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 PROVENANCE OF JEROME'S CATALOGUE OF VARRO'S WORKS By G. L. Hendrickson The existence of Jerome's catalogue of the writings of Varro was known before the discovery of the document itself from two sources: from mention by Jerome in his life of Origen (de viris ill. ch. 54) and by excerpts preserved by Tyrannius Rufinus (Apologiae in Hieron. ii. 20) from the letter which contained the catalogue. In the lively inquiry into Varro' s literary work which marked the second quarter of the last century, the loss of this list was fre- quently lamented, and all hope of its recovery had apparently been abandoned. But in the summer of 1847, L. Urlichs was drawn to the seat of Sir Thomas Phillipps, Middlehill, by the report of his notable manuscript collections, and here, among other things, he was shown and received a copy of a privately printed extract from a MS of Arras entitled, "Preface to Origen on Genesis." It con- tained Jerome's letter from which the excerpts of Rufinus were drawn, with the complete catalogues of Varro and Origen. Through his Plautine studies Ritschl had already been drawn to a closer study of Varro, and his famous monographs on the Libri disciplinarum and on the Logistorici had but recently appeared (1845). To Ritschl, accordingly, Urlichs referred his discovery, and upon the basis of the recovered catalogue Ritschl produced his great essay, "Die Schriftstellerei des Marcus Terentius Varro," Bonn, 1847 (incorporated into Rheinisches Museum for 1848 and now accessible with supplementary and related studies in Opuscula, Vol. III). The list as edited in the private publication of Sir Thomas Phillipps contains 522 books. It does not contain all of the known works of Varro, and indeed its omissions are very considerable. Ritschl, of course, saw at once the necessity of verifying the printed list by comparison with the MS of Arras, and a few years later this was done for him by August Schleicher (the well-known Sanskrit- ist and comparative philologist), who made an exact transcript [Classical Philology VI, July, 1911] 334 Jerome's Catalogue of Varro's Works 335 of the whole letter with a facsimile of the Varronian catalogue. This transcript and facsimile were published as a university Program in the autumn of 1849, and may now be found in the volume of the Opuscula referred to above. The most essential correction yielded by Schleicher's revision was the number of books for the Imagines, viz. 15, which Sir Thomas Phillipps' compositor had inverted so as to read 51. The matter rested here for some years, until in 1856 a French scholar, Chappuis, discovered the same preface to Origen in two Paris MSS. These new sources of the text (apart from other corrections which need not detain us here) revealed a new work entirely omitted in the MS of Arras, an epitome of the Imagines in four books. The corrections thus afforded were pointed out by Ritschl and discussed in Rheinisches Museum for 1857. Had Ritschl, as he did so carefully in the first instance, now footed up the total number of books in the list as thus revised, he would undoubt- edly have made the surprising observation which it has remained for Prof. Alfred Klotz of Strassburg to make — that the sum given by Jerome is four hundred and ninety books. 1 The significance of this number is revealed by a well-known chapter of Gellius (hi. 10) in which, apropos of Varro's Imagines or Hebdomades, he discusses the virtutes potestatesque septenarii numeri. At the end of the chapter Gellius complains that some of Varro's instances of the significance of the number are farfetched, such as the seven wonders, the seven wise men, and the seven against Thebes: Turn ibi addit se quoque iam duodecimam annorum hebdomadam ingressum esse, et ad eum diem septuaginta hebdomadas librorum con- scripsisse. Seventy times seven is four hundred and ninety — the total of Jerome's list! That this coincidence can scarcely be a matter of chance, everyone, I think, will agree with Klotz. But the meaning of the coincidence remains a matter for interpretation, and it is my belief that the conclusion which Klotz draws will scarcely bear the test of criticism. 1 Klotz's paper is the first in the current volume of Hermes (XLVI). Apart from the striking discovery which is his starting-point, and from his main contention con- cerning the source of the catalogue which I shall here criticize, his discussion is full of suggestiveness and advances our knowledge in many important details. 