STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. TEXTUAL NOTES ON PHAEDRUS BY J. P. POSTGATE As this paper will deal in large measure with the evidence for the text of Phaedrus which is supplied by his mediaeval paraphrasts, it will be as well to indicate at the outset the limits within which that evidence can be used. Of its utility if employed with discernment there can now be no question. M. Havet's restorations of the read- ing at App. 26. 1 lupus for lepus and of the order of verses in iii. 7. 11 ff., 21-24 to follow 10, are instances in point. Without traversing anew all the ground over which my prede- cessors have traveled I would refer to some cardinal points. In the first place a fable may be both in Phaedrus and a paraphrast and yet the paraphrast have not taken it from the fabulist. A good example is App. 13, the "Widow and the Soldier," with which the narrative in Romulus iii. 9 (Hervieux, les fabulistes Latins, II, 217) has only its subject in common. Secondly, in fables where Phaedrus was the original source, besides the involuntary alterations which must occur when a mediaeval Latinist is turning the verse of a classical writer into prose, there may be deviations and accretions due to the para- phrast himself. I imagine for example that no one will attribute to Phaedrus the insipid pendant to the catalogue of birds' characteristics that we find in the Wissemburg collection (Hervieux, op. cit., p. 188) and Romulus iv. 4, though in other lines these show a great deal of verbal resemblance to the Phaedrus fable iii. 18. Phaedrus had written, 10 ff.: "Fatorum arbitrio partes sunt uobis datae; / Tibi forma, uires aquilae, luscinio melos, / augurium coruo, laeua cornici omina." What we find in Romulus is "Tibi nitor et color et forma, maior uirtus aquilae, lusciniae cantus uocis, auguria sumit coruus," then "grunnire accepit columba, dolet ritus [manifestly corrupt], gruis ostendit semper tempus et in oliva parit turdus; fleet ula [for "ficedula"] probanda pomis, lucifero gaudet hirundo, nudus sero uolat uespertilio, gallus nouit noctis horas," in which the mediaeval stamp is unmistakable. With these cautions in mind we shall be able to use with profit the evidence of versions often so close to their [Classical Philology XIII, July, 1918] 262 Textual Notes on Phaedrus 263 original that we can without appreciable effort transform their prose into the actual iambics of Phaedrus. 1 One of the chief uses of these paraphrases is that they help us now and again to track glosses or adscripts which have ousted the genuine reading from the manuscripts and from our text. A simple instance is i. 2. 27 ff. : Furtim igitur dant Mercurio mandata ad Jovem, adflictis ut succurrat. Tunc contra deus; "quia noluistis vestrum ferre, 'inquit,' bonum, malum perferte." deus would have eluded observation but for the indications of the paraphrasts. Ademar (Hervieux, p. 139), who is closest of all to Phaedrus here (for ver. 26 he has "fugitant inertes uocesque prae- cludit metus," only putting uocesque for the fabulist's uocem), gives "furtim mittunt per Mercurium mandata Ioui ut desistat. E contra Iupiter: quia noluistis ferre bonum ut feratis malum hunc sustinete malum," etc. Wissemburg (ibid., p. 176) has "tunc rogare coeperunt : succurre, Iupiter, morimur: et contra altitonans: quia noluistis bonum sufferre, sustinete malum," and Romulus who has amplified and embroidered considerably gives (ii. 1) "tunc uoces cum lacrimis ad sydera tollunt: succurre, Iuppiter, morimur. E contra illis altitonans," etc. The three candidates for the place of honor in Phaedrus' sentence, deus, Iuppiter, altitonans, disappear directly we see what is behind, the equivalent and metrical Tonans. A parallel intrusion into the traditional text may be detected in i. 21. 