MA S TER NEGA TIVE NO. 92-80620-15 MICROFILMED 1992 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK as part of the ^ "Foundations of Western Civilization Preservation Project Funded by the NATIONAL ENDOWMENT FOR THE HUMANITIES Reproductions may not be made without permission from Columbia University Library COPYRIGHT STATEMENT The copyright law of the United States - Title 17, United States Code - concerns the making of photocopies or other reproductions of copyrighted material . . . Columbia. University Library reserves the right to refuse to accept a copy order if, in its judgement, fulfillment of the order would involve violation of the copyright law. A UTHOR : ANDERSON, ANDREW TITLE: EI-READINGS IN THE MSS OF PLAUTUS PLACE: [CLEVELAND] DA TE : [1 907] <* COLUMBIA UNIVERSITY LIBRARIES PRESERVATION DEPARTMENT Master Negative # ".« BIBLIOGRAPHIC MICROFORM TARGET Restrictions on Use: ' Original Material as Filmed - Existing Bibliographic Record A( :quisitions NYCG-PT FRN: m : EL: AU: 05 -27-92 SNR: AlC: 00:05-27-92 F 1 C : '? CON:??? FSl:? 1LC:???V 11 :? COL: EML: GEN: BSE: BKS/PROD Books FOL/BIB NYCG92~B35912 Record 1 of - Record added today + I0:NYCG92-835912 RTYP:a ST:p CC:9668 BL!:afTi DCF:? C3C:? MOD: CP:ohu L:8rig IHJ :? GPC:? BIO:? PC:s PD:1907/ REP:? CPI:? MMD: OR: POL: DM: RR: 040 NNCj-cNNC luO 1 Anderson, Andrew R. 245 10 Ei -Readings in the hss of PlautuSrh[micr of or m ) . i cBv Prof. Andrew R. An drew. 260 [Cleveland, }:ci907 J. 300 [73 I Hr, p. LDG OR[U gu 05-27-92 TECHNICAL MICROFORM DATA IMAGE PLACEMENT: lA @ JB IIB DATE FILMED:_f_7^-'^3__,_^_. INITIALS___ _l/fj^€t— HLMEDBY: RESEARCH PUBLIcXtIONS. INC WOODBRIDGE. CT r Association for information and image iManagement 1100 Wayne Avenue, Suite 1100. Silver Spring. Maryland 20910 301/587-8202 Centimeter 12 3 4 5 1 1 1 Inches I I 6 lliml TTT 7 8 9 10 11 12 13 llll[llll[llllpllll|llll|l|J||||llllll |||||||||||||||N I I I Mill 14 15 mm iiiliiiiliiiil TTT 1.0 I.I IS.6 mil 3.2 163 3.6 1.25 1.4 2.5 2.2 2.0 1.8 1.6 MflNUFRCTURED TO fillM STRNDflRDS BY nPPLIED IMAGE. INC. Extracted from the Transactions of the American Philological Association, Vol. xxxvir, 1907. V V. — Ei'Readings in the Mss of Plautus, By Prof. ANDREW R. ANDERSON, Princeton University. By the term "^/-readings " are meant those that give ei for the classical f, as deico for dico, veivo for vivo ; or for the clas- sical I, as airabeis for ciirabis. There are a few instances also where the ei is given for sounds other than those mentioned, but they are less typical, and hardly need to be illustrated by examples at present. These readings have been treated differently by different "^itors. Ussing consistently rejects the ei throughout and gives the classical orthography. Almost as consistently Ritschl in the triumvirate edition and Goetz-Schoell in the editio minor read the ei in practically every place where there is any Ms authority for it. Leo reads the ei in only a very few isolated instances, but does not read it in scores of other instances where there seems to be equally good reason for doing so. I had expected that Lindsay in his Oxford edition would exercise greater discrimination and would defi- nitely settle the matter, as his note prefixed to the Argumentum in his larger edition of the Captivi shows that he understood perfectly the principles involved. (I am pleased to acknow- ledge that it was chiefly this illuminating note which prompted my own investigation.) Unfortunately