mail: Bill Thayer	Italiano	Help	Up	Home
This webpage reproduces an article in Classical Philology Vol. 1, No. 1 (Jan. 1906), pp69‑80. The text is in the public domain. This page has been carefully proofread and I believe it to be free of errors. If you find a mistake though, please let me know!

 p69  The Septimontium and the Seven Hills

In 1896 Wissowa published an article in Satura Viadrina (Breslau), entitled "Septimontium and Subura," in which he discussed the so‑called Septimontium or second stage in the growth of the city of Rome, and gave a new explanation of the meaning and position of Subura at that early date. His views have been generally accepted by topographers, although of late dissenting voices have been heard. I have therefore thought it worth while to review briefly the history of the septem montes of Rome, and to present and discuss Wissowa's theory of the Septimontium and Subura.

In the literature of the Ciceronian and Augustan periods there are not infrequent references to the seven hills (septem montes) of the city of Rome, such as the following from Vergil (Aen. VI.783): septemque una sibi muro circumdabit arces felix prole virum, and (Georg. II.534): scilicet et rerum factast pulcherrima Roma septemque una sibi muro circumdedit arces. Cicero (Ad Att. VI.5.2) speaks of the ἄστυ ἑπτάλοφον, and Tibullus (II.5.55) writes: carpite nunc tauri de septem montibus herbas, dum licet hic magnae iam locus urbis erit. Varro (De lingua Latina v.41) says: Septimontium nominatum ab tot montibus quos postea urbs comprehendit, and (ibid. vi.24): dies Septimontium nominatus ab his septem montibus in quis sita urbs est. Aulus Gellius (xiii.14), where he is quoting the opinions of M. Valerius Messala Corvinus, consul in 53 B.C., speaks of the septem urbis montes in such a way as to show that Messala used the term as an ordinary designation for Rome in his day. We may therefore assume without reservation that in the periods under review men spoke regularly of the septem montes of the city, and the question presents itself: Which were these seven hills?

A comparison of the notices found in the historians, Dionysius, Strabo, Livy, Varro, Tacitus, etc., makes it clear that these  p70 septem montes were the hills of the city inclosed by the Servian wall, although there is nowhere an exact catalogue of all these hills in one passage. Tradition varied more or less in assigning the addition of some of the hills to this or that king, but its oldest form represented Servius as having incorporated the last two, the Esquiline and the Viminal. The chief variation in the traditional account concerned the Quirinal and the Caelian; but be this as it may, there is little doubt that under septem montes in the Ciceronian epoch were included normally the Palatine, Capitoline, Aventine, Caelian, Esquiline, Viminal, and Quirinal — five montes and two colles.

In later imperial times, however, with the growth of the city, it is evident that considerable confusion was introduced into this list, as is shown by Servius' note on the passage from Vergil (Aen. VI.783) already quoted:

In this passage the list is that of the Servian hills, except that the Janiculum has been substituted for the Capitoline; a substitution easily explained by the fact that the Janiculum had become an important part of the city in Augustus' reorganization, and that the Transtiberine region could not be entirely omitted. Servius assumes that even in Vergil's time this must have been the ordinary view of the content of septem montes, for the expression breves septem colliculos . . . . qui tamen aliis nominibus appellabantur, whatever this may mean, certainly does not refer to the seven hills of the Servian hills as originally understood.

So also the Bern. Scholiast on this same passage says: septem id est septem montes Romae Aventinus Tarpeius Caelius Ianiculus Quirinalis Viminalis Esquilinus. Tarpeius being only another common name for Capitolinus, we have here the list of the hills in the Servian city, with the substitution of the Janiculum for the Palatine. A comparison of such notes shows that the later commentators felt bound to admit the Janiculum into the list, and therefore eliminated one or another of the original names.

 p71  In the Regionary Catalogue of the fourth century the list of hills includes the following: Caelius Aventinus Tarpeius Palatinus Exquilinus Vaticanus et Ianiculensis; and in the Mirabilia of the twelfth century we find Ianiculus Aventinus qui et Quirinalis dicitur Tarpeius Palatinus Exquilinus Vaticanus et Ianiculensis, showing an increasing uncertainty of tradition.

