ἀνά τε ἔδραμεν: EXPLAINING THE LONGEVITY OF THE TMESIS

Kazanskaya Maria N.

UDC 811.14

Philologia Classica. 2022. Vol. 17. Fasc. 1

dva T£ eSpa^ev: Explaining the Longevity of the Tmesis*

Maria N. Kazanskaya

Institute for Linguistic Studies of the Russian Academy of Sciences, 9, Tuchkov per., St Petersburg, 199004, Russian Federation; subura@mail.ru

For citation: Kazanskaya M. N. ava te sSpa|iEv: Explaining the Longevity of the Tmesis. Philologia Classica 2022, 17 (1), 48-59. https://doi.org/10.21638/spbu20.2022.104

It is a well-known fact that tmesis (independent use of the preverb from its verb) as a linguistic phenomenon was progressively eliminated from Ancient Greek, so that only residual usage is attested in the language of the Classical age. However, one verb, avaSpa|iEiv, retained tmetic usage with the particle te intervening between the preverb and the verb, ava te sSpa|iE, until late Antiquity (Appian, Eunapius). It is significant that this construction (on par with the non tmetic form aveSpa^E) was used in prose, which suggests that it was part of actual linguistic usus. The article examines the reasons behind the unique longevity of this tmesis. Following an overview of the occurrences of ava te sSpa^Ev in Herodotus, Appian and Eunapius, and the comparison of the use of the tmetic and non tmetic forms, the elements of the construction are discussed. It is shown that the survival of ava te sSpa^Ev must have been influenced by the semantic development of the verb (the root no longer denotes actual running, but springing to one's feet or rapid growth), as well as the capacity of the preverb ava- to appear independently of its verb (the deontic ava). Finally, a possible shift in meaning of te (as invariable part of the expression) is discussed. While it is impossible to pinpoint one single factor that determined the singular longevity of the tmesis ava te sSpa|iE in Greek, a combination of factors seems to have contributed to its survival.

Keywords: avaSpa|iEiv, ava te sSpa|iE, tmesis, Homer, Herodotus, Appian, Eunapius, deontic ava.

Tmesis in Ancient Greek is a feature mainly associated with the language of epic poetry. It is commonly recognized that tmesis in Homer reflects a linguistic reality of Greek in its early stages, when the preverb was still very close to a preposition / adverb and had not yet become an inseparable part of the verb.1 There exist, of course, close analogies in

* An earlier version of this article was presented at the Indo-European Seminar, hosted by the Institute for Linguistic Studies of the Russian Academy of Sciences (April 10, 2020). I would like to thank the participants of the seminar for their valuable comments and suggestions. The paper and the article were written as part of the collective research project "Indo-European morphological isoglosses" supported by the Russian Foundation for Basic Research (РФФИ № 20-012-00367). I would like to thank reviewers for their highly helpful comments that needed further clarification. Any remaining errors are, of course, solely mine.

1 There is a fair amount of variation in how tmesis is defined in different studies. To quote only a few, Wackernagel 1926-1928, II 170 = 2009, 614 speaks of the original separability of preverbs; Bertrand 2014, 11 defines it as "non-agglutination of the verbal particle to the verb"; Chantraine in his Homeric grammar deliberately avoided the term "tmese" altogether, preferring to speak of flexibility in the use of prepositions, preverbs and adverbs in Homeric Greek (Chantraine 1953, 82-86). On the antiquity of the tmesis in Greek and on the issue of tmesis in Mycenaean, see Marpurgo-Davies 1985, 86-88; Horrocks 1980, 1-5; Horrocks 1997, 202; Duhoux 1998; Hajnal 2004.

the separation of preverbs in other Indo-European languages.2 However, observation of the tmesis in the Homeric poems shows that by his times it was becoming progressively less grounded in actual vernacular use, and was on its way to becoming a poetic device that the poet had at his disposal and could choose to use for metrical or stylistic reasons. But it took some time for tmesis to be eliminated from the Greek language: restricted to a limited number of constructions, it survived until the Classical age, appearing in the Hippocratic corpus, occasionally in Attic prose and Aristophanes, and, most importantly and consistently, in Herodotus.

It has been suggested that tmesis might have been used by Herodotus in imitation of Homer,3 but a thorough study of the examples and the comparison with tmesis in the Hippocratic corpus shows that tmesis must reflect a type of expression that had certain currency in the living Greek language. Furthermore, tmesis shows signs of developing a new usage, because among these prose examples, in Herodotus and other writers, it is possible to distinguish the older, inherited tmesis, used with specific particles and conjunctions, from the tmesis freshly created for emotional or colloquial contexts for expressivity. Thus, the examples in Herodotus fall into the following categories.

(1) as part of the syntactic construction where the verb in the aorist is separated from the preverb by wv; it is important to note that the function of the form in the vast majority of cases is aoristus gnomicus, and the construction is used predominantly in ethnographical passages describing customs, beliefs and practices. For example, Herodotus thus describes what an Egyptian habitually does after accidentally touching a pig: ^v tic; ^auaq aurwv naptwv uoc, aurotai Total l^aTioiai an' wv e^a^e ewutov pa; e; tov noTa^ov ("if one of them in passing touches a pig, with all his garments he cleanses himself, going into the river", 2, 47, 1).4 Similarly, crocodile hunt involves the blinding of the beast: ¿neav Se e^eAKuaGfl e; y^v, npwTov anavTwv o Bnpeur^c nnAw KaT' wv enAaae autoi tou; o^BaA^ou; ("when [the crocodile] is pulled to shore, first of all the hunter plasters mud all over its eyes", Hdt. 2, 70, 2). Priestley in her analysis of these and other examples stresses that in most cases of the gnomic aorist in tmesis with wv appears in descriptions of customs that Herodotus finds surprising; in the remaining cases it conveys a sense of immediacy5. C. J. Ruijgh drew attention to the fact that the tmetic construction with wv had a fixed collocation later in the sentence.6

(2) with ^ev... Se... particles, where the same verb appears in both parts of the antithesis, in the first part in tmesis with the verbal form separated from the preverb by ^ev,

2 Other languages where preverbs enjoyed a fair amount of independence from their verb include Germanic languages, Vedic, Avestan and Latin (Wackernagel's discussion still remains seminal, Wackernagel 1926-1928, II, 171-176 = 2009, 612-618). The fact that this phenomenon is attested in three different traditions allows us to view it as a morphological isogloss.

