LINGUIST List 28.3565

Tue Aug 29 2017

Review: Phonology: Muller, Ball (2016)

Editor for this issue: Clare Harshey <>

Date: 29-Aug-2017
From: Reza Falahati <>
Subject: Challenging Sonority
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Martin J. Ball
EDITOR: Nicole Muller
TITLE: Challenging Sonority
SUBTITLE: Cross-linguistic Evidence
SERIES TITLE: Studies in Phonetics and Phonology
PUBLISHER: Equinox Publishing Ltd
YEAR: 2016

REVIEWER: Reza Falahati, University of Ottawa



“Challenging Sonority: Cross-Linguistic Evidence”, edited by Martin J. Ball and Nicole Müller, is a collection of twenty articles focusing on one of the key topics in linguistics which has a long tradition of controversy in the field. While two of the articles provide a review on sonority related to natural language and acquisition, and two other contributions use sonority as a background, most examine sonority and the related principles in under-studied languages as well as speech from typically developing children and those with speech disorders.

Sonority has been mainly defined as the intensity or constriction degree/aperture related to a segment. The following hierarchy is proposed for ranking sounds in terms of their sonority:
vowels > glides > liquids > nasals > fricatives > stops. According to this, vowels are the most sonorous segments and stops are the least, with glides, liquids, nasals, and fricatives in between.

There are a number of principles being used to explain different phenomena in sonority-related studies. For example, the Sonority Sequencing Principle (henceforth, SSP) determines the order of consonants in the onset or coda position of a syllable by the sonority hierarchy. This principle requires that sonority must decrease from the nucleus to the edges of the syllable (i.e., onset and coda). The Sonority Dispersion Principle (SDP) posits that in a sequence of onset consonant clusters and vowel (e.g., C1C2V), the consonants and vowel should be maximally and evenly separated from each other. This means that a sequence of /pla/ is preferred over /pja/ since liquids are almost the midpoint between plosives and vowels on the sonority scale. This principle also requires that the rise of sonority from onset to nucleus be steep, and the decline from nucleus to coda be shallow.

This book starts with Chapter 1, “Sonority in Natural Language: A Review”, by Joan Rahilly, which provides a critical review of the core principles in sonority theory in natural language varieties from both phonetic and phonological perspectives. These include correlating sonority with either vocal tract configuration or loudness (e.g., Green 2003; Proctor & Walker 2012) or using a set of OT-related markedness and faithfulness constraints (e.g., Prince & Smolensky 2004; Steriade 1990). The circularity of the formalism used in OT and the categorical nature of the scales are the main points emphasized by Rahilly.

The next two contributions provide a cross-linguistic study and simply use sonority as a background. In Chapter 2, “Sonority and the Unusual Behavior of /s/”, Heather Goad examines /s/-clusters in the syllabification systems of three unrelated languages, namely Acoma, Blackfoot, and Ōgami, to see how the unusual behaviour of /s/ in Indo-European languages is displayed in these non-Indo-European languages. She states that in order to analyze the diverse phonotactic behaviours of /s/ across Indo-European and non-Indo-European languages, /s/ needs to be outside of its sonority class and we need to adopt a syllable representation for it which “functions as the coda of an empty-headed syllable in Indo-European, as an onset followed by an empty nucleus in Acoma, and as nuclear in Blackfoot and Ōgami” (p. 26).

In Chapter 3, “Relating the Sonority Hierarchy to Articulatory Timing Patterns: A Cross-linguistic Perspective”, Joana Chitoran investigates the articulatory data (using EMA) from three languages (i.e., Georgian, Slovak, & Tashlhiyt) having different syllabic organization. Georgian allows only vowels as the syllable nucleus, Slovak allows both vowels and liquids, and Tashlhiyt allows all vowels and consonants to be syllabic. The author argues that the variation existing in syllable structures across languages can be best explained by referring to language-specific properties of articulatory timing between adjacent gestures. This approach could explain both the patterns which are in agreement with the sonority hierarchy as well as the ones which show rankings opposite to the SSP. What this article has in common with earlier studies is that Chitoran still shares the idea of articulatory motivation behind the sonority hierarchy, but she puts the focus on the gestural timing coordination between articulatory gestures rather than “degree of constriction” as suggested by other researchers (e.g., Green 2003).

