In line 11 of the above occurs the word retlîche, which is emended in Denkmäler to rehtliche. The expression urcundun retliche translates the Latin testes idoneos, and should therefore not be changed. For retliche is a syncopated form of redelîche, which properly translates Lat. idoneos.

In line 29 occurs the phrase ce theru mûzzungu theru selveru samunungun, translating ad immunitatem ipsius ecclesiae. In the vocabulary of Braune's Ahd. Lb. mûzzungu is given doubtfully under mûzzunga 'mutatio.' But the reference here is to the right of a church to hold property deeded to it with the same freedom or exemption from loss as any other heir would enjoy. The word is therefore muozzunga 'immunitas' (i. e. immunity from loss on account of the claims of a co-heir): muozôn freie Zeit haben,' muoza 'licentia, facultas, fas, otium,' muozzan, maozan 'Raum haben; die Gelegenheit, Freiheit, Veranlassung wozu haben; dürfen, mögen, können, müssen.' The word does not otherwise occur in OHG., and was probably coined for the occasion by the translator.


The last two lines of the MS read:

GoRio huob dhia ahnt uhf erbibinota abollin Gebot er uhper den ehtle unht do fuer er sar enabcurnt ihn nequeo Vuisolf

This is given in Kögel's text:

Gorio huob dhia hant af, erbibinôta Abollinus,

Gebot er uper den hellehunt: dô fuer er sâr en abcrunt.

In my text (Mod. Phil. XII, 174 ff.) I write as above, except that h is retained in huf, huper. I now believe that we should take some account of ihn, and write the lines:

Gorio huob dhia hant huf, gebōt er huper den hellehunt:
Erbibinota Abollin, do fuer er sår en abcrunt hin.

This leaves Abolin (except for the capital A) as in the MS, and interchanges the second part of line 58' with the first part of line 59. Abollin is for Apollyon, and hin for in 'ein, hinein.'

In line 55 I should write hilft (not hilfit) for the MS ihlft. The syncope of the unstressed vowel in later OHG is frequent enough to make is unnecessary to correct the MS here, aside from the rearrangement of the letters.


In line 27 of the text as usually given fehda of the MS is reproduced as fehta. Since & is used in MSS as a quick way of writing et, feh&a should be expanded as feheta, with the not uncommon svarabhaktic vowel. For examples cf. Braune, Ahd. Gr., § 69.

In this connection I suggest the following ending for Hl. The second half of line 68 is to be taken as modifying the first part, giwigan meaning here worn out, destroyed.' Of course, it is impossible even to guess at the original words of the closing lines, but we may be sure that there were not many lines more, and that the substance of them was about as follows:


stiottun enti stâhhun

in hartemo wige.

starkên slegim,

unti bêdero bluot, fleotantêr bah,

sih miscta heiz ana dero wuostûn heidu.

do swancta der alto helid dat swert duruh dena helm,

[blocks in formation]

In the quotation above the words are written with the same inconsistency as in the preserved text. Hence stiottun with -ttas in muotti, but fleotantêr like sceotantero; bêdero as in line 62, but heiz (perhaps this should be heit), heidu, with -ei- as in heittu, giweit; dat as in the text; tescrôtan for zerscrôtan, since no initial z- occurs; -au- in haubit, auh as in bauga, rauba.

University of Chicago.



The Res Gestae Divi Augusti, a kind of autobiography of the official career of the first emperor of Rome, has long been known from a copy, with a translation into Greek, preserved on the walls of a temple of Rome and Augustus at Ancyra (Angora), the capital of Roman Galatia in Asia Minor. A few years ago Sir W. M. Ramsay published nearly fifty fragments of another Latin copy 1 which had been set up at Antioch, a Roman colony of Augustus in Pisidia, a division of the province of Galatia. Now Professor David M. Robinson has edited Ramsay's fragments again and two hundred and fifteen additional fragments which he copied in the course of excavations conducted at Antioch with Ramsay's permission.2

This new text of the Res Gestae is of considerable importance; for one reason because it helps to fill some of the gaps in the Ancyran copy. The new readings are given in Robinson's paper. Here it is enough to note that Augustus's use of prepositions and connectives is emphasized, and that the word auctoritate is to be read in chapter 34 where dignitate formerly was supplied. word, more later.

Of this

Robinson states it as his belief that the text of Antioch is not a copy of that at Ancyra because there are differences in the two documents, and yet he is bothered by a very definite likeness between them in the fact that his column IX begins with the selfsame words found at the beginning of col. 6 at Ancyra, and that

1 Journal of Roman Studies, VI (1916), 108-129.

"The Deeds of Augustus as Recorded on the Monumentum Antiochenum," American Journal of Philology, XLVII, 1 (1926), pp. 1-54, with seven photographic facsimiles. These plates (some of them) are none too clear, especially as they reproduce on a small scale lettering which is not by any means large on the stone. Robinson acknowledges the assistance of Mr. E. E. Peterson in the preparation of the paper. I shall refer to this paper as AJPh.

