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 byJSTOR. 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.istor.org/participate-istor/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 COLUMNA ROSTRATA OF C. DUILLIUS Bt Tenney Frank A reference to recent grammars and histories shows that the famous inscription of the Columna Rostrata (Dessau I.L.S. 65) has again fallen into disrepute. It will be remembered that Ritschl and Mommsen pronounced it an antiquarian invention of the early empire. In 1890, however, Woelfllin by an able discussion^ restored faith in it so that for a decade at least it was freely cited by historians and philologists. Woelfflin argued that so many ancient forms and phrases occurred in the inscription that it must in the main be regarded as a fairly faithful copy of a third-century original, and that the forms which were obviously of a later date could best be explained as due to the mistaken attempts of a restorer to invent archaisms in order to replace parts of the original that had been broken or rendered illegible by age. This view seems to me not impossible, though at present it has been generally abandoned for Mommsen's theory. WoelfHin's enduring contribution to the discussion lies in having pointed out the real archaic character of several phrases that up to his time were classed as examples of late and rhetorical verbiage. He noticed for instance that the use of que to the exclusion of et was characteristic of early documents, especially documents of an official character like the lex Agraria, the lex Cornelia, and the Senatus consuUum de Bacchanalibusf that the phrase closes navales^ was not redundant but a natural expression in the early day when classis meant "levy" whether of land or sea forces; that praesente .... didatored preserved in the participle a legitimate old meaning which was conserved in various other verbal forms of praeesse, "to have charge of"; that the phrase inaltod mare could have been used only before altum became a substantive and a synonym of mare, a change » Sitz. Munch. Akad., 1890, pp. 293 £f. 2 Note that que occurs three times in the four lines of the very old inscription of the Faliscan guUd, Dessau I.L.S. 3083. ' Paulus Festus, 225 : muUitudo hominum guam navium; also 56: classis clipeata; and 248: classis procincta. [Classical Philology XIV, January, 1919] 74 The Columna Rostrata of C. Dxjillius 75 that was taking place in the days of Plautus; and so on. Whoever will collect these remarkable archaisms — note the number of such excellent old forms as exemet, inaltod, marid, olorom, triresmos, navebos, numei, and na«>afed,' consider the non-appearance of such forms as qiwm, endo, and ai, which an antiquarian would naturally have selected, and remember that we have no sure case of conscious archaizing of inscriptions — should be convinced that at least the sub- stance of the inscription rests upon a genuine original. The recent aversion to the inscription seems to be based chiefly upon Norden's assertion' that in comparison with the simple and jejune epitaphs and honorific inscriptions of the following two centuries the Duillian eulogy is fulsome and rhetorical; that in fact it is too self-conscious an effort to be attributed to the Romans before the day of their first literary attempts. It must be admitted that the inscription does not harmonize with Roman character of the third century as we know it. Perhaps, however, it is so un-Roman precisely because Rome had as yet no styHstic and formal literary models for Duillius to follow. It was indeed in Sicily, among the most rhetorical of the Greeks and in their most verbose days, that DuiUius achieved his fame. He campaigned between Agrigentum and Segesta before and after his temporary naval command, and doubtless, like other Roman generals in Sicily, spent some days during the rainy season in the hospitable companionship of the gracious Hiero at Syracuse. That Roman ofiicers acquired in this way an acquaintance with many things besides the Greek drama which they at once transplanted to Rome is of course well known. And the man who, according to Cicero, found pleasure in a cortege of pipers and torchbearers could hardly have failed to note during his Sicilian campaigns the elaborate eulogies with which every petty Greek official was immortalized by his native city. In lieu of contemporary SiciHan examples, which have not survived, we may refer to the typical inscription which the Athenians had engraved in honor of Phaedrus some ten years before.^ For services that Duillius might well have smiled at Phaedrus 1 De Stilone, Cosconio, Varrone, 1895. Cf. F. D. Allen, Early Latin, p. 67, who questioned the inacrlption on stylistic grounds long before Norden. 2 DM. Syll? No. 409. 