The next eleven chapters deal with sonority and the related problems in a range of different languages. Brent Ernest Archer, in Chapter 4, “Sonority in Zulu”, a somehow different work, uses a statistical method on a corpus of Zulu to test whether there is a “tendency” in such a non-Indo-European language to sequence segments in onset and nucleus in a way that there is a maximum sonority difference between them. In order to operationalize the distribution of the syllable’s transition score, the researcher uses the following sonority sequencing proposed by Ball, Müller, and Rutter (2010): vowels: 6, glides: 5, liquids: 4, nasals: 3, fricatives: 2, stops: 1.
Based on the general results, Archer claims that the SSP cannot account for the syllable formation in human languages. The question which still stays here is how to deal with “marked” sounds such as ejectives, implosives, and clicks in Zulu and other languages with different sound sources.

Matthew Faytak in Chapter 5, “Sonority in Some Languages of the Cameroon Grassfields”, investigates the unusual phonotactics in Kom and Limbum, two Bantu languages, spoken in south-western Cameroon. Faytak’s main motivation to focus on these languages stems in having a set of peculiar obstruents in these languages called “fricativized vowels”. The unusual behaviour of this set is that they are licensed to take nucleus position to the exclusion of most of the sonorant consonants. The author initially classifies these segments as syllabified obstruents in order to show their unexpected behaviours in regard to the SSP. However, later on he provides acoustic support (e.g., visible formants) which mainly conforms to the vocalic nature of these segments. The author suggests that the principle behind the SSP should be based on a multiple spectral parameters rather than constriction degree only.

In Chapter 6, “An Investigation of Sonority Theory in Mandarin Chinese”, Li Qiang examines the segments frequency and distribution of onset consonants and vowels in a corpus of Mandarin Chinese to test the SSP predictions regarding the preferred syllabic shape. Consonant-glide combinations are considered to be a single phoneme in Mandarin. The results show that the frequency of high vowels in his corpus was the highest followed by low vowels and mid vowels, respectively. Adopting Clement’s (2009) sonority ranking for English vowels (i.e., low vowels > mid vowels > high vowels), Qiang argues that the prediction made by the SSP is not born out. It could be that the small size of the corpus (2,500 out of 87,019 possible characters) in this study may not be a good representative of the Chinese characters.

In Chapter 7, “Sonority and Syllabification in Casual and Formal Mongolian Speech”, Anastasia Karlsson and Jan-Olof Svantesson examine the SSP in two varieties of Mongolian: casual and formal speech. The syllabic structure of the language allows three consonants in the coda position. However, when these combinations violate the SSP, the two varieties use different strategies to preserve the morphological structure of words. Voice assimilation and stop lenition are the two major strategies that are used to preserve the sequence of consonants that is formed as a result of suffixation in casual speech. However, in the formal register, schwa epenthesis is the dominant strategy used to avoid violation of the SSP. Karlsson and Svantesson mainly argue that the use of these strategies in both formal and casual speech is to make “consonant clusters pronounceable”. The results of this study could also be viewed by referring to the temporal organization in articulatory gestures (see Chitoran, this volume) and the speech rate difference in careful and casual speech. The rate of production is proved to influence gestural orchestration and consequently the type and rate of deletion and other processes across languages (Barry 1985; Byrd 1994, 1996).

Samira Farwaneh in Chapter 8, “Sequential Constraints on Codas in Palestinian Arabic”, deals with the coda consonant clusters in Palestinian Arabic (PA). This language allows both complex onsets (CCVC) and complex codas (CVCC). When the sequence of the consonants violates the well-formed coda condition, vowel epenthesis is triggered to avoid illegal codas. Moreover, PA applies the vowel insertion to the coda consonant sequences where the SSP is not violated. In addition to cases like this, vowel insertion is applied in words where there is a sonority plateau (e., stop-stop, fricative-fricative) between the coda consonants in specific order. In order to account for the phonotactics of coda consonant sequences in PA, Farwaneh argues that in addition to the SSP, Obligatory Contour Principle (OCP) and markedness constraints are needed. Yip (2013) has argued that the manner and place of articulation of both C1 and C2 in addition to their order could affect the perceptibility and gestural overlap of the adjacent segment.

