Читать книгу The Latin Dual and Poetic Diction: Studies in Numbers and Figures - Andrew J. Bell - Страница 14

Table of Contents

Speaking generally, alius seems to play the part in older Latin that alter plays in Silver and later Latinity. In archaic Latin alius is often used to denote the other of two as well as of three or more. It is also found at times with the meaning of ‘any one’, like aliquis. In Golden Latinity as a rule alter is ‘one of two’, and alius is ‘the other’ of three or more. But in Virgil alter is already used for ‘the other’ of three or more; and in later Latinity and the Romance languages altro or autre is the regular word for ‘other’, while of alius few traces subsist. The Romance languages follow poetic diction, which tends of two words to choose the stronger and more emphatic; and that in this case was plainly alter.

Both start from the root ali-, which we have in the English ‘else’, and in the Greek ἄλλος (=alyos), ἀλλότριος, and reduplicated in ἀλλήλων, where the duals ἄλλω-ἄλλω passed to neuter plurals ἄλλα-ἄλλα, contracted to ἄλλᾱλα (Ion. ἄλληλα) with a single λ in the 2nd syllable, parallel to the single s in misi, as following a long vowel. We have another grade of this al- in ol or ul, that we find in olli, archaic for illi, and in ultro and ultra ‘on yon side’. We read:

alii ventosis follibus auras

Accipiunt redduntque, alii stridentia tingunt

Aera lacu. Gemit impositis incudibus antrum.

Illi inter sese multa vi bracchia tollunt

In numerum versantque tenaci forcipe massam. (Aen. 8. 449-53.)

It is clear that illi is parallel to the preceding two alii, and that it denotes a third class of smiths, who wield the hammer; and Servius’s note is: ‘illi’ quidam pro ‘alii’ accipiunt. Henry wished to read alii here, but the kinship of the words was felt by Virgil.

While autre goes back to the root ali-, not so our ‘other’; we see its root in ‘and’ and in the German andere, in the Sanskrit anyas, and the Greek ἔνιοι; it was once onþer; but the n is lost in ‘other’ just as the n of Gans and hanser is lost in goose. In Greek ἔνιοι has lost the sense of ‘others’, and is ‘some’. So in Latin aliquis has as a rule lost its old meaning of ‘some one else’ and means ‘some one’. But I read in Tacitus: ne eis quidem annis, quibus Rhodi specie secessus exsul egerit, aliquid quam iram et simulationem et secretas libidines meditatum (Ann. 1. 4.4). Our editors usually correct aliquid to aliud, against all the manuscripts, but aliquid here probably still conveyed to Tacitus and his readers the meaning ‘anything else’. So the Skt. anye ‘others’ passed to the Greek ἔνιοι ‘some’; and in Latin aliquos usually means ‘some’; but we read:

Quique sui memores aliquos fecere merendo. (Aen. 6. 664.)

While most editors change aliquos to alios, Sidgwick retains aliquos, the reading of all good manuscripts, but finds a ‘Virgilian pathos’ in his rendering ‘some hearts’ suggesting the narrow range of gratitude for human merit; but it is far more likely that Virgil still felt here in aliquos, the old meaning ‘others’. And in:

Sin aliquem infandum casum, Fortuna, minaris (Aen. 8. 578),

the meaning must be ‘some dread fate other (than his return)’; and Cicero has: vellem aliquid Antonio, praeter illum ... libellum (Brut. 163. 44), and Seneca: aliqua res extra eloquentiam (Cont. Top. 22), and Porphyrio: neminem posse aliquid agere quam quod consuerit (ad Od. 1. 1. 16).

