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 DATE OP NOTITIA AND CURIOSUM By Elmbb Tkuesdell Mbbbill The two related statistical documents of the fourth century commonly called the Notitia and Curiosum have had a history varied and interesting enough to furnish the plot for a novel of adventure. They sprang from a common source, but hardly lived long enough in an unsullied existence to recognize their own individuality. Neither, indeed, accurately represented their com- mon parent, and the craft of man immediately began its confus- ing work with them. Some details were speedily added to each by way of improvement, to bring them "up to date." A genera- tion later one was still further emended by borrowing from the other, and the resultant combination was drawn upon for other documents. The details of the history from now on till the great awakening of interest in the topography of ancient Rome that marks the fifteenth century appear to have been insufficiently investigated. Some young student may do us a considerable ser- vice by a comparative study of all the extant manuscripts of both Notitia and Curiosum in the light of the other documents more or less allied with them. But upon the approach of the fifteenth century they were no longer to be allowed to ramble about as irresponsible and anonymous vagrants. Signorili ascribed the authorship of one of them, at least, to the well-deserving Paul the Deacon. Later scholars went him several better. Flavio Biondo attributed a poor copy of the Curiosum to Sextus Rufus. Giano Parisio entered the fray, and after duly interpolating a copy already bountifully interpolated by Pomponio Leto, printed the result as the work of Publius Victor, apparently a specially con- jured-up double of Aurelius Victor. Later on this pseudo-author appears as Publius Aurelius Victor, or as Publius Fabius Victor. Then that ever-active falsifier, Pirro Ligorio, took a hand, and still further interpolated the pseudo- Victor, writing it out in an imitation of ancient majuscules, with intent to deceive. Onofrio [Classical Philologt I, April, 1906] 133 134 Elmer Teuesdell Mereill Panvinio, honest man, came into possession of this manuscript of Ligorio, interpolated it boldly in, as he himself proclaimed, "sex- centi loci," and published the result with cheerful vanity in 1558. This book was widely used, and, as the irony of fate would have it, was generally believed to represent an original of which the real Notitia and Curiosum (still known) were held to be but stupid and blundering epitomes. Thus matters continued till Sarti, in the last century, succeeded in shattering the imposing claims of mingled fraud and blunder, and restoring the modest rightful heirs to their own again. Thus far the outline history of the case, which may be sifted in detail out of the pages of Preller, Urlichs, Jordan, De Rossi, and Mommsen. If only these documents were persons, is there not here the material for a plot of the good old-fashioned sort? In consideration of the complication of the text of Notitia and Curiosum by reason of interpolations and alterations that date back to a very early period in their existence, it is not strange that scholars since the return of sanity have been at variance on many points concerning them. The original text of each is still somewhat open to question; the precise relation of one to the other, and of both to their common source, admits of further investigation; and on the question of the date to which the com- position of each is to be assigned there has been, and is still, a difference of opinion. Bunsen, for example [Beschreibung d. Stadt Rom I, p. 175) , believed the Notitia to have been com- posed before the time of Constantine, and the Curiosum no earlier than the sixth century. Becker was at first disposed to ascribe both Regionaries to the fifth century, along with the Notitia Dignitatum [Handbuch d. rOm. Alterthilmer I, p. 71) but later concluded that they were of Constantinian origin {ibid., p. 709). Preller assigned the Curiosum, in its existent form, to some time between 357 a. d. and the reign of the younger Theodosius (408-50) and the Notitia, which he believed to be merely a recension of the Curiosum, to that reign [Regionen d. Stadt Rom, p. 66). Mommsen ascribed the Notitia to the year 334, and the Curiosum to a period anywhere between 357 and the eighth century {Abh. d. sdchs. Gesell. d. Wiss. II, pp. 549 ff. ; III, The Date of Notitia and Curiosum 135 pp. 269 ff.). Jordan places the Notitia between 334 and 357 (pos- sibly between 337 and 357 ; Topog. d. Stadt Rom II, p. 9), and the Curiosum between 357 and 450 (possibly between 357 and 403; ibid. I, pp. 49 ff. ; II, p. 540). Eichter [Topog. von Jiom,^ P- "7) assigns the Notitia precisely to the year 354, and the Curiosum to the year 357, or a time not very long ("nicht allzuweit") there- after. But this last remark evidently frightens him a bit, for he hastens to say that the lack of mention by the Curiosum of the restored and renamed Pons Aurelius and Pons Probi is no sure indication that it was composed before 450 a. d. Lanciani