Occurring first in Varro we find another term, Septimontium, used in a double signification. In the De lingua Latina (v.41) Varro writes: ubi nunc est Roma Septimontium nominatum ab tot montibus quos postea urbs muris comprehendit, e quis, etc. There can be no doubt that Varro means by this to state categorically that the settlement which has now become the city of Rome was once called Septimontium, because it extended over seven hills, the same which the Servian wall afterward inclosed. Mommsen has explained the passage to mean that there were as many hills in the Septimontium as in Rome, but not necessarily the same. This would involve the change of quos to quod, and even then it would be somewhat difficult to reconcile the text with the fact that Varro proceeds after e quis with a description of the seven Servian hills. Another reference to Septimontium in this sense is found in Festus (321), where we read: Sacrani appellati sunt Reate orti qui ex Septimontio Ligures Siculosque exegerunt.

The other meaning of the word is the festival celebrated on December 11, which occurs in the following passages:

Varro L. L. VI.24: dies Septimontium nominatus ab his septem montibus in quis sita urbs est feriae non populi sed montanorum modo ut Paganalia qui sunt alicuius pagi. Here it is perfectly clear that Varro believes that the festival of the Septimontium derived its name from the seven Servian hills, and that it was a festival which was celebrated by the montani in their local organization, as distinguished from the populus as a whole. Comparison of this passage with that previously cited from Varro makes it certain that there too the writer is thinking only of the Servian hills, and not of any different list.

This Septimontium was evidently one of the more important rural festivals in imperial times. Suetonius Dom. 4, says: septimontiali sacro quidem senatui equitique panariis plebeii sportellis  p72 cum obsonio distributis initium vescendi primus fecit. Columella (II.10.8) speaks of the Septimontialis satio; and Palladius (xiii.1) writes: faba circa septimontium seri potest. Tertullian (De idol. 10), in speaking of the temptations to idolatry to which teachers are exposed, says: etiam strenuae captandae et septimontium et brumae et carae cognationis honoraria exigenda omnia. Other references to the origin of the festival are: Schol. Veron. ad Aen. II.635: septem montes unde etiam dies sacer septimontii constitutus est, and Plut. Qu. Rom. 69: τὸ δὲ σεπτιμούντιον ἄγουσιν ἐπὶ τῷ τὸν ἕβδομον λόφον τῇ πόλει προσκατανεμηθῆναι καὶ τὴν Ῥώμην ἑπτάλοφον γενέσθαι.

In the calendars the festival is mentioned three times. First in the rustic calendar, dating from the early Empire, found in mantua and called fasti Guidizzolenses (CIL I.253). In this calendar only seven festivals in the six months from July to December are noted, one of which is the Septimontium on the third day before the Ides of December, the 11th. The other occurrences are in the calendar of Furius Dionysius Philocalus, 354 A.D., where it is written Septimontia and in that of Polemius Silvius, of 449 A.D., where it is erroneously ascribed to December 1. These late references show that the festival lasted until the end of the Empire.

In the other calendars which contain this date, the fasti Maffei, Praenestini, Amiternini, and Antienses, there is no mention of this festival, but only notice of an Agonium. This was explained by Mommsen as another designation for Septimontium, but the prevalent view at present is that Agonium is only a general term, denoting a celebration in honor of some deity whose name is not given, and that the Septimontium occurred along with this Agonium on December 11, just as the Liberalia took place on March 17, along with another Agonium (Martiale).

As we know that the festival was celebrated as early as Varro's time, some explanation of its omission from these calendars must be sought, and it is doubtless found in the fact that is noted expressly by Ateius Capito, cited in Festus (245), that all sacra were divided: publica sacra quae publico sumptu pro populo fiunt quaeque pro montibus pagis curiis sacellis; at privata quae  p73 pro singulis hominibus familiis gentibus fiunt. In the calendars only the feria publicae of the first kind are mentioned, and the Septimontium, being plainly a sacrum pro montibus, would not be mentioned in the ordinary fasti of the earlier Empire. This corresponds exactly with Varro's statement, already quoted, that the Septimontium was feriae non populi sed montanorum modo ut Paganalia, etc. Over the precise meaning of montani and pagani there has been much discussion, but the evidence justifies Wissowa in summing up the matter as follows:

The conclusion would therefore be that the Septimontium was celebrated at the end of the Republic by the montani as such, although practically this embraced all the population of the city, about local centres, and that it did not become one of the feriae publicae of Capito's first class until the old and normal distinctions in such matters had been lost.

