OUP CORRECTED PROOF – FINAL, 9/2/2015, SPi Syllable and Segment in Latin RANJAN SEN Preview - Copyrighted Material 1 OUP CORRECTED PROOF – FINAL, 9/2/2015, SPi 3 Great Clarendon Street, Oxford, ox2 6dp, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries # Ranjan Sen 2015 The moral rights of the author have been asserted First Edition published in 2015 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2014943829 ISBN 978–0–19–966018–6 Printed and bound by CPI Group (UK) Ltd, Croydon, cr04yy Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work. Preview - Copyrighted Material OUP CORRECTED PROOF – FINAL, 7/2/2015, SPi 1 Philology and phonology 1.1 Approaches to Latin phonology The phonological evolution of Latin from its Proto-Indo-European roots to its separation into the Romance languages has been the subject of impressive scholar- ship by an illustrious register of researchers. Alongside works devoted to specific areas of phonology, such as Graur (1929) on geminates, Bernardi Perini (1974)on stop + liquid and final /s/, and Devine and Stephens (1977) on consonant clusters, several handbooks are wholly or substantially devoted to the phonology of Latin: Lindsay (1894), Niedermann (1997[1906]), Sommer and Pfister (1977), Meiser (1998), and perhaps most thoroughly, Leumann (1977), to name but a few. This book focuses upon isolating the precise phonological conditions for five recalcitrant developments in Latin, which have hitherto resisted explicit and comprehensive formulation. However, I furthermore seek to motivate why those conditions existed and were instigators of change in the language, accounting for the directions and idiosyncrasies witnessed in the phenomena. The role of synchronic phonological structure in guiding and constraining sound change, versus phonetic pressures alone, is much debated (}1.2); consequently a continuous thread in this book consists of an evalu- ation of a key aspect of structure in the context of the developments: the syllable. I argue that syllable structure played an important role in most of the changes investigated, by conditioning surface variants, but was on the whole a step removed from the essential core of the developments, which can be formulated with reference to phonetics alone without allusion to phonological structure. The syllable wasPreview recognized - asCopyrighted a phonologically relevant Material unit in antiquity. The Latin grammarians frequently used the term syllaba without further elucidation; for example, there are 59 occurrences in Servius’ commentary on Vergil’s Aeneid, and 76 in Quintilian’s Institutio Oratoria.1 Many facets of the syllable that are discussed today were identified in antiquity by grammarians of Latin, Greek, and Sanskrit (see Allen 1973: 27, 29–30, 32–4, 53–7): syllabification and syllable-internal phonotactics 1 Counts taken from Perseus for Servius and the IntraText Digital Library for Quintilian, both accessed on 28 May 2009. OUP CORRECTED PROOF – FINAL, 7/2/2015, SPi 2 Philology and phonology (e.g. Herodian G.G. 3.2.393–406), syllable-internal structure identifying the necessity of the nucleus (e.g. Dionysius Thrax G.G. 1.11–12, 16), the distribution of glides and high vowels according to syllable structure (e.g. Priscian G.L. 2.13), syllable weight distinctions (e.g. Dionysius Thrax G.G. 1.17–20), and the relevance of the syllable in metrics (e.g. Longinus Proll. Heph. p. 83). Latin scholarship has continued to develop these observations, invoking the syllable with reasonable regularity in analyses of phonological developments. Some recent studies have been devoted to Latin syllable structure, principles of syllabification, and the syllable’s role in certain diachronic and synchronic phenomena (e.g. Marotta 1999; Lehmann 2005; Cser 2001; 2012). Most previous research on Latin phonology has focused on building upon the insights of the Neogrammarian school of the 19th century by employing the com- parative method, and its principle of the regularity of sound change, to chart the phonological development of Latin throughout its history. Scholars primarily use evidence from (i) Indo-European cognates, (ii) orthography in Latin inscriptions, (iii) manuscripts of Latin authors, (iv) the statements of Latin grammarians in antiquity, and (v) the development of the Romance languages. The establishment of synchronic and diachronic characteristics of Latin phonology in this way remains a most worthwhile enterprise in the light of constantly emerging inscriptional evidence and new interpretation in reconstruction. Alongside this approach, however, another technique for advancing our know- ledge has developed, especially in recent decades. This method considers Latin phenomena from the point of view of the insights of contemporary phonetics, the phonetic and phonological typologies of the relevant processes, and the analyses of different phonological theories. In particular, some scholars have cast a spotlight on phonetic and phonological plausibility in reconstruction, a concept that is only occasionally considered in traditional approaches. For example, the development of w PIE voiced aspirates */b= d= g= g =/ in the Italic languages has been the subject of a great deal of debate, and given rise to variant histories, based on their ultimate outcomes in different phonetic environments in Latin, Faliscan, Oscan, Umbrian, and other languages of the Italic branch. However, by investigating the plausible non- contrastive phonetic aspects of the phonemes in different contexts and positions, Stuart-Smith (2004Preview) provides an- Copyrighted account of their development Material which is consistent with our knowledge of voiced aspirates in modern languages and phonological processes such as devoicing, fricativization and fortition. This line of attack has always been present in traditional scholarship; the com- parative method after all depends upon informal notions of phonetic similarity and plausibility. An early example of the success of employing contemporary typological evidence is Corssen (1858–9), who reconstructs for archaic Latin fixed initial-syllable stress on the basis of evidence from vowel reduction, syncope, and the typology of such processes in modern languages. But the approach gathered momentum with works aimed at reconstructing the pronunciation of Latin, notably Sturtevant (1940) OUP CORRECTED PROOF – FINAL, 7/2/2015, SPi Approaches to Latin phonology 3 and Allen (1978). A crowning achievement of this period is Allen (1973), which investigates several aspects of Latin prosody and phonetics from the perspective of contemporary linguistics; thus he investigates vowel length, syllable weight, syllable structure and syllabification, the prosodic word, and the nature and position of the accent. To illustrate the approach in one detail, Allen draws a strict distinction between vowel length and syllable weight, reporting that their confusion ‘is still unfortunately encountered in some modern handbooks’ (1973: 54). Work of a similar nature is seen in Pulgram (1970; 1975) and Zirin (1971), although the focus of these studies is firmly on the theory and reconstruction of prosodic systems, rather than on diachronic problems in Latin phonology, or on applying evidence from contemporary phonetics. Since that time, studies in Latin phonology have continued to draw upon phon- etics and phonological theory, but the main body of work continues along traditional lines. Sihler (1995) offers several comments on phonetic plausibility, usually in explanatory notes rather than the main text (as to be expected in a comparative and historical grammar of this kind). For example, in his discussion of Latin rhota- cism (intervocalic */s/ > /r/), Sihler (pp. 172–3) comments that ‘a change of s to r may appear extreme, phonetically, but is observed in many languages’, and then provides examples from Proto-Germanic to West and North Germanic, the Eretrian dialect of Ionic Greek, and Sanskrit. In the domain of synchronic typology, he states that ‘very little is known of PIE phonotactics’, but ‘still, typologically speaking, the assumed phonotactics are not exotic’ (p. 169). Such a concern for establishing the phonetic and phonological likelihood of a purported explanation through invoking parallels is, admirably, often seen, but the main difficulty is that phonetic plausibility is com- monly insufficiently