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. THE LATIN ACCUSATIVES MED, TED, ETC. By Edwin W. Fay Brugmann's note on the Latin accusatives me (d), te(d), se(d) (IF. 23, 210) develops a quite successful attack on the current theory that the final -d was picked up by the accusative from the ablative forms med, etc. This everyone will admit: the accusa- tives are, in all probability, prior to the period of samdhi fluctua- tion between med and me in the ablative. But when we are asked to derive med (ace.) fromrae + ed (cf. Avestan at quasi '-7e,' cited by Bartholomae in his Worterbuch but once with a pronoun, viz. : dwqm at 'te quidem'), the ice is too thin, I think, for cautious treaders. The truth is, in my opinion, that the -d of the accusative was picked up from the ablatival long before the ablatival -d became "inconstant" in sentence phonetics. Let us suppose for the first person pronoun, say, that we had in Italic proper representatives of the forms for accusative, instrumental, and ablative, as they appear in Sanskrit: then we should have ace. me, instrum. me, abl. med, the last still alive in the enclitic -met which, from ego- met (memet), went over, first presumably to nosmet, whence it spread to other person forms like tumet, semet. 1 In the general syncretism of ablative and instrumental in Latin, met (med 2 ) and me yielded an instrum. -abl. med; and the accusative me became, I suggest, involved in the morphological history of the homopho- nous instrum. me. And be it said in passing that the reason advanced by Brugmann for the upgrowth of the new Umbrian and Oscan accusatives Horn 'te,' siom 'se,' viz., the liability to con- fusion between ace. *med and abl. *med (unfortunately not repre- 1 1 am not oblivious of other explanations of the enclitic -met, none of which seems to me satisfactory: egomet is 'I from (of, by) myself, of my own motion,' where -met exhibits much the same semantic development that is now currently assigned to sed 'by himself >' apart,' an explanation which I personally begin to take cum graito, for sed as a pre- verb (in seditio = quasi 'Weggang') may mean 'away' and be cognate with 656s. 2 Bugge's canon as to final -d in all secondary verb endings in the Italic languages carries with it the corollary that, semantic interference apart, final -t and -d became, by samdhi, indistinguishable at an early period. [Classical Philology IV, July, 1909] 301 302 Edwin W. Pay sen ted in the extant Oscan-Umbrian remains) is no better a guess than the assumption that tiom, etc., are new forms made to avoid the confusion that existed between inherited ace. te and instr. te, etc. As regards the merging of accusative and instrumental-abla- tive into the form med, good syntactical evidence of case-rivalry may be offered, not merely for the utor category, adduced by Audouin (De'clinaison, p. 273), but specifically the evidence of verbs of "sacrificing." Nor does Audouin, by his reference to Breal, Tab. JSug., p. 63, do justice to this category when he cites {loc. cit.) Umbrian buf treif feitu 'tres boves facito' in contrast with vinu feitu 'vino facito.' It were worth mentioning that in Umbrian, by some stylistic allocation, names of animals are put in the accusative with 'facere,' other offerings in the ablative; while in Latin we find the ablative with both classes of objects (cf. Plautus St. 251, quot agnis fecerat?; Act. Frat. Arval., ap. Dessau Inscript. Lat. Select. 5034: "ture et vino in igni in poculo fecit").' Now if we take up the particularly old inscription instance of ace. med to which Brugmann appeals, we shall see, upon exami- nation, that even here, on the Praenestine fibula (Dessau 8561), in the legend Manios med fhe fhaked Numasioi, med is ambigu- ous as to its case. Here, if Numasioi is the name of a person, as, to the best of my knowledge, has been universally supposed hitherto, med must be an accusative. But after all, the fibula 1 Further material for illustrating the case-rivalry may be listed here : sacruflco, abl., Plaut. Epid. 175, "sacruficas ilieo Oreo hostiis," but "pecora rite sacruflcare," Xdvy xli. 18, "ignavam suem s.," Ovid Fasti 4. 