3 ff. : defectus annis et desertus uiribus leo cum iaceret spiritum extremum trahens, aper fulmineis uenit ad eum dentibus et uindicauit ictu ueterem iniuriam. Our suspicions are aroused by the circumstance that P (the Pitho- eanus) has ad eum uenit, but D (the scheda Petri Danielis) uenit ad eum. Of the paraphrasts, Ademar (p. 136) has "super fulmineis uenit ad eum aper et dentibus se uindicauit, ictu ueteres iniurias infestans," etc., Wissemburg (p. 169), "aper ad eum uenit iracundus 1 This is especially true of the Fabulae Antiguae of Ademar, Hervieux, op. tit., pp. 131 ff. 264 J. P. POSTGATE spumans fulmineis dentibus et uindicauit ictum ueterem," Romulus (p. 201), "aper ad eum uenit iratus spumans fulmineis dentibus et uindicauit ictum ueterem." Can we doubt what this means, viz., that in the ultimate source of all our authorities the line appeared in this form: ad eum aper fulmineis spumans uenit dentibus, and that the vigorous spumans (of which Ademar's super is a cor- ruption) should replace the useless ad eum of the Vulgate ? Compare the close parallel in Martial, a reader of Phaedrus as we know, "fulmineo spumantis apri sum dente peremptus" (xi. 69. 9). Spu- mans or spumeus is in fact almost a standing epithet of aper (Vergil, Juvenal, Martial, Claudian). Cf. also Sen. Dial. iii. 1. 6, "spumant apris ora, dentes acuuntur attritu." In i. 27. 4, a passage already corrected in part by M. Havet, a similar confrontation of our authorities will lead us to the truth: uulpes ad cenam dicitur ciconiam prior inuitasse et illi liquidam in patena posuisse sorbitionem. The variations of the paraphrasts are as follows: Ademar (p. 154), "posuit illi in scutella sorbitiunculam," Wissemburg (p. 166), "coctas sorbitiunculas in marmore [corrected to -a] fudit," Romulus (p. 209), "posuit illi in catino sorbitionem liquidam." The Greek version as we find it in Plutarch Qu. Conu. i. 614 E F is nearest to the Wissem- burg paraphrase, ol ok roiavra 7rpo/3Xij/iara nadUvres ovhtv av rrjs Alaveiev &v i] iitv ervos n \nrapov Kara Xi6ou irXaT€ias Karaxeafiivri ri]v yipavov eSet-aro o\jk evuxovfiivn , dXXd yehna Traaxovaav. Not only has marmore this, presumably independent, confirmation, but it is supported by the discrepancy in the words used of the vessel by the other authorities, patina, scutella, catinus, which shows that this was not named but described. Accordingly M. Havet, who has also some very just remarks on the unmetrical character of the Vul- gate reading in patina liquidam, has rightly restored in marmore to the text. But he has not done so well in building upon illi, for which he substitutes leui in the sense of ir^arelas, a meaning it does not possess. Textual Notes on Phaedrus 265 Illi has no more claim to respect than ad eum in the passage last discussed. Ponere without a dative is common enough of "serving up," "putting on table"; Pers. i. 53, Mart. iv. 69. 2 may serve as examples. And the word we want to represent irXarelas is patulo: cf.Verg. Aen. vii. 115, "patulis . . . . quadris," Juv. 3. 277, "patulas .... pelues"; cf. Varro L. L. v. 120, "patenas a patulo dixerunt." Phaedrus uses patulus for capacious, App. 23. 2. i. 5.6: sic est locutus partibus factis leo: ego primam tollo, nominor quoniam leo; secundam, quia sum fortis, tribuetis mihi; turn, quia plus ualeo, me sequetur tertia; malo adficietur, siquis quartam tetigerit. That rex should stand somewhere in the second line seems as certain as that fortis in the third should be changed to socius, Withof, or consors, Gow. After M. Havet's discussion in his Critical Dis- quisition (ed. pp. 187 f.), it is superfluous for me to argue this point. It would seem moreover that it could only have stood by the side of quia in the last foot but one. This leaves us nominor and leo to deal with — suspicious both of them. The first I had already con- demned and conjectured that it concealed nomine before I read the note of Bentley in which nomine hoc "on this account" is proposed. But leo which Bentley kept (quia sum leo) is not less objectionable and even more suspicious, as just above it stands another leo. And yet it hardly hides the truth. Cluo, another form of clueo, "am called," is the lightest of changes and will at once account for the appearance of nominor in the place of nomine hoc, 'cluet, nominatur' (Nonius 87. 