Early Evidence for Vowel Harmony in Tibetan
The DOI matches its Crossref record; none of the 25 sources its claims cite is checked for support yet.
Abstract
It is to Yuen Ren Chao that we are indebted—among so many other things—for the first modern phonetic description and phonological analysis of a spoken Tibetan dialect; in this pioneering work of his with a Lhasa speaker he noted a number of alternations in vowel quality that seemed to run counter to the generally regular one-to-one correlation to be observed between the vowels of the Tibetan written language and the vocalization of the modern dialects. Chao's work on that occasion was by and large limited to a description of the reading pronunciation for a text in the written language, and hence it then appeared to him possible to treat these alternations as the result of differences in tempo or even as simply fortuitous; but subsequent work with speakers of the same language, involving living speech rather than the reading of literary texts, has shown that the phenomenon which Chao noticed and so meticulously recorded in his early publication is instead part of a widespread process of vowel assimilations which at least in Lhasa Tibetan operates in certain morphemes in a fashion regular enough to merit the designation ‘vowel harmony’.
Overview
CoChat Summary
TL;DR
Miller argues that vowel harmony, well documented in modern Lhasa Tibetan, is also evidenced in early Tibetan manuscripts and inscriptions through assimilatory vowel shifts and the use of the reversed i-graph. He uses this evidence to reinterpret numeral allomorphs such as ‘15’ and ‘18’ and to defend the comparative significance of Tibetan vowel alternations.
Key findings
- Modern Lhasa Tibetan shows morphophonemic vowel harmony: high vowels are i, e, a, ü, o, u and nonhigh vowels are e, ε, a, ö, o, ə; nonhigh suffix vowels are raised after high-vowel bases, e.g. -pa ~ -pa, -šöö ~ -šüï, -cəò ~ -coo, while geminate a blocks the change. Examples include tsaàpa ‘(I) strained’ vs. tsaaqi ‘(I) will strain’, ná.pa vs. náqi, tee.pa vs. tiiqi, and šáána vs. thuuna.
- Early Tibetan manuscript and inscriptional evidence shows both complete and partial vowel assimilation, e.g. WT tshe-riñ = tshiriìn, de-ga = táqa, sru-mo = somo, skyun-po = kuupu, with shifts such as e...i > i...i, u...o > o...o, e...u > i...u, and i...o > i...u.
- The reversed i-graph (gi-gu inversé) is not merely orthographic: apart from graphic/aesthetic uses, it correlates with vowel harmony and represents a qualitative allophone of /i/ in contexts such as a-vocalization, e.g. bgyi// vs. bgyi ba, and myi vs. myï.
- In the numerals, the Mahāvyutpatti standard gives bcu lña and bcu brgyad for ‘15’ and ‘18’, while common bco lña/bco brgyad reflect spoken vowel harmony u...a > o...a; Lhasa cógsúm ‘13’ cannot be explained by Tibetan-internal harmony alone and points to an earlier *a for ‘3’, parallel to Middle Chinese *sâm, so Ulving’s dismissal of comparative significance is rejected.
Methods
- Philological and comparative analysis combining modern descriptions of Lhasa Tibetan (Chang and Shefts; Sprigg) with early Tibetan written records: Tun-huang manuscripts, Central Asian manuscripts, the 821–822 Sino-Tibetan treaty inscription, the Bsam-yas stele, and Pelliot tibétain texts.
- Classifies occurrences of the reversed i-graph in selected texts into columns I–V: I for phonological vowel-harmony contexts; II–IV for graphic, aesthetic, or line-initial uses; V for anomalous cases.
- Compares Written Tibetan forms with modern Lhasa forms and with Chinese reconstructions, especially Middle Chinese *sâm, for the numeral allomorphs.
Limitations
- The paper states that early manuscripts are often carelessly written; the 821–822 stone is badly damaged and rubbings are difficult to read; some i-graph uses are purely graphic or aesthetic; only six examples are cited for any one development; and Sprigg’s and Chang and Shefts’s transcriptions diverge enough to raise doubt whether the same dialect or language is being described.
- Evident from the text: many cited forms depend on editions and plates of varying quality, some manuscript passages are too poorly understood for glosses, and the provided text is heavily OCR-corrupted, obscuring some examples.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 0 Not valid · 23 Not checked · 51 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“it does prove possible to trace both varieties of vowel harmony——that resulting in identical vocalization, as well as that resulting in assimilated but not identical vocalization in two-morpheme words—in manuscript and other earlier Tibetan written records, notably in the precious Tibetan Tun-huang caves and other Central Asian sites.”
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“A certain amount of free variation in these developments has just been noted; in Chang and Shefts's materials it is also important to note that to some extent this variation appears to be a factor going along with if not actually correlated with difference in the speaker's sex.”
What the sources say
- Not matchedcites Chang and Shefts
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“But it is still possible to quote forms from these texts which show that assimilatory vowel shifts of the kinds described above for the modern Lhasa dialect have been a feature of the language from much earlier times”
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the use of this orthographic device in most of the Tung-huang and other Central Asian Tibetan manuscript materials, and for that matter in virtually all of the early written records of Tibetan, notably in the great Sino-Tibetan bilingual inscription of Lhasa memoralizing the Sino-Tibetan treaty of 821-822 A.D., provides valuable additional evidence for early vowel harmony in Tibetan.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Thus, for the male informant on whose speech the treatment is based, they have šúúme 'a fire', while for the same form for the female informant they record šóóme; both are to be put together with WT bžugs-me.”
What the sources say
- Not matchedcites Chang and Shefts
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates an associationSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
20 more claims
Create a free account to see all 74 claims and every source verdict.
- Probably not a substantial claim. 28
- It may state two separate claims as one. 32
- The quote only partly states this claim. 1
Citations
Cited by
Items 101–107 of 107
Vowel Harmony in Pre-Generative Phonology
Kenstowicz, M. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.25
Vowel Harmony in Papuan Languages
Klamer, M. A. F. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.75
Vowel Harmony in Contact Situations
Friedman, V. A., & Joseph, B. D. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.42
Tongue-Root Harmony [Atr]/[Rtr]
Casali, R. F. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.7
Botma, B., & Shiraishi, H. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.62
Vowel Harmony in Caucasian Languages
Butskhrikidze, M. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.73
Hualde, J. I. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780198826804.013.70
Metadata
- Authors
- Miller, R. A.
- Corresponding author
- Miller, Roy Andrew
- Venue
- Language
- Publisher
- Cambridge University Press (CUP)
- Published
- 1966-06
- Volume
- 42
- Issue
- 2
- Pages
- 252-277
- ISSN
- 0097-8507 (ISSN-L)
- 1535-0665
- OpenAlex
- W2332648645
- Cited by
- 107
- References
- 0
- Citations by year
- 2024: 82 · 2021: 1 · 2020: 1 · 2019: 1 · 2018: 15
- FWCI
- 0.70
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Language (publishedVersion)
- Topics
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Phonetics and Phonology Research
- Social Sciences › Arts and Humanities › Language and Linguistics › Linguistics and language evolution
- Social Sciences › Arts and Humanities › Language and Linguistics › Syntax, Semantics, Linguistic Variation
- Keywords
- vowel harmony, Tibetan language, Lhasa Tibetan, vowel assimilation, historical phonology, morphological alternation, Tibetan dialects
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-21, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-21 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.