Band 8, Heft s5 - Special Issue: Sound change in endangered and small speech communities; Editors: Georgia Zellou and Alan Yu

How sound change is initiated and propagates in smaller speech communities is not well understood. This paper provides an overview of the main themes, including theoretical and methodological issues, of the special collection on sound change in endangered and small speech communities.

When do the mechanisms of regular sound change fail to apply? What types of languages and situations exhibit and promote phonological stability? I consider these questions using data from the languages of Aboriginal Australia, where there has been debate on this question. I show that the standard explanations are inadequate, and possible solutions have not yet been empirically investigated. Given how many of these languages are already either no longer spoken or severely under threat, it is important to investigate these questions urgently.

Almost universally, diachronic sound patterns of languages reveal evidence of both regular and irregular sound changes, yet an exception may be the languages of Australia. Here we discuss a long-observed and striking characteristic of diachronic sound patterns in Australian languages, namely the scarcity of evidence they present for regular sound change. Since the regularity assumption is fundamental to the comparative method, Australian languages pose an interesting challenge for linguistic theory. We examine the situation from two different angles. We identify potential explanations for the lack of evidence of regular sound change, reasoning from the nature of synchronic Australian phonologies; and we emphasise how this unusual characteristic of Australian languages may demand new methods of evaluating evidence for diachronic relatedness, and new thinking about the nature of intergenerational transmission. We refer the reader also to Bowern (this volume) for additional viewpoints from which the Australian conundrum can be approached.

Crosslinguistically rare sounds may be uncommon as a result of being phonologically marked (Trubetzkoy 1939) or due to articulatory or perceptual biases (Maddieson 1998). Certain types of sound changes are often argued to have roots in articulatory and perceptual biases (Blevins 2004). But in cases where there is limited data available, such as with understudied languages, it may be difficult to find evidence for the roots of sound changes. Synchronic variation can be used to provide evidence for diachronic sound changes (Blevins 2004; Lindblom 1990; Ohala 1993), which is particularly useful when historical data is limited. In this investigation we discuss phonetic biases, including acoustic and perceptual factors, that contribute to a set of sound changes in Tlingit, a critically endangered Indigenous language of Alaska, British Columbia, and the Yukon, that resulted in a primary split-merger (Blust 2012). This investigation provides further support for including explicit discussion of synchronic variation as part of the description of understudied languages. We propose that there should be a stronger emphasis on documenting and analyzing variation within understudied languages because excluding variation potentially masks significant intralinguistic and crosslinguistic phenomena.

Laz Turkish (LT) is a nonstandard variety of Turkish mainly spoken in the North East of Turkey. LT emerged due to language contact between Turkish (Turkic) and Laz (Kartvelian). Laz has a smaller vowel system compared to Turkish and this has impacted the vowel harmony (VH) system in LT. In contrast to ST, which has a productive VH system, LT has only partially productive VH. Based on a corpus of fieldwork data collected in Rize, Turkey, this study investigates how LT displays partial harmony. The results show that (i) /ɯ, y, œ/ are rare in LT, and ST /ɯ, y, œ/ correspond to /i, u, o/ in LT respectively, (ii) open syllables in LT tend to have /i/ compared to closed syllables which tend to occur with /u/, and (iii) the smaller vowel system has led to disharmony with respect to preceding vowels, but /i u o/ still trigger harmony on following vowels, so VH does not peter out across the word.

One of the defining characteristics of tonal systems in phonological theory is the notion of tonal stability (Goldsmith, John. 1990. Autosegmental and metrical phonology . Oxford: Blackwell; Yip, Moira. 2002. Tone . Cambridge: Cambridge University Press). Tones remain stable even if their tone-bearing unit changes or is deleted. In the context of historical sound change, tones may either persist as floating tones (from an autosegmental-metrical perspective) or fuse with adjacent syllables and give rise to more complex tonal inventories. However, the process of segmental loss which ultimately conditions historical tonal change is gradual in nature. Tone-bearing units may lenite or they may be optionally realized, leading listeners to rely on coarticulatory cues or phonetic information on adjacent syllables. In the current paper, I examine how variation in the realization of clitic pronouns is conditioned by adjacent tonal cues in Itunyoso Triqui (Otomanguean) within a corpus of spontaneous speech recordings. This research examines and provides evidence for the hypothesis that tonally conditioned allomorphy arises specifically when two conditions are met: (a) there is a morphological context where prosodic units are likely to lenite (highly redundant contexts); and (b) adjacent tonal cues are most informative at this morphophonological boundary. The findings shed light not only on how phonologically conditioned allomorphy arises but also on how variable deletion is sensitive to patterns of multiple exponence during language use.

