Cross-Linguistic Phonotactics and Acoustic Profile of Tone Sandhi in Multilingual Beijing Mandarin and Zhuang Speakers

Project Abstract

This study investigates the cross-linguistic phonotactics and acoustic profile of tone sandhi in multilingual Beijing Mandarin and Zhuang speakers, with a focus on how tonal systems interact with syllable structure, segmental inventory, and prosodic context across language contact situations. Drawing on a mixed-methods design, we combine production experimental data, perception tests, and corpus-based phonotactic analysis to map how tone sandhi processes vary as a function of lexical tone, syllable type (open vs. closed), codemixing, and speaker bilingual proficiency. The project collects data from a balanced corpus of Mandarin and Zhuang speech, supplemented by controlled elicitation tasks and a perception paradigm using manipulated tonal sequences to assess the salience and boundary conditions of sandhi realizations. We first establish a comprehensive inventory of tone sandhi patterns exhibited by bilingual and multi-dialectal speakers, identifying both language-specific and transfer-driven sandhi strategies, including fission and fusion of tonal trajectories, contour simplification, and register shifts under rapid speech. Acoustic-phonetic analyses employ high-resolution pitch, duration, and spectral slope measurements, enabling the extraction of cross-language statistical signatures for sandhi conditioning such as adjacent tone influence, lexical tone compatibility, and morpho-syntactic boundaries. The study then models the perceptual thresholds for sandhi cues using signal-detection theory to determine which acoustic cues reliably signal sandhi transformations across listeners with varying linguistic backgrounds, thereby illuminating the cognitive representation of tonal strategies in bilingual processing. A sociophonetic layer examines how speaker demographics, language dominance, education, and exposure modulate sandhi productivity and acceptability, testing the hypothesis that extended bilingual use strengthens cross-language phonotactic compatibility or prompts systematic reanalysis of tone-phonology mappings. The results reveal distinct yet overlapping sandhi environments Mandarin-influenced tone adjustments predominantly occur in phrase- and sentence-final positions and within rapid sequential tones, while Zhuang-influenced sandhi exhibits more robust contour preservation with selective simplification in complex tonal sequences. Cross-linguistic patterns demonstrate partial consonant-timing coordination, with timing adaptations in tonal onset and offset aligning to syllable structure constraints, suggesting a shared prosodic framework guiding sandhi productivity. The findings contribute to theoretical debates on tonal phonology by illustrating how bilingual repertoires negotiate phonotactic constraints and by refining models of tone sandhi as dynamic, context-sensitive mappings rather than fixed rules. Practical implications include improved computational models for Mandarin–Zhuang mixed-language speech synthesis and recognition, better diagnostic tools for language maintenance in multilingual communities, and enhanced pedagogical materials that reflect authentic cross-language sandhi phenomena. The study concludes with a synthesis of cross-language influences on phonotactics and acoustic realization, proposing a typology of sandhi strategies that accounts for both universal prosodic pressures and language-specific constraints, and outlines directions for future work on broader Sino-Tibetan language contact situations.

Project Overview

What This Project Is About

---

A plain-language overview of how speakers of Beijing Mandarin and Zhuang use tone and sound patterns when they learn and speak multiple languages. The project looks at how tones change in different language contexts and how this affects the way words sound and are understood across these two languages.

---

---

The Problem It Addresses

---

There is limited understanding of how tone sandhi (tone changes in sequence) works when speakers juggle Mandarin and Zhuang, two languages with distinct tone systems. This gap matters for teaching, speech technology, and social communication in multilingual communities.

---

---

Objectives of the Project

---

1. Describe how tone sandhi operates in Mandarin and Zhuang when used by multilingual speakers.
2. Identify common patterns and differences in tone changes across the two languages.
3. Explore how phonotactics (rules about sound combinations) influence tone use and perception.
4. Provide practical suggestions for language teaching and speech technology tools.

---

---

What You Will Do Step by Step

---

1) Review relevant literature to understand existing findings on tone sandhi and phonotactics. 2) Collect speech samples from multilingual speakers of Mandarin and Zhuang. 3) Analyze tone patterns and sound sequences in selected contexts. 4) Compare patterns across languages and identify influencing factors. 5) Discuss implications for teaching and technology. 6) Summarize findings and propose future work.

---

---

Expected Outcome

Anticipated results include a clear description of how tone sandhi functions across both languages in multilingual speakers, a practical guide for educators and developers, and a basis for improving speech recognition and language learning tools. The study aims to contribute to better communication and understanding in multilingual settings.