Читать книгу The Latin Dual and Poetic Diction: Studies in Numbers and Figures - Andrew J. Bell - Страница 16

Table of Contents

For the syntax of the dual we are most indebted to Berthold Delbrück, who, starting from the view that it was primarily intended to designate natural pairs like ὄσσε, ὀφθαλμώ, χεῖρε, πόδε, μηρώ in Greek, showed how it passed on to two objects usually found paired, like ἵππω or βόε, the horses or oxen of the span, or δοῦρε, the pair of spears carried by the Homeric warrior. But when it passed on from pairs to the expression of any two objects not paired in such fashion, it was then used for the plural, and this called forth in return a use of the plural for the dual. Delbrück notes an example of this in:

καὶ γὰρ νῦν δύο παῖδε, Λυκάονα καὶ Πολύδωραν,

οὐ δύναμαι ἰδέειν Τρώων εἰς ἄστυ ἀλέντων (Il. 22. 46),

where the δύο παῖδε are not a pair, but two of the fifty sons of Priam. So in Il. 3. 246 the ἄρνε δύω are any two lambs from the flock. The plural is properly opposed to the singular, its function being the expression of more than one; and so here we have an open invasion of its realm by the dual.

The dual is often used to express several pairs, as in:

ὄσσε δ’ ἄρα σφέων

δακρυόφιν πίμπλαντο, γόον δ’ ὠΐετο θυμός (Od. 20. 348-9),

or in:

Ξάνθε τε καὶ σύ, Πόδαργε, καὶ, Αἴθων Λάμπε τε δῖε

νῦν μοι τὴν κομιδὴν ἀποτίνετον (Il. 8. 185-6),

or in:

κούρω δὲ δύω καὶ πεντήκοντα

κρινάσθων κατὰ δῆμον (Od. 8. 35-6).

In: δύω δέ τέ οἱ Θύραι εἰσίν (Od. 13. 109) we have the plural thus used, and δύω does not emphasize the duality, but gives the number of pairs.

In: λεύσσετον πάντα (Aesch. Eum. 255) we have the dual used of the chorus of twenty-four, which falls into two ἡμιχόρια; and in: καὶ περὶ τούτου πάντες ἑξῆς οἱ σοφοὶ πλὴν Παρμενίδου συμφερέσθων, Πρωταγόρας τε καὶ Ἡράκλειτος καὶ Ἐμπεδοκλῆς, καὶ τῶν ποιητῶν οἱ ἄκροι τῆς ποιήσεως ἑκατέρας κτλ. (Plato Theaet. 152 E) we have the dual used for a plurality of persons falling into two divisions, not paired nor necessarily equal in number. In Russian I find that the dual is used after the numerals two, three, four, and both. In:

αἴ κ’ ἀποκηδήσαντε φερώμεθα χεῖρον ἄεθλον (Il. 23. 413).

Lang and Leaf translate: ‘if through heedlessness we win but the worse prize’; and at first sight one might suppose that ἀποκηδήσαντε was here used of three, Antilochus and his team. We should, however, rather, translate: ‘but if through the heedlessness of you two we win’, &c., recognizing here a fine example of the Σχῆμα καθ’ ὅλον καὶ μέρη. In Latin we find:

Atridas Priamumque et saevum ambobus Achillem (Aen. 1. 458),

where it seems at first sight that ambobus is used for three. Servius’s note runs: atqui tres dixit; sed Atridas pro uno accipe, quos unius partis constat fuisse. But the union here with Priamum leads me rather to accept Atridas for one, it is true, but to take that one as Agamemnon.

We have already noticed:

μή πως, ὡς ἀψῖσι λίνου ἁλόντε πανάγρον,

ἀνδράσι δυσμενέεσσιν ἕλωρ καὶ κύρμα γένησθε (Il. 5. 487-8).

Here the Scholiast, possibly repeating Aristarchus’s explanation, explains ἁλόντε as in agreement with σὺ καὶ οἱ ἄλλοι, which he supplies as the subject of γένησθε. But plainly there is no idea of pairing necessarily implied here, and we have a use of the dual for the plural with nothing to palliate it. In:

ὅτι νῶϊν ἀνήκεστος χόλος ἔσται (Il. 15. 217),

νῶϊν is evidently not of two, but of five: Poseidon, Athene, Hera, Hephaestus, and Hermes. In:

ὡς δ’ ὅτε χείμαρροι ποταμοὶ κατ’ ὄρεσφι ῥέοντες

ἐς μισγάγκειαν συμβάλλετον ὄβριμον ὕδωρ (Il. 4. 452-3),

where we have the two armies of the Greeks and Trojans compared to two wintry streams that fling together their stormy waters, we have the two streams in the plural, the verb in the dual, and the waters in the singular. In Latin the plural octo is an old dual denoting two groups of four each, and viginti is ‘two tens’.