Marie Klopfenstein in Chapter 9, “Exceptions to the SSP: Evidence from Ottawa for a Metatheoretical Approach”, uses data from Ottawa, an endangered Algonquin language, to examine the predictions made by the SSP. Using Ball, Müller, and Rutter’s (2010) model, she presents words with onset clusters which adhere to the SSP; however, the fricative-stop and nasal-obstruent sequences are also presented as the consonant clusters which violate the SSP. Klopfenstein tries to examine different explanations proposed by phonologists when exceptions, like the ones in Ottawa, happen. Merging two classes into one, as Clements’ (1990) model does for stops and fricatives as obstruents, is one explanation. She argues that although merges like this could result in less SSP violation, they may lose some of their explanatory power in return. The author highlights the specific role of morphology in the SSP rankings since they could convey important information in a language.

Sonya Bird and Ewa Czaykowska-Higgins in Chapter 10, “Parsing Salish Consonant Clusters”, present an acoustic study on consonant sequences in Nxaʔamxcín, a Salish language, which allows two to four consonants root-initially as obstruent-obstruent, obstruent-sonorant, sonorant-obstruent, and sonorant-sonorant. Among these clusters, both root-initial and root-final obstruent-obstruent sequences are considered a special sequence where schwas are not inserted to break the sequence into smaller ones. Moreover, the root-initial obstruent-sonorant sequences behave differently from root-final sonorant-obstruent sequences by having more tendency for schwa insertion in the former context. In addition to the results regarding schwa insertion between two consonants, this study shows that schwa could also be inserted after the clusters if they are followed by a suffix-initial sonorant. This shows that the manner of articulation of the segment following (or preceding) the cluster is quite significant in this respect. Bird and Czaykowska-Higgins conclude that the SSP plays a major role in parsing the consonant clusters in Nxaʔamxcín. However, in order to get a fuller picture of the phonotactics of the consonant clusters in this language, an “interaction of phonological constraints on syllable structure, perceptibility, and articulatory timing” is needed (p. 194).

In Chapter 11, “Sonority in Gitksan”, Jason Brown deals with Gitksan, an Interior Tsimshianic language in British Columbia, Canada. This language shows very low-ranked constraints on sequencing stops and fricatives adjacent to each other in both initial and final position in different orders; however, this is not true for sonorants which cannot be next to each other. The question here is why this language allows sonority plateau (e.g., fricative-fricative) or even sonority fall (e.g., fricative-stop), but does not tolerate stop-sonorant sequences. Brown provides a perceptual account to explain such a difference.

Yolanda Rivera-Castillo in Chapter 12, “Syllable Structure in Papiamentu and the Sonority Scale”, examines the syllable structure in Papiamentu. The coda and onset consonant clusters in the language show specific behaviours: sonorants cannot be adjacent to other sonorants. Two obstruents can be next to each other if both are voiceless and have a different feature for [continuant]. Rivera-Castillo argues that the phonotactic constraints for consonant clusters in Papiamentu are governed by melodic properties and some other related constraints such as Minimal Sonority Distance & Syllable Contact Law rather than the SSP. She states that these constraints are correlated to perceptual cues.

