted in a process called spell-out. The theory assumes underspecification of vocabulary items such that the phonological information of a word is not listed together with its matching morphosyntactic properties. The theory also assumes that syntactic and morphological objects have the same types of constituent structures. In particular, Distributed Morphology attempts to make precise the claim that all derivation of complex objects is syntactic. In respect to the interface between syntax and morphology, this architecture has a clear consequence: since the only mode of combination in the grammar is syntactic, it follows that in the default case, morphological structure simply is syntactic structure. Moreover, this is a comprehensive rejection of the Lexicalist Hypothesis which claims that words and word-formation belong in the lexicon (see ch. 11, 12 and 25).
Taking a different view, in chapter 17, Chet Creider and Richard Hudson elaborate an account of inflectional morphology in Word Grammar (Source: Lingua 107 (1999): 163-187.). In their approach the relevance of the word and the lexicon is accepted. They use ideas from the Word-and-Paradigm tradition - Lexeme, Stem and Inflection - in combination with the logic of default inheritance. They apply this theory to a range of different morphological data and compare the theory with the other contemporary approaches like a- morphous morphology, distributed morphology and network morphology.
In chapter 18 'On the separation of derivation from morphology: toward a lexeme/morpheme-based morphology' (Source: Quaderni di Semantica 9(1) (1988): 3-59.), Robert Beard presents another word- based approach. A key feature of this Lexeme/morpheme-based model is the Separation Hypothesis, which assumes that the derivation of meaning and the realization of phonological marking are distinct processes in word-formation. The function (grammatically relevant <>) of derivations is not predictable on the basis of affixation nor vice versa.
VOLUME II -- MORPHOLOGY: PRIMES, PHENOMENA AND PROCESSES The main concern of this volume is the nature of morphological primes, phenomena and processes. Beginning from defining a word, morphological phenomena like difference between inflectional and derivational morphology is discussed. 4 articles concerning issues in inflectional morphology, and a series of articles on derivational morphology dealing with issues in compounding, reduplication, clitics and particles etc. are the main topic of discussion.
In chapter 19, 'Qu'est-ce qu'un mot?' (Source: Knud Togeby, Travaux du Cercle Linguistique de Copenhague, Volume 5: Recherches Structurales, Copenhague: Nordisk Sprog-og Kulturforlag, 1949, pp. 97-111) Knud Togeby asks the question 'what is a word?' He grapples with the challenge of defining the word and identifying a set of robust criteria for word recognition, a theme also taken up in chapter 20 by John Lyons.
John Lyons in chapter 20 'The word' (Source: Introduction to theoretical linguistics (1968) pp 194- 208, Cambridge: Cambridge University Press.) gives a brief account of the primary units of grammatical analysis and the way in which the terms applying to them have been defined in modern linguistics. (For him, the relationship between sentences, clauses, phrases, words and morphemes can be expressed as a unit of 'higher' rank being composed out of units of 'lower' rank.) Next, morphological phenomena are discussed. Traditionally, the two major divisions of morphology are said to be inflection and derivation. What is the difference between the two?
In chapter 21 'Inflection' (Source: M. Hammond & M. Noonan (eds.) Theoretical Morphology: Approaches to Modern Linguistics, California: Academic Press, 1988, pp. 23- 44.), Stephen R. Anderson explains the properties of inflection, emphasizing its syntactically driven nature. Assuming that inflection is outside of derivation, he claims that material introduced by inflectional rule (not lexically) on the basis of properties assigned in the syntax to the morphosyntactic representation of the word presupposes, but is not presupposed by, material that is present in the lexical form. In particular, nonregular (hence lexical) morphology as well as material which is introduced not in response to the requirements of the syntax but for semantic or purely derivational reasons, may appear in derivational forms or compounds because it is in the lexicon.
Chapter 22 'Stems in Latin verbal morphology' (Source: Mark Aronoff, Morphology by itself: Stems and Inflectional Classes. Linguistic Inquiry Monograph Twenty-Two, Cambridge, Mass.: MIT Press, 1994, pp. 31- 59.) stays with the same theme. Focusing on the role of stems in Latin verbal inflectional morphology, Mark Aronoff examines the role of stems from a functional perspective in a word-based theory of morphology (cf. chs. 7, 10, 16, 17, 18).