An anonymous grammarian tells us that aliquando is compounded of alius and quando, and is in use for the past and the future; evidently he feels that it naturally marks some other time than the present. But generally we have for ‘something else’ aliquid aliud, for ‘somewhere else’ alicubi alibi, alicunde aliunde for ‘from some other quarter’, and alias aliquando for ‘at some other time’; i.e. the loss of the idea ‘other’, constant in ἔνιοι, is only usual in aliquis. This loss seems to have come from such repetitions as we have in: sin aliquis exstiterit aliquando (Cic. de Orat. 3. 80. 21), or: verum aliquando aliqua aliquo modo alicunde ab aliqui aliquast tibi spes mecum fortunam fore (Pl. Epid. 331-2), where we feel all but the first ‘other’ superfluous. But in: hic opus est aliquot ut maneas dies (Pl. Poen. 1421) the natural translation is ‘it is well for you to wait a few days longer’. And so in Nonius’s reading of: nec nobis praesente aliquis quisquam nisi servus (Pl. Amph. 400).

In return we have a number of cases where alius seems to mean the same as aliquis or ullus; e.g.: neque maius aliud neque praestabilius invenias (Sall. Jug. 1.2), non alia ante Romana pugna atrocior fuit (Liv. 1. 27. 11), neque enim aliud ... difficilius reperient (Quintil. 4. 2. 38), quamvis non alius flectere equum sciens aeque conspicitur gramine Martio, nec quisquam citus aeque Tusco denatat alveo (Od. 3. 7. 25-8) (where alius is balanced by quisquam), Eridanus, quo non alius per pinguia culta in mare purpureum violentior effluit amnis (Geo. 4. 372-3). Probably with examples like these we are to associate such uses as: Fama, malum qua non aliud velocius ullum (Aen. 4. 174), or: mulier qua mulier alia nulla est pulcrior (Pl. Merc. 101), so that for the union aliud ullum we have above aliud used in the sense of ullum, a usage we must examine under Metonymy. So in:

Sed non ante datur telluris operta subire

Auricomos quam quis decerpserit arbore fetus (Aen. 6. 140-1),

ante quam quis is poetic for alii quam qui; and the direct prose would be: illi soli dabitur qui et.

While in archaic and classical Latin alius is usual where three are in question, it is used at times to express the other of two, i.e. for alter; e.g. in: per illam tibi copiam copiam parare aliam licet (Pl. Epid. 323-4), remedium tumultus fuit alius tumultus (Tac. Hist. 2. 68.4). In uses like: ex loco in alium locum (Plin. Ep. 10. 69), aliudque cupido, mens aliud suadet (Ov. Met. 7. 19-20), alius est amor, alius cupido (Afran. com. 23), alius seems the opposite of idem, a force heightened to mutatus in:

Non vires alias conversaque numina sentis? (Aen. 5. 466.)

Here, too, when the clause is negative, alius is equivalent to alter, as is clear when we compare: malum qua non aliud velocius ullum (Aen. 4. 174) with: Misenum ... quo non praestantior alter (6.164). In: quos alios muros, quae iam ultra moenia habetis? (9. 782) we are very close to the use of alius instead of alter for ‘second’, which we reach in: alius Latio iam partus Achilles (6.89).

This explains the idiom unus . . . alius for alter . . . alter, in: (leges) duas promulgavit, unam ... aliam (B.C. 3. 21), and: unam ... epistolam acceperam ... in qua significabatur aliam te ante dedisse (Cic. Att. 7. 12.1). Alius is not used for alter in union with another numeral, as in: altero vicesimo die (Cic. Fam. 12. 25.1); but we read: quarum unam incolunt Belgae, aliam Aquitani, tertiam ... Celtae (B.G. 1. 1.1), ab alio exspectes, alteri quod feceris (Com. Inc. 82), geminae ... portae, quarum altera ... aliam (Val. Fl. 1. 833-5), duo agmina parant quorum altero populatores invaderentur, alii castra Romana adpugnarent (Ann. 4. 48.4). We have in: quaeritur huic alius (Aen. 5. 378) alius used for compar, a use that would be striking even for alter.

