Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  Rhotics

Reviewer: Malcah Yaeger-Dror
Book Title: Rhotics
Book Author: Lorenzo Spreafico Alessandro Vietti
Publisher: Bozen-Bolzano University Press
Linguistic Field(s): Phonetics
Language Acquisition
Subject Language(s): Bavarian
Greek, Modern
Comorian, Ngazidja
Issue Number: 26.1175

Discuss this Review
Help on Posting
Review's Editors: Malgorzata Cavar and Sara Couture


The book under review provides a collection of papers presented at ‘r-atics-3’, held at the Free University of Bozen-Bolzano in 2011, as a continuation of the first two conferences on this topic, held in 2000 and 2002.The book provides insight into the patterns of variation and change of rhotics in different languages, from a variety of perspectives, shedding light on the phonetics, phonology, acquisition, and sociolinguistic variation of /r/-sounds in languages as diverse as Dutch, English, French, German, Greek, Hebrew, Italian, Kuikuro, Malayalam, Romanian, Saraiki-Punjabi, Slovak, Tyrolean-German and Washili Shingazidja; in the process, it contributes to the discussion on the unity and uniqueness of this group of sounds. Given the importance of this topic both to phonetics and the sociophonetics of language variation (Maguire and MacMahon 2011), it is not surprising that there are 13 outstanding papers in the collection: seven concerning the phonetic realization of rhotics in different languages, three concerning language change, and three on variation in the speech of bilinguals.

Chapter I is an introduction or ‘preamble’ by the editors, Alessandro Vietti and Lorenzo Spreafico (pp. 9-19). It focuses on the fact that the class of ‘rhotics’ is both acoustically and articulatorily quite diverse, as well as on the fact that in many communities, multiple rhotic variants occur simultaneously, carrying social as well as linguistic meaning for the community. The book’s chapters provide evidence for the claims made in the introduction. The editors point out that many of the papers foreground articulatory as well as acoustic evidence, reflecting recent advances since 2003.

The first article, by van ‘t Veer (pp. 23-40), considers different rhotic acquisition patterns for two toddlers (ages 1-2 and 1-4) of Quebec French -- a boy and a girl. She focuses on KR clusters. Unfortunately the author does not inform us of the toddlers’ social background, their parents’ rhotic patterns, the extent to which they interact with speakers from outside their immediate families, or where exactly in the province the study was carried out. Quebec rhotics have been documented to vary by (a) dialect region in Quebec (e.g., Santerre 1978, Sankoff and Blondeau 2007), and (b) the major changes taking place in the rhotic system of the province (Sankoff and Blondeau 2007, this volume), so giving information on /r/ realization in clusters without any of that other information is less helpful than one might wish. Moreover, given the article’s inclusion in a section on bilingualism, it is surprising that the author never clarifies even the extent to which the children’s parents (much less the children themselves) are bilingual. The study codes rhotic tokens for duration, voicing, manner, and place of articulation, to support Flege’s claims, but without the absent socio-demographic information, it is very difficult to draw conclusions.