3 Thus, Smyth 1920, 367, §1652. Aly 1921, 268, who viewed tmesis as a mannerism that Herodotus had supposedly picked up from Hecataeus and that carried distinctly epic connotations. Aly's purely stylistic and literary approach to tmesis in the Histories was rightly contested by Wackernagel (1926-1928, II, 173 = Wackernagel 2009, 615).

4 The text of Herodotean passages in this article follows the latest Oxford edition of the Histories by N. Wilson (2015).

5 Priestley 2009, 126-146 for her analysis of the eighteen examples of tmesis with wv in Herodotus (which she calls type 1 tmesis); cf. the conclusion: "As Stein declared in the late nineteenth century, type 1 tmesis often lends a sense of suddenness or immediacy to an action. However, in Herodotus at least, this is not its only function. Herodotus tends to use type 1 tmesis to mark out things that his audience would in all likelihood have found unexpected, amazing, or unbelievable" (Priestley 2009, 146).

6 As cited by van Ophuijsen, Stork 1999, 119 n.3; this particularity was not specifically noted by Priestley in her study of the construction.

and in the second part with the verbal part omitted. Thus, enumerating the Persian losses in the battle of Salamis Herodotus says: ¿v Se rw novw tout« ano ^ev e0ave o axpaxnyoq Apia^iyvn^ o Aapeiou, Hep^ew ¿wv aSeA^eoc;, ano Se aAAoi noAAoi te Kal ovo^aarol nepaewv Kal Mr|Swv Kal rwv aAAwv au^axwv, oAiyot Se rive; Kal 'EAAr|vwv, "and in this battle died Ariabignes, son of Dareios and Xerxes' brother, and also died many other men, worthy of notice, among the Persians and the Medes and other allies, and a few of the Greeks" (Hdt. 8, 89, 1). Similarly, the legend of the destruction caused by field mice is recounted thus: ¿rnxuGevrac vukto; apoupaiou; Kara ^ev ^ayeiv rouc ^aperpewva; aurwv, Kara Se ra ro^a, npoc Se rwv aaniSwv ra oxava ("[they say that] field mice, pouring in [the camp of Sethos' enemies] at night, devoured their quivers, devoured their bows, and the straps of their shields besides", Hdt. 2, 141, 5). Priestley in her analysis remarks on the aptness of this construction for lists (especially, for listing casualties or destruction).7

(3) tmesis used for expressivity in direct emotional speech, for example, when Amasis, enraged at his repeated impotency, threatens his wife Ladike: 6 yuvat, Kara ^e ¿^ap^a^ac, Kal eari roi ouSe^ia ^nxav^ ^ ouk anoA.wA.evai KaKtara yuvatKwv naaewv, "Woman, you have bewitched me, and there is no way that you will not perish a death most terrible of all women" (Hdt. 2, 181, 3). Although relatively rare in Herodotus (only Hdt. 2, 181, 3 and 7, 12, 1), it is found in iambographers and Aristophanes, and must reflect a reality of the spoken language.8

(4) finally, there is one isolated verb, avaSpa^eiv, that is used by Herodotus repeatedly in tmesis where the verb is separated from the preverb by the particle re in two types of contexts, of a person suddenly springing to his feet from a sitting or lying position (Hdt. 3, 78, 1; 7, 15, 1; 7, 218, 1), and of exponential growth of cities or states (Hdt. 1, 66, 1; 7, 156, 2). All examples occur in the narrative; more importantly, this type of tmesis seems to have persisted after Herodotus, for we find examples of tmetic usage of this verb in late prose (Appian and Eunapius). AvaSpa^eiv seems to be the only verb to have been used with tmesis for such a long time, and yet the reasons why this particular tmesis enjoyed such unique longevity, long after tmesis as a linguistic phenomenon was eliminated from the living language, have not been well explained. Aly set the examples with avaSpa^eiv apart, but viewed them only as another instance of Herodotus imitating epic style;9 Wackernagel in his discussion of tmesis in Herodotus did not pay specific attention to this group of examples;10 Priestley (who classified this group of examples as type 3b tmesis) describes the occurrences, without attempting to explain the usage.11

This paper attempts to explain the singular longevity of ava re eSpa^ev: first the occurrences of this tmesis in prose will be analyzed with special attention to the semantics of the verb, and then the elements that seem to have ensured its continued use will be discussed.

7 Pristley 2009, 163: "Herodotus uses the device as a listing technique for destruction".

8 On colloquial or emotional tmesis, see Wackernagel 1926-1928, II, 173 (= Wackernagel 2009, 615); Willi 2003, 250 (on examples from Aristophanes); Priestley 2009, 169-170 speaks of the hostile tone that is evident in all the contexts where this type of tmesis occurs.

9 Aly 1991, 269.

10 Wackernagel 1926-1928, II, 173 (= Wackernagel 2009, 615).

11 After stressing (in my view, overstressing — see below) that ava r' eSpa^ev appears in combination with a Te... Kai... construction, Pristley notes: "The frequency of type 3B and the regularity of the structural conditions in which it is found strongly suggest that this is an inherited linguistic structure. Any effect that the tmesis may have once had was probably no longer felt" (Pristley 2009, 175).

