Glotta

The noun panna is documented mainly in Latin graffiti, in an area covering Hispania, Gaul, Germania Superior, Noricum, and Britannia. In Roman times, this lexeme (the ancestor of English pan and German Pfanne) is only known from epigraphic sources. This fact, combined with the word's territorial distri-bution, has led to a debate about its etymology, with the predominant hypoth-esis being that panna is a corruption of Classical Latin patina or a uox peregrina-or more specifically, a Celtic loan. This debate, which has been active since the beginning of the 2Oth century, is still unresolved, and references to the matter are scattered across many publications. In the present paper, we present the status quaestionis regarding this lexicological issue, reconstruct the history of the word panna from antiquity to the present day, and correct the pertinent article of the Thesaurus linguae Latinae, published in 1984. We conclude that interpreting the form panna as a vulgarism-and possibly as a regionalism-is more plausible than viewing it as a Celtic loan.*

The adjective pi rho upsilon mu nu omicron sigma at the root/base' has no commonly accepted etymology, but its syntax, semantics and pattern of attestation are suggestive of an archaism. After reviewing the problems with existing analyses, I submit that it reflects *k(w)rp-mn-o- derived from the Indo-European root *k(w)rep-(Greek pi rho epsilon pi omega 'to be conspicuous, Arm. erewim 'to appear'), for which I propose to reconstruct a proto-meaning 'to protrude'. Moreover, I argue that pi rho epsilon mu nu omicron nu 'stump, base' is related to pi rho upsilon mu nu omicron sigma vos, reflecting *k(w)rep-mn-o-m. In section 1, I discuss the syntax and semantics of pi rho upsilon mu nu omicron sigma in Homer. Then follows a review of the problems with existing etymological interpretations (section 2) and a discussion of the phonological and semantic issues raised by the proposed reconstructions of pi rho upsilon mu nu omicron sigma and pi rho epsilon mu nu omicron nu (section 3). Finally, in section 4 it is proposed that kappa alpha rho pi omicron sigma 'wrist' < *k(w)rp-o-also derives from the same root.

This paper investigates the origin of the Old Latin amasso-type formations amassim, c & umacr;rassis, seruassit, etc.), which correspond in meaning and function to sigmatic future indicatives such as faxo, respexis, rupsit, etc. and subjunctives such as faxim, dixis, surrepsit, etc. Although there is general agreement that these forms are of analogical origin, most previous accounts fail to cogently explain the occurrence of geminate -ss- following a long vowel. Re-evaluating the distribution of intervocalic sibilants in pre-Classical Latin, including the treatment and nativization of Greek loanwords with intervocalic s, this paper argues against deriving amasso from the present stem amaplus the -s- of faxo, respexis, rupsit, with some exceptional treatment of -s-. Instead, it proposes that amasso reflects a morphological reformation based on the `t-stem' (amat-) of the perfect passive participle. By analogy with forms such as faxo (interpreted as underlying /fact-so/), speakers created /amat-so/, which developed regularly into amasso. This analysis not only resolves the formal anomaly of the amasso type but also underscores the synchronic productivity of the participial t-stem in Latin verbal and nominal morphology.*

The retention of -ss- in the Aeolic dialects implies that Proto-Aeolic had s-stem dative plurals with -ss- (&veso-ot). Yet the Lesbian lyric poets only have forms with a single -s- (&vesot). In an influential study, Morpurgo Davies argued that the &vesotv-type is native to Lesbian and used these forms as crucial evidence for the simplification of -ss- in Proto-Aeolic. However, a closer look at the epigraphic and literary evidence strongly suggests that &vesotv in Lesbian lyric is a poetic borrowing from Ionic and that Proto-Aeolic maintained -ss- in the dative plural of the s-stems. These findings have important implications for the history of Proto-Greek, Proto-Aeolic, and our understanding of the poetic dialect of Sappho and Alcaeus. *

This paper examines the Latin word bacar, which is said by ancient glossaries to denote a type of wine vessel similar to a bacrio. Variants such as bacario/ bacarium further reinforce the association with wine containers. The discus-sion then turns to the etymological relationship between bacar and ba(c)ca, a term cited by Varro as a Hispanic word for `wine'. In this connection, par-ticular attention is paid to the potential link between bacar and the Iberian word baikar, as suggested by their phonetic resemblance. Significantly, baikar is attested exclusively on small drinking vessels. Taken together, the informa-tion from Latin sources regarding the Hispanic origin of ba(c)ca `wine' and the semantic appropriateness of bacar `wine vessel' in relation to the attesta-tions of Iberian baikar support the idea that all these words ultimately belong together.*

The noun panna is documented mainly in Latin graffiti, in an area covering this lexeme (the ancestor of English pan and German Pfanne) is only known from epigraphic sources. This fact, combined with the word's territorial distribution, has led to a debate about its etymology, with the predominant hypothesis being that panna is a corruption of Classical Latin patina or a uox peregrina - or more specifically, a Celtic loan. This debate, which has been active since the beginning of the 2Oth century, is still unresolved, and references to the matter are scattered across many publications. In the present paper, we present the status quaestionis regarding this lexicological issue, reconstruct the history of the word panna from antiquity to the present day, and correct the pertinent article of the Thesaurus linguae Latinae, published in 1984. We conclude that interpreting the form panna as a vulgarism - and possibly as a regionalism - is more plausible than viewing it as a Celtic loan.*

The unique wording of the famous verse aurea prima sata est aetas (Ov met. i,8g) is due not just to a quirky neologism as B & ouml;mer would have it. Instead, the form sata alludes to Saturnus, the god of the Golden Age, based on a famous folk etymology transmitted by Varro and others.