414; cf. Plautus Poem. 451 "s. ullum turis granum;" immolare: in view of the original meaning of this verb= 'to pour meal on' (cf. particularly Henzen Acta Frat. Arval., p. 93, 94), the ablative of the thing offered is especially cogent evidence for the vigor of the ablative construction with verbs of "sacrificing ; " examples, to say nothing of literary instances which may be turned up in the lexica (see particularly Ateius Capita, ap. Macrob. iii. 10. 3 ; and cf. Horace Carm. i. 4. 12), are: (Acta, cf. Dessau, 5047) "immolavit suovetaurilibus maioribus .... verbeces immolavit xvi;" Dessau 108, "immolatio Caesari hostift" (on a calendar) ; ibid. 112, "tres libertini hostias singulas immolent." Further note adolere mactare with accusative ; operari with general objects in the accusative, but a cognate instrumental, sacris; litare with the accusative, but cf. "sacra litate bove" (Ovid) ; " hostia maximft parentare " (Cicero Leg. 2. 54) ; " ea mihi cottidie | aut ture aut uino aut aliqul semper supplicat, | dat mihi coronas" (Plautus Aul. 23). In Greek the case rivalry is much less in evidence, but Homer may be cited for ohov in contrast with avivStiv (iSari. The examples of fiijpir with the dative are ques- tioned. On Sanskrit hu- with the instrumental see Speijer as cited below, p. 305, n. 3. The Latin Accusatives Med, Ted, Etc. 303 may have been a votive offering, and Numasioi the name of the divinity attested by the three following inscriptions: D. 3147, Nujmisio Mar[tio] | M. Terebonio C. 1. | donum dat liben[s] meritod. D. 3148, .... onius Q. f . | Numisio Martio | donum dedit | meretod. D. 3149, Marti sive | Numiterno | Achilles Aug. lib. | proa, et Ulpia Nice | eiu[s] d. d. | 1. d. d. d. Cf. Dessau's note: "idem numen in tribus his titulis N. 3147- 3149 significari apparet, praeterea inauditum." But I see no good reason why the Numasios of the Praenestine brooch may not be identical with this "Numisios inauditus," and the fibula a votive offering to him. Nor do I see that we run any risk of being anachronistic in supposing that, in the early period of the Prae- nestine fibula, a soldier wore a brooch, even if we lack literary evidence for the history of the period. The literary evidence for the brooch (Trepovrj) worn by the Homeric warriors is complete. And the brooch was quite specifically a part of the Athenian uni- form for war. Thus Lycurgus, describing a state of panic, when the old men had to be formed into a "home-guard," writes (Con. Leoc. 40) : ISelv r\v «a#' okrjv rrjv ttoXiv iirl yijpw^ 68

vio$ rjdrjKT) top o-refyavov, though I think the origin of /acere = 'sacruficari' is to be sought in the locution sacrum facere, with ellipsis of sacrum. Here let us glance at an early inscription "in margine vasculi aenei prope Nemi repertum" (D. 3236) : "Cn. Q. et med Diana." Here there is no doubt but the object is an offering to a divinity, but we are left in doubt whether med is ace. with <^"dedit"^>, 2 or 1 The following citation from Guhl u. Koner's Das Leben d. Griechen u. ROmer 1 , p. 656, is in point : " fibulae .... bei den Mannern haupts&chlich zur Verknfipfung der Enden des Sagum und Paludamentum auf der rechten Schulter .... unter alien Schmuckgegendstanden diese Fibulae vorzugsweise haufig an einst bewohnten Statten sowie auf Schlaehtfeldem aufgefunden " Be it added here that a fibula pre- sented to Numasios may have been intended for use in dressing his statue for war. 2 The ablative is not to be excluded even if the verb in ellipsis be " dedit." Thus in "aidiles vicesma parti Apolones dederi " (D. 3216), Htibner (" R6m. Epigr." in Mailer's Hdbeh. I 2 , p. 691) construes parti as ablative. For the ablative, cf. also Vergil Aen. vii. 245 : " h6c pater Anchises auro libabat ad aras " Geo. ii. 192 : " qualem pateris libamus et auro." In Homer the dative ( = instrumental) of the vessel used in an offering, ■*■ 196, i\ 137.