Lat. disertus revisited

Lat. disertus Revisited 1. Segmentation of diser- and its underlying verbal root In the most recent etymological dictionary of Latin, de Vaan includes disertus ‘skilled in speaking’ as well as disertim ‘clearly, plainly’ under the entry of the verb serō ‘link, join’, together with sermō ‘speech, talk’. 1 The most natural assumption is that he has identified disertus as the perfect passive participle of dis-serō ‘set in order; set out in words’, a prefixed verb of serō. In fact, this hypothesis was presented in earlier works. 2 Even some Roman writers suggested the same etymological connection, 3 for example, Varro (L.L. 6.64), Cicero (de Orat. 1.240 and Div. 1.105, though indirectly), and Paulus ex Festo. 4 Despite the semantic accordance among the forms, however, a phonological discrepancy is non-negligible, that is, the geminate -ss- of disserō vis-à-vis the single -s- of disertus and disertim. This issue is duly pointed out by Bader and Leumann, 5 although de Vaan provides no explanation. Degemination is excluded since the preceding -i- did not undergo the compensatory lengthening (i.e. *-iss> -īs-?) that was caused by the change. The so-called mamilla-rule (a geminate simplification from *mammílla ‘nipple’, a derivative of mámma ‘breast’, triggered by accent shift) 6 is also unlikely here because dissértō, the frequentative form of dísserō, keeps its original -ss-; a change like *dissértos > disertus is thus without basis. 7 In this light, it is better to regard the connection of diser- and disserō, found in Varro, Cicero, Paulus ex Festo etc., 8 as a case of folk etymology 9 – we will come back to this issue later (§3) – and the single -s- of diser- as original; the non-application of rhotacism to this intervocalic -s- is most likely due to a dissimilatory effect triggered by the following -r- as with miser ‘miserable’. 10 VAAN (2008), p. 557. E.g. NIEDERMANN (1925), p. 70; (1953), p. 116; WALDE & HOFMANN (1938-1954), Band 1, p. 356. 3 As shown in ERNOUT & MEILLET (2001), p. 177; MALTBY (1991), p. 191. 4 LINDSAY (1913), p. 64. 5 BADER (1962), p. 268; LEUMANN (1969), p. 547. 6 See WEISS (2020), p. 169. 7 See also ERNOUT & MEILLET (2001), p. 177. 8 See BADER (1962), p. 268; ERNOUT & MEILLET (2001), p. 177. 9 See LEUMANN (1977), p. 179. 10 See HARTMANN (1913), p. 159; SOMMER (1914), p. 78; JURET (1921), p. 124, 235; (1938), p. 57; LEUMANN (1977), p. 179; ERNOUT & MEILLET (2001), p. 177; SIHLER (1995), p. 172. 1 DE 2 Latomus 81, 2022, p. 591-598 – doi: 10.2143/LAT.81.3.3291217 © Société d’études latines de Bruxelles – Latomus, 2022. Tous droits réservés. 592 KANEHIRO NISHIMURA How can we then interpret the diser- of disertus and disertim as well as of the undoubtedly related adverb disertē ‘clearly, plainly’ from an etymological standpoint? If it is unadvisable to pick out -ser- to associate with serō, another feasible segmentation of diser- remains, that is, dis-er-. This option is not new at all, having been suggested already by several scholars. 11 The view shared among them is that -er- came from *-ar- via vowel reduction in a word-medial, closed syllable. Regarding the origin of *-ar-, in turn, two theories have been proposed, and both are mentioned in Leumann’s article, 12 which particularly focused on diser-tus/-tim/-tē. The first possibility considered by Leumann is that *-ar- has to do with ar‘put together’ of artūs ‘joints; limbs’ and that disertus underlyingly means ‘set apart’. 13 Yet Leumann argues instead for a second possibility, namely, that artus ‘tight’ lies behind dis-ertus, and artus in turn continues *arctus, originally a p.p. of arcēre ‘shut up, enclose’, in sync with Hartmann’s idea. 14 In view of co-ercitus, p.p. of co-erceō ‘shut up together’ (< *co-arceō), however, we expect to have not *arctus but *arcitus as an original form (thus, †dis-ercitus with vowel reduction). And yet Leumann argues that co-ercitus is analogical to other verbs of the second conjugation (cf. monitus with -i-, p.p. of moneō ‘advise’). 