Chiara Meluzzi in Chapter 13, “A New Sonority Degree in the Realization of the Dental Affricates /ts dz/ in Italian”, investigates different realizations of dental affricates in the Italian variety of Bolzano. The Standard Italian language contains both voiceless and voiced dental affricates /ts/ and /dz/, while German has only voiceless affricates /pf/ and /ts/. The acoustic results of the study show a new affricate variant (almost 25% of the tokens) labelled as “mixed” where a voicing distinction was observed between the plosive and the fricative parts of the affricate. The results show that there is an interaction between “mixed” affricate realizations and social factors. The presence of a gap between the occlusive and the fricative parts in the “mixed” affricates is regarded as an evidence for the in-between status of mixed affricates. The point which remains unclear here is the way the author has interpreted affricates like /ds/ having a new sonority degree. It is not obvious whether Meluzzi is treating the emerging affricate as a new phoneme or a combination of a plosive and a fricative.

Martin J. Ball and Nicole Müller in Chapter 14, “Sonority and Initial Consonant Mutation in the Celtic Languages”,examine the initial consonant mutation (ICM) in three Celtic languages (i.e., Welsh, Breton, and Irish) to see how this sound change conforms to the predictions made by the SSP and the SDP. Their findings for Welsh show that all the onset clusters with two consonants (except /s/ + C sequence) respect the SSP. The 3-consonant clusters mainly violate the SSP. In fact, this is mainly due to authors’ interpretation of /gw/ as a plosive-glide sequence rather than a labialized velar. In general, mutation reduces sonority distances between the initial consonant(s) and the following vowels in all three languages. Ball and Müller claim that such a phonological change does not support the predictions made by the SDP.

The next five chapters (i.e., 15-19) explore language acquisition and impaired speech in the context of sonority. Jessica A. Barlow in Chapter 15, “Sonority in Acquisition: A Review”, evaluates the assumptions made by the SSP by reviewing the studies on syllable structure acquisition by adults and children with speech delay. According to the reviewed studies, CV syllables are acquired first, followed by CVC and CCV(C)/CVCC. Obstruent onsets are acquired before sonorant onsets by children. This supports the SSP since obstruent-nucleus provides a maximal rise in the onset sonority compared to sonorant-nucleus (Steriade 1993). However, the acquisition of either obstruents or sonorants in the coda position does not support the SSP predictions. Barlow states that phonotactic restrictions as well as different structural representations should be considered to provide a full picture of syllable acquisition.

Mehmet Yavaș and Elena Babatsouli in Chapter 16, “Acquisition of /s/ Clusters in a Greek-English Bilingual Child: Sonority or OCP?”, in a longitudinal study address the acquisition and reduction of #sC clusters in a Greek-English bilingual child. The main goal of this research is to test the opposite predictions made by the SSP and the OCP. According to the SSP, the /s/ + stop clusters are more marked than /s/ + nasal/liquid/glide clusters, hence they are expected to have lower accuracy in children’s production than the less marked clusters. The OCP[continuant], however, predicts that the /s/ + liquid/glide clusters are more marked than /s/ + stop/nasal clusters and children are expected to produce them less correctly than the unmarked clusters. The results show that /s/ + stop/nasal clusters appear earlier in the production of the child than other clusters. This supports the predictions made by OCP [continuant].

In Chapter 17, “The Influence of Sonority on Cluster Acquisition by Egyptian Arabic Children Aged Two to Three Years”, Mona Maamoun investigates the acquisition of clusters in CVCC Egyptian Colloquial Arabic (ECA) words by typically developing children aged 2-3, to examine the predictions made by the SSP. The general results of the study showed that the highest rate of correct responses happened in clusters with a sonority plateau, followed by clusters respecting the SSP.

Chapter 18, “Sonority and Cluster Reduction in Typical and Atypical Phonological Development in Farsi”, by Froogh Shooshtaryzadeh, investigates the word-final and word-medial (i.e., abutting) clusters in 10 Farsi-speaking children with a functional phonological disorder (PD) and typically developing children (TD). She considers different phonological processes such as cluster reduction, metathesis, fusion, and voicing/devoicing in the production of PD and TD children to see how the SSP or Articulatory Ease Principle (AEP) can account for developing phonologies. The results show that the reduction of coda clusters is mainly motivated by AEP and/or Continuity Preservation Property (CPP), a constraint which favours the deletion of the final segment in a cluster. However, the extent of influence of each factor varies across PD and TD groups.