336 G. L. Hendrickson His view is stated briefly and positively: the catalogue which Jerome reproduces is derived immediately from the prefatory book of the Imagines, where it stood in connection with the passage from which Gellius quotes. That the catalogue is ultimately from Varro's own hand Ritschl had already inferred, but Ritschl thought rather of some such work as the libri de sua vita as its source. Klotz's reason, however, for considering the Imagines as the necessary source lies in the fact that, in Varro's continuous and prolific literary activity, the point of time obviously did not exist before and could not recur again when the sum of his writings was exactly (ad eum diem) seventy hebdomads. But cogent as this reasoning may seem, a moment's reflection will show that it betrays a hint, at least, of the logical circle. In effect he says: Varro is the author of the catalogue. It contains four hundred and ninety numbers. Therefore it must have been compiled at the point of time and in the place when and where Varro reports that he had composed four hundred and ninety books. The defect of this argument lies in the assumption that Varro is the immediate compiler of the list. But how if he be not ? How if the words of Varro in the preface to the Imagines have merely set a problem for some later scholar to solve? But before taking up the question of the possible source of the catalogue, let us examine more carefully the assumptions upon which Klotz's hypothesis rests. They are these: (1) That Jerome has given us the total number of books in the catalogue which was before him. (2) That since his interest was directed upon the number rather than upon the names of works, his curtailment of the list consists in adding together under one title two or more works which stood adjacent to each other and were related either in content or form. (So, for example, the item Antiquitatum 45 consists, as Klotz thinks, of 41, the true number, +4 de gente Romana [not named in the catalogue], while in the case of libri singulares 10, the compression of a series of //.ovofiifiXoi into a single item is mani- fest.) (3) That all of the books in the catalogue were written before the prefatory book of the Imagines. (4) That all other works not expressly mentioned in the catalogue or accounted for by the method of curtailment above described were written afterward. Taking up these assumptions in order, let us note first the language Jerome's Catalogue of Varro's Works 337 of Jerome with reference to his curtailment of the catalogue. He concludes his enumeration with these words: Et alia plurima quae enumerare longum est. Vix medium descripsi indicem et legentibus fastidium est. Now, it surely is a very difficult feat of interpretation to make this language coincide with Klotz's theory of curtailment only by suppression of titles. The only natural meaning which the words can bear is, "I have copied off scarce one-half of the list," and so Ritschl understood it (Op. hi, 485, 487), allowing, of course, for a large margin of pardonable exaggeration on Jerome's part. Vix medium surely can only mean half through, from beginning to end, as when (for instance) Geta says in the Phormio vix dum dimidium dixeram, intellexerat. We should have to be under some extraordi- nary compulsion of circumstance (such as Klotz doubtless feels) to grant for a moment that vix medium indicem could at one and the same time cover the half, i.e., of the titles, and the whole number of books (which the hypothesis demands). But the point need not be pressed to a decision on the basis of this phrase alone; the words of Jerome which immediately precede settle the matter beyond possi- bility of cavil. As has been said, and as Klotz recognizes, the interest of Jerome is in bulk or numbers (to contrast the zeal of a former age with the sloth of his own 1 ), and after the long list of the works which he does mention he continues, et alia plurima quae enumerare longum est. Surely the significance of this can only be that merely as a matter of bulk or numbers many more volumes {alia plurima sc. volumina 2 ) remain to make the count (enumerare) complete. We must conclude, therefore, that Jerome's list contained not only titles but also volumes which he does not record. With regard to the second of Klotz's assumptions, that Jerome has curtailed the Varronian catalogue by conflation of numbers, with consequent suppression of titles, it is obviously sound with reference to the item libri singulares 10. His explanation of the item Antiquitatum 45 may be allowed as plausible. But the fact that 1 Ut intellegamus nos Epimenidis dormire somnium, etc. s Cf. the introductory words: et quia non (?) otiosum est apud Latinos Grae- corum voluminum indicem texere, de eo qui Latine scripsit aligua commemorabo, etc. 338 G. L. Hendrickson Jerome has not done this in some other cases where it would have been very natural, as, for example, in the two items de scaenicis actionibus 3 and de actis scaenicis 3, would suggest caution. That the logistorici, with the surprisingly large number of 76, should be a pocket from which to