— If the verb in ellipsis be " fecit," i. e., sacrum fecit, we may compare from the Acta Fr. Arc. (D. 5039), "in mensa sacrum fecerunt ollis " where, obviously, ollis is a plain ablative of means ; but, as later in the ritual they prayed to these ollae (ollas precati sunt, cf. Wissowa, Realencycl. s. v. "arvales fratres," II, p. 1476) "sacrum fecerunt ollis" carries the additional force of "ollas consecraverunt," which accounts for the subsequent worship of the jars. That, last of all, these were broken and thrown away is nothing but a corollary to their consecration : having been made temporarily the depositaries of divinity the jars were thrown down the slope to prevent The Latin Accusatives Msd, Ted, Etc. 305 acc.-instrum. with ,' and et might be a graver's mistake for the abbreviation fe of his copy (cf. No. 3471 on p. 307 below). 1 It remains to consider whether fecit actually reached a general meaning like 'donavit' 1 or 'supplicavit' (cf. Plautus Aul. 23, supra, p. 302), with the construction "alicui aliqu& re," 2 and we may begin with general analogies, first by quoting Plato (Euthyphro 14 C) : to dveiv hcopelcrdai icm tois deols. So Vergil (Aen. iv. 453) designates the tus and vinum Dido was offering as dona; further note in the New Testament, Matt. 5:23: iav ovv Trpoo-epy<; to 8&poi> aov tirl to OvaiaaTripiov A similar generalization of the root yap took place in Sanskrit, where yajanti 'colunt, deco- rant' (deum aliqua re) passes over to the sense of 'donant.'* Passing to take a glance at the particular usage of facere "sacruficare" in the inscriptions, it is not certain that "facere alicui aliqua re" is found; but it will be worth our while to look at a classified list of examples 5 which lead to a provisional conclu- sion in favor of the admissibility of this construction. And the theory on which the list has been constructed accords with the notion already advanced that facere = sacruficare has developed any subsequent pollution, including, doubtless, the pollution of accidental breaking : "ollas precati sunt et osteis apertis per clivum iactaverunt." Did not the "ollarum precatio" involve, at the end, a deprecatio, a deconsecratio implicit in iactatio per clivum? !Cf. mactat (1) 'immolat,' (2) 'donat' (so Lewis and Short, s. v. I and II, A). 2 "Alicui aliqua re parentare" is Latin of record. 3 Cf. Speijer's Sanskrit Syntax, §45 Rem., where the 'proper' construction and the definition of yaj- are laid down ; and note the statement relative to hu- with the instrumental (as well as the accusative) of the offering. 4 The passage cited by the smaller Petersburg Lexicon is MBh. 13. 31. 10: samyag yajanti ye cestlh ksanta danta jitendriyah satyaih dharmam ksitim gagca tan namasyami yadava, to which the commentary runs : " satyaii dharmam ca yajanti : pujayanti ; ksitim gacca yajanti ('bona bovesque sacrificant ') : brahmanebhyah prayacchanti ('sacerdo- tibus donant')." Another passage, not cited in the lexica, is Acvaldyana Grhyasutra 1. 1. 4: "samidham evg 'pi craddhadhana adadhan manyeta yaja idam iti namas tas- msi ['qui] sarmenta tamen credens [in ignem] imponens [fuerit], putet sacrufico (= dono) hoc, honor [sit] illi [deo]'": here we may deny the sense of dono, if we will, but the passage will still have a value as exhibiting the range of offerings. 6 The list lays no claim to being exhaustive and is based on a somewhat cursory inspection of the inscriptions given by Dessau. Nor have I had access to the Corpus so as to make the best use of the materials offered by Dessau. 