Chapter 23 'Paradigm economy' (Source: Journal of Linguistics 19 (1983): 115-128.) treats another aspect of inflection: paradigms (cf. ch. 8). Andrew Carstairs discusses how, in any inflected language, the inflectional resources available in some word-class or part of speech are distributed among members of that word-class. He argues that inflectional class systems depend crucially on the distinction between suffixal inflection and stem alternations: stem alternations and affixal alternation interact in ways that favour certain patterns and disfavour or completely exclude others.
In chapter 24 'On rules of referral' (Source: Language 69(2) (1993): 449-479.), Gregory Stump puts forward a proposal (in another word- based approach) for a formal theory of rules of referral to account for syncretism in inflectional morphology. This theory affords a precise account of a range of rule interactions involving rules of referral. It furnishes a simple explanation for the fact that syncretisms donot always encompass whole words, for the fact that some referrals are bidirectional, and for the fact that two or more referrals may participate successively or simultaneously in the definition of a single instance of syncretism.
Then there follows a group of articles on derivational morphology.
In chapter 25 ''Remarks on Nominalization'' (Source: R. Jacobs & P. Rosenbaum (eds.) Readings in English Transformational Grammar, Waltham, MA: Ginn, 1970, pp. 184-221.), Noam Chomsky argues in favour of the Lexicalist Hypothesis which hypothesizes that syntactic operations only apply to syntactic constituents and hence cannot be employed in derivational morphology. Focusing on the contrast between derived nominals (which are created by derivational morphology in the lexicon), and -ing gerundives (which are the result of syntactic inflection), he claims that the former are similar to underived words and the latter are comparable to syntactic phrases. Word formation is firmly put in the lexicon; words are treated as syntactic atoms whose internal structure is unavailable to syntax.
The next two articles deal with compounding.
In chapter 26 'On the creation and use of English compound nouns' (Source: Language 55 (1977): 810-842.), Pamela Downing deals with the semantic properties of noun + noun compounding in English. She illustrates that the constraints on English N+N compounds cannot be characterized in terms of absolute limitations on the semantic or syntactic structures from which they are derived. Rather, the data examined here reflect tendencies for compounds to be based on permanent, non-predictable relationships of varying semantic types, depending on the nature of the entity being denoted.
Rochelle Lieber's article (ch. 27) 'Argument linking and compounds in English' (Source: Linguistic Inquiry 14(2) (1983): 251-286.) proposes an analysis of compounds with a Lexicalist framework. The Argument- linking Principle allows us to formulate a set of predictions about possible and impossible combinations of stems within each compound type and the productivity of the compounds in the language. She argues that a compound type containing an argument-taking stem will never be as productive as compound types containing no argument- taking stems. Synthetic compounds are analyzed by referring to the theta grid of the verb (cf. ch. 13).
In chapter 28 'Nimboran position class morphology' (Source: Natural Language and Linguistic Theory 11 (1993): 559-624.), a different kind of phenomenon, positional class morphology, is introduced. Sharon Inkelas explores the Papuan language Nimboran where verbal morphemes are inflexibly ordered and morphemes with the same ordering properties are in complementary distribution (cf. ch 42). He labels these verbal morphemes ''positional class morphemes'' and exemplifies new support for the theory of level-ordering. Examining a large corpus of data, he offers a formal theory of position class morphology.
In chapter 29 'Reduplicative constructions' (Source: Joseph Greenberg (ed.), Universals of Human Language, Vol. 3: Word Structure, Stanford: Stanford University Press, 1978, pp. 297-334.) Edith A. Moravcsik surveys the form, meaning and distribution of reduplicative constructions within and across languages. She comes to the conclusion that: i) phonological properties determining which part of a string be reduplicated in cases of partial reduplication are restricted to 'canonical form'-type properties like consonantality, vowelhood, and linear precedence among the segments and boundaries. ii) languages usually use reduplicative patterns - i.e. quantitative form differentiation - for the expression of meanings that have something to do with the quantity of referents. iii) reduplicative constructions express a more specific meaning than their unreduplicated counterparts.