exhume works of such character as the de antiquitate litter arum ad Accium and the de utilitate sermonis is certainly improbable. If conflation had taken place at this point they would much more naturally have followed the related works which stand next — the de lingua Latina or the de sermone Latino. Concerning the third of Klotz's positions, it is obviously neces- sary for him to show that all of the books in the catalogue were written before the prefatory book of the Imagines. The main question here touches the relation between the date of the Imagines and the libri rerum rusticarum, which the catalogue includes . Concerning this latter work our information is specific: annus enim octogesimus admonet me ut sarcinas conligam (i. 1.1.), that is, the year 37 B.C., the eightieth of Varro's age. Only a little less specific is Varro's own testimony concerning the completion of the Imagines. It has already been cited above (p. 335) : se quoque iam duodecimam annorum hebdoma- dam ingressum esse. The natural meaning of this phrase is obviously the beginning of the twelfth hebdomad, that is, the seventy-eighth year of Varro's age, which yields the year 39, as Ritschl assumed without question. But Klotz urges with apparent plausibility that the words are indefinite, and he sees no reason why the range of their meaning may not be extended to a point of time which shall include the eightieth year. That this in fact is the case, Klotz says (p. 9), is shown by the inclusion of the rerum rusticarum in the catalogue. The logical flaw in this argument (for it assumes the point to be demonstrated) need not be urged, since it is true that, if it were impossible to fix more definitely the limits of fluctuation in Varro's phrase, we should have to admit the possibility, at least, of Klotz's contention. But consider for a moment: Varro is dealing with the mysterious significance and coincidences of the number seven, that is, the hebdomad. Applying it to his own life at the moment of writing, he observes that he has just finished a hebdomad and has entered upon another. Now as a matter of fact we have no assurance that Varro was not fibbing to make his case good, but his case Jerome's Catalogue of Vaeeo's Works 339 has no point unless he either then was or pretended to be exactly, as he says, at the beginning of another hebdomad. If we are at liberty to stretch the phrase as we like, it is obvious that the statement is at once removed from all relation to seven, and so loses all value as an illustration of coincidence upon seven. To be sure, Varro has already passed the point of coincidence, but he has just passed it — iam duodecimam annorum hebdomadam ingressum esse et ad eum diem septuaginta hebdomadas librorum conscripsisse. It will be seen that in the two co-ordinate sentences iam and ad eum diem aim to indicate a point of time with precision. Tarn is not to be rendered with "already"; its meaning is represented rather by the English "just" — the German "eben" — a usage which is abundantly attested. 1 I have dwelt upon this point because it enables us to prove beyond reasonable doubt that the passage of the Imagines, which in Klotz's view gave occasion for the catalogue, must have been written some two years earlier than the libri rerum rusticarum. But the latter work is enumerated in the catalogue. We must conclude, therefore, that the catalogue did not stand in the Imagines. Furthermore, Jerome's list includes an epitome in four books from the Imagines; but it need scarcely be said that the existence of such an epitome before the completion and publication of the work epitomized is highly improbable. With regard to the libri discip- linarum (also in the catalogue), which Ritschl placed in the year 33 B.C. on the strength of an allusion in the elder Pliny, the evidence is not sufficiently decisive for the purposes of an argument of this kind. Ritschl's conclusion has been challenged, though as Schanz thinks wrongly. Finally, as the fourth of Klotz's positions, it is necessary for him to assume that all other works of Varro not expressly mentioned in the catalogue were written after the Imagines, unless their absence can be satisfactorily accounted for by the theory of curtailment of the original list by conflation of numbers and suppression of titles. 1 Ingredior is inceptive in meaning and iam is used with it frequently to mark precisely the point of inception: iam ingredientem in navem .... retractum esse (in Verr. 5.96). Neronem iam ingressum iuventam commendavit patribus (Ann. 3.29), "who had just arrived at man's estate" (Ramsay). Similarly with inceptive verbs: iamque rubescebat Aurora (Aen. 3. 521), marking the precise moment at which Italy was first seen by the Trojans. 