Bryn Mawr Classical Review

BMCR 2016.08.42 on the BMCR blog

Bryn Mawr Classical Review 2016.08.42

Giuseppe Pezzini, Terence and the Verb ‘To Be’ in Latin. Oxford classical monographs.   Oxford; New York:  Oxford University Press, 2015.  Pp. xvii, 355.  ISBN 9780198736240.  $135.00.  

Reviewed by Luca Rigobianco, Università Ca’ Foscari Venezia (


The book deals with the question of two phenomena related in some way to the verb ‘to be’ in Latin, namely the so-called prodelision or aphaeresis of e- in est and es (e.g. Ter. Eun. 471 ex Aethiopiast [= Aethiopia est]) and the prosodic omission of final -s between a short vowel and the initial consonant of the following word(e.g. Ter. Haut. 15 dicturu(s) sum). Pezzini, through an accurate philological and linguistic analysis, shows that the two phenomena, seemingly unrelated and mostly overlooked in previous studies, depend fundamentally on the syntactic features of the verb ‘to be’ in Latin and, more precisely, on its possibility of being cliticized under certain conditions.

The book consists of six chapters. After a first introductory chapter (pp. 1-25), the second chapter (pp. 27-97), based on an earlier article by Pezzini,1 is devoted to the collection of evidence about the use of the contracted forms -st and -s attached to the preceding word instead of, respectively, est and es (a phenomenon also attested in the Sabellic languages, as pointed out by Pezzini). Pezzini specifies the phonetic contexts in which the phenomenon seems to occur, i.e. after words ending in a vowel, in elidible -m and in short vowel + -s, and then he considers the different sources which attest the use of the contracted forms -st and -s (with the use of -s being much less evident). These forms are directly attested in manuscripts (especially, but not exclusively, of Plautus, Terence, and Lucretius) and inscriptions. Furthermore, they can be assumed on the basis of both metrical reasons and problematic uncontracted forms in the manuscript tradition of a text. Traces of these forms can also be detected in the linguistic reflections of certain ancient grammarians (Marius Victorinus,2 Consentius,3 and Velius Longus).4

The third chapter (pp. 99-139) is divided into two sections. In the first section, Pezzini shows that the contracted forms -st and -s cannot be explained as either abbreviations or the outcomes of a phonetic process, as the terms prodelision and aphaeresis suggest. -st and -s are, rather, clitic forms, as evident from both their phonological features and the univerbation to the preceding word. A paragraph of this first section focuses on the doubtful pattern -est < -ĭs + est (e.g. Ter. Haut. 1019 consimilest [= consimilis est] moribus [sc. natus]) instead of the expected -ist, on which Pezzini, after taking into account different explanations, suspends his judgment. The second section of the third chapter consists of an analysis of the spread of the contracted forms -st and -s in different types of text across different historical periods.

In the fourth chapter (pp. 141-191), Pezzini analyses the variation between contracted (-st, -s) and uncontracted (est, es) forms in Terence’s corpus, considering various factors which could possibly account for this variation (metrical constraints, semantics of the verb ‘to be’, and stylistic factors). The semantics of the verb ‘to be’ seems to be one of the most relevant factors. Specifically, Pezzini, in accordance with the observations of Soubiran5 and Fortson,6 shows that the contracted forms -st and -s are preferred when the verb ‘to be’ is used as an auxiliary, are optional when the verb ‘to be’ is used as a copula or in idiomatic expressions (such as opus est), and are avoided when the verb ‘to be’ has a locational or existential meaning. In addition, the use of the contracted forms also depends on the syntax and word order: the contracted forms -st and -s do not occur after monosyllables, after syntactic breaks, and when est and es break a strong syntactic bond (such as that between a name and its modifier).

The fifth chapter (pp. 193-234) addresses the phenomenon labelled by Butterfield as ‘sigmatic ecthlipsis’,7 i.e. the prosodic omission of final -s between a short vowel and the initial consonant of the following word, usually attributed to a weak pronunciation of final -s. In this regard, Pezzini shows that sigmatic ecthlipsis has little supporting evidence, contrary to what is usually assumed. Specifically, on the basis of a thorough analysis of Terence’s corpus, Pezzini demonstrates that the apparent cases of sigmatic ecthlipsis can be explained for the most part as cases of iambic shortening (e.g. Ter. Haut. 617 sătĭ(s) contemplata) or as cases of cliticization of forms of the verb ‘to be’ beginning with s-, such as sum, sim, sis and sit (e.g. Ter. An. 736 ŏpŭ(s) sit). After a final summary chapter (pp. 235-248), the book ends with three useful appendices (Evidence for Contraction in Terence; Omission of Final -s in CIL I2; Lines Potentially Involving Sigmatic Ecthlipsis in Terence), a reference list, and three indices (General Index; Index of Words; Index Locorum Potiorum).

From a methodological point of view, Pezzini’s book stands out for the philological accuracy in the analysis of the texts on which linguistics observations are based, the constant attention to the complex issue of the relationship between speech and spelling, and the judicious application of the quantitative and statistical analysis to the data.

The book goes significantly beyond the state of the art and convincingly clarifies the considered phenomena (the so-called prodelision and sigmatic ecthlipsis). For this reason, its results will necessarily have to be taken into account in the editing of Latin works, for restoring texts more consistent with ‘real’ Latin.

As a final remark, I wish to point out that a discussion of the modern spelling convention for prodelision (’s versus st, cp. e.g. Ter. Eun. 426 tute’s [=tute es], Ter. Ph. 162 aegrest [=aegre est]) would have been interesting. In particular, the sameness of the two phenomena calls for the use of the same spelling convention. In this regard, the use of the apostrophe also for st (e.g. aegre’st instead of aegrest) would be preferable to the use of s without apostrophe (e.g. tutes instead of tute’s), because of the possible misunderstanding of such a form.


1.   Pezzini, Giuseppe (2011), ‘Contraction of EST in Latin’, Transactions of the Philological Society 109, 327-43.
2.   GL 6.22.14ff.
3.   GL 5.389.30ff., 5.402.24ff., 5.403.19ff.
4.   GL 7.80.20ff.
5.   Soubiran, J. (1966), L’Élision dans la poésie latine (Paris).
6.   Fortson, B. W. IV (2008), Language and Rhythm in Plautus (Berlin and New York).
7.   Butterfield, D. J. (2008), ‘Sigmatic Ecthlipsis in Lucretius’, Hermes 136, 188-205.

Read comments on this review or add a comment on the BMCR blog

Read Latest
About BMCR
Review for BMCR
Support BMCR

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010