To start with the obvious, &vaSpa|etv serves as a suppletive aorist form for avatpexeiv and is used in a variety of meanings. LSJ regroups the usages of avatpex« according to the meaning of &va-: (I) "to run back" (hence, "return", "retrace", "have recourse to"), and (II) "jump up and run, start up" (hence, "spurt up", "shoot up (of plants, and metaphorically of cities and people)", "soar aloft (of great themes)")12. Tmesis of the preverb ava- occurs in two uses, both belonging to group II ("start up, spring to one's feet" and "shoot up (of plants)"), and invariably involves the particle te. It is also restricted to two personal forms, 3Sg (ava Te SSpa^ev) and 3Pl (ava Te SSpa|ov).

(1) "spring to one's feet". Thus, Herodotus uses the verb, as the usurper Smerdis and his brother realize that the palace is under attack:

"As they saw the eunuchs in an uproar and shouting, they sprang back to their feet and, as they discovered what was happening, turned to arms." (Hdt. 3, 78, 1).

The expression stuxov... ¿v pouAfl Sxovte; in the previous sentence suggested that the brothers had been sitting as they held their counsel (cf. the adverb ndAiv). Similarly, in book VII the verb in tmesis is used of Xerxes awaking with a start from his dream:

Ssp^nc |ev nspiSs^c Yevo^svo; tfl o^i &vd te SSpa|e ¿k t^c KoiTr|C Kal ns^nsi aYYeAov Apta^avov KaAeovta,

"Xerxes, exceedingly frightened by this vision, leapt up from his bed and, lo, he sends a messenger to summon Artabanus" (Hdt. 7, 15, 1).

The narrative here is extremely dynamic, as the pairing of the aorist &vd ts £Spa|e with the historical present ne|nei shows.13 It is worth noting that Herodotus specifies that Xerxes leapt up from his couch (in the other two examples, the point of rest is not mentioned). Finally, the same swift springing to action from a state of repose appears in the description of Phoceans' discovery of Xerxes' forces:

"The Phoceans thus discovered that [the Persians] had arrived. As they were going up, the Persians remained hidden, for the whole mountain was covered in oak-trees. The day was windless, and as there was much rustling, as was natural with leaves scattered underfoot, the Phoceans sprang up and began arming themselves, and immediately the Barbarians appeared.." (Hdt. 7, 218, 1).

Herodotus does not specify what the Phoceans were doing when the rustle of leaves made them aware of the enemy's presence. But it is evident that it was the usual calm

12 See LSJ 1996, 124 s.v. avatpexeiv. Priestley 2009, 171 labels, rather misleadingly, the usage of ava-5pa|ietv of "jumping to one's feet or starting up" as the "literal sense" of the verb, as opposed to ava5pa|ietv used of the growth of cities. However, both usages should be considered figurative (see below, n. 24).

13 Cf. van Ophuijsen, Stork 1999, 204: "ne|nei 'sends': coordinated with avd... £5pa|ie by te. Kai and followed by the imperfect sAsys with the historical present invited comparison with 3.36.4 ava5pa|i«v sGee

its main use seems to be to express the immediacy of Xerxes' response".

camp activities. It is also worth noting, how quickly the narrative progresses after ava re eSpa^ov ol OwKeec, with the imperfect ¿veSuvov highlighting the fact that they only had time to start arming themselves (cf. Se eiSov avSpa; ¿vSuo^evou; onAa, ¿v Gw^art ¿yevovro "as they saw men arming themselves, [the Persians] were amazed" at the beginning of the next sentence), and the adverb auriKa emphasizing how close the Persians actually were and the immediacy of their appearance.

The same usage of avaSpa^etv is found in Appian. Once, to describe springing from one's seat under the influence of a strong emotion:

evGa S^ ^sya ^o^aa; o Avxwvio; ava te eSpa^e x^c sSpa; auv opYfl Kal nepl x^c apx^C ¿neGeiaZev auxot;

"Then Antonius with a loud shout sprang up in anger from his seat and began invoking the gods against them, concerning the office [of the tribune]" (App. B. C. 2, 5, 33).

It is worth noting, how structurally close this description is to Herodotus' depiction of Xerxes' reaction to his dream. This is not to suggest in any way that Appian was imitating Herodotus in this passage: the contexts are too dissimilar (even as regards the emotion, as Xerxes springs up from fear, Antonius from fury). Rather, the formal similarity points to a persisting narrative model, which, incidentally, might have contributed to the preservation of the tmetic construction ava re eSpa^ev. The second example appears in the description of Sertorius' superstition: there was a white fawn in the region that he considered his omen, and if she was not in sight, he would abstain from battle, thinking that it would not be successful, and engage in battle, when she appeared:

S' Wf9r| Sta Spu^wv Spo^w q>epo^£vr|, ava te eSpa^ev o Zepxwpio; Kal euGu;, wanep auxfl npoKaxapxo^evoc, ^Kpo^oAiaaxo ¿c xou; noAe^iou;,

"When finally [a doe] was sighted galloping through the woods, Sertorius would spring up and straight away, as if he was serving her the first and best parts of sacrifice, would shower the enemy with javelins" (App. B. C. 1, 13, 110).