28). The word was archaic in the time of Phaedrus and was used here as in Seneca's jeu d 'esprit on the emperor Claudius 7 "exprome propere sede qua genitus cluas," to ridicule the pretensions of a monarch. I may add that it escapes the objection which Bentley brings against nominor "non eo ceteris feris excellit quod nominetur sed quod sit leo." For cluere often hardly differs from esse, as, e.g., in Lucretius. For the sake of euphony we must arrange the words thus, ego primam tollo, nomine hoc rex quia cluo. Before leaving the passage I may note that we can see from the paraphrasts that an early (corrupt) variant for ualeo was uelox. 266 J. P. POSTGATE i. 6.1: Vicinis furis celebres uidit nuptias Aesopus et continuo narrare incipit: uxorem quondam Sol cum uellet ducere, clamorem ranae sustulere ad sidera. The indications of the paraphrasts point to the uicinis which M. Havet has restored for uicini; for / might easily fall out before /. But there are signs also, as indeed he suggests, of a lacuna, the paraphrasts agreeing in adding interueniens (Wissemburg) or its equivalent cum interuenisset (Romulus), (interuenit dicens, Ademar), in addition to which narrare seems to desire an object. Perhaps we should add after 2, conuiuis interueniens talem fabulam. For talem fabulam compare talem fabellam (i. 2. 9). The beginning of Aesop's speech to the wedding guests also appears to have been lost, for Ademar (p. 134) has "Audite gaudia uestra" and Romulus (p. 198) " 'audite' inquit 'gaudia uestra,' " which seems to be a curt abbreviation of "uos uestra quae sint audiatis gaudia." i. 31. 3 ff.: columbae saepe cum fugissent miluum et celeritate pennae uitassent necem. As M. Havet follows Bentley in changing fugissent to the compound ecfugissent, and as both Heinsius and Bentley altered uitassent to euitassent, it does not seem superfluous to observe that neither change is necessary. Vitassent can express the sense required, as Horace's "uitaui denique culpam / non laudem merui," A. P. 267, is enough to show; and the line thus gives the meaning of effugissent. Fugissent marks a prior stage "fled from," as in Ovid Metamorphoses i. 506, "sic aquilam penna fugiunt trepidante columbae / hostes quaeque suos." At the end of this poem the indications of the paraphrasts show a line has been lost as M. Havet rightly argues. His suggestion that after tunc de reliqtlis una "merito plectimur" we should add huic spiritum praedoni quae commisimus is based on Wissemburg (p. 176), "sed digne et bene patimur qui nos predoni commisimus" (cf. Romulus p. 205, "sed digne hec patimur Textual Notes on Phaedrus 267 que nos tali commisimus"), but he has taken spiritum from Ademar (p. 139), "sic merito agitur qui nostrum spiritum tali credidimus inimico." But in this fable Ademar follows Phaedrus much more closely than Wissemburg and Romulus, who in fact introduce a third actor in the Hawk whom the columbae choose as king. I agree therefore with Riese in basing the supplement on his paraphrase and proposing nostrum hosti tali quae credidimus spiritum. For hostis in a similar connection besides the passage of Ovid quoted above see Phaedrus i. 28. 10; iv. 2. 18, "hostis callidi." The occurrence of credentes in 10 is no objection to credidimus here but rather an argument in its favor. hi. 8. 9 f . : ergo ad patrem decurrit laesura inuicem magnaque inuidia criminatur filium. filium has naturally been questioned, as it offends against the rule of Latin usage by which these words of relationship are referred to the immediate subject. For contraventions of the rule there appears to be always some special reason, as in Lucan ix. 134 f . Sextus Pompey is allowed to say "nee credens Pharium tantum potuisse tyrannum litore Niliaco socerum iam stare putaui," because socer and gener of Caesar and Pompey had practically passed into proper names. But none can be discerned here. Why M. Chauvin's futilem has been admitted into M. Havet's text I do not know. Beyond a superficial resemblance to filium it has nothing to recommend it; bellulum, M. Havet's own suggestion, is somewhat better. But we want a substantive, not an adjective, and so trossulum, another conjecture of his, is on the right track, though in itself quite improbable. I propose pupulum, a diminutive of pupum, which I take from Catullus 56. 5 where it is conjoined with puella. Lipography is a recognized cause of corruption in the text of Phaedrus, e.g., i. 1. 7, querevis; iii. 7. 3, occucurrit. If pupulum became pulum, any scribe could make it into filium, which he had a few lines above. iv. 18 (19). 