As observed by Hyllested and Cohen (2007), no inherited Greek lexeme con-tains an initial sequence epsilon upsilon-, alpha upsilon-, omicron upsilon- followed by a labial consonant; this leads the authors to propose a sound change which operated in the early stages of Greek: *Vu > u /#_C[+lab], which explains forms such as upsilon phi alpha iota nu omega 'weave' for expected dagger alpha upsilon phi alpha iota nu omega < *h(2)(e)ub(h)-nie/o-. The authors propose that a similar rule might have operated in word-internal position, citing the family of kappa upsilon phi omicron sigma 'hunchbacked' < *keub(h)-. In the present paper, I will seek to evaluate the claims of Hyllested and Cohen (2007). A discussion of kappa upsilon phi omicron sigma and a critical investigation of the possible counterevidence leads to the preliminary conclusion that nothing speaks against the existence of a rule *Vu > (u) over bar /C_C[+lab]. After this, I gather additional possible examples of this rule and discuss each instance in detail. The evidence is found to strongly support the proposed rule.*

The paper derives Latin ringor 'show the teeth', not from *ureng- 'twist, bend, but from *reig- 'stretch'. This is more phonologically consistent with the reconstructed participle *rictus, and has semantic support in Old Irish reflexes as well as in the usage of the Latin verb rigeo, which several times describes facial expressions. In light of this, the paper discusses the meaning of Oscan recta(s) in a mixed language curse tablet (Cm 15), suggesting it may be a hypercorrect spelling for expected ricta(s). Lastly, it discusses a relevant textual crux at Ovid, ars 1.515 lingua ne rigeat, which is emended to lingua ne rigeas*.

The paper discusses the sigma chi eta & micro;a x alpha theta' omicron lambda omicron nu alpha oXov x alpha iota & micro;epsilon rho o & varsigma; and the "accusative of respect" in Greek and Anatolian. It argues that, irrespective of whether the Greek are calques of the Anatolian constructions or were boosted in frequency due to language contact, the use of the sigma chi eta & micro;o in Greek curse tablets is not likely to be due to the influence of the Greek epic language.*

This article examines code-switching and script choice in the Letters of Pliny the Younger, simultaneously demonstrating textual criticism's broader relevance to the study of Roman code-switching. It argues first, through a systematic reappraisal of the evidence, that the manuscript tradition allows us to trace the use of Greek script to antiquity This is particularly true for Books i-g, while dependence on less reliable early printed editions complicates any attempt to evaluate alphabet choice in Book io. Second, by reevaluating numerous proposed switches in the Plinian corpus, the paper shows that many of the code-switches in Latin script identified by previous scholars are better classed as loanwords, while Greek script is used for the most securely classed switches in the Letters and exclusively for phrases longer than a single word, verbs, adjectives, and quotations. Together, these two facts suggest that script is more relevant than previously thought for distinguishing code-switches from borrowings in the Epistulae*

This study delves into the morphosyntactic analysis and interpretation of the which has been the subject of numerous controversies over the years. Traditionally, when not emended, it has been interpreted as a sigmatic future derived from the perfect stem ysya- of the verb yiyvo mu at. The absence of the contraction which is typical of sigmatic futures, particularly with formed from roots ending in -a, complicates this interpretation. Against this background, the study presents a new proposal suggesting that ysyaovrrat be considered as a perfect subjunctive, formed by adding the thematic vowel and primary endings to the perfect stem ysya-. From a syntactic point of view, it is a prospective subjunctive without particle in the main clause, which the poet created in accordance with the Homeric parallel of the passage in which it appears.*

In this paper, I offer a detailed discussion of the rare Latin word stiria f. 'icicle', analyzed both in a synchronic and in a diachronic perspective. Building on the traditional root etymology of Lat. stiria, I (i) adduce new textual, semantic, and morphological arguments supporting the connection of this Latin word with PIE *stieh(x)- 'to stiffen (intr.)' and (ii) specify the final laryngeal in this root as *h1. I further argue against the long-standing view according to which Lat. stiria f. 'icicle' is etymologically related to Lat. stilla f. 'drop'. Instead, Lat. stilla f. is argued to be a cognate of Gk. sigma tau iota lambda eta f. 'drop ( tiny, small thing; a bit)' and thus a derivative of a different PIE root, namely, *stei- 'to drip, run'.

Latin subjunctive complement clauses, particularly in Plautus' Archaic Latin, can either be introduced by a conjunction or consist only of a subjunctive (fac ut uenias type and fac uenias type). This paper aims to explain the contexts in which these constructions are chosen and the reasons for those choices. To that end, all subjunctive complement clauses in Plautus were collected, anno-tated and analysed. The analysis was conducted in accordance with several criteria. The results of the analysis demonstrate that neither the register of a character's speech and their social standing (sex, type and rank) nor the rela-tive position of the main and subordinate verbs influences the choice of con-struction with or without a conjunction. That being said, two criteria are extremely significant: (i) the distance (in number of words) between the main and subordinate verbs (the longer the distances, the greater the probability of conjunction use), and (2) the use of a main verb with deontic (quasi-modal) semantics and well-defined pragmatic functions ("Imperatives", "Desiderative verbs", and "Evaluative expressions"). As a result, we can conclude that in Plautus the subjunctive complementation without a conjunction displays for-mal characteristics of proximity (the absence of the conjunction and the shorter distances between both verbs), which iconically reflect the cognitive and functional proximity between both verbs within the construction (quasi-modal semantic integration and functional specialization).*