So far it is easy enough to follow the tradition, but there is another series of references which complicates the problem seriously. In Festus (348) we read as follows:

This list contains the Palatium, Cermalus, and Velia, the three parts of the Palatine, the Fagutal, Oppius and Cispius, the three parts of the Esquiline, the Caelian, and the Subura. Paul. Diac. in his epitome (341) writes:

 p74  There is also a mutilated gloss on the Festus passage which may be emended to read as follows:

Now, it is at once clear from these citations that Labeo interpreted the festival of the Septimontium as applying to a list of hills entirely different from those of the Servian city, and belonging to an earlier stage of the city's existence, when its extent was much narrower. In other words, Labeo and Varro are absolutely at variance with each other, and it is entirely impossible to reconcile their statements by any exegetical jugglery.

Further, there are eight hills named instead of seven, and it is self-evident that some error is lodged here; for no antiquarian would be so foolish as to cite the names of eight hills to explain Septimontium. Various measures have been resorted to in order to reduce the number to seven. Niebuhr struck out Subura; Bunsen, Caelio; Huschke changed the order and, inserting in, reads Subura in Caelio. None of these methods commends itself with any great force. It is evident that this list is arranged on no topographical or historical basis, and the only question arising is how to make seven out of an apparent eight.

Leaving this for a moment, and going back to the two explanations of Septimontium, we have to compare a note of Servius, already cited, on Aen. VI.783:

There seems to be no doubt that breves septem colliculi refers to the hills that are given by Festus, and that the divergence in view between his source and Varro had continued until a late period, when antiquarians were divided between the two opinions and a third which was a combination of the two. All of which leads us surely to the conclusion, that in the Ciceronian period there was a festival called the Septimontium, celebrated presumably by the inhabitants of the city about local centres, but  p75 whose origin was already so uncertain that learned antiquarians like Varro and Labeo published diametrically opposite explanations. This is the more surprising as it was a matter which concerned the topography of their own city. Such divergence of opinion on such a subject would certainly seem to be strong evidence for the remote antiquity of the Septimontium itself.

Now it has been generally assumed by recent topographers that we may distinguish two stages in the growth of the city between the earliest settlement on the Palatine and the city that was marked out by the so‑called wall of Servius Tullius, and these two stages have been called the Septimontium and the City of the Four Regions. The evidence for the first of these is wholly derived from Labeo's explanation of the festival of the Septimontium, and from topographical possibilities and probabilities. The third stage is based on Varro's description of the four regions into which the city was divided — a division which is thought to antedate the Servian extension. As already stated, Wissowa's monograph contains the most careful presentation of this view, but it has recently been called in question on two grounds: that of too great artificiality, and that of too little basis of evidence. His argument in outline is this:

To this argument of Wissowa's the objections may at once be made that two occurrences of Septimontium as the name of a period are not enough to warrant us in believing in the truth of the tradition; that, if there had been such a period, we should have certainly had other references to it; and, most important of all, that it is impossible to admit that so learned an antiquarian as Varro should have been in doubt about such a matter. This last is a serious objection, and therefore, leaving it for the moment, it will be well to examine the questions connected with the names of the montes themselves.

Three passages need to be taken into account, those already quoted (p73) from Festus and Paulus, and a third from Lydus, whose reputation for obscurity is here ably sustained. De mens. 118 Bekker:

The uncertainties and possibilities of the text of Lydus are numerous, but it appears probable that he is attempting to reproduce the view contained in the passage from Servius previously cited, and that in his first list he intends to give the names of the seven hills of the republican city, Τιβούρτιον standing for Caelian and Πραινέστιον for Quirinal, although this is very problematical. In the second list he had, as a matter of fact, given the Servian hills over again, except that Βελινήσιος (Velia) has displaced the Viminal. His evidence may be regarded, therefore, as valueless for the actual lists, but as valuable for its bearing on three points: (1) that he, too, recognized Septimontium as a name of one stage in the city's growth (ἡ περίοδος τῆς πόλεως ὅτι ἐπὶ ϛ´ λόφους τὰ τείχη τῆς Ῥώμης ἐκτέταται); (2) that there had been almost always two interpretations of the application of the term Septimontium; and (3) that any definiteness of distinction in the names of the two sets of hills had long since died out. We are left, therefore, to the testimony of Labeo alone for the names of the earlier list.