Once again, the resemblance of the narrative context in which ava re eSpa^ev appears in Appian and Herodotus is striking: as in Hdt. 7, 218, in the context of a military campaign, a sign (the good omen) makes Sertorius pass swiftly from his state of inactivity to action. It should be noted, though, that both here and in B. C. 2, 5, 33 the resemblance with Herodotus' usage is formal: Appian does not owe the expression to Herodotus, he only inherited the type of narrative context where it could be used. It is also important that we find contexts where Appian used avaSpa^etv to denote springing to one's feet without tmesis, which suggests that in his times the tmetic form was one of the possible variants that carried specific connotations.14

(2) "shoot up (of plants)", used metaphorically of rapid growth of cities, states, but also individuals. Herodotus has two examples of avaSpa^etv with tmesis in this sense. The first occurs in the conclusion of the account of Lycurgus' reforms, as Herodotus formulates the effect that the reforms had on Sparta:

14 Thus, Kal o Mapioc;, ¿nsirs i^Kouasv, Ü9' ^Sovrjc; dvsSpa^sv wc; afrroc; op|ir|a«v ¿nl to spyov, "and Marius, when he heard it, sprang up with joy, as if he was going to rush to the deed himself" (App. B. C. 1, 8, 72).

"Thus changing their constitution, they turned to eunomia (good laws), and after Lycurgus' death having built him a shrine revere him greatly <up to now>. And as their land was good, and the number of men not small, they shot up immediately and flourished" (Hdt. 1, 66, 1).

The combination of avd ts SSpa|ov and euGevfGraav shows that, when speaking of Sparta's rapid growth and prosperity, Herodotus is primarily using a metaphor from vegetative growth. A similar combination of verbs used of plant growth, appears in the description of Syracuse's ascent to greatness under Gelon:

[...] o Ss ta; ZuprKouaa; ¿Kpdtuve, Kal r|adv ol ndvta al ZupfKouaai. Al Ss napautiKa &vd t' SSpa|ov Kal spAaatov,

"Gelon, on his part, reigned over Syracuse, and Syracuse was everything for him. And the city shot up straight away and grew" (Hdt. 7, 156, 1).

While euGeve« denotes a thriving state in general (whether that of plants or of animals), pAaatetv is restricted to the sprouting or burgeoning of plants15. The combination of avd t' SSpa|ov and SpAaatov would suggest the image of a plant that grows rapidly in height, but also, through its many sprouts, in width. In both cases avd ts £Spa|ov is used of development of the cities at a much higher rate, following a change in conditions (change of constitution for Sparta; Gelon's personal governance over Syracuse), the verb highlighting the change of pace.

A similar usage of avd ts £Spa|ov occurs in the Roman times, in Eunapius' description of young Porphyrius' study under Cassius Longinus in Athens:

tux&v Ss t^c npoanKouan; naiSeia;, &vd te sSpa|e toaoutov Kal ¿neSwKev, w; — AoYYivou |sv r|v &Kpoatf; — Kal ¿Koa|ei tov SiSdaKaAov ¿vto; oAiyou xpovou,

"having received due education, he shot up to such an extent and even more, that — for he attended Longinus' classes — he even became a jewel in his teacher's crown in a short while" (Eunap. Vit. soph. 4, 1).

Here avd ts £Spa|e is used of Porphyrius' unusually rapid development (also unexpected, seeing that he was not a Greek): while it stems from the same transfer of avaSpa|etv denoting vegetative growth to a different sphere (growth of states or in this case, of individual), Eunapius' use of avd ts £Spa|e is manifestly independent of Herodotus. Although this is the only occurrence of the verb avaSpa|etv in the Lives of Philosophers and Sophists, Eunapius seems to have used the verb in his other writings, possibly, both with and without the tmesis.16

15 Chantraine 1968-1977, I, 178, s.v. pXaaravw notes on the word family: « Tous ces termes s'appliquent à la croissance végétale, et notamment aux bougeons ; ils se sont prêtés à un emploi métaphorique ».

16 The verb àvaÔpa|i£Îv of children coming of age occurs in the fragments of Eunapius' Universal History, preserved in excerpts: naiÔ«; Ôè aûrwv Kal toù oîkstikoù npoc; te r^v eÛKpaaiav rwv àépwv àvéÔpa|iov Kal napà r^v rçXiKÎav rçpr|aav. •• "their children and the children of their servants grew swiftly due to the temperance of our climate, and reached manhood before coming of age.." (Eunap. fr. 42 Blockley). Concerning the use of àvéÔpa|ov without tmesis, it is, of course, possible that the Byzantine compiler modified Eunapius' original expression (as is always the case with fragments); but it is equally plausible that Eunapius may have had both the tmetic and non-tmetic variant at his use.

These examples not only demonstrate a remarkable consistency in the use of ava te eSpa|e(-ov) throughout the ages (seven centuries separate Herodotus from Appian and nine from Eunapius), but also show that the expression was very naturally used in prose, with no indication that it stood out in any way from the general style of the narrative: thus, it cannot be regarded as a poetic feature or an imitation of epic style. It is also worth noting that this tmetic use of ava re eSpa|e cannot be traced back to Homer.17 Although there is one occurrence of avaSpa^etv with tmesis in the poems, in the simile likening Diomedes at the sight of Hector to a man awed by a torrent, its usage is very different:

"As when a man, helpless, on his way through a great plain, stands still before a swift river, flowing into the sea, and seeing it roaring with foam, runs back." (Il. 5, 597-599).

The preverb ava- in this case is used in the sense "back; backwards" which obviously sets this example apart from the occurrences of ava re eSpa|e in later prose. It is also the only example of this particular tmesis in Homer, and an author wishing to use tmesis in imitation of Homeric style would probably have chosen one that occurs multiple times in the poems (in particular, as part of a formula — e. g., ¿k t' ovo^aZev). Thus, despite the formal resemblance, it is difficult to speak of continuity between Homeric and prose usages of avaSpa^etv with tmesis.

Finally, prose writers seem to have had both the tmetic and non tmetic variants of avaSpa^etv at their disposal. This can only be inferred for Herodotus who uses the personal forms (3Sg. and 3Pl.) of avaSpa^etv only with tmesis; however, in three examples the verb appears in the form of participle or infinitive without tmesis.18 But, as has been shown above, Appian and Eunapius seem to have used both variants.