Alius is at times constructed with the ablative, as though it were a comparative, e.g.: alius Lysippo (Ep. 2. 1. 240), alium sapiente bonoque (Ep. 1. 16. 20), expertis alia experiri (Liv. 5. 54.6), si accusator alius Seiano foret (Phaedr. 3, Prol. 41). True, the ending -ius hardly seems the same as that we have in maius -oris, or in plus (=plouios) -pluris. Hence Sommer (302.1) thinks the comparative force inherent in the root ali-, just as minor-minus, with no proper comparative ending, gets its comparative force from the root mi- ‘to lessen’. But Brugmann believes that it is from words like alius that the ending -ios gets its comparative force. We have this ending in medius, δεξιός, and σκαιός (older σκαϝιός, Latin scaevus). No doubt δεξι- and σκαϝι- are old locatives like ali- ‘on yon side’. Tertius has this same comparative ending, and illustrates the use of alius for one of three, as well as for one of two; for if alius is originally a comparative, the latter will be its primary force.

We find, then, that alius, often a comparative in force and probably primarily in form as well, had in old Latin already passed on to a use for one of three as well as for one of two, a use it retained there to a considerable degree. Through its association with words like quis or ullus, it had lost this comparative force at times, and had with it lost all reference to a dual or plural, i.e. was absolutely singular. We may note its omission in:

Impastus ceu plena leo per ovilia turbans,

Suadet enim vesana fames, manditque trahitque

Molle pecus mutumque metu (Aen. 9. 339-41),

where the last phrase seems equivalent to: et mollis pecoris alia mandit trahitque alia.

We read in Servius: alter enim de duobus dicimus, non de tribus (ad Buc. 3. 34), and when we find it used of three and not of two, we seem to have entered on a new period in its history. When we read in Cato: (vinum) in dolium infundito, ... transfundito in alterum dolium, post dies XX in alterum dolium transfundito (R.R. 112.2), it is not quite certain that it is three objects, and not a succession of pairs with which we have to do. But in:

Hoc erat, alma parens, quod me per tela, per ignes

Eripis, ut mediis hostem in penetralibus, utque

Ascanium, patremque meum, iuxtaque Creusam

Alterum in alterius mactatos sanguine cernam? (Aen. 2. 664-7),

the use of alter for one of three is plain. But we have also:

Tum geminas vestes auroque ostroque rigentes

Extulit Aeneas . . .

Harum unam iuveni supremum maestus honorem

Induit (11. 72-3 and 76-7).

And in like fashion we find: unum exuta pedem (4.518) and: unum exserta latus (11. 649), where we have, not, it is true, alter for unus, but unus for alter. And thus we find alter, too, like inter, advancing from two to three, and apparently receding from two to one; for it is replaced by unus.

Properly, in designating the members of a pair we should have alter ... alter, as in: alter istinc, alter hinc adsistite (Pl. Rud. 808). But often, as with Castor and Pollux, only one need be expressed, as in: in altera parte fluminis (B.G. 2. 5.6); or the other is designated by a different word, as in: summus ibi capitur meddix, occiditur alter (Enn. Ann. 328); and so alter takes on the meaning of ‘second’, as in: erus ... et erus alter (Pl. Capt. 1005), becoming a preferred competitor of secundus that may imply inferiority, as we see in: haec fuit altera persona Thebis, sed tamen ita secunda ut proxima esset Epaminondae (Nep. Pelop. 4.3). We find it used for second in a series of three; e.g.: primus ... alter ... tertius (Aen. 5. 310 ff.) or: una ... alter ... extremus (5.563 ff.). Hence we get unus et alter ‘one or two’ and unus aut alter ‘one or perhaps two’, which in later Latin comes to mean the same as Cicero’s unus et alter. While alius is not used as an ordinal numeral in union with other ordinals, alter is often thus used, as in: litteras quas mihi altero vicesimo die reddidit (Cic. Fam. 12. 25.1). In: alter ab undecimo tum me iam acceperat annus (Buc. 8. 39) Servius is quite positive that the thirteenth, not the twelfth year, is meant, the twelfth being too far from puberty.