Martin J. Ball, Nicole Müller, and Chris Code in Chapter 19, “Sonority and Aphasia”, first review some studies dealing with paraphasic and jargonaphasic data to see how sonority could account for such impaired speech. The authors show that the patterns found in subjects’ errors are against the predictions of both sonority theory and the SDP. They then focus on studies of lexical and non-lexical speech automatisms in English, German, and Cantonese. The data from English and German were analyzed both in terms of consonant cluster phonotactics and syllable structure while the analysis of the data on Cantonese was restricted to syllable structure. The results show that the preferred syllable for all languages is CV. As for the consonant classes the C position is primarily filled by obstruents, followed by nasals. Contrary to the results for paraphasia and jargonaphasia, the nonword/non-lexical automatism results support the SSP. In general, it is argued that sonority should be considered as a tendency rather than a hard-wired phonological law. The authors conclude that the alternative interpretation of the same phonetic change in favour of or against the SSP justifies such speech disorders at the level of phonological organization rather than phonetic planning and implementation.

Mark J. Jones in the last chapter, “Motivating and Explaining the Structure of Segment Sequences”, reviews some studies which list the problems with sonority as an efficient factor in explaining segmental sequences and syllable structures. These include the circularity in its definition (Ohala 1992), the problem of intensity dealing with allophones, the extra-syllabic nature of some segments, and the diverse syllabification of the same set of segmental sequences across different languages (e.g., V.CCV vs. VC.CV). Jones argues that an acoustic-auditory recoverability approach, by referring to signal modulation and auditory dimension, can account for the attested patterns in a better way than the SSP.


The main goal of this edited volume is to investigate sonority and some related principles (e.g., SSP, SDP, OCP, and markedness) by focusing on under-studied languages as well as language acquisition in children with normal and disordered speech. The secondary aim is to find whether sonority is located in the speaker’s competence, such as in phonology, or whether it is situated in phonetic planning, a component which is intermediate between phonology and a motor implementation component (Ball, Müller & Code, this volume). In general, this work is successful in examining sonority by studies which cover a wide range of research questions and apply various methodologies. The diversity in the work is mainly due to specific focus of the book and contributors’ field of expertise which ranges from phonetics, language acquisition, phonology, communication sciences and disorders, to speech pathology. For example, this volume provides a full picture of sonority by employing acoustic (e.g., Bird and Czaykowska-Higgins), articulatory (e.g., Chitoran), and perceptual (e.g., Brown) perspectives to study this phenomenon. Moreover, providing data from children with disordered and normal speech opens a new window for testing the predictions made by sonority.

This edition could be improved if it had gone through a more rigid proofreading and correction process since some of the authors are not native English speakers. Moreover, the IPA was not used correctly by some authors in some articles.

In general, the studies presented in this book show conflicting results. Most of the works (e.g., Archer; Ball & Müller; Meluzzi; Qiang; Shooshtaryzadeh; Yavaș & Babatsouli) do not support the predictions made by sonority-related principles. However, there are some contributions which employ sonority principles in addition to some other constraints in order to get a full interpretation of their results (e.g., Bird & Czaykowska-Higgins; Farwaneh; Klopfenstein). The lack of support for sonority in some of the chapters in this book could be due to adopting different classifications for the segments, especially the marked segments, in some studies. For example, /gw/ is classified as plosive-glide sequence rather than a labialized velar in Chapter 14. This could have major consequences on the interpretation of the results. Moreover, Miller (2012) has argued that the sonority scale should encompass two smaller scales: one based on the degree of constriction and the other based on sound source. Adopting such scale could provide us with results which reflect sonority in a finer way across languages.

This book also attempts to address the question regarding the location of sonority. Is it part of speaker’s phonology, or is it part of finer phonetics? There are some studies in the field which have shown that this bipolar dichotomy (i.e., phonology vs. phonetics) is not a very clear-cut division. They mainly assume that such distinction should be considered as one continuum and many of the phonological processes could be explained by including some other factors such a perception and markedness in phonetics-phonology interface (see, for example, Jun 1995; Kochetov & Pouplier 2008).

