All posts by Ranjan Sen

York U talk

I’m giving a paper at the York University, Toronto, DLLL Lecture Series in Linguistics:

The Talking Dead:
Fine phonetic detail in Latin sound change
(abstract below)
Vari Hall, York University

Thursday 6 Nov 2014, 5.15pm
Ross S562


Evaluating the predictions of phonetically based theories of sound change (e.g. Blevins 2004) against structure-based explanations has rendered the fine-grained reconstruction of older languages a crucial enterprise. The sporadic Latin ‘inverse compensatory lengthening’ (Hayes 1989) changed long vowel + single consonant (VːC) into short vowel + geminate (VCC) – liːtera > littera ‘letter’ – in the 3rd-1st centuries BC to judge from inscriptional evidence, and can be straightforwardly explained structurally by syllable weight preservation. Is this the best explanation?

The rule can be distilled into three phonetically guided processes, supporting Kavitskaya’s (2002) phonologisation model of compensatory lengthening. A clear diachronic VːC > VCC occurred in ‘high vowel + voiceless obstruent’: high vowels are intrinsically the shortest, and vowels are commonly shorter before voiceless obstruents than other consonants. Therefore, the phonologically long vowels which were phonetically shortest by nature, in the environment where they were phonetically shorter still, became phonologically short, by phonologisation of that duration. The concomitant lengthening of the consonant can be explained by the hypothesis, supported by several Latin phenomena, that closed-syllable vowels in Latin were longer than their open-syllable counterparts (Sen 2012), contrary to near-universal expectations. Therefore, the short phonetic duration of high vowels before voiceless obstruents resulted in their reanalysis from long vowels in open syllables to short vowels in closed syllables, a structural context to which their longer-than-expected phonetic duration could be attributed. As the only segment which could be causing the closure, the following voiceless stop was realised as a geminate with minimal aerodynamic difficulty.

The process can be explained by a phonetic account of diachronic phonology, rather than invoking structural constraints on change such as ‘weight preservation’.


Blevins, J. (2004). Evolutionary Phonology: the emergence of sound patterns. Cambridge: Cambridge University Press.

Hayes, B. (1989). ‘Compensatory lengthening in moraic phonology’. Linguistic Inquiry 20: 253-306.

Kavitskaya, D. (2002). Compensatory lengthening: phonetics, phonology, diachrony. New York; London: Routledge.

Sen, R. (2012). ‘Reconstructing phonological change: duration and syllable structure in Latin vowel reduction’. Phonology 29 [3]: 465-504.


My joint paper with Joan Beal ‘(W)ho, w(h)en, w(h)ere, and w(h)at? The eighteenth-century pronunciation of “wh”‘ will be presented at the 18th International Conference on English Historical Linguistics (ICEHL18) in Leuven, Belgium, 14-18 July 2014.

Library – Katholieke Universiteit Leuven

Unfortunately, neither Joan nor I can attend the conference. However, we are very pleased that Christine Wallis, the Research Assistant on our BA/Leverhulme-funded research project, will be in Leuven to present the paper. Thanks, Christine!

BA/Leverhulme Grant

‘The Eighteenth-Century English Phonology Database’ funded!

Prof Joan Beal (PI), Dr Nuria Yáñez-Bouza (University of Manchester, Consultant) and  I (CI) have been granted a BA/Leverhulme Small Research Grant from 1 June 2014 for 9 months to construct an Eighteenth-Century English Phonology database.

Watch this space for more details (and see Beal & Sen 2014 for a discussion of the value of such a resource together with a pilot study)!!/image/social.JPG