24 f.: canes confusi subito quod fuerat fragor repente odorem mixto cum merdis cacant. The editors and critics who have altered mixto appear to have attacked the wrong word. We should simply remove the m from 268 J. P. POSTGATE odorem and take cacant in its absolute use. According to the The- saurus this is the only passage where the verb has an accusative in this sense. I take this opportunity of saying a word in defense of my con- jecture in v. 17 of this fable accepted by Dr. Gow for the Corpus Poetarum text. The Commonwealth of Dogs, learning of the misbehavior of their ambassadors and timentes, rursus aliquid ne simile accidat, odore canibus anum sed multo replent. The aliquid simile should be mentioned in the previous line, but that contains nothing about it in the form we have it in the MSS : rumor legatos superiores prodidit. There too there has been lipography, meatus; also confusion of c(C) and g(G). Cf. prol. 8 fugotae the MSS, fueaiae Gronovius. The resulting gatus was readily tinkered up into legatos, the word which the context suggested. The verbal cacatus is not found else- where, but for the plural, cf. "ad meos haustus" (i. 1. 8). iv Epilogus (v. 5) : Adhuc supersunt multa quae possim loqui et copiosa abundat rerum uarietas; sed temperatae suaues sunt argutiae, immodicae offendunt. quare, uir sanctissime, Particulo, chartis nomen uicturum meis, Latinis dum manebit pretium litteris, si non ingenium, certe breuitatem adproba, quae commendari tanto debet iustius quanto poetae sunt molesti ualidius. The text of this little piece is with the exception of a single word at last correct. But that one corruption stultifies it completely. Phaedrus, commending brevity and promising Particulo immortality from compositions which possess the quality he commends, winds up by saying that this brevity is laudable in the same proportion as poets are a nuisance. For the intolerable poetae Nauck proposed loquaces; but this leaves the tradition unexplained. It is not difficult to find the word which has been ousted from the text by its stupid explanation. To go on saying the same thing, to " harp " on the same thing is cantare, e.g., Plaut. Trin. 289, "haec dies noctesque tibi canto Textual Notes on Phaedrus 269 ut caueas," the vain repetition is cantilena, Cic. De or, i. 105, "qui non Graeci alicuius cotidianam loquacitatem sine usu neque ex scholis cantilenam requirunt," and the tiresome iterator is cantor, as in Plautus Pseud. 366 "cantores probos!" (qui eadem semper clamant, the Thesaurus). Appendix 14. 6 f. quos ultra paulo uilla spkndidi diuitis erat acceptura uirginem e matris sinu. M. Havet makes a fair point against ditis splendida, Bothe's con- jecture (accepted in the Corpus), when he says that a house cannot be "per se splendida nisi propter marmor uel citreas in foribus fascias," but his own sponsi diuitis, which he thinks produced splendidi diuitis through doubling of the di, has little likelihood. It has not, I think, been observed that it is quite needless to say to whom the uilla belonged; anyone would assume that it was that of the rich suitor. I propose splendens ditiis. Ditiae, a shorter equiva- lent of diuitiae to which it is usually corrupted, is found in Plautus and Terence {Haul. 194). Splendens ditiis is like Catullus' "tota domus gaudet regali splendida gaza," (64. 46) or Vergil's "at domus interior regali splendida luxu instruitur" (Aen. i. 637). I imagine that splendens became splendida and was then assimilated to diuit- (i)is; but there may have been lipography as M. Havet thinks. Ovid (Fasti iv. 136) uses diuitiae in a similar way, of the costly orna- ments on the statue of a goddess. App. 17. Premente partu scrofa cum gemeret iaeens, accurrit lupus et obstetricis partibus se posse fungi dixit, promittens opem. suspectum officium repudiauit malefici et "satis est" inquit "si recedis longius." After this line, for which Wissemburg (p. 180) has "expono, frater, fetum, inquit, secura si tu recesseris" and Romulus (p. 206) exactly the same words though in a slightly different order, Wissemburg proceeds "obsecro ut des mihi honorem: et tu olim matrem habuisti" (these five words in the margin) "ille autem ut recessit statim profudit sarcinam" and Romulus "obsecro da mihi honorem. fuit etiam et tibi mater, ille autem ut recessit, frustra petendo fraudulenter fatigatus ipsa statim profudit sarcinam." 