This is followed by chapter 30, which deals with metathesis.
In chapter 30 'Metathesis as a grammatical device' (Source: International Journal of American Linguistics 35(3) (1969): 213-219.), Laurence C. Thompson and M. Terry Thompson discuss metathesis in the Austronesian language Rotuman and the Straits Salish language Clallam.
The final chapter in this volume ends with chapter 31 'Clitics and particles' (Source: Language 61(2) (1985): 283-305.). In this, Arnold M. Zwicky investigates typological properties of clitics and particles, phenomena whose existence complicates decisions about the status of the word. He distinguishes clitics from inflectional affixes and independent words, and for this purpose he makes use of various tests like phonological tests, syntactic tests, accentual test etc.. He shows that in many cases items labeled 'particles' have been treated as clitics. As most of the 'particles' in the literature are simply words, he argues that treating words with idiosyncratic distributions as acategorial 'particles' is wrong. Finally, he depicts a class of discourse markers: a grammatical category of items which are often classified as 'particles' but which turn out to be independent words rather than clitics of any sort.
VOLUME III -- MORPHOLOGY: ITS RELATION TO PHONOLOGY This volume focuses on the relation of morphology to phonology. Contra the doctrine of separation of levels, in which sound structure used to be described before grammatical structure, hence separating grammatical information when performing phonemic analysis, a series of articles in this volume pioneer the idea that morphological and phonological analysis can be intertwined in a linguistic description.
Chapter 32 is Leonard Bloomfield's classic paper on 'Menomini morphophonemics' (Source: Travaux du circle linguistique de Prague 8 (1939): 105-115.), in which he demonstrates that some phonological alternations require a morphological trigger. Simple words and the members of compounds in Menomini can be analyzed into morphologic elements which vary greatly in different combinations. He distinguishes these morphophonemic alternations from morpholexical variations and describes the internal sandhi or morphophonemics of the language. He sets up each morphological element in a theoretical basic form, and then states the deviations from this basic form which appear when the element is combined with other elements.
In chapter 33 'A problem in phonological alternation' (Source: Language 15 (1939): 1-10.) also, Morris Swadesh and CF. Voegelin discuss that phonological alternations are motivated by morphological factors (also see ch 6 for the discussion of IP). They consider Tübatulabal, a Uto-Aztecan language of California, as a striking illustration to account for irregular or ''non-patent'' phonological alternations.
Two decades later, writing on phonotactics in early generative grammar, Morris Halle discusses the limitations which the language places on the occurrence of distinctive feature complexes in the sequence. In chapter 34 'Sequential constraints' (Source: Morris Halle, The Sound Pattern of Russian, The Hague: Mouton, 1959, pp. 55-75.), he has shown that there are phonological regularities governing the phonological realization of morphemes. Phonological representations contain a minimum of specified features and the automatic distribution of features is governed by the following three types of rule: i) the morpheme structure rules deal exclusively with the feature composition of individual morphemes. ii) the morphological rules, which are part of transformational level, require reference not only to the feature composition of morphemes but also to the morpheme class to which the latter belong. iii) the phonological rules assign features on the basis of purely phonological criteria; they require reference only to features and phonological boundaries.
Halle's idea was further articulated by Richard Stanley in 'Redundancy rules in phonology' (Source: Language 43(1) (1967): 393-436.), in his proposal for a theory of lexical redundancy (ch 35). How should the linguist deal with those phonological properties of words and morphemes that are entirely predictable? That is the question that Stanley set out to answer. He gives particular attention to the formal nature of morpheme structure rules and to the use of blanks in representing redundancies. However, later he replaces morpheme structure rules with a new formal device called 'morpheme structure conditions' and provides strong motivation for the preference of the latter one.