306 Edwin W. Fay from the locution "sacrum facere," with subsequent suppression oi facere; and, as opem tulerunt continued to exist alongside of opitulati sunt, so, I think, [sacrum] facere continued alongside of sacruficare. A. alicui deo aliqua rS sacrum facere : tibi [= Iovi (Iunoni)] hoc (hac) bove mare (feminft) pulchro (-a) sacrum fiat .... tibi (Apollini) viiii popanis .... sacrum fiat (Dessau 5050, vss. 106, 122, 141 = Acta. Sacr. Saec. celebratorum a. u. c. 737, d. Iun. 1-3). We find the active, with the merely accidental omission of the receiving god, in the following, also a (refact) inscription of the time of Augustus (D. 112) : sive quis <(Augusti numini> hostift sacrum faxit The full example wanted, but with sacruficium in place of sacrum, is found in Dessau 5050, 115: s[a]crificium fecit deis [I]lithyis libeis viiii, .... pth6ibus viiii .... Ilithyia .... tibi viiii pthoibus sacrum fiat. B. 1. sacrum facere alicui deo, re sacra omissa : D. 3527 Herculei [ sacrum | C. Marci C. 1. Alex, fecit .... 3505 Bonae deae | conpoti | sacrum | fecit 1 | Antonia .... 3956 deae Telluri | sacrum | M. Aurelius | Threptos | fecit 3988 deo amabili | sacr. | Aelia Ehorte | fecit 8065 d. m. | sacrum | Deanaeet | memoriae | Aeliae | . . . . | fecerunt. 2. idem quod 1, verbo facere in breve coacto : 3276 Deanae sacrum | . . . . | — , fecer. 3432 Herculi Iovis sacrum | C. Voltilius Alexander fee. 3433 Herculi | Primigenio | sacrum | . . . . | Paetus f . 3460 Herculi | barbato | sacrum | M. Hellius | Secundus | | f cf. 3459 Herculi \ lapidari | Alman[tji | censes | p. (=posuerunt). 8066a Dianae sacr. | Quintae | parentes fecer. 3. idem quod 1, verbo facere omisso : 3995 Vi divinae | sacrum | C. Vettius C. f . | Gavolus (in latere) Vi divinai [ sacrum | (in alio latere) Vi divinae | sacrum | 2 5996 C. Hosidius Mar | cianus cum suis [ Pantheo | sacrum 4. idem quod 1, verbo sacrum omisso : 'Note the juxtaposition of sacrum and fecit. 2 In this inscription the nominative Gavolus attests [fecit] ; in the shorter legends sacrum may be nominative as in 3989: "lucus | sacer | deae | Satrianae." The old spelling divinai reveals the utter inconsequence of the archaisms in the inscriptions. The Latin Accusatives Med, Ted, Etc. 307 3101 Iunone Loucinai | Diovis; castud 1 facitud 6127 Q. Lainio Q. f. praifectos pro trebibos fecit 2 3463 Herculi | iussu, Silvani | dei | Cn. Sentius Corpus | fee. 3 3471 dibus | Herculi | e.t. | Silvano | f. e. 4 | Primus | . . . . | v. s. 1. m. 50 far, and no trace of the ablative of a thing presented, ex voto* with sacrum facere. In most cases the object on which the inscription was found is the speaker, and scholars naturally supply aediculam, vasculum, etc. (better perhaps were me[d] ) and interpret fecit as i-rroiei or as faciendum curavit. The formal ambiguity between dative and ablative goes further to obscure the apparent cases of the ablative of the thing offered which I am about to present: C. facere c. ablativo ret sacrae donatae : D. 3747a Numisi | Aug. s. | C. Iulius | Chryso|gonus | voto fecit. Here voto may be dative = in votum, or ablative = Hid rS quam voverat fecit." 3626 Onesimus Aug. lib., | proa, | fecit imaginibus et | Laribus cul- toribus | Fucini. 1 The interpretation of castud as an adverb = caste is doubtless correct, however much it would suit the contention of this study to render it as " re casta." s This very old inscription was found on the lip of a bronze vessel, and may be compared with No. 3236 above (p. 304). If we supplied med, it would be uncertain whether med were accusative or ablative. 3 In this inscription and the next, it may be that some word like aediculam is to be supplied with fecit: that depends on where the titulus was mounted, whether on an aedicula, or on an object suitable for an ex voto. 4 The mode of writing et and/e[cit] on this inscription throws light on the emenda- tion of et tofe proposed for No. 3236 (see p. 304). 