A. v. Premerstein discovered the proper location in the text of Ramsay's fragment containing part of the word auctoritate, Hermes, 59 (1924), PP 95-107.

AJPh., pp. 22 and 49.

these words do not even begin a sentence, but belong to the second line of chapter 32. Robinson has not settled the question raised by the differences and the peculiar similarities between the two copies.

We have here essentially a problem in textual criticism. Its importance lies in the significance of the Res Gestae-a document unique in Roman history, from the hand of Augustus himself. Obviously, if the text at Antioch is a copy from the stone at Ancyra, its readings are useful only where the Ancyran text is lost; but if it is not a copy, all its readings are valuable for the purpose of getting closer to what Augustus actually wrote.

It will be convenient to make a list of the possible relationships between the two copies. (a) Ant.5 is a copy of Anc.

(b) Anc. is a copy of Ant. (c) Both are copied directly from a manuscript sent to Ancyra from Rome, or from copies of that manuscript prepared at Ancyra for the use of various cities of the province. (d) Both are similarly derived from a manuscript sent to Antioch. (e) Each is a copy of a separate manuscript sent directly from Rome to Ancyra and to Antioch respectively. Later the question of their archetype at Rome will be considered.

What is the evidence? In all, four copies of the Res Gestae are known, the two Latin copies at Ancyra and at Antioch, a Greek translation on the walls of the same temple which has preserved the Latin copy at Ancyra, and a second Greek copy at Apollonia of Pisidia in Galatia. The copies, then, are all confined to the province of Galatia. It is a rather remarkable fact that among the many thousands of inscriptions found the Roman Empire over no trace of the Res Gestae has been discovered save in these three Galatian cities. Augustus, we know, directed that the Res Gestae be set up before his tomb at Rome. From the contents Dessau is of the opinion that Augustus intended the account only for Rome, and not for the provincials. Certainly there are many details of merely local importance. Dessau, then, would attribute the making of the copies in Galatia to the personal devotion of a Roman governor. The explanation is very reasonable, though it

Ant. for the inscription at Antioch; Anc. for the inscription at Ancyra. Suetonius, Augustus, 101.

'H. Dessau, Geschichte der Römischen Kaiserzeit, Berlin, 1924, 1, 484 and 585.

may not be conclusive. This argument however looks toward Ancyra, the capital, as the immediate source of all the known copies. And, whether we posit a devoted governor or not, it would seem natural for the different cities to get their copies from the provincial capital. Again, the copies at Ancyra, Greek and Latin, are on the walls of the shrine of the provincial koinon, that is, a religious center of the province. What more natural than to find other cities obtaining copies from their central shrine?

The internal evidence from textual readings must now be considered. First, that of the Greek translations. The Greek copy at Ancyra is rather complete, and it has long been recognized as distinctly a translation with a Roman and Latin flavor. Whether this translation was prepared at Rome or in the province has not been determined. It is fairly literal, following often the word order of the Latin; but in chapter 17 the adverb o [éπara], which there means "from that time on or "for the future," is not found in a corresponding Latin word. It may be an indication that the stone-cutter of the Latin text omitted a word which was given in the manuscript he used. In that case the Greek translation could not have been made from the Latin as it is on the stone at Ancyra. But this is conjecture, and the evidence to follow it up is slight. The copy is Apollonia is fragmentary, yet from five different sections of the document parts of some thirty-four lines, which are considerably longer than the lines at Ancyra, have been preserved." They show a word for word correspondence with the Ancyran copy, and so leave no room for doubt that the translations are one. Moreover in both texts an essential word vaòv which translates templum, is omitted in 11, 21, 10, though there follows after the next word ȧyopáv an enclitic connective re, which seems to presuppose vaov. This omission is hardly due to the translator's manuscript. It proves a very close relation between the Greek copies. But in 3, 6, 14 of the Ancyran inscription Tv is read, while at Apollonia rou is found, and editors prefer the latter reading.10 And in the Ancyran text 6, 10, 5 iorópnoev has been abbreviated by leav


E. Diehl, Res Gestae Divi Augusti, Bonn, 1918, p. 5 and note.

• Mommsen's edition of the Res Gestae, Berlin, 1883; Domaszewski, Philologus, 70 (1911), 569-70.

10 See Diehl, loc. cit., and Cagnat, I. G. R., III, 159, p. 69, loc. cit.

« VorigeDoorgaan »