340 G. L. Hendrickson That in this way the early work de antiquitate litterarum ad Accium is accounted for by Klotz, as having been absorbed into the seventy- six books of logistorici, I have indicated above, suggesting at the same time a doubt as to the probability of such an explanation. This explanation, however, cannot be employed to account for the absence of the libri de poetis, as Klotz frankly confesses: "Nicht zu erklaren weiss ich dass auffallende Fehlen des Werkes de poetis (p. 15)." Its proper place in the catalogue is indicated by the treatise de poematis, on both sides of which are series of works of three books each, a number too small to have absorbed a work which consisted certainly of more than one book. Concerning the date of the de poetis, I do not find in Ritschl nor in the manuals of literary history any positive suggestion, and perhaps none is available. But the Brutus of Cicero gives us a terminus ante quern which admits of no doubt. The fundamental investigation upon which this conclusion rests is Leo's memorable chapter on the life of Plautus in the Plautinische Forschungen. Some supplementary evidence I adduced in my paper on a pre-Varronian chapter of Roman literary history in the American Journal of Philology for 1898 (Vol. XIX, pp. 289 ff.). It appears, then, that Jerome's list omits at least one important work in several books which could not reasonably have been omitted from a catalogue of Varro's own making. To gather up, now, our conclusions in brief summary, this much would seem certain: (1) Jerome's source cannot have been the Imagines. For if a list of Varro's writings had stood there it must have consisted of four hundred and ninety numbers — no more, no less. But Jerome's list contained other works which he does not cite, and not only other titles, but other volumes — alia plurima. 1 (2) The list of Jerome was compiled at a date subsequent to the completion of the Imagines, since it contains one work certainly, the libri rerum rusticarum (and probably others), which was written at a later time. (3) The list omits one work, the libri de poetis (and > In this connection it should be said that the Imagines, so far as we can discern, was a book of popular nature, combining portraits with a descriptive text and a metrical elogium for each portrait. Apart from other considerations it would seem highly improbable that such a work should have been considered a suitable place for minute autobiographical matter, such as a detailed and complete index of the author's works. Jerome's Catalogue of Vaebo's Works 341 perhaps more), which was already written and published at the time of the completion of the Imagines, nor can its absence be explained by the hypothesis of conflation of titles. From these considerations the conclusion is inevitable that the list is not from the hand of Varro — at all events not from the time nor the place which Klotz assigns. That, as Ritschl believed (p. 491), Varro himself in his autobiography (a work presumably of his latest years) had either drawn up a complete list of his own works or had furnished such information concerning them as to afford the basis of a catalogue need not be denied. As regards Jerome's catalogue — to distinguish it from an hypotheti- cal list of Varro's own making — it would seem rather that the words of Varro had set a problem for some later Trivaicoo<; or literary biographer; that, undertaking to make a complete list of Varro's writings, he had used as a point of reference Varro's own testimony. He thus put together a number which should yield 1 a total of four hundred and ninety volumes down to the publication of the Imagines, and then with indication of this division line he continued and added others, which were still numerous. That errors of chronology (as in the case of the rerum rusticarum or the de poetis) or carelessness of double inclusion (as with the de valetudine tuenda 2 ) must be charged to the compiler of the list will scarcely invalidate such an hypothesis. Perhaps someone better versed in the pinacographic literature of i Is it not possible that some trace of this method of construction may be dis- cerned in the entry de valetudine tuenda 1 ? The surprising character of this item Ritschl noted (p. 440) : "es beruht gewiss auf irgend einem Versehen oder Zufall, sel es des Hieronymus oder schon seiner Quelle," because it violates the manifest rule of the compiler not to enumerate separately and with title works of one book. The other fj.ov6fii(3\(H are put together without titles into the item singulares 10. But furthermore, unless all signs fail, this work must be identical with the work which is very accurately cited in the learned note of Probus ad Eel. 6.31: Varro in logistorico, quem inscripsit Messalla de valetudine. That is, not only is the entry of a twv6f3i- fl\os unique in our catalogue, but also this particular work, as being a logistoricus , has already been included presumably in the 76 books assigned to the logistorici as a whole. What, then, shall we say ? Is it purely oversight or accident, as Ritschl held ? Or did the compiler need another number to make up his total ? A similar question might be raised in regard to the surprisingly small number of libri singulares, viz. 10. One has only to compare other lists of famous polygraphs to see how surprising it really is, and for my part I should hesitate to subscribe to Ritschl's conclusion (p. 441): "daraus folgt aber fiir uns dass .... diese Form (sc. /uop6/3i|8Xos) also eine sehr untergeordnete Rolle in der Schriftstellerei des Varro spielte." 