he did not uniformly apply the principles that he himself had previously stated with such clearness. E.g. in the first eight plays he has often not admitted a genuine {i.e. diphthongal) ^/-reading into his text, as Ci. 62^, where P reads ei for the imperative of ire, whereas in Ep. 600 on the testimony of A he has accepted preimnni. In the last twelve plays he generally follows the ei where the Mss give it for an original diphthong, but not elsewhere. Yet in Mcr. 282 ei. et P has been rejected, so also in Mer. 294 deiceres A. Nevertheless in Mer. 471 veivo A has been admitted. Evidently, adhuc sub iiidice lis est. 74 Andrew R. Anderson, [1906 -. Plautus occupies a peculiarly felicitous position for an investigation of this sort. He belongs to a period when the rules governing the use of ei and i are well-defined and cer- tain — a state of affairs that by no means existed for Terence or Lucilius. Furthermore there is in his Mss abundant material to serve as a basis for reasonably sure generali- zation, there being in all 365 instances of the ei in his Mss, 242 in th^ ^mbrosianus (A), and 123 in the Palatini (BCDEJVT> The purpose of this investigation will be to test these Ms readings, to determine which of them are false and un-Plautine, and which are genuine and represent Plautine orthography. I shall try also to determine whether these genuine ^/-readings go back in direct line of succession to Plautus himself, or whether they were reintroduced after his text had been more or less thoroughly modernized, and to treat other questions germane to the subject. The fact that the total number of ^/-readings in our Pala- tine Mss for all the plays is less than the total number in A for only a portion of the plays at first occasions surprise. But the comparison is not a just one. In order to be just the comparison ought to give the number of ^/-readings in that ancestor of all our P Mss that was contemporaneous with A. For between that time and the time of our P Mss it seems to have been a general practice with copyists to change ei to /, and in many places this was done even where the e and the i belonged to different syllables and corruptions resulted of which illustrations are given below : — Am. 13 369 Cap, 940 Ep. 626 Men, Arg. 2 234 735 Mer, 840 Mo. 650 / B* « B^ (dat.) si BD sed Gruter uti codd. ut ei r ulmis BJ ulmeis B^ Ei e corr. post ras. B^ (dat.) ire hi B^CD^ ei rei Gruter inarrabo B* ei narrabo B^ Ubi qui d- B, Ubiquid C, Ubi qui (ex que) d- D, Ubique id post Scioppium Gruter. quasi CD quas ei (ex quasi) B t Vol. xxxvii.] Ei-Readi7igs in the Mss of Plautus, 75 Pe, 318 846 Poe, 714 Ps, 242 1107 Ru, 562 763 Tru, 559 fames ift eire B {inire) CD^ fame sine ire D^ {sineire) T fame ; sine ire leg. eolaphum icit BCD eolapho me icit Acidalius philippi BCDphilippei Pylades placidis BCD placide is Camerarius habente in omen B, habenti nomen CD habenty ei nomen leg. iectas-QQiy EIECTAS A pugnis BCD pug7ieis A Camerarius perditum sit BCD, se it Camerarius The fact that these places where the change of ei to i has resulted in corruptions are so numerous shows how general the practice of changing ei to i must have been, and m all probability if we had the ancestor of our P Mss con- temporaneous with A, it would give a number of ^/-readings at least equal to that given by A. In deaHng with the main question there are three sources that help us to determine the usage for Plautus' time : — 1. Plautus