This study examines Spanish phonemic stops in the speech of 15 ethnically Bora bilinguals (8 males, 7 females) living in the Peruvian Amazon, within the broader context of examining gender-based phonetic variation in small speech communities. We target pronunciation of intervocalic phonemic /p t k b d g/ extracted from sociolinguistic interviews. The acoustic analysis focuses on consonant duration and relative intensity of each target phoneme. The results reveal clear gender-based variation, with males adopting more lenited variants of certain phonemic stops than females. We discuss these findings in light of gender-based research on phonetic variation in communities undergoing sound change. More generally, our study contributes to the literature on language variation and societal contacts in small speech communities in Amazonia.

Suzhounese exhibits an unusual place anteriority contrast between rounded and unrounded dorso-palatal high front vowels, postalveolar fricative vowels, and apico-alveolar apical vowels. This arrangement is vulnerable to loss under intensifying contact with Standard Mandarin. Using acoustic and tongue ultrasound data, we investigated the phonetic implementation of place in the Suzhounese fricative and apical vowels and the similarity of place targets with the apico-alveolar and alveolo-palatal fricative consonants /s/ and /ɕ/, and the front vowel /i/. Two age cohorts differing in their acquisition history and usage of Standard Mandarin were investigated. The younger cohort, who had earlier and likely more intense exposure to Standard Mandarin, exhibited a subphonemic shift in the fricative vowels toward less anterior, more /i/-like constrictions. Due to this shift, implementation of place targets among vowels and consonants was less uniform for the younger cohort, particularly in acoustic terms, but articulatory target uniformity among the vowels at each place was robust for both cohorts regardless of the degree of shift. We discuss possible contact-based mechanisms for the observed changes, as well as implications of the observed interactions between L1–L2 transfer and L1-internal structural cohesion.

We conducted a perception experiment in the field to examine the synchronic consequences of a tonal split in Risiangku Tamang (Tibeto-Burman). Proto-Tamang was a two-tone language with three series of plosives and two series of continuants. The merger of its continuants provoked a split of the original two tones into four, two high and two low, which combine pitch and phonation features. The quasi-merger of the voiced and voiceless plosives left sporadic remnants of initial plosive voicing in low tone syllables. A previous production study has shown that speakers use pitch and phonation features concomitantly to distinguish high from low tones, while producing initial plosive voicing only marginally with low tones. The present perception study establishes the preeminence of the pitch cue, but also confirms the effective use of the two older cues in tone identification. An apparent-time analysis shows the phonation cue to be less used by younger speakers, in keeping with the historical evolution. The use of the residual voicing of plosives, instead of decreasing with younger speakers, is shown to increase. This result could be explained by an increased contact of the young generation with Nepali, a toneless Indo-Aryan language with a four-way initial plosive contrast.

Kalasha, an endangered Dardic (Indo-Aryan) language, contrasts a rich set of rhotic vowels , a vowel type, which is found in less than 1% of the world’s languages. The acoustic and articulatory correlates of rhotic vowels, and their development and geographical distribution in Kalasha and other Indo-Iranian languages are still poorly understood. The current study brings together typological data on retroflex approximants and flaps in 192 Indo-Iranian language varieties, and phonetic data on rhotic vowels and retroflex approximants in endangered Dardic (Kalasha and Dameli) and Nuristani (Kamviri and Eastern Kataviri) languages. The phylogeography of retroflex approximants and flaps indicates that rhotic vowels are prevalent in those areas of South Asia where retroflex approximants are in abundance. Specifically, the development of rhotic vowels in Kalasha may have been amplified by the presence of retroflex approximants in neighboring Nuristani languages. We show that phonetically the rhotic sounds in the two Dardic languages are produced with a bunched tongue shape, whereas the retroflex approximants in Nuristani languages are produced with the raising of the tongue tip.

The effect of ability in a tonal language on shifting cue weights in a non-tonal language is explored through a production and perception experiment carried out with speakers of Kuy (Katuic, Austroasiatic) in Ban Khi Nak, Sisaket Province, in Northeast Thailand. Specifically, the realization of a modal-breathy voice quality contrast is analyzed through observing the effects of language ability and usage-related sociolinguistic factors. The results show increased usage of F0 and CPP cues in production and of F0 cues in perception of the voice quality contrast with greater ability and usage of a tonal language. Tonal language ability and usage also affects other acoustic correlates of voice quality, but does not uniformly weaken them. Degree of integration into Thai society shows mixed effects, only somewhat correlating with decay of voice quality cues. The results have bearing on the role of larger national and regional languages in restructuring of phonological contrasts and in phonological attrition, and provide insight into the relationship between multilingualism and sound change.

