Stress placement in English loanwords into Mirpur Pahari (MP) is used to explore whether a usage-based approach can inform the incorporation of external factors (e.g., exposure to a donor language; here, English) into formal phonological analysis of loanword adaptation alongside internal factors (e.g., phonology of the recipient language; here, MP). Stress placement in English loanwords into MP shows across- and within-speaker variation between conformity to MP stress rules (formalized in classical Optimality Theory; OT) and retention of stress on the syllable that is stressed in English; this is a challenge for most theories of loanword adaptation. In our hybrid approach, variable adaptation patterns in individual speakers’ loanword realizations in production data from twelve MP speakers in the UK are correlated with degree of exposure to English, operationalized as vocabulary size, and a unified formal account is sketched through usage-based weighting of constraints in Stochastic OT.

We investigate acoustic correlates of the laryngeal contrast in intervocalic stops in three speech communities – homeland Polish speakers and first- and second-generation heritage Polish speakers in Toronto. Using a sample of 1,187 tokens extracted from conversational speech, we show that the employment of some parameters signaling laryngeal contrast differs between the homeland and heritage speakers. Some parameters are suppressed (i.e., closure phonation and duration) and some are amplified (i.e., vowel and release duration). We interpret these processes in light of proposals by Flege and Bohn (2021) and Polinsky (2018) regarding contact-induced interaction (i.e., assimilation versus dissimilation between languages). We propose that the interaction between systems occurs at a low level of granularity, that is, at the level of individual phonetic properties of a single phoneme.

Brazilian Veneto (BV), an understudied variety of Veneto spoken in several areas of Brazil, has only one rhotic phoneme, which has been described as alternating between [r] and [ɾ] (e.g., Frosi & Mioranza, 1983). I investigate whether rhotic variation in BV is conditioned by prosodic factors (i.e., position in the syllable, position in the word, and stress) through an analysis of rhotic duration based on data from a naming task. I hypothesize that stronger prosodic positions yield longer rhotics (i.e., trills). The results overall confirm this hypothesis, with longer rhotics favored in stressed, word-initial singleton onsets. Participants’ productions are also analyzed qualitatively, revealing that rhotic variation involves not only the production of trills and taps, but also approximants and fricatives. Potential effects of contact with Portuguese on BV rhotic variation are discussed.

This chapter examines how the front rounded vowels /y/ and /ø/ are realized in two varieties of Chinese Korean: Hunchun and Dandong. In Seoul Korean, these vowels have mostly diphthongized, but variation is reported in other dialects. In a large-scale study including acoustic measurements from the two aforementioned heritage varieties, two homeland varieties (i.e., Seoul and Northern Hamgyeong), and local Mandarin varieties, as well as self-reports of language background, we examine the potential influence of ancestral dialect, prestige dialect, and contact language on the realization of heritage varieties. Results show more monophthongal realizations of the vowels in question in Dandong, Hunchun, and Northern Hamgyeong Korean than in Seoul Korean. We also find that Dandong speakers show less diphthongization of /y/ than Hunchun speakers, whereas the two groups do not differ in the production of /ø/. We attribute this difference to influence from Mandarin, which has a more dominant community-level presence in Dandong than in Hunchun.

This study investigates the effect of changes in voice onset time (VOT) on heritage speakers’ perception of Korean intervocalic stops (i.e., /p, t, k/), and compares their results to those of Korean monolinguals and second language (L2) learners of Korean who are L1 speakers of American English or Mandarin Chinese. A discrimination task using five synthetic /C1V1C2V2/ stimuli that differed in VOT of C2 was created to test inter-group differences. While the L2 learners display categorical awareness of VOT variation, Korean and heritage speakers perceive the two consonants to be the same for most stimuli regardless of VOT values. This unexpected lack of attention to VOT variation among heritage speakers suggests that they may switch their language mode to Korean and activate Korean phonology in discriminating non-phonemic VOT differences. However, their responses are not uniform or robust, with some showing a pattern similar to that of L2 learners, revealing strong individual differences among heritage speakers.

We examined the production and perception of voiced versus voiceless obstruents by thirty-three adolescent heritage speakers of Polish and Russian. First, a word list task was used to elicit the production of voiced and voiceless plosives in word-initial position. Voice onset time (VOT) values for both sets of stops were compared to the values reported for monolingual Polish and Russian speakers. To investigate the perception of phoneme contrasts in the heritage language, we used an auditory phoneme discrimination task that contained ten minimal pairs of real and nonce words with contrasting voiced and voiceless obstruents. The results showed an almost perfect perceptual discrimination and a separation of voiced and voiceless plosives in production; however, the VOT values differed significantly from those of monolinguals. Both groups showed convergence of VOT values towards the values reported for fortis and lenis stops in the majority language (i.e., German), leading to a non-native accent in the heritage language.

This study addresses the following questions: (1) What kind of unique phonological characteristics do heritage Japanese learners share? (2) Does L1 or L2 Japanese share these characteristics? (3) Do these characteristics correspond in oral and written production? (4) What could affect phonological transfer? Longitudinal/cross-sectional data to identify phonological transfer from English to heritage Japanese were collected from diary entries/compositions as well as oral and written tests by heritage Japanese learners growing up in Australia, compositions and written tests by Japanese monolinguals in Japan, and essay translations by L1 English speakers learning Japanese as an L2 in Australia. Possible phonological transfer features in heritage Japanese were examined to see if they occur in the L1 and L2. Correspondences between oral and written transfer features were also investigated, and quantitative analyses examined factors associated with phonological transfer. Findings indicate that phonological transfer is acquisitional, possibly influenced by literacy, the mental lexicon, and both sociocultural and individual factors.