himself, 2. Inscriptions of Plautus' time or earlier, 3. Comparative grammar, which as far as our purpose i^ con- cerned is both based on and supplementary to the two sources already mentioned. I. Plautus himself has fortunately left us two passages which leave little room for doubt that he distinguished between the diphthong ei and the monophthong l\ 1305 Immo edepol una littera plus sum quam medicus. Gr. Turn tu Mendicus es ? La. Tetigisti acu. The one letter by which mendicus exceeds medicus is n. So that for the i in mendicus, which is long, Plautus must have used not ei but i. Manifestly it would be hypercritical to draw any distinctions between the i in mendicus and any other long i in Plautus. Ru, La. y6 Andrew R. Anderson, [1906 .L. Trm, 262. AsT. Comprirae sis eirara. Tru. Earn {or eram) quidem hercle tu, quae solita 's, comprime, TnpudSns, quae per ridiculum rustico suades stuprum. AsT. * Eiram ' dixi ; ut excepisti, dempsisti unam litteram. This is not indeed the reading of the Mss, which in this passage are badly corrupt, but eiram, the restoration of Gep- pert, according to which I have read the passage above, has won the acceptance of scholars. It makes no difference for my purpose whether we read earn ( = erajn, mistress, as though Truculentus had understood Astaphium to say eraifi^ not eiram\ or whether we actually read eram. According to this interpretation, eira, wrath, would have one letter more than eray mistress, and would at this time have been spelled with ei-. Derivations of eira which do not recognize an original diphthong for the first syllable do not disprove this orthography, but are rather themselves proved wrong by this passage. Cf. Walde, Lateinisches etymologisches II or- terbuch, s.v., ** Ira . . . wohl zu ai. isanyati * treibt an,' isnati ds., isyati *setzt in Bewegung, erregt* mit lat. *eisa, eira eigentlich * Erregung, Erregtheit ' beriihrt sich am nachs- ten av. aesma- *Zorn,' gr. olarpo^ * Wut,' olfxa * stiirmischer Angriff, Andrang ' (wenn aus *ol(Tfia). . . ." If these passages from Plautus are put together, it may be safely concluded that the orthography of our poet differen- tiated between the monophthong i and ei, which originally had been and probably still was a diphthong. Cf . Marx, hidex Scholaruniy Greifswald, 1891, xviii. 2. Latin inscriptions up to and including the time of Plau- tus confirm the distinction just given. The most important inscription differentiating the two is the so-called SC de Bae- chanalibus {CIL. I, 196)01 189 B.C., five years before Plautus* death. It is only after his time that we find ei used for the monophthong l as well as for the originally diphthongal ei. 3. Comparative grammar, on the basis of the two sources already presented, together with evidence gleaned from other sources, has determined that the ei of the third century B.C. must have one of the following pedigrees : — I V Vol. xxxvii.] EuReadings in the Mss of Plautus, yy a, Indo-Germanic ei, as in deicere, b, Indo-Germanic ai or ^/ which, standing in unaccented syllables had been weakened to ^/as in the relative quei < quoi 2Jid. the dative singular of the third declension, as virtutei< *viriuiai, €. Indo-Germanic eu>ou> oi>ei, as in leiber, free. - d Analogy, but it must go back ultimately to one of the afore- mentioned pedigrees, as nobeis, redieit. Accordingly, in discriminating between the various ^/-read- ings of the Plautine Mss, we are to observe the following rule : if any of the readings can be proved to come directly or analogically from any of the previously me7itio7ied sources acknowledged by comparative grammar, it may be defended as representing Plautine orthography, otherwise it cannot. Class I. This class contains those that may be defended as representing Plautine orthography, classified according to origin. a. From Indo-Germanic ei I. In the root of the verb deicere (dicere) A {i.e. from the Ambrosianus) Men. 243 591 Mer. 268 281 294 300 465 467 484 bis 512 bis 516 529 554 760 763 Poe. 474 1 23 1 1233 P {i.e. in the Palatini or any one of them) Ci. 603 Ps. 1323 In the root of the verb eicere Qcere) A Mi. 205 P J//. 28? Probably in Cap. 797 iecero is a corruption of eicero. In the root of the verb eire {Ire) A Ep. 79? Afen. 513 Afer. 303 Mi. 1422 Poe. 347 992? Ps. 326 330 349 1 182 Ru. 518 584 1018 Tru. 301? P As. 108? 480? 486 676 Au. 458 694 Ba.ii'j^} 1181? Cas. 212 Ci. 62^ Cu. 487 491 611? Ep. 714 Men. 435 617 736 875 Afer. 282 689 747 749 bis 787 Mi. 521 812 1085 Mo. 336 693 852 969 Ps. 349 Tru. 714? 2. 78 Andrew R, Anderson. [1906 4. In the root of the noun leitus (Jttus) Ru, 1019 A 5. In leis (Tis) Mer. 281 A 6. In catameitus (catamltus) Men. 144 P 7. In suppeilo (suppVd) As. 815 P Leis, catameitiiSy suppeilo, have been included in Class I only by way of conservatism, as the quality of the ei is not definitely known. It is not impossible that they should all be put in Class II. 8. In the voc. sing. masc. mei (nft) A Mer. 503 525 P Men. 182 361 676 Mi. 1330 Ru. 867 (dat.?) Lindsay, however, LL. 427, quoting Charisius, GL. I, 159 K; I, 561 K ; Diomedes GL. I, 331 K ; Velius Longus, GL. VII, 7 7 K, prefers to take mi from mie, thus making it monoph- thongal. 9. In the dat. sing, mei {ini) P Ba, 942 Mo. 194 Ru. 867 (voc?) It is hardly necessary to distinguish here whether the form co!iies from *mei or *moi. In the adj. meirus (nfirus) Ru. 593 A In the neg. nei (ni) P Men. 849 Poe. 865 Ru. 811 7f/. 315 In the root of the verb sino in the perf. subj. act. Tri, 521 sciris (for seiris /) P 13. In the form veis (vis) used as 2d sing. pres. indie, act. of vo/o A A/en. 266 Mer. 287 484 510 687 769 776 Roe. 414 437 Rs. 47 324 P Cas. 964 In the loc. sing. A HEIC Mer. 307 468 773 a. 498 £p. 567 Men. 238 239 241 460 1049 Mer. 311 406 489 518 519 526 531 606 694 784 Roe. 351 659 910 1 2 15 Ru. 1014 Rp. 521 Mer. 266 785 786 Rp. 545 e Mer. 594 NISEI Ru. 1012 PEREGREI Mo. gs7' Charisius, GL. I, 212 K, holds that the form should be peregre. 10. II. 12. 14 SEI SEIC SEIN t jA. Vol. xxxvii.] Ei-Readings in the Mss of Plautus. 79 P heic Men. 375 Roe. 713 sei Au. 699? a. 652? Mer. 155? Ps. 1324 1325 1334 Ru. 950? Tn. 595? Tru. 40? quase (for quasei?) St. 648 septimei Pe. 260 utei Pe. 476? b. From Indo-Germanic oi 1. In nom. sing. masc. of rel. pron. quei {quX) A Ep. 607 Men. 243 Roe. 469 993 P Men. 451 bis 2. In nom. plur. masc. of 2d decl. A Ba. 942? Me7i. 259 260 573 Mer. 262 263 bis 318 bis 778 Pe. 295 394 Poe. 345 bis? 689 988 Ps. 1 1 79 P Cap. in Arg. Acr., Men. in Arg. Acr., Men. 18 bis 19 29 620 1082 1120 Poe. 714? Tru. 99 100 3. In dat. abl. plu. of 2d decl. A Ba. 927 Ep, %^ Men. 202 258 289 290? 459 553 Mer. 299 554 787 bis 790 bis Mi. 13? Pe. 293 Poe. 407 579 1216 Ps. 343 415 y?/^. 764 P Cu. 612 J/^;/. 105 Mi. 165? J/t7. 154 ^^,. 