Baba Malay encodes a coarse-refined contrast in speech styles, with the vowel [ɛ] being traditionally restricted to the use of refined word forms. Coarse forms ending with [al], [aɾ], and [as] contrast with refined forms that end with [ɛ], which are used when the speaker wants to be perceived as being particularly refined. A comparison of natural conversational speech with careful performative speech supports the claim that speakers of endangered languages are not “monostylistic”. Production data from eight speakers across different age groups show that the [ɛ] forms more commonly occur in the careful performative speech of younger speakers, as compared to when they converse more informally. No such trend is established for older speakers. The production data is considered together with the findings of a previously-executed matched-guise experiment, which shows that younger speakers in particular are reanalyzing the refined forms as being emblematic of community identity. In this study, younger speakers appear to be using [ɛ] to index an ideal speaker of Baba Malay in their careful performative speech. The research highlights that speakers of endangered languages, and even members of the younger generation, do have access to multiple styles.

This paper examines word-initial engma deletion in Bininj Kunwok. Loss of initial consonants is a well-documented historical process in many Australian languages (Blevins, Juliette. 2001. Where have all the onsets gone? Initial consonant loss in Australian Aboriginal languages. In Jane Simpson, David Nash, Mary Laughren, Peter Austin & Barry Alpher (eds.), Forty years on: Ken Hale and Australian languages , 481–492. Canberra: Pacific Linguistics), but there has been no systematic analysis of initial consonant loss as a synchronic variable (Fletcher, Janet & Andrew Butcher 2014. Sound patterns of Australian languages. In Harold Koch & Rachel Nordlinger (eds.), The languages and linguistics of Australia , 91–138. Berlin: De Gruyter Mouton). In the case of Kunwok, word-initial velar nasal deletion (e.g. nganabbarru ∼ anabbarru ‘buffalo’) has been described as having regional distribution and is a prominent feature of speakers from the western and southern peripheries of the dialect chain, but variable in speakers from the central region (Evans, Nicholas. 2003. Bininj Kunwok: A pan-dialectal grammar of Mayali, Kunwinjku and Kune . Canberra: Pacific Linguistics). This study tests the distribution of the word-initial engma for both linguistic conditioning and sociolinguistic factors, and arrives at three conclusions. First, that preceding environment is a contributing factor. Second, that morphological class is a categorical conditioning factor. And third, that the phenomenon is spreading and there is evidence of language change in progress. This paper also takes into consideration community perspectives, noting that the sound change in progress is accompanied by a change in perceptions.

Dialects of the South American language Mapudungun are claimed to display a dental-alveolar contrast across four manners of consonant articulation: stops, nasals, laterals, and fricatives. Such a full, symmetrical system of distinctions among coronals is typologically unique and, as such, is predicted to be unstable. This paper’s survey of contemporary data, however, shows that, despite lexical contrast being marginal and dentals being morphologically restricted, the distinction is highly salient to native speakers of the more vital dialects. A careful examination of the pattern’s historical roots and diachronic stability, furthermore, allows us to reconstruct it throughout the 400-year textual record. Indeed, the early descriptions and transcriptions shown that, instead of contracting, the contrast expanded, by borrowing the alveolar fricative /s/ from Quechuan and Spanish. The historical and articulatory data shows that while /t̪ n̪ l̪ θ/ are laminal, /t n l/ are apical. Incoming /s/, however, does not follow the pattern, being laminal and prompting a reorganization of featural contrasts among fricatives. As a result of erosion of native fluency under Spanish contact, loss of the dental-alveolar contrast has become commonplace, although there is much variation across speaker, dialect, and manner of articulation. Crucially, dialects which had only voiced fricatives until the borrowing of /s/ seem to have added voicing as a new contrastive feature, helping to preserve the coronal contrast among fricatives, even where vitality is reduced.

This paper considers the typical focus of analysis in a sound change study across generations: the speech community. I argue that changes in social practices across generations may mean that generational comparisons can be problematic, and these issues are particularly pertinent in small and endangered language communities. Using data from Scottish Gaelic (Celtic, ISO = gla), a minority endangered language of Scotland, I exemplify the challenges posed by the speech community construct via an examination of lateral production across generations of speakers. Gaelic traditionally contrasts three phonemic laterals, but analysis shows that this might be changing. There are two possible directions for sound change in the Gaelic lateral system: results show that younger speakers produce some palatalised laterals as palatal glides without laterality. Meanwhile, the remaining laterals are less acoustically distinct among younger generations, suggesting the possibility of future merger. The construct upon which any apparent- or real-time study of sound change relies is the speech community as a unit of analysis for change. While there are differences among the groups of speakers analysed here, I argue it is potentially problematic to consider this to be a form of sound change due to differences in social practices among generations surrounding Gaelic usage and socialisation. Ultimately, I advocate for a socially-informed approach to sound change study which sympathetically takes local social structure into account.