This much can gathered from the occurrences of ava re eSpa|e in prose. Before turning to the question of its unique longevity, there is one piece of testimony that should be mentioned concerning the views of ancient grammarians on the construction. The scholia Londinensia to Dionysius Thrax19 contain a section nepl 'IaSo; that lists features that the

17 As has been rightly recognized by Priestley 2009, 170 and 172-173 (contrary to Aly 1921, 269 who viewed tmesis of ava xe eSpa|ie as heritage of the epic style in Herodotus). However, for Priestley the main point of difference between Homeric and Herodotean usage is not the semantics of the verb, but the fact that in Homer xe stands alone, and is not used as a part of xe. Kai... construction as is typical for Herodotus (Priestley 2009, 172). However, as will be argued below, the role of xe in ava xe eSpa|ie appears to be different.

18 Tatixa Se eina; ¿Ad|ipave xa xo^a 6; Kaxaxo^euawv auxov, Kpoiao; Se avaSpa|iWv eGee e^w, "With these words [Cambyses] reached for his arch to shoot him, but Croesus, springing to his feet, ran out" (Hdt. 3, 36, 4); ¿v xauxr|ai xfjai npoaoSoiai x^; laxn; Aeyexat paatAea 9r|eu|ievov xpl; avaSpa|ieiv ¿k xoti Gpovou, Seiaavxa nepl xf axpaxtf, "during these attacks of the battle, it is said that the king, observing [the military action], jumped up from his throne three times in fear for his army" (Hdt. 7, 212, 1); Seuxepr| Se ^|iepr| ano Tfj; ¿|пpraioc A9r|vaiwv ol 9ueiv uno paaiAeo; KeAeuo|ievoi 6; avepr|cav ¿; xo lpov, wpwv pAaaxov ¿k xoti aTeAexeo; oaov xe nr|xuaiov avaSeSpa|ir|K6Ta, "on the second day after the burning of Athens, as those who were sent by the king to make a sacrifice entered the sanctuary, they saw an offspring from the stump that had shot up to about an elbow's length" (Hdt. 8, 55).

19 These scholia are preserved in two manuscripts, the codex Londinensis Add ms 5118 and the codex Matritensis 81; Alfred Hilgard named this group of scholia scholia Londinensia after the former manuscript as being of superior quality (see Hilgard 1901, XXXII-XXXVI).

compiler considered typical of the Ionian dialect, where avd ts eSpa^e is cited to illustrate the separation of compound words:20

Ta auvGeta pf|ata SiaAuei Kal |eta^u ti autwv Aa|Pdvei, olov &vd t' sSpa|ov,

"[This dialect] dissolves composite words and puts something in the middle, as for example,

&vd t' sSpa|ov" (schol. in Dion. Thrac. vol. I, 468 Hilgard).

This is the final entry of the section. Alfred Hilgard, probably influenced by other examples from the section that are for the most part, though not exclusively, taken from Homer,21 surmised that olov avd t' eSpa^ov must refer to II. 5, 599. However, this cannot be right: were the scholiast thinking of the Homeric simile, he would have retained the form avd t' £Spa|e, or even cited the whole hemistich avd t' £Spa|' omaa«. Moreover, Homeric poems offer a plethora of other tmeses, many repeated multiple times, and the scholiast choosing avd t' eSpa^ov of all examples would be difficult to explain. It is thus much more likely that his reference was Herodotus22 and that in his eyes the expression was Ionic, not specifically epic. And while it might be tempting to use this association with the Ionian dialect to deduce that avd ts eSpa^e was a relic preserved in one dialect and that it later came to be incorporated into the Koivf, thus surviving until late antiquity, this would not be productive: it would not explain why this tmesis, among all other tmeses that existed in the Ionian dialect, was so singled out.

The explanation of the longevity of the construction avd ts eSpa^e would seem to lie in its constitutive elements, i. e. the semantics and the inner form of avaSpa|etv (where the tmetic usage warrants a separate analysis of the preverb and the root), and the function of the particle Te.

(1) The inner form of avaSpa|etv. To begin with the obvious, in neither sense in which avaSpa|etv appears in tmesis (whether it is used of springing to one's feet or of shooting up of plants) does the root signify running in the literal sense.23 This is evident in the case of avaSpa|etv used of rapid growth, but it is no less significant in the

20 Interestingly, the separation of preverb from its verb that constitutes the classical type of tmesis for modern scholars, occurs in the section nepl 'IdSo; under two rubrics: aside from the entry cited above, al avaatpofal Kal ta uneppata'Iwv«v elaiv [...] uneppata Ss tauta vfnioi, o'l Kata pou;'Ynepiovoc'HeAioio iqaGiov, avtl KatfaGiov (schol. in Dion. Thrac. vol. I, 467-468 Hilgard). The term T|fjauc occurs as well, but is applied to a different phenomenon — a syntagm that had not yet undergone univerbation: al t|faeic 'Iwv«v elaiv, olov neAono; vfaov avtl tou neAonovvraov (schol. in Dion. Thrac. vol. I, 468 Hilgard).

21 Thus, in the list of linguistic features considered Ionic by the compiler, the replacement of imperative by an infinitive form is illustrated by AGrvai| ¿nitetAai (Il. 4, 64), and the replacement of genitive by nominative by ol Ss Suo aKoneAoi (Od. 12, 73). The examples may be continued.