15 Further, he considers artāre ‘restrict’ without -i- an iterative older than ex-ercitāre ‘train, exercise’ with -i- to defend his i-less *arctus. 16 Such analyses seem to me arbitrary: 17 the appearance of -i- in p.p.’s of some second-conjugation verbs is so stable that it was in all likelihood established at a rather early stage (thus, we regularly find, e.g., monitus, but no trace of a †montus like his ar(c)tus); and artāre looks like a factitive denominative verb of artus rather than an iterative (cf. nouus ‘new’ and nouāre ‘make new’), hence its comparison with ex-ercitāre is unfruitful. Therefore, it is better to reconsider the other option Leumann put aside without any argumentation, namely, the theory that associates *-ar- with artūs ‘joints; limbs’ as well as ars ‘skill, art’ and arma ‘arms, weapons’, whose ar- (< Proto-Indo-European *h2er-) 18 is diachronically continued in Greek ἀραρίσκω ‘fasten, fit together’ and ἀρτύω ‘arrange, prepare’. 19 Indeed, artus ‘tight’ adduced by Leumann as evidence for his arcēre-theory is likely to be in the 11 See K. BRUGMANN apud SOMMER (1914), p. 78; JURET (1921), p. 235; BADER (1962), p. 268; LEUMANN (1969), p. 547-548; (1977), p. 179; ERNOUT & MEILLET (2001), p. 177. 12 LEUMANN (1969). 13 LEUMANN (1969), p. 547. 14 HARTMANN (1913), p. 159. This etymology is also tentatively presented by WEISS (1996), p. 358. 15 See also NUSSBAUM (1994), p. 178-180, n. 57. 16 LEUMANN (1969), p. 547-548. 17 See also JURET (1921), p. 235. 18 LIV2, p. 269-270. 19 See, e.g., K. BRUGMANN apud SOMMER (1914), p. 78; BADER (1962), p. 268. LAT. DISERTVS REVISITED 593 same lexical cohort. 20 Yet Ernout & Meillet, at least tentatively supporting this interpretation (“dis + artus” for « disposé ou qui dispose avec art » or « qui divise bien »), hedgingly add the comment « On n’a pas de certitude ». 21 Although they do not clarify the reason for their hesitation, it possibly comes from semantic issues; while the prefix dis- usually implies distance or separation, the verbal root ar- describes the process of connecting things to each other. This oxymoronic quality has perhaps hindered the etymological segmentation *dis-ar- from being widely accepted. 22 However, such semantic incongruity does not actually exist from the beginning. To prove this we will look at things from a different angle. Among the cohort of diser-forms, disertim is the earliest attested one, going back to Liuius Andronicus. This adverb as well as disertē is also used by Plautus, while the adjective disertus appears slightly later, namely, from Terence onwards. Therefore, it may be instructive to restart our discussion not from disertus but from disertim. Indeed, this approach will lead us to a better understanding of the diser-forms in terms of semantics as well as morphology. 2. The original meaning of diser- and its derivation Liuius Andronicus employs disertim ‘clearly, plainly’ (a frozen ti-stem adverb) in his literary activity – translating Greek works into Latin. We can thus examine the nuance of the word more closely based on the original passage in Greek. Surprisingly, this simple step has been neglected in previous studies. The adverb is used by the “half-Greek” poet 23 in tuque mihi narrato omnia disertim ‘and you must tell me all the things disertim’, known to be a translation of Odyssey 1.169 ἀλλ᾽ ἄγε μοι τόδε εἰπὲ καὶ ἀτρεκέως κατάλεξον. Clearly, ἀτρεκέως ‘exactly, truly’ is the archetype of disertim. This passage represents the scene in which Telemachus calls for Athene, disguised as Mentes (the leader of the Taphians), to tell who (s)he is, where (s)he is from (line 170), and so forth. 