Syed’s paper, ‘Acquisition of English [r inverted] by adult Saraiki (a southern dialect of Western Punjabi) Pakistani learners,’ analyzes the extent to which a rhotic is learnable, based on the distance between the (Saraiki) L1 sound, and the new (British) L2 sound; the results are viewed through the prism of Flege’s Speech Learning Model (hereafter SLM) (1995). The introduction includes a lucid discussion of Flege’s theory, as well as a clear description of the speakers’ native /r/ (a trill, with a variable vowel-onglide in the initial position). Syed divides the study population into three groups based on the extent of L2 [English] contact. Syed first reports on the perception of the British-r by the L1 Saraiki monolinguals and shows that the listeners hear the tokens as /l/ or /w/; I wish that he had verified whether specific acoustic information was more likely to be heard as an alternative pronunciation. In an AX test, subjects recognize the retroflex-r presented as ‘the same’ as their own in an /ara/ context approximately 73% of the time, but they also recognize it as /w/ 60% of the time, and as /l/ 40% of the time. Given the studies of Docherty and Foulkes 2001, or Ogden 2009, presenting evidence that speakers in certain regions of the UK actually produce a [w]-like /l/, without further evidence we might assume that the listeners may be accurately reflecting what they hear. Following Flege’s hypothesis, if the primary requirement is that the speakers be able to distinguish between their own and the foreign /r/, there is at best 27% chance that they will learn the English-r; further complications are caused by the British-/r/’s phonetic similarity to their own /w/ and /l/. Production and perception tests of 90 adult learners of English support Flege’s hypotheses; discrimination of the L2 /r/ from [w] improves for the ‘UK’ learners more than for the ‘student’ learners, who in turn are better than the ‘inactive’ learners, but even the UK-based group was not very good at articulating a British-/r/ as distinct from a native L1 /r/. The article is clear and supports Flege’s theory. As a sociolinguist, I might wish that in future studies the author could also consider the intergroup attitudes of the speakers; this article provides an impressive study and assessment of Flege’s theory. Future work should take into consideration evidence that trained L1-English phoneticians do not necessarily have more reliable perceptions than the listeners in the study (Yaeger-Dror et al 2009).

In the third study, ‘On rhotics in a bilingual community’ (pp. 57-80), Spreafico and Vietti carry out Ultrasound Tongue Imaging (UTI) studies of rhotics in an Italian and Tyrolean-German region. The study analyzes articulatory variation by bilinguals from this region and draws conclusions relative to bilingual theories such as Flege’s. Previous studies in the region have stated that Tyrolean speakers use a uvular ‘Bavarian’ /r/ in their German, but an alveolar flap in their Italian, while the Italian speakers are more likely to maintain a flap in both. The authors then describe their Word List Style (WLS) corpus of 19 local bilingual speakers, divided into four groups according to the extent to which they make use of their L2. The authors conclude that the balanced bilingual speakers differentiate between the two languages’ /r/, but the articulatory position differs from that of monolinguals of either language, maintaining a single articulatory position for /r/ in both word lists. The two monolingual speakers contrast with each other and the bilingual group, with one using only dorsal and the other only alveolar positions. The authors conclude that the simultaneous bilinguals distinguish the two r’s consistently, but not by a great deal, and not with any assistance from tongue shape; they suggest that speakers in bilingual regions may differ from those who acquire two languages independently and who do not live in a region where both languages are ‘native’. They propose that while speakers who migrate from one language area to another may attempt to acquire non-overlapping systems, those raised in bilingual families may prefer to merge two systems to demonstrate their local-bilingual bonafides. This study is very informative and raises very interesting questions for our understanding of theories of language learning. Despite the importance the authors attribute to language attitudes, unfortunately there is little actual discussion of the speakers’ (linguistic and cultural) identity.

Part II of the text analyzes ‘phonetics and phonology’ of rhotics. Each chapter analyzes the rhotics from a different speech community. ‘Articulatory coordination in obstruent-sonorant clusters and syllabic consonants: Data and Modeling’ by Philip Hoole, Marianne Pouplier, Štefan Beňuš and Lasse Bombien (pp. 81-96) uses electromyographic (EMA) data to analyze consonant cluster onsets with rhotics in French and German, and syllabic rhotics as used by speakers of Slovak. The first analysis poses the question of why the likelihood of such clusters differs radically in the languages studied. The second proposes a theory for why syllabic rhotics are typologically relatively rare, and proposes a hypothesis for what factors favor the emergence of such syllabic consonants. The paper emphasizes that analysis of the coordination of articulatory overlap and systematic patterning may clarify how sound systems develop. Analysis of rhotics permits systematic cross-language comparisons of articulatory coordination in the targeted communities.