22 If one were to choose one context as the scholiast's source, Hdt. 1, 66 would be a plausible possibility. It is worth noting that Herodotean forms do appear in the section nepl 'IdSo; among the examples that illustrate phenomena proper to the Ionian dialect: one of the most striking examples is tw « avtl tf; au SifGoYYou Kexpntai, olov Gw|ia avtl tou Gau|ia "[the dialect] uses « instead of the diphthong au, as for example, Gw|ia instead of Gau|ia" (schol. in Dion. Thrac. vol. I, 468 Hilgard); the form Gw|ia (and the notion itself) are, of course, associated with Herodotus' Histories, while Homer used the form Gau|a.

23 Priestley's discussion of the semantics of &vaSpa|etv is strangely worded: "Three times in the Histories avd te £Spa|e/£Spa|ov is used of people in a literal sense ('leapt up', 'jumped up and ran'), and twice it is used of cities in a metaphorical sense, derived from the verb's use in relation to plant growth ('shot up')" (Priestley 2009, 171). Actually, in neither case is it possible to speak of literal usage, and Priestley is forced to misstate the verb's meaning when she writes 'jumped up and ran': more often than not, no running is involved (cf. Xerxes' jumping up from his couch, see Hdt. 7, 15, 1), but when it does take place, it is expressed by another verb (e. g. &vaSpa|o>v sGee £^«, Hdt. 3, 36, 4).

case of springing to one's feet: it is, in fact, for this reason that the expression Kpotao; Se avaSpa|wv eGee e^w (Hdt. 3, 36, 4) is not pleonastic, with the participle denoting Croesus' springing to his feet, while his running out of the room is expressed by eGee. It is easy to trace the origins of the use of avaSpa|etv for rapid (vegetal) growth: it must surely have appeared as an expressive synonym for the more neutral verb avepxo|ai, and actually both verbs (used of trees and, metaphorically, of children) are attested already in Homer. Thus, as Odysseus likens Nausicaa to a palm:

A|Aw S| noxe xotov AnoAAwvo; napa pw|w ^oiviKo; veov epvo; avepx6|evov ¿vor|aa,

"Such did I see, beside Apollo's altar, a young spring of palm, rising [into the air]" (Od. 6, 162-163).

The point of comparison is youth and slenderness, and the verb avepxo|ai aptly highlights how stately and dignified Nausicaa appears to Odysseus, and how upright was her posture. On the other hand, as Thetis talks of Achilles, she twice likens him to a young shoot, using the verb avaSpa|etv:

^ x' ¿nel ap T£Kov ulov a|U|ova xe Kpaxepov xe e^oxov ^pwwv- o S' dvsSpa|ev epve'i iao;...

"I who gave birth to a son, excellent and valiant, oustanding among the heroes: and <rapidly> he grew, as a palm tree" (Il. 18, 55-56; the comparison is repeated at Il. 18, 437).

In the eyes of a loving and grieving mother, Achilles grew up almost too rapidly; hence, aveSpa|ev that highlights Thetis' pride and grief.24 Thus, in avaSpa|eiv was chosen for this type of semantic contexts not because the idea of literal running was somehow involved, but because it denoted a faster movement than epxo|ai.25

(2) The independence of ava. The preverb ava- has a history of replacing compound verbs meaning "to rise", both in the indicative, and even more significantly, in the imperative forms. Thus, ava with apocope replaces the verb avopvu^ai in the following passage (the non-prefixed verb is used in the first phrase of the sentence):

£fax', 6pxo S' snetxa |eveпт6Ae|oc noAunoixr;, av Se Aeovx^o; Kpaxepov |£vo; avxiGeoio, av S' Aia; TeAa^wviaSn; Kal Sto; 'Ene^;...

"Thus did he speak, and then rose Polypoites, staunch in battle, and the mighty strength of the godlike Leonteus rose as well, and rose Ajax, son of Telamon, and the divine Epeios" (Il. 23, 836-838).26

24 The comparison was well known, and allusions to it are found until late antiquity: cf. [...] ouxw xaxiaxa |iev x^v 'EAAaSa yAwxxav eiAkuoe xaxiaxa Se ei; &K|i^v KaGanep epvo; xi xwv euGaAwv aveSpaiev. (Heliod. Aeth. 2, 33, 3).

25 Incidentally, this was noticed by ancient scholars, as the scholium on Od. 6, 163 shows: veov epvo; dvEpx6|Evov• vewaxl dvEpx6|Evov. o|oiov xw "oS' aveSpa^e epvei iao;" [.] npoaKEiTai Se xw veov xo dvEpx6|Evov, iaov xw avaTpexovTi. xo yap xaxew; iov xpexei, "young shoot, rising upwards: i.e. that had risen recently; same as 'he grew swiftly, similar to a young shoot' [.] the word 'rising up' refers to 'young', and is similar to 'rapidly growing'. For that which moves quickly, runs" (schol. in Od. 6, 163 b2 Pontani).

26 The same construction wpvuxo... av Se... occurs at Il. 3, 267. Monro 1891, 164 §177 treats it as an instance of ellipsis, while Wackernagel 1926-1928, II, 177 (= Wackernagel 2009, 620) speaks of the use of a simple verb in the first clause, and of the preverb (without the rest of the compound verb) in the second clause as a syntactic pattern typical of Homer, thus avoiding the term ellipsis.