In:

εἴπερ γάρ κ’ ἐθέλοιμεν Ἀχαιοί τε Τρῶες τε

ὅρκια πιοτὰ ταμόντες ἀριθμηθήμεναι ἄμφω (Il. 2. 123-4),

ἄμφω is used not of two individuals, but of two nouns of multitude.

So in:

Se satis ambobus Teucrisque venire Latinisque (Aen. 7. 470).

In prose we often find ambo for bini as opposed to singuli, as in: oculi vel ambo vel singuli (Cels. 6. 6. 14). In an enumeration of warriors Virgil has Assaracique duo (Aen. 10. 124) but in the next verse is: germani Sarpedonis ambo, where ambo is merely a poetic variety for duo. In return we find him using duo for bina in: duo quisque Alpina coruscant gaesa manu (Aen. 8. 661-2), where the dual would be used for a number of pairs. From such uses it is easy to explain Virgil’s use for twelve of now bis seni in: pueri bis seni quemque secuti (Aen. 5. 561), now bis sex in: bis sex thoraca petitum perfossumque locis (11. 9-10). When we compare: tum pendere poenas Cecropidae iussi, miserum, septena quotannis corpora natorum (6.20-2) with Theseus’s words in Euripides:

σώσας κόρους

δὶς ἑπτά, ταῦρον Κνώσσιον κατακτανών (H. F. 1326-7),

we see a use of septena corpora for bis septem corpora, which we shall consider further in a following chapter.

The plural is often used for the dual. By this I do not mean to assert that the plural is often used in Greek or Latin to express two objects. For two objects not forming a pair the plural, not the dual, is the proper idiom. Indeed, even when the plural is used for a natural pair, like oculi or humeri, we are not sure that there ever was a time when it was not properly so used. An examination of the inflexions of the dual makes it probable that the dual appeared later than the plural, and in natural language was never constant in use, but always sporadic. True, in Sanskrit it is used with much regularity and constancy, and I am told the same is true of Gothic; but these are both book languages like mediaeval Latin. When I read in Homer: δύω κανόνεσσ’ ἀραρυῖαν (Il. 13. 407), I have no right to say that κανόνεσσι is a dative plural used for an older dative dual κανόνοιιν: Homer has no such dual. Of natural pairs ὄσσε is the only one he uses always in the dual; he never uses πόδε, though he has ποδοῖιν; χεῖρες is far commoner with him than χεῖρε, and in:

καί ῥ’ ἀπομόρξατο χερσὶ παρειὰς φώνησέν (Od. 18. 200),

‘and with her two hands she wiped off her two cheeks and spake’, Ohler explains that when two duals meet thus in a phrase, for both of them a plural is used. Ohler seems to think there was a time when all natural pairs found expression by the dual only; but there seems little ground for this belief.

Gottfried Hermann thought δύω was a mark of the later use of the dual, as opposed, to the natural use; that ἵππω meant the team, but δύω ἵππω any two horses from the herd. But uses like δύω Αἴαντε (Il. 18. 163), Ἀτρείδα δύω (Il. 1. 16) make me feel that the effect of δύω, as of ἄμφω, is merely to emphasize the duality, just as una emphasizes ambo in: una ambo abierunt foras (Ter. Eun. 702).

But words like θύραι, or foras, or equae, which were once duals, but are now plurals, seem certain examples of the use of the plural for the dual. Words like Μοῦσαι, or νῖκαι, or mensae, or animae, may never have been duals, but may be merely later plurals formed on the analogy of θύραι or equae. But elliptical plurals like Αἴαντες (Il. 7. 164) for Ajax and Teucer, Castores for Castor and Pollux, δεσπόται for ‘the master and mistress’, fratres for ‘brother and sister’ may be asserted as uses of the plural for an older dual.

Some editors have suspected such a dual in: patres natosque videbit (Aen. 2. 579). But Leda is certainly dead by this date, as is probably Tyndareus; even if he is still thought of as living, the idea that patres here is for Helen’s father and stepmother hardly seems to suggest the height of domestic felicity here implied. We have seen how liberis turned the preceding coniuge to the plural in Cicero (Att. 8. 2.3); probably in like fashion natos here has changed patrem to patres, which is here for Menelaus.

But no editor of Virgil, to my knowledge, has suspected an old dual in: vidi Hecubam centumque nurus (Aen. 2. 501). Heyne’s note is: centum nurus latius dictum; quinquaginta enim erant filiae, totidem filiorum uxores seu nurus. But κουράων ... δώδεκ’ ἔσαν τέγεοι θάλαμοι (Il. 6. 247-8) clearly implies that Priam and Hecuba had only twelve daughters. Servius has five explanations to offer, each more improbable than the preceding; he begins: aut finitus est numerus pro infinito ὑπερβολικῶς, and ends by referring the centum to the following per aras, though he finds that in v.523 aras has become haec ara, which would be a transition from a hundred to one. (Aras in v.501 is a plural for a singular, following the analogy of altaria.) It is small wonder that he adds: est autem haec plena adfectu et dolore repetitio. Only his first explanation has any probability in itself, and it seems barred by the quinquaginta thalami in v.503. But we read:

infelix qua se, dum regna manebant,

Saepius Andromache ferre incomitata solebat

Ad soceros (Aen. 2. 455-7),

which Virgil evidently hoped would explain to his readers his parallel use of nurus less than fifty verses after. For soceros and nurus are correlatives, and soceros is certainly for socerum et socrum, as Heyne saw. In nurus we have a plural used for an old dual, which stands, not for one pair, but for fifty pairs, each of a filius and a nurus. In such a case the masculine form is usually preferred, but here we have the close association with Hecuba to determine the gender. Then from the story told by Deiphobus in the sixth Aeneid it is clear that it was in effect a ballroom into which the Greeks were so rudely intruding.

Another plural used like the Greek elliptical dual is: geminosque Triones (Aen. 1. 744 and 3. 516), where there is general agreement among scholars that it is the Great and Little Bear that are meant. Septentriones is shortened to Triones, as is Acroceraunia to Ceraunia (Aen. 3. 506). In: septem subiecta trioni (Geo. 3. 381) we have a tmesis that shows how easy this shortening was; and in Cicero we have for the Little Bear: Minor Septentrio (N.D. 2. 111. 43). In: egressi superant fossas (Aen. 9. 314) fossas seems a similar elliptical plural for: vallum et fossam; we have already noticed geminos sub rupe Quirinos (Juv. 11. 105).

Bentley mentions as a crux for grammarians: fortissima Tyndaridarum (Hor. Sat. 1. 1. 100). Is Tyndaridae, of which we have here the genitive, a nom. pl. masc. of Tyndarides, and so a dual in form merely, but really a plural in meaning, being ‘the four children of Tyndareus’? or is it the plural of Tyndarida, the Latin form of Tyndaris, and so dual in meaning as well as in form, being the two daughters of Tyndareus? Bentley favours the former view, notwithstanding the fortissima, which must have made the phrase seem feminine in meaning to the Roman reader. Lucian Müller, favouring the same view, quotes dulcissime rerum (Sat. 1. 9.4) as a parallel for the lack of agreement in gender. Wickham goes so far as to warn the student against connecting fortissima in meaning with liberta (v.99), with which it seems in agreement; for it is surely her exploit Horace is lauding here. When we recall the old Roman use of cratera for κρατήρ, and that of Ancona for Ἀγκών, we shall see that the Roman would naturally use Tyndarida, the Greek accusative form, as the nominative for Τυνδαρίς. This seems a case where from the Latin Scholia we may learn how the Romans themselves understood the phrase. Porphyrio’s note is: At hunc, etc., Eleganter, quia Clytaemnestra Tyndarei filia fuit, quae Agamemnonem maritum securi percussit; Acron’s: Tyndaridarum, Graecarum mulierum, pro Tyndaridum; significat autem Clytaemnestram aut Helenam. Nam Clytaemnestra Agamemnonem, Helena Deiphobum interfecit, et ambae Tyndarei filiae. Acron’s scholium plainly points to Tyndaridarum as the gen. pl. of Tyndarida, used for Τυνδαρίς. But Tyndaridae here seems used to designate the class of women who kill their husbands, a class that finds its prototype, according to Acron, in either Clytaemnestra or Helen. Fortissima of course belongs to liberta, to whom Horace awards the palm for vigour in her class; divisit medium, making a clean and clear cut. Tyndaridarum, then, is neither a masculine plural nor a feminine dual, but a feminine plural.

With the Homeric use of Αἴαντες for ‘Ajax and Teucer’ we may associate the use of Ἕλληνες for ‘Hellen and his people’ and Teutones for ‘Teuto and his people’ noticed by Hirt (Gr. L.- und F.- Lehre, p. 213). He further cites as a parallel the Skt. śvaśurās, ‘the father-in-law and his connexions’; and Reichelt notices a like use in Avestan (Av. El. p. 221). Along the same line of development seem the Skt. compounds ending in -adi, as in deva indradayas, ‘the gods beginning with Indra’ (Whitney, Skt. Gr. 1302D). In οἱ περὶ Φαβρίκιον (Plut. Pyrrh. 20) for Fabricius, we have the opposite. So Herodotus has οἱ ἀμφὶ Μεγαρέας καὶ φλειασίους, meaning the same as οἱ Μεγαρέες καὶ φλειάσιοι, used immediately after in 9. 69. Out of the Attic use of οἱ ἀμφὶ Πρωταγόραν for ‘Protagoras and his school’ develops the use of this phrase for Protagoras himself, and from οἱ περὶ Ἡράκλειτον ‘the school of Heraclitus’, or οἱ περὶ Ἀρχίαν πολέμαρχοι (Xen. Hell. 5. 4.2) ‘Archias and his colleagues’, seems to develop the phrase for Fabricius cited above.