1033, to which should be added the analogical fonns nobeis, vobeis, found in A Mer. 273 Poe. 402 643 664 678 1213 1216 121 7 1253 1274 1277 4. In the root of the verb veiso {vlso) A Ru. 567 c. From Indo-Germanic ai 1. In the dat. sing, ending of 3d decl. A Ep. 229 Men. 263 519 Re. 330 624 P ^<2. 1060? O'. 133 Tru. 551, but perhaps here it would be better to follow Bugge and read muli erei, in which case this instance would justly be transferred to Class II. 2. Termination of pres. pass. inf. A Mer. 769 777 778 Pe. 297? Roe. 1301 P Mi. 884 1163? Poe. 710? Ru. 684 1012 1292 So Andrew R. Anderson. [1906 3. Termination of imperative pass. pres. 2d plur. A Mer. 782 P Mo, 22 4. Dat.-abl. plur. of ist decl. A Ep, 517 Men. 570 1133 bis Mer. 479 bis Foe. 868 976? Ru. 763 772 5. Endings of perf. act. (a) Indie, ist sing. A Ba. 530? Men. 535 591 1139 Mer. 391? 500 P/?^. 386? 750? 1176? 1378 Ru. 217 bis? St. 497 P Am. 926? As. 582? Ci. 547 ^//. 1 131 {b) Indie. 2d sing, (ultima) A Ci. 296? Mer. 754? (/) Indie. 3d sing. A Mer. 530 Foe. 1283? There is no doubt that this termination was always long in Plautus exeept where the law of breves breviantes operated. Yet it cannot be regarded as certain that it was also diph- thongal. Cf. CIL. XIV, 4123 FHEFHAKED (for -eid?), and CIL. I, 32 DEDET (for -eit). Yet it is not to be for- gotten that a more reliable inscription, CIL. I, 196, gives fuit and censuit. {d) Infinitive (penultima) A Mer. 269 PERIEIsse While the ei may here, as in the 3d sing., be defended on analogical grounds, cf. CIL. I, 196 ADIESE ADIESENT ADIESET, and Sommer, Handbuch d. lat. Laut. u. Formen- lehre, p. 628, Anm. 2, it seems to have been confined to flexions of the verb ire, or at least to perfect forms in which the terminations were preceded by /. Cf. CIL, I, 196 COMVOVISE and similar forms. d. From Indo-Germanic eu in kiber (Tiber), free, and its compounds. A Foe. ^20 1218 1240 Ru. 217 [Total in A 182 Class I j Total in P . . . . . 98 I Total in A and P . . . 280 Vol. xxxvii.] Ei'Readings in the Mss of Plautus. 81 Mk Class II. In this class are put the forms in which ei seems to be used for a monophthong, and which are therefore un-PIautine. 1. In dil- < disl- A Foe. 4g4 DEILIDAM? P Ru. 820 deiligentia 2. In -im- < -ism- A £/>. 600 preiMUM 3. In -IS of accus. plur. 3d decl. /-stems < *'ins. A Ci. 244 A/en. 219 231 237 Mer. 281 513 786 Fe. 182 325 Fs. 140 St. 349 607 682 Tri. 236 P £/>. 447 Afo. 47 Ru. 409 583 4. Diphilus P Cas. 32 Mo. 1149? 5. filia Foe. 1239 A Cf. C/Z. I, 54 FILEA 32 FILIOS(nom.) 187 Fl bis Either from I.-G. dhei (hochstufe) * iactere^ which appears in Latin femina, felare, and which in filia has undergone the change of 'e{i) to i through the influence of the / of the following syllable, see Walde, s.v., or perhaps better from I.-G. dhi (tiefstufe) with Buck, AfF. XVII, 270; Solmsen, KZ. XXXIV, 4 ; Brugmann, IF. VI, 93, 3) ; also Brugmann, Kurze vergl. Gram. I, p. 73. 6. miles Foe. 1372? A Cf. CIL. I, 6^ MILITARE etc.; Gr. o-fuX-o? and its com- pounds ; Skt. mildti. Hardly to be connected with Skt. midha, * praeda,' and Gr. /aio-^os. 7. m'llle St. 587 A Probably to be derived with Sommer, IF. X, 216, from *sm't- gzhli, * eine Tausendheit' So very nearly Fay, IF. XI, 320. 8. propmo St. 425 A ; cf. Gr. TrpoTrtVu* 9. VIVO and its derivatives A Mer. 471 Foe. 1187 bis Cf. CIL. I, 33 VITA. From I.-G. u 10. Suffixes -Ico- -~ino- -wo- -isco A AMEICVS Foe. 1213 MORTICEINE/'.'. 283 82 Andrew R. Anderson. [1906 Vol. xxxvii.] Ei-Rcadings in the Mss of Plautus, 83 II. P ame (ici) Cas. 435, ameica Ru, 351, quomq' mei sciam (conquein