Familiar is the reciprocal force of alius repeated, as in: ceteri verbo alius alii adsentiebantur (Sall. Cat. 52.1), ut ipsi inter se alii aliis prodesse possent (Cic. Off. 1. 22.7). When alter is thus repeated, and only two are in question, it sometimes has this force, as in: ut alter alterius iudicium non modo reprehendat, sed etiam rescindat (Cic. Cluent. 122. 43), but sometimes not, as in: consules primum religiones, deinde alterum alterius mors et comitia ... impediunt (Liv. 41. 16.7). But for more than two it is commonly reciprocal, as in: ut nemo memoria dignus alter ab altero videri nequiverint (Vell. 1. 16.5), cum alter alterum indignaretur imperare (of four) (Capitol. Alb. 1.2), omnes rediere ... inconsideratae dementiae alter alterum arguentes (Amm. 31. 15. 15).

As early as Lucretius we find the union alius . . . alter in: ex alio terram status excipit alter (5.835), hic odor ipse igitur, naris quicumque lacessit, est alio ut possit permitti longius alter (4.687-8). But this confusion was not felt in some phrases. We feel how different from altero die or alio die used for it in: servolos rogitabam ... item alio die quaerebam (Ter. And. 89) ‘on the next day’, is alio die in: mox quasi alio die studebat (Plin. Ep. 3. 5. 11) ‘as on an ordinary day’, or: confecto negotio bonus augur ... ‘alio die’ inquit (Cic. Phil. 2. 83. 33) ‘not to-day’. By altero die we mean ‘on the morrow’, ‘on the day after’, as in: altero die pervenit Caesar (B.C. 3. 30.6), altero die quam a Brundusio soluit (Liv. 31. 14.2). But in the Itin. Anton. (Plac. re. R. 30) alia die de natale domini is ‘the day after his master’s birthday’, and alia die (Pallad. 9. 8.6) ‘on the following day’.

We have alter for ‘the neighbour’ in: qui nihil alterius causa facit et metitur suis commodis omnia (Cic. Leg. 1. 41. 14), cave ne portus occupet alter (Hor. Ep. 1. 6. 32). Just as we have alterum tantum for ‘as much again’, we have alter ego for ‘a second self’ in: te me esse alterum (Cic. Fam. 7. 5.1), and so alter idem in: est enim is quidem tamquam alter idem (Lael. 80. 21). But it is obvious that this may be carried too far; and just as from ad and salto we have adsulto, from ad and alter we have adulter, the ‘too neighbourly’ man. We read in Festus: et adulter et adultera dicuntur, quia et ille ad alteram et haec ad alterum se conferunt (Paul. p. 22). Probably the alter in: fruitur nunc alter amore (Tib. 1. 5. 17), quam vacet alterius blandos audire susurros (Prop. 1. 11. 13) has much the same significance; for in late Latin alterare is used for ‘to spoil’, much in the same sense as adulterare.

We noticed how alius is used as the opposite of idem, and in Virgil advances to the sense of par or compar. But in Horace we have alter used as the opposite of idem in: quotiens te speculo videris alterum (Od. 4. 10.6). In late Latin instead of alius ... alius we have alter ... alter as in: altera substantia divinitatis, altera humanitatis (Vinc. Ler. 13. 19), et aena et lignea et fictilia simulacra, et alterius alteriusque materiae (Prosper, in Ps. 113.4). So for more than two alter becomes usual, as in: alter adulescens decessit, alter senex, aliquis praeter hos infans (Sen. Ep. 66. 42),

Et nunc ex illo forsan grege gentibus alter

Iura dat Eois, alter compescit Hiberos,

Alter Achaemenium secludit Zeugmate Persen (Stat. Silv. 5. 3. 185-8),

cum alter maneret in Capitolio, alter in Palatio, alter . . . alter ... alter (Lampr. Hel. 304). Hence we have the mixture in: altera detur si similis tellus, aliaeque ... exsurgant rupes (Sil. 12. 72-4), altera nox aliisque gravat plaga caeca tenebris (Stat. Theb. 8. 16).