Research on bilingual populations finds that phonetic transfer between languages is common, especially in similar sounds. Yet, phonetic studies that include bilingual speakers of Indigenous languages remain rare. This paper analyzes phonetic transfer in Diné Bizaad (Navajo), a Dene language i ndigenous to North America. Specifically, I examine the evidence for intergenerational change and convergence with English in the release period durations of Diné affricated stops [kx, tx]. The analysis is based on tokens of Diné [kx, tx] and English [k h , t h ] extracted from recent interviews with 51 Diné-English bilinguals. Release durations of targeted sounds are measured, compared with earlier analogous measurements, and statistically analyzed to assess age effects. Participants, especially younger speakers, are predicted to produce Diné [kx, tx] with shorter releases than earlier measurements, and with releases that align more with similar English sounds. Data confirm that speakers pronounce Diné [kx], but not [tx], with shorter durations; younger participants also produce [kx] with the shortest releases, and with durations closest to English [k h ]. Results suggest that perceptual similarity between velar sounds leads to phonetic convergence, but alveolar sounds are sufficiently distinct to prevent transfer. Together, findings attest to the relevance of minority language contexts to theories of bilingual phonology.

In Hoocąk, flaps become nasal after nasal vowels. Descriptions spanning 150 years suggest that the process has changed in several ways. Its output has evolved from what was probably a nasalized flap to a nasal stop. Flap-nasalization previously counterfed progressive vowel nasalization but now feeds it. Also, the morphological domain of flap-nasalization may be narrowing. We present the first acoustic phonetic study of the process, showing that in Kenneth Miner’s 1974–5 fieldwork recordings, the output of flap-nasalization ranges from flap-like to stop-like durations.

Our study investigates the nature of tonogenesis in Sylheti, which was triggered by the merger of aspirated and unaspirated consonants in the language. We propose that the tonogenetic factors contributed by the four-way laryngeal contrast condition a three-way tonal contrast in Sylheti depending on the voicing and syllabic position of the sound. We take into consideration both monosyllabic and disyllabic sets of words and claim that Sylheti has three lexical level tones: high, mid, and low. We have built linear mixed-effect models of f0 and duration to examine the acoustic correlates of tone contrast in the language. We conclude that Sylheti has a three-way tonal contrast distinctively affecting pitch.

We present a preliminary quantitative analysis of (s) variation in Pirahã. Previous research noted a gender difference in (s) use in which women use [h] in environments where men use [s]. Using word list data, we analyze (s) rates among three men and three women. Our results indicate that women do utilize [h] more than men, but neither use one variant categorically. Furthermore, (s) is more phonetically variable, and varies in more environments, than previously described. We argue that (s) is undergoing lenition from [s] to [h], and that this change is led by women. This suggests the “female lead” found in many sound change studies is not restricted to the West. The pattern is even more striking because social and cultural factors operate very differently in the Pirahã context. We evaluate whether current explanations for the female lead in sound change are applicable in Pirahã.

This paper investigates the status of rhotics, /uː/, and the diphthongization of /eː/ and /oː/ in New Braunfels German, a variety of Texas German, a set of moribund dialects spoken in Texas. Boas, Hans C. (2009a. The life and death of Texas German . Durham: Duke University Press: 158) notes that “retroflex /ɻ/ appears … [mainly] in borrowed words.” He further assumes that New Braunfels German “differs from other German American dialects such as Pennsylvania German and Wisconsin German in that English sounds almost never appear in native German words” (160). The results discussed here indicate that this statement no longer holds true. Today, perceptual analysis and phonological measurements of Texas German Dialect Project data reveal that both American English retroflex approximants and diphthongs [eɪ] and [oʊ] appear frequently. Moreover, /uː/ is sometimes centralized/fronted with short/mid-long quality. Language attrition and language death, two mutually reinforcing variables, appear as the driving forces for these phenomena.

We investigate apparent-time variation of the low falling (T21) and low level (T11) tones of the Black Lahu language of Yunnan, China. Linear mixed-effects modeling of spontaneous speech shows that both tones have a higher F0 trajectory among younger speakers and in certain phonetic environments. Since F0 is known to lower with increasing age, for comparison we also analyze variation in the high rising tone (T45) and find no evidence of generational difference. This suggests that the effect of age on low tones is not due to physiological change across the life span. We leave open the question of whether this result reflects a change in progress or a stable sociotonetic difference between older and younger speakers. This study contributes towards two underrepresented areas of sound change research: (1) sociotonetic approaches to tone variation in naturalistic speech styles, and (2) engagement with Indigenous scholars who are cultural insiders in small rural minority language communities.