2 See the preceding note. 342 G. L. Hendeickson antiquity may be able to adduce a parallel or analogue to the division of the catalogue such as I have suggested. It would seem, however, that it might have arisen very naturally from the auto- biographical item which Varro recorded in the Imagines. But we cannot disguise the fact that the material for constructing a positive theory of composition is very meager. We can only affirm with confidence that Jerome's list contained more volumes than the number adduced, and that therefore some division line within the complete catalogue afforded Jerome a natural stopping-point at which to break off his enumeration. We may now contemplate the list, essentially as Ritschl did, as the work of a subsequent biographer or literary antiquarian. We shall grant readily that it depends on information derived from autobiographical sources (such as this very item of the seventy hebdomads) and so may include titles which no one but Varro himself was in position to afford, that is, works which were written but never published. 1 With reference to the possible time of the compiler's labors a slight but not wholly negligible clue is afforded by the two titles satyrarum and satyrarum Menippearum. For in the light of our present knowledge of the history of the word satura, and of the recognition of a form of literature bearing this name, it is doubtful whether we can credit its use to Varro at all. Certainly its employ- ment as a title would imply a currency of usage which we can scarcely place earlier than the second half of the first century a.d. 2 To be sure, the word had come into use in the later years of Varro's life, when the libri de sua vita were presumably written, but inasmuch as the Menippean satires go back in their origins to his early years I venture to doubt whether the name saturae was ever attached to them by Varro himself. It represents rather, I suspect, an inter- pretative classification of a later time. Nonius cites them uniformly by specific titles (e.g. Varro Marcipore), and not once, I think, uses satura or saturarum. Even the passage of Gellius, upon which the usual designation rests, is less valuable as evidence than is com- 1 Gell. loc. cit.: aliquam multos, cum proscriptus esset, direptis bibliothecis suis non eomparuisse. ' See Marx Proleg. in Lucil., p. xv extr., and the writer's paper in the last number of this Journal (pp. 129 ft). Jehome's Catalogue of Varro's Works 343 monly assumed : Menippus cuius libros M. Varro in safaris aemulatus est, quas alii cynicas, ipse appellat Menippeas (ii. 18; cf. xiii. 31). From this earlier editors constructed the title saturae Menippeae. Bucheler with characteristic reserve uses only Menippeae. In the language of Gellius, however, saturis need be only the descriptive terminology of his time, 1 and grammatical congruence would then yield inevitably cynicas and Menippeas. But another form of title is suggested by such citation as Macrobius' Varro in Menippeis suis (v. 20, 13) or Probus' Varro in cynicis (ad Eel. vi. 31), which I take it would in themselves be interpreted most naturally as standing for Menippea or cynica, following the analogy of Empedoclea or Aratea. From the statement of Gellius we should then conclude that Menippea was Varro's own term, and we should infer that cynica was an interpretative addition, or perhaps a popular title of later time. The form Menippea would apply very accurately to Varro's statement in the pages of Cicero of his relation to his cynic predecessor — Menippum imitati, non interpretati — and it would account most adequately for the various allusions to him as Varro Menippeus. The designation and classification of his Menippean writings as saturae then followed upon the general recognition of a category which could receive them, and so they appear in the preface of the elder Pliny and in Quintilian's survey of Roman satire. Yale University 1 Just as Apuleius (Flor. ch. 10) uses saturas to characterize and classify the Greek writings of the Cynic Crates of Thebes, pupil of Diogenes and predecessor of Menip- pus. On the text of Apuleius see Susemihl I, 30, n. 87.