In order to further distinguish properties that are actively controlled by the speaker from properties which are the by-products of a mechanical execution, Solé (2007) has suggested using different speech rates. According to her, variation in the temporal duration of segments triggers the temporal change of controlled effects whereas they have no impacts on mechanical effects. This could be a good line of research for further studying where sonority is located (see, Falahati 2013, for more discussion).

Another promising direction for research is to develop a comparative study of sonority from both articulatory and acoustic-perceptual perspectives. With the rapid development of advanced technologies in the field, collecting articulatory data from speech with different rates and styles as well as acoustic-perceptual information could shed more light on this topic. In such an attempt, the language-specific gestural organization and timing as well as morphological characteristics of each language should be taken into account.


Ball, Martin. J., Müller, Nicole & Rutter, Ben. 2010. Phonology of communication disorders. New York. Psychology Press.

Barry, Martin. 1985. A palatographic study of connected speech processes. Cambridge papers in Phonetics and Experimental Linguistics, 4, 1-16.

Byrd, Dani. 1994. Articulatory timing in English consonant sequences, PhD dissertation, University of California, Los Angeles.

Byrd, Dani. 1996. Influences on articulatory timing in consonant sequences. Journal of Phonetics, 24, 209-224.

Clements, George Nick. 1990. The role of sonority cycle in core syllabification. In J. Kingston & M. E. Beckman (eds.), Papers in laboratory phonology I. Cambridge: Cambridge University Press, pp. 283-333.

Davis, Stuart & Shin, Seung-Hoon. 1999. The syllable contact constraint in Korean. An optimality theoretic analysis. Journal of East Asian Linguistics, 8, 285-312.

Falahati, Reza. 2013. Gradient and categorical consonant cluster simplification in Persian: An ultrasound and acoustic study (Doctoral dissertation). Ottawa: University of Ottawa.

Green, Antony D. 2003. Extrasyllabic consonants and onset well-formedness. In C. Féry & R. Van de Vijver (eds.). The syllable in optimality theory. Cambridge: Cambridge University Press, pp. 238-253.

Jun, Jongho. 1995. Perceptual and articulatory factors in place assimilation: An Optimality Theoretic approach, UCLA dissertation.

Kochetov, Alexei & Marianne Pouplier. 2008. Phonetic availability and grammatical knowledge: an articulatory study of Korean place assimilation. Phonology 25: 399-431.

Miller, Brett. 2012. Sonority and the larynx. In S. Parker (ed.). The sonority controversy. Amsterdam: Mouton de Gruyter, pp. 257-288.

Ohala, John. 1992. Alternatives to the sonority hierarchy for explaining segmental sequential constraints. Papers from the parasession on the syllable. Chicago Linguistics Society, 319-338.

Prince, Alan. & Smolensky, Paul. 2004. Optimality theory: Constraint interaction in generative grammar. Oxford: Blackwell.

Proctor, Michael & Walker, Rachel. 2012. Articulatory bases of sonority in English liquids. In S. Parker (ed.). The sonority controversy. Amsterdam: Mouton de Gruyter, pp. 289-316.

Sole´, Maria Josep. 2007. Controlled and mechanical properties in speech: A review of the literature. In M. J. Sole´, P. S. Beddor, and M. Ohala (eds.), Experimental Approaches to Phonology. New York: Oxford University Press. pp. 302-322.

Steriade, Donca. 1990. Greek prosodies and the nature of syllabification. New York: Garland Press.

Yip, Jonathan. 2013. Phonetic effects on the timing of gestural coordination in Modern Greek (Doctoral dissertation). Ann Arbor: University of Michigan.


Reza Falahati completed his PhD in Linguistics at the University of Ottawa in 2013. Then he worked as a Postdoctoral Research Fellow in Italy, SNS, for two years. His research interests are mainly articulatory phonology, phonetics, language acquisition, speech perception, and discourse analysis.

Page Updated: 29-Aug-2017