When we see the clearer and more definite force of alter in classical Latin, we need not be surprised that it proved the victor. And yet it was not always felt clear or forcible enough to give the sense ‘one of the two’, and we have often alteruter used for this. But in such a case, when it is repeated, it is expressed by alter only, as in: ne ... alteruter alterum praeoccuparet (Nep. Dio 4.1), aut etiam alterutrum, nisi terminet alterum (Lucr. 1. 1012).

Sommer feels that -ter in alter is the same as -ter in aliter. Alter is for the old aliteros where, because of the two morae following ali-, the i has fallen out, while in aliter, the old neuter singular, there followed but one mora, the short syllable -ter; and he compares validus with valdē, where the i is lost before the final -dē, the equivalent of two morae. Aliter is the same kind of adverb as we have in dulce ridentem ‘laughing a sweet laugh’, a cognate accusative neuter of the adjective. Osthoff accounted for the -ter in breviter as being contracted from itere, breviter being primarily brevi itere (Woelff. Arch. 4. 455); but Delbrück pointed out that the ending -iter in obiter and pariter must be connected with the -ter in inter, subter, and propter. Probably propter was originally propiter, and perhaps behind inter lies an older initer (cf. ἐνί); but the proclitic use of these words as prepositions would naturally lead to shortening as well as loss of accent, giving us propter and inter. Cette for *cedite Sommer thinks was shortened very much as was our ‘good-bye’; he calls it an allegro form. Aliter looks like a starting-point for such adverbs as celeriter. Very clear becomes the sense of atque in union with it in: omnia plena pacis aliter, ac mihi Calvena dixerat (aliter) (Cic. Att. 14. 9.3), and we see at once the force of aliter et in: si aliter est, et oportet (aliter esse) (Att. 11. 23.1). Mela’s aliter a ceteris agunt (1.57) shows the way to Cledonius’s velocius equus ab equo ‘one horse runs faster than another’.

Iterum, unlike aliter, has assumed the ending of the acc. sing. masc.; perhaps we may connect it thus: semel et iterum pervenire ‘to reach one meal and then a second’. Its root is the same that we have in ibi (=thereby) and ita; and iterum rogo is: ‘I ask that second thing.’ The other form itero, found in inscriptions, would stand naturally for iterum in: ac primo quidem decipi incommodum est, iterum stultum, tertio turpe (Cic. de Inv. 1. 71. 39).

Ceteri we are constantly using in et cetera, not seeing, as the Romans, too, failed to see, that ceteri was a contraction for caeieteri, the Roman equivalent of καὶ ἕτεροι ‘and the others’. For ἕτεροι the Doric ἅτεροι seems the older form, being for sm̥teroi ‘the one party of the two’ (cf. ἕν for sem). In common use it is ‘the remaining majority’ as opposed to reliqui ‘the remaining few’; but, as we shall see, there are many uses of it which show this to be a later and acquired meaning. We have cetera for alia in: ceu cetera nusquam bella forent (Aen. 2. 438-9) and alia for cetera in:

Inde alias animas quae per iuga longa sedebant

Deturbat, laxatque foros (6. 411-12),

and in:

Obstipuere animis alii, sed Troius heros

Agnovit sonitum (8. 530).

The spelling in inscriptions is at times caeteri. The form ceterus is not in common use, and the meaning makes this natural. In caetera multitudo and caeterum triticum the use of the nominative is easy, but caeterus vir would be impossible. Still we have masculine collectives; and Gesner1 cites: ceterus ornatus domi Pompeiis emptus est (Cato R.R. 22.3), which I do not find in Keil’s edition. Usually alius is used for it, as in: et alius exercitus ratibus iunctis traiectus (Liv. 21. 27.6). Ceterum the conjunction, and cetera are got from the same construction as we have in: cetera Graius (Aen. 3. 594).

1 Gesner did not receive this into his own text of the Auctores de Re Rustica, but notices it there as a reading got from the notes of Politian, and adopted in the editions of Jenson and Jo. Gymnicus.