LINGUIST List 28.4484

Thu Oct 26 2017

Review: Achinese; Kinyarwanda; Sundanese; Zulu; Quechuan; Phonology; Typology: Bennett (2015)

Editor for this issue: Clare Harshey <clarelinguistlist.org>


Date: 15-Jun-2017
From: Amanda Dalola <dalolamailbox.sc.edu>
Subject: The Phonology of Consonants
E-mail this message to a friend

Discuss this message


Book announced at http://linguistlist.org/issues/26/26-1766.html

AUTHOR: Wm G. Bennett
TITLE: The Phonology of Consonants
SUBTITLE: Harmony, Dissimilation and Correspondence
SERIES TITLE: Cambridge Studies in Linguistics 147
PUBLISHER: Cambridge University Press
YEAR: 2015

REVIEWER: Amanda Dalola, University of South Carolina

REVIEWS EDITOR: Robert A. Coté

SUMMARY

“The Phonology of Consonants: Harmony, Dissimilation, and Correspondence” by William G. Bennett provides a novel optimal theoretic analysis of the phenomenon of dissimilation (the avoidance of similar sounds/segments) by reconceptualizing it instead as the avoidance of surface correspondences. Calling on recent work in Agreement by Correspondence (ABC), this title successfully demonstrates that dissimilation is a possible outcome of the Surface Correspondence Theory (SCT). The book first develops the SCT by articulating an exhaustive list of its predictions for dissimilatory processes and then tests them in a series of case studies featuring data from typologically diverse languages. The discussion culminates in a typological survey of 154 dissimilation patterns taken from a sample of 134 languages, which can be accessed in a downloadable and searchable electronic database hosted on the book’s accompanying website.

Chapter 1, “Introduction,” develops the previous research on the ABC theory (Walker, 2000a, 2000b, 2001; Hansson, 2001/2010, 2007; Rose & Walker, 2004) and presents the main tenet of the book—that dissimilation should be viewed not merely as the avoidance of similarity but rather as the response to more stringent conditions attached to it. It reasons that harmonizing consonants are required to correspond only because they are first similar in some way; dissimilar consonants, then, remove the need to correspond because they were never similar in the first place. This rationale is used to show that consonant harmony is rooted in similarity rather than proximity, which explains its ability to act at a distance.

Chapter 2, “The surface correspondence theory,” provides an optimal theoretic account of the author’s Surface Correspondence Theory of Dissimilation (SCTD), which serves as the basis of the analysis. Based on ABC, the essence of the SCTD is that a correspondence relation holds over the consonants in an output form, and that these correspondences are evaluated through two types of markedness constraints: CORR constraints, which assign violations for non-correspondence between consonants which share some particular feature, and CC-Limiter constraints, which assign violations when there is correspondence between consonants that do not meet some further condition specified by each constraint. The text goes on to visualize the complete list of possible interactions resulting from various types of correspondence, yielding forms that may exhibit assimilation, assimilation + harmony or dissimilation.

Chapter 3, “Kinyarwanda: the effects of domain edges, and the adequacy of a single SCorr relation,” is a case study of the two long-distance consonant interactions found in the language: sibilant harmony (retroflexion agreement within the stem) and Dahl’s Law (voiceless voiced stop dissimilation in prefix if stem begins with voiceless stop). Using the same concept of correspondence, the data is able to account for both processes: sibilant retroflexion harmony can be accounted for by ABC, while Dahl’s Law dissimilation can be explained as avoidance of penalized correspondence.

Chapter 4, “Sundanese: complementary assimilation and dissimilation,” is a case study of Sundanese’s complex pattern of interlocking [r]~[l] alternations: L-assimilation turns /r/ into an infix [l] when following a root-initial /l/, while R-dissimilation turns an infix /r/ into [l] when it precedes an/r/ in the root. In terms of position, L-assimilation can only happen if /r/ and /l/ are the onsets of two adjacent syllables, while R-dissimilation can happen only if two /r/s are not onsets of adjacent syllables. Correspondence accounts for these phenomena as follows: where it is possible, L-assimilation can be explained by ABC, and where it is not, R-dissimilation can be explained as an avoidance of penalized correspondence.