In the next chapter, ‘Articulating five liquids: A single speaker ultrasound study of Malayalam,’ by James M. Scobbie, Reenu Punnoose and Ghada Khattab (pp. 99-124), Malayalam was chosen for analysis because there are five liquids which are consistently distinguished, two of which are rhotics [a flap and a trill], two laterals [bright and dark], and a ‘palatal approximant’, which seems to be similar to the Central and South American ‘assibilated r’ studied by Proctor (2011). Both acoustic (static) and ultrasound (dynamic) evidence are presented clearly, along with velocity information. Tricolor (red, blue and green) figures provide clear evidence for the non-overlap of assibilated [green], rhotic [red] and lateral [blue] gestures, or for temporal change in a gesture from V1 [green] to consonant [red] to V2 [blue]. All the figures are clear and support the points made in the text. The articulatory study also lends itself to comparison with studies using other ultrasound systems, like those of Proctor (2011) which compared Russian and Spanish articulation of liquids. The authors make suggestions for future studies of a more diverse set of vowel environments, as well as for adaptations of their ultrasound equipment to permit greater analytical focus for future rhotic studies. At the moment, however, this study provides the most nuanced analysis of rhotic and lateral articulation in a Dravidian (or any other) language.

‘The many faces of /r/’ by Mary Baltazani and Katerina Nicolaidis (pp. 125-144) permits analysis of both acoustic and articulatory (electro palatal -- EPG -- artificial palate of 62 sensors) evidence for the alveolar flap-r in Greek, comparing five speakers’ rendition of words with /r/ in different prosodic and phonological positions (in an identical carrier phrase); this article has a ‘laboratory phonology’ perspective. The research design can be seen as both an advantage (i.e. “Let’s nail the most ‘careful’ pronunciation first!”) and a disadvantage (i.e. “How you gonna allow for sociophonetic variation when the speaker has a mouth full of hardware?”). There appears to be an ‘intrusive’ vowel between the /r/ and either a consonant or silence, even in initial position, often referred to as an epenthetic vowel (Warner et al 2001). The authors measured the duration of the rhotic and environmental vowels, as well as the degree of constriction, the level of emphasis, and the first two formants of the rhotic, under various stress, and coarticulatory conditions. Even in a carrier phrase, more than half of the tokens were produced with incomplete constriction, providing clear evidence that the Greek rhotic has indeed ‘many faces’. Reduced contact was less likely to occur in Cr-clusters, where the /r/ is found to assimilate to the neighboring consonant, but is more likely to occur word initially or in -rC clusters where the duration is longest. Note that the degree of durational variation implies that the Greek ‘flap’ manner of articulation is somewhat compromised. While this study provides a point of departure for future studies of coarticulation in Greek, there appears to be fairly considerable variation across phonological or stress contexts, as well as inter-speaker variation. Results will be of relevance to socio-phoneticians as well as laboratory phonologists. For example, the paper by Sankoff and Blondeau demonstrates that in other communities as well, gestures for rC and for Cr differ consistently and lead to differing articulatory patterns. While the Sankoff/Blondeau paper emphasizes the sociophonetics of the variation, this paper provides an interesting ‘laboratory phonology’ perspective on such variation.

Comparing phonological positions, the next chapter, ‘Another look at the structure of [ɾ]: Constricted intervals and vocalic elements’, by Carmen-Florina Savu (pp. 145-158), analyzes both the structure of the tap and the adjacent vowel segments which appear in Romanian as well as in Greek and Polish; the author finds the flap is propped up with ‘epenthetic’ vowels on both sides. As with the previous paper, durations of both the flaps and environmental vowels are measured along with the vowels’ formants. The five speakers produce words and nonsense words (ara, iri, etc.) in isolation; vowel segments and durations are measured using Praat. Unfortunately, despite the plethora of normalization programs now available, non-normalized vowels for all speakers are averaged and plotted onto the same figures, although the epenthetic schwa-like vowel could vary radically with the speakers’ articulatory tract characteristics. The author concludes from her evidence that the intrusive/epenthetic vowel segment can be thought of as coarticulated with the nearest vowel in the word spoken. It would have been advantageous for the two adjacent studies to be cross-referenced, at least for the durational measures of V, r, and epenthetic-vowel segments, despite the fact that it may be difficult to compare isolated words with words in carrier phrases. The author concludes that all flap tokens are flanked by vowels, whether or not the rhotic occurs in a consonant cluster. If there is no adjacent vowel in the word, a minimal epenthetic neutral vowel is added, even if the rhotic itself is phonologically treated as the primary vowel in the word; for the latter, acoustic plots of Serbian and Czech phrases including /r/ vowels are included.