The usage of avd replacing the imperative (cf. English "up!") to urge someone to rise from a state of inactivity is also attested from Homer on. There are four examples of this kind in the Homeric epics, invariably in combination with aAAd: aAA' ava mS' sti Ketao ("but get up, and lie no longer", Il. 18, 178); aAA' ava, tdxa vwiv spi; Kal xepal Yevrtai ("but get up, lest strife for us comes to fists", Od. 18, 13); aAA' ava tdxa aatu nupo; Sn'i'oio Geprtai ("but up you get, lest the city be burnt down by hostile fire", Il. 6, 331); aAA' ava el |e|ovdc ys Kal o^e nep ula; Axaiwv / Teipo|evouc ¿pueaGai uno Tpw«v opu|aYSou ("but up you get, if you still wish — late as it is — to free the sons of Achaeans that are beset by the din of the Trojans", Il. 9, 247-248)

In all these examples ava is used independently from the rest of the phrase, even on the level of pronunciation, as ava is left unabridged, even before a word beginning with a vowel.27 As regards the semantics, Luz Conti has defined its usage here as deontic.28 This usage of ava survived in Classical Greek (and, it may be surmised, later as well). Although the construction seems to have been prone to corruption in the manuscripts,29 it is attested as an exhortatory exclamation several times in tragedy: thus, in Sophocles, the chorus exhorts Ajax to action, aAA' ava ¿^ eSpdv«v "but up now, <rise> from your seat" (Soph. Aj. 194a), and in Euripides, Hecuba in her monologue exhorts herself, ava, SuaSai|ov "up, you unhappy one" (Eur. Tro. 98). In the Alcestis, aAA' ava is used by Admetus to his dying wife: aAA' ava, ToA|a... "but bear up, have courage" (Eur. Alc. 277): this appeal, characterized by L. P. E. Parker as "a Homeric, military-style exhortation" (Parker 2007, 111, ad Alc. 277), is interesting in that it shows a certain development in usage, for what Admetus wants is not so much for Alcestis to literally get up from her bed, but rather that she should fight for her life and not surrender to death.

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

These two types of expressions seem to show that ava- enjoyed a greater degree of independence than, for example, rata- (for which no similar deontic usage is attested, cf. English "Down!"). This was not, strictly speaking, a prerequisite for the tmesis avd ts £Spa|e; however, the continued use of deontic ava might have contributed indirectly to its survival.

(3) Particle ts. The choice of the particle ts as the element that separates the preverb from the verb would not, at first glance, seem to be significant: it is however, an invariable part of the expression.30 Jessica Priestly in her examination of avd ts £Spa|e in Herodotus focuses on the fact that in every context where it appears, avd ts £Spa|e is followed by another verb introduced by Kai, which leads her to view it as an instance of the ts. Kai. construction.31 I would like to suggest, however, that the placement of the enclitic particle

27 This is particularly stressed by Schwyzer, Debrunner 1988, II, 424 (§ B.v.ß.7); see also LSJ 1996, 98 s.v. avd.

28 Conti 2015, 40-41. On syntactic independence of dAA' ava, cf. "fungiert hier avd (mit Anfangsbe-tonnung ava) als unabhängige Aussage, die syntaktisch nicht mehr in den Satz integriert ist, dessen Ausdruckskraft es verstärken soll" (Conti 2015, 40).

29 Finglass 2008, 199 (ad Soph. Aj. 194a) lists Aesch. Cho. 963 and Eur. Suppl. 44 as two contexts where exhortative ava may be reconstructed behind the manuscript reading; on the second context, see also Parker 2007, 111 ad Alc. 277.

30 It is interesting to compare it with the fixed usage of wv when the tmesis is used with the empiric or gnomic aorist, Cf. Wakernagel's remark: "We shall perhaps understand the basis of this peculiar type of tmesis, when we work out the origin of the particle wv (ouv), which remains completely obscure" (Wackernagel 2009, 616 = Wackernagel 1926-1928, II, 174).

31 Priestley 2009, 174-175. This assumption leads her then to conclude that "the formulaic nature of dvd T£ s5pa|iov / sSpa^s makes it seem likely that the tmesis itself has little, if any, narrative function. The

xe (that, following Wakernagel's law, is placed in the second syntactic position32) after the separated preverb and not after the verb itself (which would give àvéôpa^év xe Kai...) seems to influence slightly the semantics of xe in the xe. Kai. construction. While retaining its basic connective meaning, xe, as it appears (unexpectedly) after the preverb, highlights the point in the narrative when the course of events starts to develop in a new and even unexpected way. This semantic nuance is perceptible in all the usages of ava xe eôpa^e that have been examined above, especially when compared to the non tmetic form àvéôpa^e, which also denoted rapid upward motion, but one that does not alter the course of events. Thus, in ö ô' àvéôpa^ev epveï iaoç ( "he grew <rapidly>, becoming like a palm tree", Il. 18, 55-56; cf. Il. 18, 437), young Achilles' growth, though unusually rapid, was in no way surprising or disruptive;33 in Aiyexat ßaaiAia ö^eu^evov xpk; àvaôpa^eîv £K xoù öpovou, ôeiaavxa nepi xfl axpaxtfl, "it is said that the king, observing [the military action], jumped up from his throne three times in fear for his army" (Hdt. 7, 212, 1), Xerxes' leaping up from his throne multiple times had no effect on the course of the battle.

To conclude, although we may never discover one single factor that determined the singular longevity of the tmesis ava xe eôpa^e in Greek, a combination of factors seems to have contributed to its survival: the semantic development of àvaôpa^eîv, the possibility that ava- had to appear independently of the verb in other contexts (especially its deontic usage), and the effect of an unexpected turn of events introduced by the separation of the preverb by the particle xe. It is also significant that the tmetic and the non tmetic forms coexisted in Greek language until late Antiquity, and an attentive examination and comparison of their occurrences allows us to capture the subtle difference in their use.

References

Aly W. Volksmärchen, Sage und Novelle bei Herodot und seinen Zeitgenossen: Eine Untersuchung über die volkstümlichen Elemente der altgriechischen Prosaerzählung. Göttingen, Vandenhoeck & Ruprecht, 1921.