Chapter 5, “Quechua and Obolo: the role of syllable edges,” presents a parallel case study of Quechua laryngeal dissimilation (between epenthetic glottal stops and glottalized consonants) and Obolo nasal agreement (between nasals and stops in the same syllable). The phenomena are discussed in tandem because they are both analyzed as arising from the same surface correspondence relation interacting with other language-specific limits on phonotactics. While Obolo represents a one-direction (R-to-L) harmony pattern, it does so uniquely through the interaction of faithfulness constraints, without the need to posit any constraints on directionality. The examination of Quechua reveals that the SCTD is also able to handle indiscriminate dissimilation patterns that are not overtly edge-conditioned, by using CC-Limiter constraints to prohibit initial correspondence.

Chapter 6, “Chol and Ponapean: complete identity effects,” begins by examining the Complete Identity Effect (CIE), in which consonants participate in an identical-or-else-dissimilated system (Suzuki, 1999). The majority of the chapter is dedicated to analyzing the CIE as it surfaces in Chol, where two ejectives are tolerated in the same root only if they are identical. The discussion then pivots into one of identity in just some respects, or the Sufficient Identity Effect (SIE), taking as an example the phenomenon of labio-velar agreement in Ponapean. In this laxer, agreeing-or-else-dissimilated system, consonants must only be partially identical to be excused from dissimilating. The SCTD accounts for both of these systems via the free ranking of CC-IDENT constraints.

Chapter 7, “ Zulu labial dissimilation,” presents a case study of long-distance dissimilation occurring in passive verb forms, in which a root containing a medial or final [labial] consonant is combined with the passive suffix /-w/, causing the underlying root labial to surface as a (pre-)palatal consonant. The SCTD is able to account for this process via the interaction of CORR constraints and CC-Limiter constraints. The chapter also evaluates several alternative analyses based on the Obligatory Contour Principle (OCP), a prohibition against the co-occurrence of similar consonants. Because the segment-level implementations of the OCP predict too much dissimilation, and autosegmental OCP makes dissimilation too restricted, the author concludes that the SCTD must not be viewed as a subset of the OCP, but rather as a separate and superior theory for accounting for the attested patterns of dissimilation in Zulu.

Chapter 8, “Segmental blocking effects in dissimilation,” explains the SCTD’s ability to account for blocking effects, by calling on data from liquid dissimilation in Yidiny, Latin and Georgian to demonstrate two plausible interpretations of blocking patterns: blocking-by-bridging and double assimilation. In blocking-by-bridging, dissimilation fails because extraneous correspondence with the blocker segment resolves violations of the Limiter constraint, while in double assimilation, the different types of CC-Limiter constraints and different domains of CORR constraints interact to produce double dissimilation systems with various properties. The chapter ends by pointing out the unresolved nature of the relationship between segmental blocking and intervention, and proposes its solution may lie in a more articulated theory of directional asymmetries.

Chapter 9, “Typological survey of dissimilation,” presents an observed typology of long-distance dissimilation, compiled from a cross-linguistic survey of 154 potential cases and informed by an additional 100+ other reported dissimilatory patterns not technically viewed as cases of synchronic long-distance dissimilation. Six main typological findings are found: (1) the typology of dissimilation is unexpectedly limited and asymmetrical, (2) the mismatch prediction is borne out clearly for structural factors, (3) the mismatch prediction is borne out less clearly for features. (4) segment-adjacent and long-distance dissimilation are not the same, (5) dissimilation is not the reverse of assimilation, and (6) dissimilation is not about markedness.

Chapter 10, “Closing remarks,” concludes the work by restating its main premise: the SCTD is able to derive dissimilation from the same surface correspondence relation posited for long-distance consonant harmony; it is based on a more precise formalization of the surface correspondence relation as an equivalence relation, which partitions surface consonants into correspondence classes. It is unlike previous/competing theories, e.g. the OCP, because it does not need to posit any constraints that explicitly forbid similar consonants. The chapter ends with a discussion of three unresolved issues of the SCTD: segment-adjacent dissimilation, directionality, and correspondence structure without the occurrence of alternations.