270 J. P. POSTGATE These additions indicate that in their common source stood two more lines which may be thus restored : honorem mihi des: et tibi mater fuit. ille ut recessit, sarcinam effudit statim. For sarcinam effundere, which is Phaedrus' phrase, see iii. 15. 6. Whether this less abrupt dismissal of the Wolf is from the hand of Phaedrus or an early imitator it is not altogether easy to say. Honorem dare is not a very satisfactory phrase here, but the second line seems unexceptionable . App. 24. Odiosa cornix super ouem consederat: quam dorso cum tulisset inuita et diu e. q. s. It is true that adjectives and adverbs are sometimes coupled in Latin; but I think few will dispute that the et here adds nothing but inconcinnity to the normal expression "quam cum inuita diu dorso tulisset." Now inuitus is one of the "epithets" most commonly transferred in Latin poetry when a part of the body is in question. So inuita manu (Ovid Pont. i. 2. 126) ; inuita dextra (Lucan i. 378) ; inuitis pedibus (Prop. ii. 25. 20); inuito collo (Ovid Met. ix. 606). And here Phaedrus should have written inuito. When I had hit upon this correction I learned from M. Havet in § 60 of his critical disquisition, p. 192, that this intruding et involves an elision to which there is no other parallel in Phaedrus — an independent and welcome confirmation. App. 26: cum uenatorem celeri pede fugeret lupus. M. Havet seems to be right in contending that celeri pede must refer to the man and not to the wolf and that the cum persecutorem fugeret of Ademar (p. 149) and the cum persecutionem fugeret of Romulus (p. 224) cast doubt on the genuineness of uenatorem. His correction "cum celeri urgentem pede uirum fugeret lupus" provides celeri pede with the necessary participle, but is not otherwise felicitous. No noun is required, as it is a common practice of silver Latin to use participles in the Greek manner for relative clauses. So Phaedrus himself ii. 7. 10 spoliatus, "the one who had been stripped"; v. 1. 7, "resides et sequentes otium" = qui otium sequebantur. The para- phrases point to persequentem. Feras persequi is found in Ovid. But Textual Notes on Phaedkus 271 as a piece of Latin writing I should prefer se sequentem. The se, to which there is no objection on the ground of euphony, might easily fall out before sequentem and persequentem would seem an obvious amendment. In 1. 9 of the same piece uenit sed abiit hac ad laeuam; et dexteram demonstrat nutu partem. I conjectured and Dr. Gow accepted the conjecture for the second edition of the Corpus that for nutu we should read nictu. This is confirmed not only by the oculis of 1. 10, but by the "adsignat oculis" of both Ademar and Romulus here. App. 27. At the end of this ironical composition in which the mutual insincerities of a pair of lovers are transfixed the paraphrasts have an addition which seems to have a basis in the lost original. Romulus (p. 218) has "sic uerbis se deluserunt" and Wissemburg after a strange insertion "de bonis meis ipse diris" 1 goes on "verbis sermonem [apparently for sermonum] inuicem se luserunt," which apparently came from ita illi uerbis deluserunt se inuicem ADDENDUM When the above paper was written in the early summer of 1917, I was unable to refer to the works of G. Thiele on the Mediaeval Fabulists, Der illustrierte lateinischer Aesop in der Handschrift des Ademar and Der lateinische Aesop des Romulus. This will explain why the paraphrasts are cited only by the edition of Hervieux. The correction of nictu in App. 26. 9 has been made independently by L. Rank in Mnemosyne (1912). The University of Liverpool 1 Does this represent some corruption of "hanc vocem audio" for which there is no equivalent in the paraphrase ? It is by no means the only senseless thing in the rendering which concludes with a sentence the construction and source of which are unfathomable to me. In their reproduction of " libenter inquit" as "benigne respondit," both Wissem- burg and Romulus are faithful to their blundering original.