76 Tenney Frank received the honors of the gilded crown, a seat at the public tables and at public games, a bronze bust, and an inscription of a hundred periphrastic lines setting forth not only his own deeds but those of his father and grandfather as well. A brief extract will perhaps serve to illustrate the redundant and padded eulogies current in Duillius' time : (irK(]va\et,ai> rots TrXeoiwrt rrjv dciKaTTav, and so on to the dreary end. Though this is of course not the model of the columnar inscrip- tion, it reveals the style of the honorific tributes which met the Roman in every public square in Sicily. The frequently criticized boast "rem navebos primos c " reminds one of x«tpoTOJ'r;^€is ■KpSnos of line 45; the fulsome style of "clasesque navales primos ornavet pa cumque eis navebos claseis Poenicas omnis" is cast in the style of TXavKirov Kadei\ribTO$ Kvdvov Kal Karayayov- Tos ivrevdev ra TrXoia riiv re TriXtP '{Ka^tv Kal airrbv T\avKeT-qv Kal to. irXota TO. ixer' avrov, and "enque eodem mac " when there was but one year of office is reminiscent of phrases like iirl NiKioy IJ,€V apxoPTOS CTparqybi viro tov Sijuov xeipoTovi)del$ iwl Tr)v ■trapaffKevriv dls irdvT N.H. xxxiv, 22. 2 Livy, xxvi, 27. ' In rostris .... vidimus, Servius in Georg. iii, 29. The earlier writers who saw it say simply in foro: Pliny, N.H. xxxiv, 20; Quint, i, 7. The fragments were found near the arch of Severus, that is, near the new rostra. « Pliny, N.H. xxxiv, 30. The Columna Rostrata of C. Duillius 79 The forms claseis and naveis occur by the side of clases and navales, and it is usually supposed that these words contain a mixture of old, pseudoarchaic, and classical forms. Obviously the single consonant s suggests an early date; the diphthong ei, on the other hand, could not belong to 260 B.C., nor does it accord well with the first decades of our era, for spurious ei does not octur in accusative plural endings of vowel-stem nouns as late as the date of the final restoration of this inscription.^ The forms claseis and naveis are, however, quite possible for the period about 150 b.c, that is, for the period of the posited restoration of this inscription. The single consonant persisted through most of the second century, as witness suma^ of the Gracchan period. Our earliest inscriptional example of the spurious ei* dates from 146 b.c, but inscriptions are so rare for the two decades pre- ceding this date that we are not in a position to claim that it did not occur several years earlier. The fact that Turpilius could con- tract flagitieis into jlagitis^ would seem to indicate that unaccented ei had become phonetically equivalent to i before the middle of the century. The ending es of clases and navales are probably due to the imperial restorer or stonecutter, but we have no right to exclude a hypothesis that these forms may also date from the second-century restorer. This es of the accusative plural of vowel stems does not as a matter of fact appear in inscriptions till the end of the century,* but there are two facts to consider with reference to it. Inscriptions dating from the middle of the century offer very few instances of just these forms from which to draw conclusions; and, secondly, since the es of vowel stems came into use through a confusion between vowel and consonant stems, it is highly probable that it appeared sporadically all through the second century b.c, since such con- fusion occurs in the other case forms then.* If this is so, the single consonant of clases accords with the ending es and the whole word may possibly be attributed to the earlier restorer. ' Lommatzsch, Archiv, XV, 129. ' C.I.L., I, 551. ^faxseis, C.I.L. I, 542. * Sturtevant, Contraction in Case Forms, p. 6. 5 omnes, C.I.L. I, 577, 105 B.C. « See Stolz, pp. 204 and 211. 80 Tennet Frank The form consol fits the second century and that alone. In 260 B.C. the word was probably written cosol, while consul was of course the form in the empire. The ns of Cartaciniensis must belong to the same writing as consol. Poenicas seems also to be the work of the earlier restorer. The word was doubtless spelled with oi in 260, while at the time of the imperial restorer oe had long since passed to u. The uniform adoption of ae in praesente, praedad, and aes is usually attributed to the imperial restorer or stonecutter, but it may also be due to an earlier restorer since the diphthong ae occurs in inscriptions as early as the S.C. de Bacch., 189 b.c. With our hypothesis macistrtos need not be taken as an invention of an imaginative imperial restorer. The dramatists of the second century often treated w-stem nouns as belonging to o-stems, and the confusion may be very old. Navaled and dictatored can now be accepted as correct third-century forms since the appear- ance of osoled^ has lent them respectability.^ No real difficulty is presented by the fact that the former should end in id, for vowel and consonant stems were frequently confused in early Latin. According to Hodgman,' Plautus alone presents a score of examples of such confusion in the ablative singular forms. Finally, by positing an intermediate restoration and thus account- ing for several hitherto inexplicable forms, we may accord fuller faith to such old words as maximos and navebos, which have been viewed with needless skepticism in the past. Brugmann's explanation of the variation between u and i* leads one to expect maximos as older than maxumos. Whether both forms (in i and in u) occurred in the inscription of 260 b.c. we cannot say positively, but there is no sound reason for doubting that they did. I should be inclined to attribute ximos of line 3 to the original inscription, and umos of line 9 to the original or to the second-century restoration. Navebos has also been rejected as an unsafe form, but Sommer* is certainly right in accepting it. The ending is orthodox and is paralleled by protrebibos of C.I.L. IX, 4204, while the e is supported by the spelling tempestatebus in I, 32. The word evidently came down unchanged from the original inscription. Exfociont remains then the only 1 Notizie, 1900, 499, and C.I.L. I, 2, No. 19. 