24 As the context of the passage shows, Telemachus does not demand any elegance or grandeur in speech from his addressee but truth and/or honesty. 25 Line 174 καί μοι τοῦτ᾽ ἀγόρευσον ἐτήτυμον, ὄφρ᾽ ἐὺ εἰδῶ ‘And tell me this also truly, that I may be 20 See DE VAAN (2008), p. 55-56. ERNOUT & MEILLET (2001), p. 49, state that « la graphie arctus [for artus] n’a aucune autorité ». See also WALDE & HOFMANN (19381954), Band 1, p. 70, for its folk-etymological origin of the spelling with -c-. 21 ERNOUT & MEILLET (2001), p. 177. 22 Cf., e.g., WALDE & HOFMANN (1938-1954), Band 1, p. 356; DE VAAN (2008), p. 557 as above in n. 1. 23 WARMINGTON (1936), p. ix. 24 Regarding the authenticity of lines 171-173 (as well as 174), compare LEVET (1976), p. 188 and PERCEAU (2002), p. 273. Note also different treatments of line 174 between BÉRARD (1924), p. 14 and ALLEN (1917), p. 7. 25 See LEVET (1976), p. 124-125 (« non-déformation », « sincérité »), 129, and passim. 594 KANEHIRO NISHIMURA certain of it’, 26 where Telemachus uses ἐτήτυμον ‘truly’ as an quasi-equivalent of ἀτρεκέως in line 169, may also support this interpretation. 27 The meaning ‘truly’ clearly differs from the nuance of the oft-cited gloss ‘skilled in speaking’ for the adjective disertus. 28 Therefore, Athene answers in line 179 as follows, echoing his word: τοι γὰρ ἐγώ τοι ταῦτα μάλ᾽ ἀτρεκέως ἀγορεύσω ‘Therefore I will frankly tell you all’; 29 the choice of ‘frankly’ for ἀτρεκέως in the translation thus fits in the context. 30 According to Leumann, whether disertim of Liuius Andronicus means ‘truly’ (« genau ») or ‘in detail’ (« ausführlich, in allen Enzelheiten ») cannot be determined from the context, 31 but a close look at the Greek original reveals that the former is more suitable. 32 Nonius also says that disertim is equivalent to plānē ‘clearly, obviously’ (“emphasizing the correctness of a statement” [my emphasis] as per OLD s.v.) 33 and palam ‘openly’, and cites several passages, including those of Liuius Andronicus and Plautus. 34 Based on this analysis, the truth of a speaker’s statement and her/his sincerity is likely to be the semantic foundation of disertim. One might claim that the meaning ‘truly’ has not yet been explained in terms of the form’s components, dis- (denoting separation) and ar- (connection, in contrast). With regard to this issue, I suggest to examine the semantics of ars ‘skill’, a ti-stem noun (cf. gen. sg. artis, pl. artium) derived from our root ar-. Skill or art may involve either positive or negative nuance depending on context; it may be simply admired in some cases, but its evaluation can be mixed in others, especially if those who exercise their skills and abilities are tactically aiming for benefits for themselves (including evil ones): cf., e.g., Odyssey 6.148 (Odysseus speaking to Nausicaa for the first time) αὐτίκα μειλίχιον καὶ κερδαλέον φάτο μῦθον ‘so at once he made a speech both winning and crafty’, 35 in which Odysseus is determined to MURRAY & DIMOCK (1998), p. 25. The adverb ἀτρεκέως is used passim in Od. In 4.486, e.g., we find exactly the same sentence as in 1.169; the context there is that Menelaus asks Proteus for true stories about the fates of the Greek leaders who fought in Troy and then left for their hometowns. 28 A rearrangement of lines 169-176 proposed by PERCEAU (2002), p. 274, does not affect my interpretation. 29 MURRAY & DIMOCK (1998), p. 27. 30 Regarding the etymology of ἀτρεκής ‘exact, precise’, which is the derivational basis of ἀτρεκέως, the details still remain unclear. Cf. LEVET (1976), p. 140; LIV2, p. 635, n. 1; CHANTRAINE et al. (2009), p. 129; BEEKES (2010), p. 165. 