be- lieves the Notitia to date from 334, and the Curiosum from or after 357 [Ruins and Excavations, p. vii). And finally, our own Professor Platner ( Topography and Monuments of Ancient Rome, p. 5) says: "The Notitia dates from 334, the Curiosum from about 357 A. D.," though he could hardly believe it possible, I should judge, that the date of the Curiosum could be in any case earlier than 357. He does not take space, however, to assign any reasons for his conclusion. Now it will be observed that all this variation of opinion about the dates of Notitia and Curiosum tends to narrow decidedly as we come down the years from Bunsen's time to our own. This is doubtless to be ascribed partly to the natural growth in judg- ment that one age ought to show as above its predecessors, and partly also to an equally natural tendency, but one not always, I fear, safely founded, to insist on the possibility of fixing an abso- lute date for everything that must have had an absolute date. The mind can not be happy without dates. The Notitia, then, appears to be finally penned up between at most 334 and 357, and only the Curiosum to roam more widely afield, being itself limited, however, by an earlier date beyond which it can not retreat, namely, the year 357. The necessary limitations of space in this journal preclude the examination and discussion of the arguments that led the authorities whom I have cited to the con- clusions adopted by them. I may remark, however, that those who assign the Notitia definitely to the year 334 appear to follow Mommsen therein (I must not include in this statement Professor Platner, of whose reasons I know nothing), and those who assign 136 Elmeb Tkuesdell Merrill the Curiosum to 357 precisely, or to a period not far removed therefrom, seem to be actuated mainly by a shuddering dread of the huge vagueness that lies beyond, and therefore cling as closely as possible to the boundary-stone they are sure of, without having a confident reason for their action. In venturing to set forth briefly my own opinion on this vexed question of date I must con- fine myself to the minimum sufficiens of argument, and must repeat many things that are well known. The general identity of Notitia and Curiosum shows clearly that they are either derived the one from the other, or both from a common original. Let us assume for a moment the former of these two alternatives. Now, the Notitia contains some forty particulars, great or small, that are not found in the Curiosum, while the Curiosum contains only ten that are not found in the Notitia, and at least three of these, as I shall remark below, are clearly glosses of a later date than the proper topographical items (the earliest known manuscript of the Curiosum is of the eighth century). Furthermore, the order of particulars is several times different in the two documents, and in general (though it can not be made out for every instance), where this variation occurs, Curiosum appears to be right and Notitia wrong. Under our provisional assumption, then, the Curiosum must be viewed as an abbreviated, and, in point of order, corrected, edition of the Notitia, with a very few later interpolations, or the Notitia as a considerably interpolated, and occasionally dislocated, edition of the Curiosum. Now, of these two possibilities, can any student of the natural history of such texts, and of the times in which they were written, have the slightest hesitation in affirming that the latter, and not the former, alternative is thus far the only possible one? The natural tendency (even as illustrated by the immediately later history of the texts themselves) is toward inter- polation, not toward purging — is toward inaccuracy, not toward accuracy. Notitia must be a redaction of Curiosum, rather than the reverse. (See also a different form of argument in Jordan Topog. II. 1.) But Notitia can not be a redaction of Curiosum; for by no means all of the particulars in the latter are of late origin, and in The Date of Notitia and Cubiosum 137 a so richly interpolated edition as Notitia there would be no suf- ficient reason to account for the omission of most of these details peculiar to Curiosum, if Notitia were descended from Curiosum. Moreover, one of these details certainly could not have been omitted from Notitia, if it had stood in its source. In the "first Appendix" Curiosum reads: Obulisci VI in circo maxima duo minor habet pedes LXXXVIIS maior habet pedes CXXIIS. Notitia, on the other hand, reads merely: Obolisci V in circo maximo unus altus pedes LXXXVIIIS (so cod. A; but cod. S agrees upon the one obelisk). The Notitia, then, can not con- ceivably be descended from the Curiosum. We are therefore driven to adopt the second of our original alternatives, that the Notitia and Curiosum are each descended from a common source. I should like to take space to show that the two lines of descent are pure — that is, that there was no cross-fertilization until a period later than that to which we can trace the texts backward; but that may for the present purpose be assumed. If, now, the Notitia and the Curiosum