2 Brugmann, K. Verg. Or., 383. » Class. Rev., XVI, 298. * Grd. I, p. 224. 6 P. 385. The Columna Rostrata of C. Duillius 81 unexplained word, and this may be a stonecutter's error induced by the occurrence of the unusual o of the ending. If this hypothesis is correct, then, the main body of the inscription goes back to 260 b.c, while a restorer of about 150 B.C. supplied at illegible places the following forms in the orthography of his day: ns of consol and Cartadniensis, eis Poe of 1. 8, ae in 1. 9, eis in 1. 12, and possibly es in 1. 7 and ae of 11. 14 and 17. Most of the forms that are certainly of the second century fall close together in the center of the present inscription at 11. 6-9, so that it seems not improbable that the original was spaced line for line exactly as the extant one is and that it was rather badly marred in one place before the first restoration. The final restoration was of course made in the early empire. Perhaps the curatores tdbularum publicarum appointed by Tiberius in the year 16 a.d. to restore drjfidcria. ypafinara (Dio 57, 16) had the work done, or Tiberius himself in the following year when he rebuilt the temple of Janus which Duillius had erected (Tac. Ann. II, 49) and which may have been mentioned in the last portion of our inscription. That the final restorer took some pains to follow his copy with care is proved by the fact that he corrected the erroneous M to in navebos, and this fact should guard us from attributing many modernizations to him. The only changes that one is compelled to attribute to him are the use of i-longa, which is purely orthographic, and the apparently mistaken o of exfociont. Regarding the form of the original monument only a word need be said. It is likely that the statue mentioned by the elogium of Duillius in the Augustan Forum^ stood on top of the columna, as in the case of the old columnar statue of Minucius,^ and the columna rostrata of Augustus.' Since the authorities of the imperial time who had seen the new column do not mention the statue, we may assume that it had met with some accident or had been removed as an offense to the eye before this time. We are told that old tufa and terra-cotta statues were not in favor in Gate's day. Of the inscription, fulsome as it is, something has apparently been lost, possibly as much as fifteen lines if the base — which we ' See Dessau, I.L.S. 55. It was discovered in 1890. 2 Represented on coins; see Grueber, III, PI. 26, 15. » See Grueber, III, PI. 69, 18. 82 Tennet Frank need not assume — had the normal proportions of statue bases, that is, length to width about three to two. The first line probably contained the name and title of the hero, as in the elogia of the Augustan Forum; and since Florus assumes (erroneously, of course) that Duillius. had himself erected the column, the inscription may have failed to men- tion who had authorized the erection. If it was suggested by the Greek inscriptions of the day, the lost portion at the end probably mentioned the hero's benefactions, as for instance the gift of the temple of Janus' (Tac. Ann. II, 49), and gave a full list of the honors voted him by the senate and people,^ not failing to note that his was the first naval triumph,' and perhaps adding the fact that huic per- missum est, ut ab epulis domum cum tibidne et funali rediret.* Indeed the prominence given by the elogium and by so many authorities to these vain honors can be plausibly explained if we assume that this peculiar inscription, which followed the pompous Greek fashion of recording such trivialities,* was their source. At any rate Silius Italicus, who gives a fairly full list of the traditional details, seems to have had the monument and the text of the inscription in mind (Pun. VI, 663-69). He is describing a fictitious panel of a temple door representing the deeds of Duillius: Aequoreum juxta decus et navale trophaeum. Rostra gerens nivea surgebat mole columna; Exuvias Marti' donumque DuiHus, alto Ante omnis mersa Poenorum classe, dicabat. Cui, nocturnus honos, funalia clara sacerque Post epulas tibicen adest; castosque penatis Insignis laeti repetebat murmure cantus. Bryn Mawr College > Cf. the inscription of Phaedrus cited above, especially 1. 30. 2 Cf. the inscription of Phaedrus, 11. 71-91. ' Primus .... navalem de Sicul. et classe Poeniea egit, Acta Triumph. * Elogium; cf. Cic. Cato 44, Liv. Epil. 17, etc. 6 There is a possibility that the cpttfae mentioned were the Epulae Jovis to which Duillius was perhaps granted continued admittance on the analogy of the Greek ff£Ti)