EVALUATION

“The Phonology of Consonants: Harmony, Dissimilation, and Correspondence” is a well-argued, innovative and unparalleled treatment of consonant dissimilation as explained by the interaction of correspondence relations. Its progression builds logically and is easy-to-follow, and the structure within each chapter is lucid and predictable. An added benefit to this collection is that any of the case study chapters are coherent on a near-stand-alone basis (together with the theoretical prefacing of the SCTD in Chapter 2), should one choose to select just one for study.

Perhaps the only area of this treatise that requires more explanation is certain details surrounding the compilation of the typological survey of dissimilation. The twelve factors under examination are defined in Chapter 9, but not all of the descriptions make it clear how each categorization might have been determined. For example, the plausibility category is meant to gauge whether or not a described case of dissimilation both exists in the synchronic grammar as it has been reported and qualifies as an instance of long-distance dissimilation. One situation given to exemplify non-plausibility is that of a historical alternation that wasn’t very widespread diachronically, and is no longer attested synchronically (Modern Greek rhotic dissimilation). The absolute nature of this example is fairly straightforward, but nowhere is it explained if all examples marked with the “diachronic” label in the database of excluded cases represent this same statistical tendency. Assuming that they do not and that some processes may still be weakly present in their respective synchronic grammars begs the question of how widespread a phenomenon must be in order to be considered worthy of a theoretical account. Other reasons for non-plausibility are given in this category, which the author himself openly flags as having been “assessed subjectively” (p. 328), but the presence of a singular soft spot in an otherwise methodologically rigorous account is somewhat unexpected.

A great strength of this book lies in its ability to ground its in-depth theoretical analysis in an expansive collection of real-word data that illustrates different types of dissimilation phenomena in typologically diverse languages. That the typological supplement and dissimilation database have been made accessible on the publisher website is an added bonus for educators looking to quickly illustrate the theory in a lesson or problemset, or corpus linguists looking to find quantitative tendencies among the set of included or excluded cases. As such, this book represents an excellent tool for introducing and reinforcing both theoretical concepts and empirical practices in the discussion of dissimilation and assimilation.

Aimed at those with a fluency in OT and an interest in assimilatory and dissimilatory processes, “The Phonology of Consonants: Harmony, Dissimilation, and Correspondence” is a comprehensive and game-changing addition for phonologists and advanced students working within the OT framework.

REFERENCES

Hansson, G.O. (2001). Theoretical and typological issues in consonant harmony. PhD
thesis, University of California, Berkeley.

Hansson, G.O. (2007). Blocking effects in agreement by correspondence. Linguistic
Inquiry, 2, 395-409.

Hansson, G.O. (2010). Consonant Harmony: Long-Distance Interaction in Phonology.
Berkeley, CA: University of California Press. Revised version of Hansson (2001).

Rose, S. & Walker R. (2004). A typology of consonant agreement as correspondence.
Language, 80, 475-531.

Suzuki, K. (1999). Identity avoidance vs. identity preference: The case of Sundanese.
Presentation at the annual meeting of the LSA. Los Angeles, January 7.

Walker, R. (2000a). Long-distance consonantal identity effects. In Proceedings of
WCCFL, 19, 532-545.

Walker, R. (2000b). Yaka nasal harmony: Spreading or segmental correspondence?
In Proceedings of BLS, 26, 321-332.

Walker, R. (2001). Consonantal correspondence. In Proceedings of the Workshop on
the Lexicon in Phonetics and Phonology, 73-84. Edmonton: University of Alberta.


ABOUT THE REVIEWER

Amanda Dalola is an Assistant Professor of French and Linguistics at the University of South Carolina. Her research interests include phonetics, sociophonetics, phonology, lab phonology, history of French and technology in the L2 classroom.

Page Updated: 26-Oct-2017