are both descended from a common source, to what period is that source, in its latest traceable form, to be assigned? I do not raise here the question as to the period of its original compilation, and its original form, nor the question whether the two appendices formed part of the document in its original form. These questions are too far- reaching for the necessary limits of this paper. As I have pointed out, a comparison of Notitia and Curiosum shows that, assuming there has been no cross-copying, the latter contains at most but few additions to the common source. If, now, either Notitia or Curiosum mentioned but one or two build- ings of Constantine's time, we might possibly be justified in regarding these, even in the Curiosum, as interpolations, made later than the time of the common origin of the two Regionaries ; and, since the Notitia contains so many additional items, the occurrence of these one or two identical Constantinian items in both documents would be no proof that they stood also in the common source. The latest form of the source might well be, then, of an earlier date than the reign of Constantine. But both 138 Elmer Teuesdell Mekeill Notitia and Curiosum enumerate four structures of Constantine's time — the Thermae, the Basilica, an Arch (in Reg. XI, and therefore not the arch near the Colosseum), and a Portico. In consideration of the scanty indications of interpolation in the Curiosum, these identical items, though few, yet appear to be too many to allow us to consider them as interpolations. I agree, therefore, with those who consider it proved that the source, in its latest form, was of Constantine's time. Now, of the four buildings mentioned, the Basilica alone can be dated within reasonable limits. Aurelius Victor (De Goes. 40. 26) says it was built by Maxentius (therefore between 306 and 312), but dedicated (or re-dedicated?) by the senate to Con- stantine. (In the Notitia it is called basilica Constantiniana, both under Reg. IV and in the first Appendix; in the Curiosum it appears as basilica noua under the Region, but as basilica Constantiniana in the Appendix. In the source, therefore, both appellations probably stood combined — basilica Constantiniana et noua, or basilica Constantiniana uel noua; cf. the similar pair of formal and popular designations, Romanum and magnum, of the Forum in the heading of Reg. IX.) The Thermae were apparently not begun till at least 312 (Aur. Vict. De Caes. 40. 27), and must have taken some years to complete, as they were very extensive; but they might well have been mentioned by name before their actual completion. The Arch, if the lanus Quadrifrons of the Velabrum (so Bunsen, Nissen, Jordan, and others) was apparently not begun till at least 312 (Nissen Da^s Templum, pp. 210 ff.). Of the Portico nothing further is known. The "source," accordingly, must have assumed its final form no earlier than 312, and probably a year or two later, to allow for the mention by name of Thermae and Arch, and perhaps for the double designation of the Basilica. But, on the other hand, it could not have reache'd its final form by the time the Arch of Constantine near the Colosseum was built, and the Temple of Romulus finished and dedicated to Constantine, for in its evident plan to include the recent structures of Constantine's reign it could hardly have passed over these prominent monuments. The temple was dedicated to Constantine in, or after, 315, according The Date of Notitia and Cubiosum 139 to current notions, since its dedicatory inscription mentioned that emperor with the title of Maximus, and his first use of that title has been assigned to the year 315, on the basis of Eckhel's inves- tigation of the evidence from coins (Eckhel VIII, p. 94). It seems more reasonable to assume this earliest possible date for the dedication, since the temple was partly built by Maxentius, who had been dead already three years (De Rossi, in Bull. Crist. 1867, pp. 66 ff.) ; and 815 is surely the date of the inscriptions on the arch near the Colosseum. The "source," then, took its final shape certainly before 315, and as certainly after 312 — perhaps in 314. Both Notitia and Curiosum are, therefore, productions of a date later than 314. As to the terminus ante quern, let us deal with the Curiosum first. Of all the specifications peculiar to that document alone, there is only one that clearly refers to an erec- tion of later date than the source. That is the specification of the second obelisk in the Circus Maximus, already mentioned; and this obelisk was set up in 357 A. d. (Amm. Marc. xvii. 4. 