5 1 have no convenient list, for the Roman cult, of ex votos. I have made the fol- lowing slight collection from Dessau : candles to Aesculapius (3851) ; silver cup to Apollo and the nymphs (3876a), cups to Jupiter (3182), to Hercules (3429) ; cymbals (tambourines, dish, scissors) to Magna Mater (4106) ; ears (models of) to Aesculapius (3853) ; a fur (skin) to Diana (3260) ; lamp to Bellona (3806) ; mirror to Venus (3182) ; gold necklace to Aesculapius (3855) ; rings and jewels to Iris (4422) ; golden serpents to Mercury (3192) ; tankard and vase to Jupiter (4317) : also note how in the Plutus (840 ff.) of Aristophanes Dicaeos came to offer to Aesculapius ( ?in lieu of the proper vasa= ffKevdpia) his old cloak (rpi@tivi.ov) and clogs (4/j.pdSia) . 6 Probably for [sacrum} fecit, but possibly for [donom] fecit = donavit. Cf. D. 3679 P. Tullius Felus | fecit oficinator | insule Vitaliane donum fecit — where, with donum fecit, we might supply the object presented in the ablative. This construction of voto fecit does not differ from that found in CIL XIV. 3525 : aditus ambitus [sit] funerio | sacriflcio faciundi | causa. 308 Edwin W. Fay Here Dessau's annotations are: "Imagines in titulo .... sine dubio sunt imagines domus divinae. Lares et imagines Caesaris saepe una colebantur Onesimus in aedicula sacraverat donoque dederat collegio cultorum dei Fucini imagines Caesaris et Larum." The interpretation of imaginibus in this inscription depends on a question of fact: (1) utrum Onesimus imaginibus fecit (fac. cur.);" (2) "an aediculam iam factam imaginibus donatis ornavit (orn. cur.)." For the second inter- pretation of the facts we might appeal to the next inscription in Dessau, viz.: 3627 P. Flavius . . . . j cultoribus domus divinae j et Fortunae Aug. Lares | Augustos d. d. [Thanks to a recent acquisition by our University library of a set of the Corpus, I now have access to the inscription as pub- lished in Corpus IX 3887. The marble plaque containing the titulus is about the size of the average sheet of foolscap, and is quite as likely to have been fastened under or over a niche in the wall of an aedicula already built as above the entrance to such an aedicula. (For a concreter realization of aedicula and niche cf. Fig. 106 in Mau-Kelsey's Pompeii, Its Life and Art'.) The situa- tion, as I imaginatively reconstruct it, was something like this: Onesimus, procurator of the Fucinus drainage-works, presented (on retiring?) to the keepers of the works — organized somewhat grandly as cultores (dei) Fucini- — ■ \with~\ a set of imperial busts for a niche in their official dwelling or board-room. The busts may have been in miniature, and there is nothing to show that a separate aedicula was built for them nor, if there was a separate aedicula, that it was the gift of Onesimus. As for the construction, the double dative as defended by Mommsen is scarcely less unique than the combination of the ablative of the thing presented with the dative of the receiver, which is the construction that I would see here.] At the end of our examination, we cannot claim to have found a more conclusive proof of the construction "facere( = donare) alicui deo aliquft re votiva" than we started with in A above ("sacrum facere alicui deo aliqua re"). But how few of the The Latin Accusatives Med, Ted, Etc. 309 inscriptions reveal to us the object of the formula [sacrum] facere; and should we supply, as the probable object of most of them, me or hoc, we still hang in an ambiguity as between ablative and accusative. Still, we are in safe bounds grammatically when we construe "Manios med fhe fhaked Numasioi" (D. 8561) as "Manius hac r§ votivft <^sacrum^ fecit Numisio ." But there remain unquestionable early instances of med fecit where the sense i/ie eiroki 1 is indubitable. There can be no doubt that in "Novios Plautios med Romai fecid" (D. 8562) we have an indubitable artist's signature to a work of art; and there are artists' signatures also in several of the following numbers (8563-69). 