Eccentric C-V timing across speakers of diaspora Tibetan with and without lexical tone contrasts Christopher Geissler 1 , Jason A. Shaw 1 , Mark Tiede 2 , Fang Hu 3 1 Yale University, USA 2 Haskins Laboratories, USA 3 Chinese Academy of Social Sciences, China christopher.geissler@yale.edu, jason.shaw@yale.edu, tiede@haskins.yale.edu, hufang@cass.org.cn Abstract This paper investigates the relative timing of onset consonant and vowel gestures in Tibetan as spoken in the Tibetan diaspo- ra. According to the coupled oscillator model of articulatory timing (Browman & Goldstein 2000, Nam & Saltzman 2003), the most readily-available coupling modes among gestures are in-phase (synchronous) or anti-phase (sequential) timing, with competition among these modes also giving rise to a stable timing pattern. The model predicts that other timing relations, i.e. ”eccentric timing”, are possible but not as readily avail- able. Data gathered using electromagnetic articulography (EMA) shows relative C-V timing consistent with either com- petitive coupling or eccentric timing. Competitive coupling is a plausible explanation for CV syllables in a tone language (Gao 2008), but acoustic analysis showed that some speakers do not produce a pitch contrast corresponding to tone. In the apparent absence of a tone gesture, we conclude that these speakers exhibit eccentric C-V timing. Keywords: speech prosody, articulation, tone, Tibetan, Articu- latory Phonology, Electromagnetic Articulography 1. Introduction Languages differ in how articulatory gestures are coordinated in time, including the relative timing between consonants and vowels, i.e., C-V timing. Previous work has observed that in lexical tone languages, such as Mandarin (Gao 2008), Thai (Karlin 2014), and Lhasa Tibetan (Hu 2016), the lag between consonant and vowel gestures (C-V lag) is longer than in non- tonal languages, such as English (Löfqvist & Gracco 1999). Additionally, C-V lag is longer for tonal syllables than non- tonal syllables in Mandarin (Zhang et al. 2019). These obser- vations support treating tone as an articulatory gesture (Gao 2008; Niemann et al. 2011) which can be organized with other gestures as competitively-coupled oscillators (Browman & Goldstein 2000, Nam & Saltzman 2003). In the present study, we focus on Tibetan as spoken in the diaspora enclaves in South Asia and around the world, outside of traditional Tibetan-speaking regions. Speakers raised in these communities are exposed to a mixed linguistic input comprising speakers of tonal and non-tonal Tibetan dialects, along with extensive contact with other languages. Of these diaspora-raised speakers, some produce a tone contrast while others do not. This fact makes diaspora Tibetan speakers a unique case in which to test the effect of tone on articulatory timing. 1.1. Coupling relations Pairwise coupling relations are a prominent way to account for the timing of gestures in Articulatory Phonology. By hypothe- sis, any relative phasing of gestures is possible, but two cou- pling modes are considered to be intrinsically stable: in-phase (synchronous) and anti-phase (sequential) (Saltzman & Byrd 2000). Language-specific timing patterns can arise from the interaction of these two coupling modes, as in the case of competitive coupling. Competitive coupling has been invoked to explain the partial overlap in onset consonant clusters (Nam & Saltzman 2003, Goldstein et al. 2000) as well as the differ- ence in C-V timing between languages with and without lexi- cal tone (Gao 2008). A schematic depiction of coupling graphs and their associated gestural scores is shown in Figure 1. The first two diagrams depict (a) in-phase coupling and (b) anti- phase coupling using common examples of a CV syllable and a CC sequence, respectively. In Figure 1(c), competitive cou- pling is illustrated with anti-phase coupling of consonant and tone gestures and in-phase coupling of both with a vowel ges- ture. While in-phase, anti-phase, and competitive coupling should be most readily available to speakers, cases of other timing relations are possible as well. Other timing relations are called eccentric timing (Goldstein 2011) and represent any timing values besides those already mentioned. Importantly, eccentric timing allows a pairwise timing between two ges- tures similar to that found in cases of competitive coupling, as illustrated in Figure 1(d). (a) (b) (c) (d) Figure 1: (a) In-phase and (b) anti-phase coupling of articulatory gestures. (c) Competitive coupling of consonant, vowel, and tone gestures can result in simi - lar C-V timing as (d) eccentric timing. 1.2. Predictions As a language with lexical tone, Tibetan may be expected to show similar C-V lag to other tone languages. Following Gao (2008), a tone gesture with competitive coupling would predict a C-V lag value of ~50 ms. However, Tibetan as spoken in