In the next chapter, ‘New insights into American English V+/r/ Sequences,’ by María Riera and Joaquin Romero (pp. 159-172), the authors’ focus, as in the immediately preceding papers, is on the evidence for coarticulation between vowel and adjacent consonants. The review of the literature is somewhat sketchy – including neither discussion of coarticulation (e.g., omitting Gay et al. 1974, and subsequent Haskins work), nor any of the vast literature on the extreme variation of ‘rhoticity’ in this position by American speakers (e.g., by Feagin 1990, Hagiwara 1995, Lawson et al. 2011, or Becker 2014, among many others). While the authors consider the six ‘native’ rhotic-dialect speakers as a uniform group, thirty years of phonetic studies make that claim hard to believe. In addition, all speakers had been living in Spain at the time of the recordings and were recorded there. Despite the early Haskins work and recent ultrasound studies, speed of articulation is assumed, based on the duration of pauses between sentences. Acoustic measurements were made for the transition area between V and /r/, with plots for the separate formants (F1-F3) for ONE speaker. Unfortunately, the readers are not told which speaker, nor why the unnormalized formant data for only one speaker is presented, nor even if in each case the speaker is the same for each formant plot, and no comparisons are offered between these purportedly ‘same’ speakers. Under these obviously non-optimal conditions it is difficult to evaluate the importance of this particular study.

‘/r/ in Washili Shingazidja’ by Cédric Patin (pp.173-190) analyzes variation in /r/ in the Bantu language, where the rhotic is a flap intervocalically and a trill in most other positions sampled. One native speaker of the dialect, resident in France for the last 10 years, and presumably bilingual, was recorded reading a word list. he author presented acoustic evidence that there is an alveolar trill in initial position, before a vowel with a high tone, or after a consonant, and a flap in all other positions.

‘Prosodic factors in the adaptation of Hebrew rhotics in loanwords from English’ by Evan-Gary Cohen (pp.191-205) would perhaps have been better-suited for another publication. The underlying assumption, that any foreign-origin lexicon is actually from English, is flawed, given that native speakers and linguists alike usually assume a generalized ‘foreign language’ source, and most immigrants are not from English-speaking countries. More critically, the author makes claims for the articulatory manner and position of Israeli Hebrew /r/ which are not supported by the phonetic research which he cites (e.g., Bolozky and Kreitman 2007, Schwarzwald 2004), and for which he presents no quantitative evidence; consequently, this paper appears to be inappropriate for a volume intended for a phonetically-trained audience.

The third part of the book, entitled ‘Language variation and change,’ includes three papers. The first one, by Antonio Romano (pp. 209-225), ‘A preliminary contribution to the study of phonetic variation of /r/ in Italian and Italo-Romance,’ provides interesting material. However, citations from Italian sources are not glossed. Apparently, Word List Style readings were analyzed; this is immaterial if there is no interest in relative prestige of different variants and if there is no interest in change in real time. Given that both are of interest to the author and are discussed, the decision to limit the corpus without even a prima facie account for that limitation is unfortunate. While the article may provide a point of departure for future studies of variation in Italian, it does not provide an in depth discussion of changes occurring in Italian.