The fact that certain morphological rules have a close affinity with phonological rules goes back to the proposal by (Whitney 1889, Bloomfield 1933). Such interaction between morphology and phonology got articulated by Paul Kiparsky as the theory of Lexical morphology and phonology (chapter 36) (Source: The Linguistic Society of Korea (ed.), Linguistics in the Morning Calm, Seoul: Hanshin Publishing, 1982, pp. 3-91.). Kiparsky proposes that in this theory, word-formation takes place in the lexicon. The lexicon is stratified on the basis of the properties of various affixes. Morphological rules and phonological rules belonging to the same stratum apply in tandem.
Kiparsky's approach, although insightful it was, had its own drawbacks, many of them stemming from the fact that affix-driven stratification all too often did not yield reliable results. Hence, Heinz J. Giegerich in chapter 37 'Principles of base-driven stratification' (Source: Heinz J Giegerich, Lexical Strata in English, Cambridge: Cambridge University Press, 1999, pp. 53-99.) mounts a rescue for lexical phonology and morphology in the form of a revised theory with base-driven rather than affix-driven lexical stratification. This model predicts that the 'Continuity of Strata Hypothesis' must be true on both the phonological and the morphological side.
In chapter 38 'A prosodic theory of nonconcatenative morphology' (Source: Linguistic Inquiry 12(3) (1981): 373-418.), following Harris's (1941, 1951) notion of long components, John McCarthy provides a more successful approach to the description of nonconcatenative morphology which employs CV skeletal tier positions as templates for canonical morphological forms. This is an extension to morphology of Goldsmith's autosegmental phonology model. McCarthy justifies this theory by an analysis of the formal properties of the system of verbal derivation and aspect and voice inflection in Classical Arabic. He also discusses reduplication and the extension of this treatment to non- Semitic languages.
In chapter 39 'Théorie de l'apophonie et organization des schèmes en sémitique' (Source: Jacqueline Lecarme, Jean Lowenstamm and Ur Shlonsky (eds), Research in Afroasiatic Grammar, Papers from the Third conference on Afroasiatic Languages, Sophia Antipolis, 1996, Amsterdam: John Benjamins, 2000, pp. 263-299.), Philippe Ségéral explores the regularities that lie beneath the surface in the apophony exhibited as part of the nonconcatenative morphology of Semitic languages. In an analysis inspired in part by government phonology and in part by autosegmental phonology he shows that there exist phonological regularities in vowel alternations across tense-aspect paradigms in Akkadian.
In chapter 40 'Re reduplication' (Source: Linguistic Inquiry 13(3) (1982): 435-482.), Alec Marantz presents an account of reduplication that is based on McCarthy's proposals for template morphology. He claims that reduplication is best analyzed as the affixation of a skeletal morpheme to a stem. Such an analysis explains the otherwise puzzling interaction of reduplication with certain phonological processes. The fact that reduplication processes can generally be characterized by a fixed consonant-vowel shape, a fact captured in the identification of reduplicating morphemes as C-V skeletal, provides considerable support for McCarthy's autosegmental representation of words on different tiers including a phonemic melody and a C-V skeleton. Given that reduplication is simply affixation, this article also supports Halle's (1979) interpretation of the phonological cycle and Lieber's (1980) morpholexical theory dealing with the interaction of reduplication and phonological processes.
In chapter 41, 'Faithfulness and identity in prosodic morphology'(Source: Rene Kager, Harry van der Hulst and Wim Zonneveld (eds), The Prosody- Morphology Interface, Cambridge: Cambridge University Press, 1999, pp. 218-309.), focusing on reduplication, infixation, root-pattern morphology and constraints on the canonical shape (e.g. of the minimal word) McCarthy and Prince use Optimality Theory to argue for a Prosodic Morphology approach, which claims that morphological templates are defined in terms of authentic units of the prosodic hierarchy (e.g. syllable, foot).
[Part 2 of this review appears in a subsequent issue http://www.linguistlist.org/issues/16/16-1352.html see the Editors' note at the beginning of this Part.]
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Kalyanamalini Sahoo has extensively worked on morphosyntactic investigations in the South-Asian language Oriya, including applicational fields like computational morphology. She received her Ph.D. from Norwegian University of Science & Technology, Trondheim, Norway, in the year 2001 and is primarily interested in computational morphology and syntax.