There are in this list three difficulties — one purely formal, namely the presence of eight instead of seven names, and two questions of topographical fact, namely the explanation of Subura and Fagutal. It is now generally admitted that Fagutal was the name applied to the slope of the Oppius toward the Velia, the present situation of the well-known church of S. Pietro in Vincoli. In earlier times it was sometimes regarded as the valley between the Oppius and the Cispius, but that was when the valley of the Subura was also thought to have had a place among the montes. This identification of the Fagutal may be regarded as an accepted fact, and Labeo's list contains, therefore, the three parts of the Palatine — Palatium, Cermalus, and Velia — which are thoroughly vouched for by manifold evidence, the three parts of the Esquiline — Oppius, Cispius, and Fagutal — and the Caelius and the Subura.

The Subura in historical times was one of the best-known  p78 districts of Rome, not altogether to its credit. It was the valley stretching from the Forum up between the Oppius and the Cispius, and can not by any subterfuge have been called a mons. Another suggested explanation is that Subura was first part of one of the hills, and then became the designation of the valley at its foot; but this would be an unexampled and unnatural transfer.

Not to linger over less probable hypotheses, Wissowa argues as follows:

While it is not certain to which part of the Caelian we should attach the name Subura, whether to the western part (with Richter) or to the eastern part (with Wissowa), so far as his explanation of Caelio-Subura goes the latter's argument seems probable,  p79 well supported by evidence, and altogether acceptable, provided we may believe that there was a difference between the name Subura in historical times and earlier. He says:

On the whole, this theory seems to me the only one that can lay any claim to probability, or to being a reasonably satisfactory explanation of the situation. The principal difficulties — and the only ones that are of special importance — are, first, the discrepancy between Varro and Labeo, and the apparent ignorance on the part of the former of the original content of Septimontium as given by the latter. If Varro was not ignorant of this, he must have deliberately refused to admit its truth, in which case we should, I suppose, be bound to accept his view as against that of Labeo in a matter of Roman antiquities. To be taken in connection with this objection, however, is the increasing confusion in regard to the whole matter of the names and number of the hills, so that a difference in the first century B.C. does not seem quite so strange as it otherwise might. Anyone who had accepted the canonical account of the founding of Rome, as it had taken shape in the time of Cicero, would probably find it hard to divert his thoughts from the history of the addition of the seven Servian hills, as we find it given in Livy. The silence of other Roman writers about the Septimontium does not seem a valid objection in the face of reasonable evidence, no matter how confined in compass. I see no other way, therefore, than to admit that Varro was mistaken in this instance — or else to throw out the whole theory.

 p80  The second objection is that of the confusion in the meaning of Subura, and the transfer of the name in historical times to the valley. This difficulty can be explained, apparently, in no other way than that which Wissowa has worked out; and here again, if one rejects the possibility of this explanation, he must perforce reject the whole theory.

Personally, I think that the weight of present evidence is strongly in favor of this view, and that it is the only working hypothesis as yet advanced which enables us to correlate the known facts of early Roman topography. This it does so well that there is not sufficient ground for refusing to accept it.

The history of the "Seven Hills" is therefore this: After the city had extended its limits beyond the Palatine, it included seven hills, or separate points of hills, the three of the Palatine, Palatium, Cermalus, and Velia, the three of the Esquiline, Oppius, Cispius and Fagutal, and the eastern or western point of the Caelius, Sucusa (Subura). A festival, the Septimontium, was established in honor of the existing city, celebrated by the montani separately. The same name was given to the city itself, probably before the institution of the festival.

After the Servian wall was built, in historical times, the old Septimontium was explained as referring to the seven hills now inclosed within this wall, viz.: Palatine, Capitoline, Aventine, Caelian, Esquiline, Viminal, and Quirinal. This became the canonical content of the term, recognized by the people in general and even by historians, the only exceptions being a few antiquarians like Labeo; although it seems that a dim idea was floating about that originally some other hills had been meant. But the septem montes of Latin literature were those just enumerated.

After Augustus had reorganized Rome and divided it into fourteen regions, of which the Transtiberine district formed one, and a very important one, it seemed manifestly improper to omit the Janiculum from the list of the seven hills, and therefore, as time went on, we find this hill substituted for one of the Servian, as, e.g., for the Capitoline in the note of Servius. In the earlier Middle Ages the Vaticanus also appears.

Western Reserve University.

Images with borders lead to more information. The thicker the border, the more information. (Details here.)
UP TO:		Antiquary's Shoebox	LacusCurtius	Home
A page or image on this site is in the public domain ONLY if its URL has a total of one *asterisk. If the URL has two **asterisks, the item is copyright someone else, and used by permission or fair use. If the URL has none the item is © Bill Thayer. See my copyright page for details and contact information.