Bertrand N. On tmesis, word order, and noun incorporation in Homeric Greek, in: A. Bartolotta (ed.) The Greek Verb Morphology, Syntax, and Semantics. Proceedings of the 8th International Meeting on Greek Linguistics, Agrigento, October 1-3, 2009. Louvain, Peeters, 2014. 11-30. Blockley R. C. (ed., tr., comm.) The Fragmentary Classicising Greek Historians of the Later Roman Empire: Eunapius, Olympiodorus, Priscus and Malchus. Vol. II: Text, Translation, and Historiographical Notes (ARCA Classical and Medieval Texts, Papers and Monographs, 10). Liverpool, Francis Cairns, 1983. Chantraine P. Dictionnaire étymologique de la langue grecque: histoire des mots. Vol. I-IV. Paris, Klincksieck, 1968-1977.

Chantraine P. Grammaire homérique. T. II: Syntaxe. Paris, Klincksieck, 1953. Conti L. Zum adverbialen Gebrauch von ava bei Homer. Glotta 2015, 91, 27-45. Denniston J. D. The Greek Particles. 2Oxford, Clarendon Press, 1954.

effectiveness of the structure seems to lie in the strong, rapid action of the verb and in the suspense or emphasis created by the xe. Kai construction (Priestley 2009, 174).

32 See Denniston 1954, 515-518, on the placement of xe and the deviations from the rule of second syntactic position. Most of the cases discussed concern postponement of xe; he does not discuss xe in tmesis as a case of irregular position of xe.

33 Similarly, ouxw; xaxew; aveSpa^ev ^ noAuxeAeia... "thus quickly did the expenditure increase." (Plut. Mar. 34, 4), where &veSpa|ev appears in Plutarch's conclusive statement on the change of Roman lifestyle, following the information of how drastically the cost of Marius' house at Baiae went up. As the expression sums up what had been illustrated by numbers in the first part of the phrase, there is no element of surprise or disruption that would warrant the use of the tmetic ava xe £Spa|e.

Duhoux Y. Autour de la tmèse grecque: situation dialectale à l'époque mycénienne ; datation de l'épopée. In: L. Isebaert, R. Lebrun (éds) Quaestiones Homericae: Acta Colloquii Namurcensis. Louvain — Namur, Peeters, 1998. 71-80.

Finglass P. J. (ed., comm.) Sophocles Ajax, edited with Introduction, Translation, and Commentary. Cambridge — New York, CUP, 2011.

Hajnal I. Die Tmesis bei Homer und auf den mykenischen Linear B Tafeln — ein chronologisches Paradox? In: J. H. W. Penney (ed.) Indo-European Perspectives: Studies in Honour of Anna Morpurgo Davies. Oxford, OUP, 2004. 146-178.

Hilgard A. (ed.) Scholia in Dionysii Thracis artem grammaticam. Leipzig, B. G. Teubner, 1901.

Horrocks G. Homer's Dialect, in: I. Morris, B. Powell (eds) A New Companion to Homer. Leiden — New York — Köln, Brill, 1997. 193-217.

Horrocks G. The Antiquity of the Greek Epic Tradition: Some New Evidence. PCPS 1980, 26 (206), 1-11.

LSJ 1996 — Liddell H. G. Scott, R. A Greek-English Lexicon, revised and augmented by H. G. Jones, with the assistance of R. McKenzie and the cooperation of many scholars. Oxford, Clarendon Press, 1996.

Monro D. B. A Grammar of the Homeric Dialect. 2Oxford, Clarendon Press, 1891.

Morpurgo Davies A. Mycenaean and Greek Language, in: A. Morpurgo Davies, Y. Duhoux (eds) Linear B: A 1984 Survey. Louvain-la-Neuve, Cabay, 1985. 75-125.

Parker L. P. E. (ed., comm.) Euripides, Alcestis: With Introduction and Commentary. Oxford, Clarendon Press, 2007.

Priestley J. Tmesis in Herodotus. Glotta 2009, 85, 118-178.

Ruijgh C. J. Autour de 're épique': études sur la syntaxe grecque. Amsterdam, Adolf M. Hakkert, 1971.

Schwyzer E., Debrunner A. Griechische Grammatik auf der Grundlage von Karl Brugmanns Griechischer Grammatik. Bd. II: Syntax und syntaktische Stilistik. 5. Aufl. München, C. H. Beck, 1988.

Smyth H. W. A Greek Grammar for Colleges. New York — Cincinnati — Chicago, American Book Company, 1920.

van Ophuijsen J. M. Stork, P. Linguistics into Interpretation: Speeches of War in Herodotus VII5 & 8-18. Leiden —Boston — Köln, Brill, 1999.

Wackernagel J. Lectures on Syntax, with Special Reference to Greek, Latin, and Germanic / ed. with notes and bibliography by D. Langslow. Oxford, OUP, 2009.

Wackernagel J. Vorlesungen über Syntax, mit besonderer Berücksichtigung von Griechisch, Lateinisch und Deutsch. Bd. I-II. Basel, Birkhäuser, 1926-1928.

Willi A. The Languages of Aristophanes: Aspects of Variation in Classical Greek. Oxford, OUP, 2003.

Wilson N. (ed.) Herodoti Historiae. T. I-II. Oxford, Clarendon Press, 2015.

Wyatt W. F. Homeric Hiatus. Glotta 1992, 70 (1/2), 20-30.

Received: 02.12.2021 Accepted: 17.04.2022

ἀνά τε ἔδραμεν: EXPLAINING THE LONGEVITY OF THE TMESIS Текст научной статьи по специальности «Языкознание и литературоведение»

Аннотация научной статьи по языкознанию и литературоведению, автор научной работы — Kazanskaya Maria N.

Похожие темы научных работ по языкознанию и литературоведению , автор научной работы — Kazanskaya Maria N.

Текст научной работы на тему «ἀνά τε ἔδραμεν: EXPLAINING THE LONGEVITY OF THE TMESIS»