31 LEUMANN (1969), p. 549. 32 Pace JURET (1921), p. 235. The latter meaning ‘in detail’ may be attributed to κατάλεξον in the Greek original (whence comes omnia in L. Andronicus, semantically supplementing narrato); for the function of the Greek verb, see PERCEAU (2002), p. 270272, 277-278. 33 In the same vein, as it seems, the translation ‘expressly’ is chosen by WARMINGTON (1936), p. 27, for the passage of L. Andronicus. 34 LINDSAY (1903), p. 819. 35 MURRAY & DIMOCK (1998), p. 231. 26 27 LAT. DISERTVS REVISITED 595 deploy his own skill in speaking quite strategically (by starting to liken his addressee to Artemis); note also that κέρδος, the derivational basis of κερδαλέον, occasionally means ‘cunning arts, tricks’, especially in the plural. In the same vein, the ablative singular of ars, namely, arte can be used as a quasi-adverb meaning ‘craftly, cunningly’. In light of the above analysis of disertim ‘truly’, the semantic contrast between this prefixed form (as well as disertē) and the simplex arte is obvious. Such an antonymic difference is most reasonably ascribed to the presence/absence of the prefix dis-, to whose functions we now turn our attention. The stereotypical definition of dis- is a prefix denoting separation, as mentioned above, e.g., dis-soluō ‘break up’ and dif-fugiō (with consonant assimilation) ‘scatter (intr.)’. On the other hand, the same morphological element may function as a negator of the content represented by the main body of a word, e.g., dif-ficilis ‘difficult’ (↔ facilis ‘easy’), dis-similis ‘dissimilar’ (↔ similis ‘similar’), dis-pār ‘unequal’ (↔ pār ‘equal; a match’), and dis-pliceō ‘displease’ (↔ placeō ‘please’). This use clearly comes along with words denoting a quality of someone/something, particularly adjectives and some verbs involving personal evaluation, and seems to be related to its different function of representing reversal or cancellation of a previous process or action (cf. Eng. un-do with a negative prefix), e.g., disiungō ‘unyoke’ (↔ iungō ‘yoke’); these types of usage may ultimately be continuations of the notion of separation (cf. Eng. off the mark, out of order, etc., where distance entails negative nuance). 36 In this line of reasoning, we can determine with confidence that disertim ‘truly’ originated in the negation (dis-) of craftiness and cunningness (ar-ti-) in speaking. What happened then, following the formation of disertim? 37 As mentioned above, we have a synonymous adverb, namely, disertē. Note that the co-existence of two adverbs ending in -tim and -tē, respectively, from the same derivational basis is not a common phenomenon. 38 Cautim (Plautus +) ~ cautē (Terence, Accius) ‘cautiously’ (cf. cautus, p.p. of caueō ‘beware’) is at least one such case, 39 but cautē is remarkably rare. The statement of Paulus ex Festo 40 disertim pro With regard to the historical background of dis-, see discussions in DE VAAN (2008), p. 172; BEEKES (2010), p. 327; DUNKEL (2014), Band 2, p. 162, n. 7. Its connection with ἀ- (seemingly a privative) of ἀτρεκέως (and the possible influence of ἀ- in translation from Greek to Latin) remains uncertain (see references in footnote 30). 37 LEUMANN (1969), p. 549-550, suggests that several Roman writers including Plautus followed L. Andronicus in using disertim, the view which perhaps leads us to infer that this Greco-Roman poet coined the word. This possibility is hardly proved or disproved. 38 See also the pattern, e.g., strictim (PLAUT. +) ~ ad-strictē (CIC. +) ‘tightly’ with considerable time lag between their first attestations. 39 We can add a few more cases like coniūctim (CAES. +) ~ coniūctē (CIC. +) ‘jointly’, but their relatively recent appearance in written texts makes it difficult to compare such examples with disertim ~ disertē in terms of historical background. 