‘The spreading of uvular [R] in Flanders’ by Hans Van de Velde, Evie Tops, and Roeland Van Hout (pp. 227-248) proposes a trajectory of change for Flemish in the 20th and 21st centuries. The analysis is based on a study of three diachronic corpora of Flemish recorded between 1925 and 2009, and considers the shift toward ‘acceptability’ of the uvular R. Section 2 reviews a century’s worth of dialect studies on rhotics in Flemish and Dutch. Section 3 reviews in greater detail dialect studies of 4000 speakers carried out in Belgium since the 1920s, analyzing the speech of Middle Class Urban speakers as well as the more traditional NORMS (nonmobile-older-rural-male speakers). Both of these dialect geography studies found that, except for the very eastern tip of the area, only a few speaker-locations on the southern border actually had shifted to the uvular rhotic by the early 1950s. The authors also include the results of a study by the second author from the late 1990s, which coded speakers for age, as well as region; that ‘trend’ study found the younger speakers more likely to produce uvular [R] than older speakers. The presentation is clear and well-supported by the data in the tables, as well as on the useful maps.

‘Instability of the [r]~[R] alternation in Montreal French,’ by Gillian Sankoff and Hélène Blondeau (pp. 249-265), is a follow-up on their “Language” paper (2007), showing change in real time, based on comparison of interviews from the 1971 Montreal study, and with interviews with the same speakers in 1984, while taking into consideration the more extensive recordings available from 1995 as well. Earlier publications demonstrated that younger members of the ‘panel’ who had not yet begun to shift toward [R] in 1971 have for the most part adopted it by 1984, while the older speakers were much less likely to adopt the newer variant.

There are two main conclusions from the data from the Montreal study of phonological variation. First, when tracing change in real time, most individuals are going to retain the variant of their youth, while a few will shift toward the newer variant. Second, where change does occur across the lifespan, it is much more likely to occur in a gradient context (like vowel shift), than in a binary change (like the shift from [r] to [R]).

The present analysis focuses on data from the follow-up corpus recorded in 1995 to permit a more nuanced analysis of variation as well as change in real time for two of the panel members. For this study they choose two speakers from the youngest age group who still retained a high percentage of the [r] in 1971, when most of their age-mates were already [R]-ful. The speaker moving up the social ladder shifted radically toward [R] usage in all contexts, while the speaker whose socioeconomic status was declining did not advance toward the newer dialect, but did not shift to the older [r]. The paper makes clear that a full study of sociolinguistically shifting features will require the analysis of the situation of use, as well as of socioeconomic and temporal factors.


The book is intended to serve as a point of departure for the research community working on variation in the phonetic properties of rhotics. Experimental phoneticians, phonologists, sociolinguists and historical dialectologists can all learn from the papers in the monograph, which, while highly technical, present their conclusions clearly and succinctly. These studies provide exemplars of how research on rhotic variation can be carried out. Most of the papers are very informative and will be useful for future researchers by providing a template for how to carry out a given research path, as well as by analysis of specific results. The publisher (and the authors) should be commended for having permitted multicolored figures to be published, since the figures in most of the papers are much more ‘decipherable’ than black and white figures.

On the other hand, it is frustrating that even in a text with a high degree of overlap among the papers’ themes, where cross-referencing would be helpful, there is almost no attempt to cross-reference between the papers in order to take advantage of convergent conclusions, even when the authors are discussing the same or similar speech communities. In addition, researchers without funding to attend the meeting could not be included in the volume. For example, while the UTI work presented is laudable one might wish that young researchers like Mike Proctor or Eleanor Lawson (e.g., Proctor 2011; Lawson et al. 2011) had been represented in the publication. Similarly, the acoustic work presented in many of the papers is quite useful and interesting, but it is a shame that the work of Heselwood, Plug and others carrying out important research in this domain (e.g., Heselwood et al. 2010; Heselwood and Plug 2011) was not represented. One hopes that future conferences in this series will not only provide a forum for the high quality of analysis and discussion found in this volume, but will perhaps also allow for papers by authors whose work is crucial to the field, but who are unable to attend. This monograph is a valuable resource for future students of variation in rhoticity.


Becker, K. 2014. (r) we there yet? The change to rhoticity in New York City English. LVC 26:141-168.