2 We also have med feced 3 as an artist's signature on the Vasculum Dresselianum (D. 8743), but I shall undertake elsewhere to show that this object, while perfectly genuine, was a "latebra scribendi deliciarum causa facta" and, as the work of a "scurra urbanus" who either was, or was connected with, a competent grammaticus, may be dated anywhere between Lucilius, say, and Quintilian. (Cf. Am. Jr. Phil. XXX, 121 f.) At a later time, also, the literal sense of fecit (or = faciendum curavit) is equally clear, e. g., in 3642 Grenium | Clodi Komani | Hermes ser. fee. 3679 P. Tullius Pelus | fecit oficinator ] insule Vitaliane | donum fecit. 4 And now to come back to our starting-point, the case-rivalry of ablative, i. e., instrumental, and accusative with the verb facer e (1) 'Troielv' (2) 'sacruficare' — fading in precision till it reaches 'We actually have epoei, cf. Dessau, ad num. 8604. 2 One of these, in three different versions, may be noted for the object te: 8566a [Rjetus Gabinius C. s. | Calebus fecit b Eetus Gabinio C. s. Calibus fee. te c Retus Gabinio C. s. Calebus feci te. 3 As to ted endo (anglice, 'to thee-ward'), the -d may be (Ritschl's) -d cognate with -8c 'ward.' In aruorswm ead, exstrad, suprad, of the S. C. de Bacch. the -dmay come from ad, postponed as in the Italic dialects. Tor the double ad in aruorsum ead, cf. Sp. con migo ' cum mecum.' 1 4 One wonders whether donum fecit may not be a reinvigoration of simple feci t = quasi 'donavit.' 310 Edwin W. Fay (ava-)n0evai, donate: we have found that it is consistent with general semantics and general syntax, as well as with the special syntax of facer e in the Italic languages, to interpret "med" in "Manios med fhe fhaked Numasioi" (D. 8561), and in "Cn. Q. et med Diana" (D. 3236), as instrumental of the thing presented; while in early artists' signatures like med fecid we see that med is certainly accusative. But in med sacrum fecit, where med might syntactically be ablative (from instrumental), the liability to inter- pret it as accusative, appositive to sacrum, was very great. So much for the syntax of record. It is not in disaccord with the syntax of morphological reconstruction as follows: me func- tioned in early Latin both as accusative and instrumental; med was a blend of instr. me and abl. med; but, in the blending, ace. me was involved along with instr. me. The scurrile locution sua rS facere (Petronius 47. 4; 66. 2 1 ) — start- ing, perhaps, in the jest of some Kabelais, Aristophanes, Petronius of the slums (or of the beau monde) — applies the construction of [Terrae?] aliqua re [sacrum] facere to one of the euphemistic senses of facere (cf . Petron. 62. 4). This low phrase can have nothing in common with Plautine tua re feceris (pace Lindsay ad Cap. 296; Buecheler ap. Lowe ad Petron. 62, 47, 4), wherein tua re (= abs te) 2 is a simple separative, in an archaic locution developed before the syncretism of ablative, instru- mental, locative made the addition of differentiating prepositions neces- sary. The same idiom survives in med ri fert, and I am entirely in accord with Lindsay's rejection (Syntax of Plautus, p. 3) of the other explanations now in vogue. For the semantic development from the separative — rather than, with Lindsay, from the associative — observe the locutions a me facit, slat, est, listed in the Thesaurus I, p. 22 C. University of Texas 1 Readers of this passage will recall the digestive virtue now ascribed to whole wheat ( = autopyrwm) bread. Nor will they feel it necessary to cut out causa as a gloss on re, but rather retain it as a verbal redundancy (cf. the Thesaurus III. 700. 62), such as occurs in autopyrum de suo sibi. 2 Cf . abs te in Rudens 1101, and Sonnenschein's note.