40 LINDSAY (1913), p. 64. 36 596 KANEHIRO NISHIMURA diserte dixerunt antiqui ‘The ancients said disertim for diserte’ also implies that disertim and disertē did not co-exist at least for a long time. The replacement seems to have been on-going for Plautus; as indicated above, he uses disertē, too, as in Am. 578 satin hoc plane, satin diserte ‘(I’ve said) this clearly enough, truly enough’, where disertē appears to be a near paraphrase of the preceding plānē, an adverb used by Nonius to explain the meaning of disertim, as discussed above. 41 Let us now direct our attention to more details of morphology. First of all, there is a close correlation (though not a derivational relation) in Latin between a tim-adverb (based on a PIE ti-stem noun) and its corresponding tus-form, namely, a p.p. of a verb (< PIE *-to-), e.g., partim ‘in part’ ~ partus (pariō ‘give birth’), statim ‘immediately’ ~ status (stō ‘stand’), passim ‘in a scattered manner’ ~ passus (pandō ‘spread out, extend’), and sensim ‘slowly’ ~ sensus (sentiō ‘feel’) – the latter two with a regular phonological change. On the other hand, we have another pattern – clearly a derivational one – between a p.p. in -tus and an adverb in -tē, e.g., rēctus (p.p. of regō ‘direct’) ~ rēctē ‘correctly’; this is due to the morpheme -ē (orig. the instrumental ending *-eh1 in PIE) that derives an adverb from an adjective in -us, -a, -um (e.g. longus ‘long’ ~ longē ‘far’). Based on these facts, it is clear that forms in -tus played a pivotal role in the existence or formation of adverbs (cf. cautim ~ cautē vis-à-vis cautus [caueō ‘beware’]). In our case, the verbal root ar- ‘put together’ derived artus ‘tight’, which in turn was the morphological basis for artē ‘tightly’; †artim is not found, however. At first glance, disertē ‘clearly, plainly (in speaking)’ might seem to be a prefixed form of artē, but the latter rarely refers to verbal activities. Hence, these two forms must be lexically separate from the diachronic standpoint. We can thus assume that disertim (< *dis- + *ar-ti-) caused disertus to come into being on the model of partim ~ partus etc. and then disertē was derived like rēctē (← rēctus). Alternatively, the pattern cautim ~ cautē, though it is rare (as mentioned above), may have caused disertē to emerge analogically, and then disertus was backformed from it. In either case, 42 it is quite certain that disertus appeared in the Latin vocabulary as a word particularly focused on the act of speaking, since disertim ‘truly (in speaking)’ underlies the whole morphological process. 3. Later reassociation of dis-er- with dis-serō Note finally that ar- is not productive anymore as a verbal root in Latin. It is identifiable only in certain relic nominal forms like, again, artus ‘tight’, most Cf. LEUMANN (1969), p. 550. It is the pattern -tim/-tē ~ -tus that most likely caused disertus to be unlike an i-stem adjective iners, gen. inertis ‘unskillful’ (cf. WEISS [2020], p. 336) with the same verbal root (*-ar-) preceded by a negative prefix (in-) likewise. 41 42 LAT. DISERTVS REVISITED 597 likely a p.p. in origin, which lost an underlying (finite) verb (like Gk. ἀραρίσκω). The same is true for the composite dis-er-. For this reason, and also because dis-er- had developed a specialized meaning, 43 Latin speakers may have unconsciously sought to fill the gap and ended up associating dis-er- with dis-serō, which also has to do with verbal acts, particularly arranging words in a proper way. 44 The meaning ‘skilled in speaking’ (< (*)‘honest in speaking’) of disertus thus arose and is already seen in Terence (Eun. 1011). 