14). The precision of manner in which the new item is included, with the careful specification of height to the half-foot, joined to the fact that no other structures clearly erected later than the "source" are gathered in (not the Temple of Romulus, nor the Arch near the Colosseum, nor even the Equus Constantini of 334; see below) strongly suggests that the mention of the new obelisk was added to the MS of the Curiosum very shortly after its erection, and that as an isolated, contemporary fact, not as the result of a formal revision of the document, which would certainly have involved more changes. But there are three other phrases found in the Curiosum alone that, to judge from their character and tone, must have been of a considerably later origin than 357. They have a dis- tinctly mediaeval flavor and air. Two of them are lexicographical glosses; viz., the addition of the words quod est putea to lacos, and of quod est sicessos to latrinae publicae, in the second Appendix. The third is mythological, reminding one of the MirahiUa: under Reg. XIV, where the Notitia, doubtless repro- ducing accurately the "source," reads merely Herculem cuban- tem, the Curiosum reads, Herculem sub ierram medium cubantem 140 Elmer Truesdell Merrill sub quern plurimum auri positus est. Such glosses as these are likely to have crept into the text — or into the margin, at least — nearer the date of our earliest MS (s. VIII) than 357 A. D. If it could be proved that the writer of that MS found them in the margin of his immediate exemplar, we might, in our assignment of the date of origin of the Curiosum, simply disregard them, and, in the lack of other evidence, fall back upon 357, or a date very soon thereafter, as the date of the document as a whole. But, though trivial, and even foolish, they do not stand on an essen- tially different foundation as parts of the MS from the mention of the second obelisk. That mention was indubitably not due, as I have pointed out, to a formal revision of the document, which would surely have led to more far-reaching changes. It is a gloss, like any other. That it is a valuable, and a datable, gloss is our good fortune. But to assign the birthday of the document to the date of that particular gloss, merely because its date only happens to be known within probably a few months, is to employ language in an arbitrary and misleading way. And as to the five other items or words peculiar to the Curiosum (Kegg. II Dacicum, III ludum magnum, VIII et Mineruae, IX Neptuni, and Mineruam Calcidicam; see above for Keg. IV basilicam nouam), it is impossible to say whether they were added (prob- ably as isolated glosses) before or after 357. Jordan was inclined to think that the Curiosum was composed into its extant form (exclusive of the three late glosses mentioned above) before 403 A. d., on the ground that otherwise the re- newed gates of the Honorian wall would naturally have been mentioned (of course in the Appendix). That might be true, if the Curiosum had ever been subjected to a formal revision and correction. So would some other things have then been included. But the absence of these, and the presence of but very few par- ticulars apparently or certainly not found in the "source," are conclusive against the occurrence of any such formal scholastic, or official, or quasi-official, revision. Therefore nothing about the date of the document can be concluded from the absence of mention of these gates. The same may be said about other sim- ilar suggestions. The Date of Notitia and Cvbiosvm 141 All, then, that can be reasonably affirmed about the date of the Curiosum is that a certain statistical document, having an origin in official sources, assumed, in 314 A. D., or at a date not more than a year removed from that in either direction, the form from which, before, or very soon after, 357 A. D., a copy was made; and this, with the gradual accumulation of a very few desultory glosses (one of which can be assigned to an origin not earlier than 357 A. d., and probably at most but a very little later), was the ancestor, not many generations removed, of our Curiosum of the eighth century. Now we are prepared to deal with the question of the Notitia' s date. Here again, as for the Curiosum, the terminus post quern is 314 A. D. (or plus or minus a year). And again, like the Curi- osum, of the variations peculiar to this particular document (and there are forty of them, great and small) one can be dated with a fair degree of probability. In the eighth region (Forum Ro- manum) the Notitia mentions in immediate connection with the senate-house, an equus Constantini (of course an equestrian statue). Similarly, the anonymous compiler (s. VIII) of the itinerary preserved at Einsiedeln mentions as on the right hand after passing down the Via di Marforio, and under the Arch of Severus, and just before turning around the corner of S. Adriano into the Subura, a cauallus Constantini (Urlichs Cod. Vrb. Bom. Topog., p. 71). And the same Einsiedeln MS, in the accompanying collection of inscriptions, gives an inscription in basi Constantini, datable as of 384 A. d., from the specified name of the consul ordinarius; and this inscription occurs in immediate connection with that on the Arch of Severus (ibid., p. 63). Now, it is of course not absolutely certain that the basis Constantini of the inscription is that of the cauallus Constantini of the Itiner- ary; but the lack of mention by either document of other Con- stantinian monuments in the immediate neighborhood makes the probability very decided that the two items refer to the same thing. Furthermore, the