Bolozky, Shmuel and Rina Kreitman. 2007. Uvulars in Israeli Hebrew. Presented at International Conference on Hebrew Language, Literature and Culture, Sydney, Australia.

Chambers, J. and P. Trudgill 1998. Dialectology. Cambridge: Cambridge University Press.

Docherty, G. and P. Foulkes 2001. Variability in (r) production. In H. van de Velde and R. van Hout (eds), ‘ratics. 27-43. Brussels: ilvp.

Feagin, C. 1990. From r to rless in three generations. In J. Edmondson, C. Feagin, and P. Mühlhäusler (eds.), “Development and Diversity”: Linguistic Variation across Time and Space, 129-146. Arlington, TX: SIL.

Flege, J. 1995. Second language speech learning. In Winifred Strange (ed), Speech perception and linguistic experience, 233-273. Baltimore: York Press.

Gay, Thomas, T.Ushijima, H. Hirose, F.S. Cooper 1974. Effects of speaking rate on labial consonant vowel articulation. “Journal of Phonetics” 2:47-63.

Hagiwara, R. 1995. Acoustic Realization of American /r/. UCLA Working Paper in Phonetics 90.

Heselwood, B., L. Plug, and A. Tickle 2010. Assessing rhoticity using auditory, acoustic and psychoacoustic methods. In B. Heselwood and C. Upton (eds.) Proceedings of Methods in Dialectology XIII. Frankfurt: Peter Lang.

Heselwood, B. and L.Plug 2011.The role of F2 and F3 in the perception of rhoticity: Evidence from listening experiments. Proceedings of the XVIIth Congress of Phonetic Sciences, 867-870.

Ladefoged, P. and I. Maddieson 1996. The sounds of the World’s Languages. Oxford: Blackwell.

Lawson, E., J. Stuart-Smith, J. Scobbie, M. Yaeger-Dror, M. Maclagan. 2011. Liquids. In Marianna di Paolo and Malcah Yaeger-Dror. (eds) Sociophonetics. A Student’s Guide. NY: Routledge.

Maguire, W. and A. McMahon 2011. Analysing Variation in English. Cambridge: Cambridge University Press.

Ogden, Richard 2009. An Introduction to English Phonetics. Edinburgh: Edinburgh University.

Proctor, M. 2011. Towards a gestural characterization of liquids: Evidence from Spanish and Russian. Lab Phonology 2:451-485.

Recasens, Daniel and Maria Dolors Pallarès. 1999. A study of /ɾ/ and /r/ in the light of the “DAC” coarticulation model. Journal of Phonetics 27(2): 143–169.

Sankoff, G. and H. Blondeau 2007 Longitudinal change across the lifespan. Language 83:560-588.

Santerre, L. 1978. Les /R/ montréalais en régression rapide. Protée 2:117-130.

Schwarzwald, Ora R. 2004. Consonant clusters. In Perspectives on language and language development, ed. by Dorit Ravid and Hava Bat-Zeev Shyldkrot, 45–60. Dordrecht: Kluwer.

Warner, Natasha, Allard Jongman, Anne Cutler, Doris Mücke 2001 The phonological status of Dutch epenthetic schwa. Phonology 18: 387-420.

Yaeger-Dror, M., Paul Foulkes, Dom Watt, Tyler Kendall, Dan Johnson, Jillian Oddie, Philip Harrison. 2009. Perception of ‘r’, a cross-dialect comparison. LSA Presentation.
Malcah Yaeger-Dror is a member of the (variationist) sociolinguistics community, and works closely with corpus and forensic linguists who share her belief that the larger the corpus, the more you can learn about variation and change within a given speech community. She has carried out sociophonetic research on English, French, Spanish and Hebrew vowels, r, and prosody. She also advocates the study of how intergroup attitudes may influence sociophonetic choices: speakers’ fluctuating attitudes towards who they want to be, how they wish to relate to specific interlocutors, & the (multiple) ways they wish to present themselves in specific situations.