45 Kanazawa University. Kanehiro NISHIMURA. BIBLIOGRAPHY T. W. ALLEN (1917), Homeri Opera recognovit brevique adnotatione critica instruxit T. W. A. Tomus III Odysseae libros I-XII continens. Editio altera, Oxford (OCT). F. BADER (1962), La formation des composés nominaux du latin, Paris. R. BEEKES (2010), Etymological Dictionary of Greek. With the Assistance of L. VAN BEEK, Leiden. V. BÉRARD (1924), Homère. L’Odyssée. Tome I. Chants I-VII. Texte établi et traduit, Paris (CUF). P. CHANTRAINE et al. (2009), Dictionnaire étymologique de la langue grecque. Histoire des mots, Paris. M. DE VAAN (2008), Etymological Dictionary of Latin and the Other Italic Languages, Leiden. G. E. DUNKEL (2014), Lexikon der indogermanischen Partikeln und Pronominalstämme. Band 1: Einleitung, Terminologie, Lautgesetze, Adverbialendungen, Nominalsuffixe, Anhänge und Indices. Band 2: Lexikon, Heidelberg. A. ERNOUT & A. MEILLET (2001), Dictionnaire étymologique de la langue latine. Histoire des mots. Retirage de la 4e édition augmentée d’additions et de corrections par J. ANDRÉ, Paris. G. P. GOOLD (1958), A new text of Catullus, in Phoenix 12, p. 93-116. F. HARTMANN (1913), Die Behandlung der lateinischen Wortfamilien im Unterricht, in Glotta 4, p. 144-165. A.-C. JURET (1921), Manuel de phonétique latine, Paris. — (1938), La phonétique latine, 2e édition entièrement refondue, Paris. B. A. KROSTENKO (2001), Cicero, Catullus, and the Language of Social Performance, Chicago. Cf. HARTMANN (1913), p. 159. A manuscript variant dissertus in Catullus 12.9 for disertus – a reading preferred by KROSTENKO (2001), p. 242, 251-252, n. 39 to differtus (cf. GOOLD [1958], p. 93-94) – is perhaps a result of such reinterpretation. 45 I owe a great debt of gratitude to A. D. YATES, an anonymous reviewer, and an editorial committee member for their useful comments and suggestions. All remaining errors are of course my own. The production of this paper was supported by JSPS KAKENHI Grant Number JP20K00608. 43 44 598 KANEHIRO NISHIMURA M. LEUMANN (1969), Lat. disertus, in J. BIBAUW (ed.), Hommages à Marcel Renard. Vol. 1, Bruxelles, p. 547-550. — (1977), Lateinische Laut- und Formenlehre. Neuausgabe, München. J. P. LEVET (1976), Le vrai et le faux dans la pensée grecque archaïque. Étude de vocabulaire. Vol. 1. Présentation générale. Le vrai et le faux dans les épopées homériques, Paris. W. M. LINDSAY (1903), Nonii Marcelli De conpendiosa doctrina libros XX, Leipzig (BT). — (1913), Sexti Pompei Festi de verborum significatu quae supersunt cum Pauli epitome. Thewrewkianis copiis usus, Leipzig (BT). LIV2 = H. RIX et al. (2001), LIV. Lexikon der indogermanischen Verben: Die Wurzeln und ihre Primärstammbildungen. 2., erw. und verb. Aufl., Wiesbaden. R. MALTBY (1991), A Lexicon of Ancient Latin Etymologies, Leeds. A. T. MURRAY & G. E. DIMOCK (1998), Homer: Odyssey. Volume 1. Books 1-12, 2nd ed. revised, Cambridge, MA (LCL). M. NIEDERMANN (1925), Historische Lautlehre des Lateinischen. 2. Auflage, Heidelberg. — (1953), Précis de phonétique historique du latin. Troisième édition, Paris. A. J. NUSSBAUM (1994), Five Latin Verbs from a Root *leik-, in HSPh 96, p. 161-190. S. PERCEAU (2002), La parole vive. Communiquer en catalogue dans l’épopée homérique, Louvain. OLD = P. G. W. GLARE (2012), Oxford Latin Dictionary. Second edition, Oxford. A. L. SIHLER (1995), New Comparative Grammar of Greek and Latin, New York. F. SOMMER (1914), Kritische Erläuterungen zur lateinischen Laut- und Formenlehre, Heidelberg. A. WALDE & J. B. HOFMANN (1938-1954), Lateinisches etymologisches Wörterbuch, 3. neubearbeitete Auflage, Heidelberg. E. H. WARMINGTON (1936), Remains of Old Latin. Volume II. Livius Andronicus, Naevius, Pacuvius, Accius, Cambridge, MA (LCL). M. WEISS (1996), An Oscanism in Catullus 53, in CPh 91, p. 353-359. — (2020), Outline of the Historical and Comparative Grammar of Latin, Second edition, Ann Arbor.