similar collocation of the equus Con- stantini of the Notitia, and the cauallus Constantini of the Itin- erary, with the similar lack of mention in the Notitia of any more adjacent Constantinian structures, establishes a like strength 142 Elmer Teuesdell Merrill of probability that they refer to the same statue. Therefore the conclusion is tolerably certain that the equus Constantini of the Notitia was set up in 334 A. D. So far the common argument, though differently expressed. Now Mommsen was of the opinion that the Notitia must have assumed its present form in precisely the year 334 A. D. He argued that it could not have been before that date, because of the mention of the equus Constantini; and it could not have been after that date, since a certain Chronicle which is surely of that year copies (so he held) from the Notitia. But Jordan pointed out that the Chronicle does not necessarily depend upon the the Notitia (Topog. II, pp. 30-39) and although I should not argue in precisely the same manner, I agree in the conclusion reached, and need not rehearse my own argument here. There- fore Mommsen's determination falls to the ground, and we are still left to discuss the terminus ante quern. Now, though it is of course conceivably possible that all the other thirty-nine variations of the Notitia crept into it between 314 and 334 A. d., yet this is by no means likely. There is no indication of a formal revision and enlargement, any more than in the case of the Curiosum; and if even desultory glosses were being added to the text in considerable numbers during that vicennium, there would appear to be great probability that such important monuments as the Temple of Romulus and the Arch near the Colosseum would, one or both, have found their way in also. But they did not. The indication would seem to be, therefore, that at least the greater part of these thirty-nine vari- ations are of later date than the year 334. The mention of the equus Constantini in the Notitia is to be explained in the same way as that of the second obelisk of the Circus Maximus in the Curiosum. The equus was a new, and therefore striking, thing when the owner of the Notitia MS, prompted by novelty, jotted it down in his margin. The Temple and the Arch were already familiar, and therefore commonplace, things, and he was not set- ting about a formal review and improvement of his document. The other thirty-nine variations may have accumulated through- out a considerable term of years. It is by no means necessary to The Date of Notitia and Curiosum 143 believe, as most, if not all, recent students have done, that these variations were all in before 357, the date of the erection of the aforesaid second obelisk. Two men, we know, possessed at that time MSS from the Constantinian "source." There were very likely a dozen others equally fortunate in this respect. But why should we suppose that every one of these men was so devoted to the improvement of his record as to pull out his MS, when that obelisk went up, and correct the Appendix in accordance with the fresh event? At least one, we know, did so. His MS gives us our Curiosum. At least one did not do so. His MS gives us our Notitia. How the others acted, if there were others, we can not tell. Their MSS have not come down to us. Similarly, the absence of mention of the Honorian gates, and of the rechristened bridges of the fifth century, can furnish no reason for concluding that the Notitia had assumed its present form before the time of those structures. Nor can the num- ber of MS generations between the source of 814 ( ?) A. D. and the lost Speyer MS of the eighth or ninth century be deter- mined. I have, to be sure, some impressions on this point, and on the date and origin of certain of the additions in the Notitia to the items of the source, but there is not space to discuss them here. The only reasonable conclusion from our present evidence, as it appears to me, concerning the date of the Notitia is quite as vague as that concerning the Curiosum. The Regionary com- monly called the Notitia had a common source with the Curiosum in a statistical document, which assumed, probably in 314 A. d., or within a year of that date in either direction, the form from which, before 334 A. d., or at most very soon thereafter, a copy was made, which was later interpolated from a gradual accumu- lation of glosses, one of which can be assigned to the year 334, or to a time very soon thereafter. When all the remaining glosses were accumulated, and whether within a single generation of the MS or not, can not now be determined. But at most probably only a few generations separate the Constantinian "source" from the Notitia of the (lost but copied) Speyer MS of the eighth or ninth century. 144 Elmer Teuesdell Meeeill It is of course possible that the archetype of either Notitia or Curiosum may have been, not a copy of the MS of 314 A. d., but that actual MS itself; but in this case the copy which served as the archetype of the sister document must have been made before the process of interpretation had fairly begun. Trinity College, Hartford