This introduction to the volume provides background on common themes that appear across its chapters. First, it contextualizes the notions of heritage languages and heritage speakers before providing reasons why research on heritage sound systems has lagged behind work on other linguistic areas. It then offers an overview of what we have discovered to date about heritage sound systems, while also pointing out gaps and topics that merit further exploration. Motivated by the discussion in the first portion of this piece, the latter half covers the breadth, goals, and novel aspects of the current volume before providing brief summaries of the empirical studies carried out in each of its seventeen chapters.

We shed light on the question of how narrow information (F) and contrastive focus (CF) are intonationally and syntactically realized by heritage speakers (HSs) of Peninsular Spanish (PS) who have German as their second L1, and compare their data to those of monolingual speakers (MSs) of PS. Results from a production experiment show clear differences between the groups with respect to preferred syntactic strategies and, consequently, the intonational realization of focal pitch accents. The preferred strategy of HSs is stress shift, followed by p-movement and simple clefts, for both focus types. Conversely, MSs mostly use different strategies for each focus type; that is, pseudo-clefts and p-movement for F, and simple clefts and focus fronting for CF. Interestingly, stress shift is not a relevant option. The attested differences support the view that the interface between discourse on one hand, and syntax and phonology on the other, is challenging for bilingual speakers (Sorace, 2011).

We present an analysis of phrasal prosody, with an emphasis on focus-marking, for heritage speakers of Samoan in Aotearoa New Zealand. The analysis is based on recordings of four speakers doing a picture-description task designed to elicit different focus positions and types, from an earlier study of home country Samoan (Calhoun, 2015). All speakers showed features of phrasal prosody similar to those found for home country Samoan; however, there was considerable variation between speakers. We relate this to the language background of the speakers, and their attitudes and beliefs toward their heritage language. In particular, there were differences between generation 1.5 and 2 speakers, relating to their engagement with and beliefs about their university Samoan language classes. This shows the importance of these factors in the acquisition and maintenance of prosodic features, similar to other more-studied language features.

The functional load hypothesis predicts that phonemes with lower functional load are more likely to merge than higher functional load phonemes. Using Cantonese spontaneous speech data from the Heritage Language Variation and Change Corpus (Nagy, 2011), this chapter addresses the functional load hypothesis in a heritage language context by comparing F1/F2 production patterns of two vowel pairs (i.e., /y/~/u/ and /a/~/ɔ/) across thirty-two speakers based on generational group (i.e., Gen0, Gen1, and Gen2) and dominant language (i.e., Cantonese versus English). An analysis of Pillai score values (Nycz & Hall-Lew, 2015) shows a significant decrease in the phonetic distinctiveness of /y/~/u/ for English-dominant speakers (i.e., all Gen2). The /y/~/u/ pair also has a lower functional load. Dominant language, however, was not a significant Pillai score predictor for /a/~/ɔ/. Overall, the results support the functional load hypothesis in a heritage language context and show how it can complement accounts of heritage languages based on dominant language transfer.

This chapter takes an individual-differences perspective on the dual sound systems of American heritage speakers (HSs) of Mandarin Chinese. Based on detailed socio-demographic data and production data on segmentals and suprasegmentals, we build holistic demographic and phonetic profiles for HSs, as well as native speakers and late learners, to explore how different aspects of their two languages (i.e., Mandarin, English) may develop in relation to each other and how individual variation in production may be related to socio-demographic factors. Using multiple factor analysis (MFA), we describe the range of these profiles, identify clusters of variation defined by different socio-demographic factors, and argue that some factors (e.g., age of arrival, language(s) spoken at home) have more predictive power for phonetic profiles than others. Overall, our results suggest a significant, if limited, link between socio-demographic factors and production, but only in Mandarin. We conclude by discussing the advantages and disadvantages of group-based and individual-centered approaches.

This chapter tracks the development of laterals across three generations of Punjabi–English bilinguals living in England. These speakers are hypothesized to speak a Punjabi-influenced contact variety of English that is typically called "British Asian English." In this study, we aim to understand the processes of phonetic and phonological transfer that led to the formation of British Asian English, and how phonetic variation is subsequently adapted and modified by a community. Our study finds that first-generation (Gen1) speakers produce phonetically similar laterals across languages and word positions, suggesting that they have a single crosslinguistic category. In contrast, second- (Gen2) and third- (Gen3) generation speakers show clear acquisition of allophony in English, yet these patterns do not resemble the system reported for the local monolingual accent. Gen3 speakers further show the greatest phonetic distinctions between their English and Punjabi. The results suggest that the English of younger speakers is developing into a distinctive accent that bears similarity to that produced by other British Asian speakers across the UK.

American Israelis are an understudied but regionally important population in Israel, who use American English as a heritage language. Most heritage language situations previously studied investigate low-status heritage languages rather than high-status heritage languages which function as languages of wider communication. Does the majority language (i.e., Modern Hebrew) influence the minority language (i.e., American English) in this unusual case, as predicted by previous research? This question is investigated through a picture-naming task comparing the speech acoustics of stop production in American English heritage speakers, American olim (i.e., immigrants), and native Hebrew speakers. Results reveal a heritage accent in Modern Hebrew rather than American English, with crosslinguistic influence from the minority language to the majority language. This unexpected result is explained using Flege and Bohn (2021)’s Revised Speech Learning Model (SLM-r), which argues that phones in a bilingual’s phonetic system are linked, allowing for, and even predicting, this type of crosslinguistic influence.