The Notitia Dignitatum — JRS 10:131‑154 (1920)

§ 1. The document (or rather two documents) which has come down under the title Notitia Dignitatum is well known to all students who have concerned themselves, however incidentally, with the government of the Roman empire in the fourth and fifth centuries. It sometimes strikes one that it is referred to in a way that betrays a defective realization of what it was. Some of those who quote it appear to think that it was compiled for the purpose of giving information to the Roman public as to the organisation of the civil and military services and belongs to much the same class of work as e.g. the Synecdemus of Hierocles or the Notitia of Polemius Silvius. It was, of course, nothing of the kind, and it has been known since the days of Pancirolus precisely what it was; but even by some who were fully aware of its character and purpose deductions have been drawn which its character and purpose exclude. For students of Roman Britain it is particularly important to have a full grasp of the general questions connected with the Notitia, since it contains a great deal of the little evidence we have for the fortunes of that country at the beginning of the fifth century.

§ 2. The primicerius notariorum was one of the highest officials of the second class (i.e. those who had the rank of spectabilis). In A.D. 381 an imperial law elevated him above the vicarii and placed him in the same group as the proconsuls.1 The sphere of his duties is thus defined:

omnium dignitatum et amministrationum notitia tam militarium quam civilium.
scholas et numeros tractat.2

Claudian in his Epithalamium on the marriage of Palladius and Celerina describes his duties in the following verses:

p132 The first duty of the primicerius was to prepare and issue the codicilli (tabulae honorum), given to every higher official, illustris, spectabilis, or clarissimus, from a Praetorian Prefect or Master of Soldiers down to a provincial governor, on his appointment to his post. His second duty on which Claudian enlarges was one that in modern states devolves upon a ministry of war. He kept the complete survey of the distribution of the military forces in the provinces.

The procedure in the case of appointing the higher officials seems to have been as follows. When the emperor had made an appointment, the chartularii of the sacrum cubiculum prepared a formal brief to that effect and forwarded it to the primicerius notariorum.4 The primicerius (like the quaestor) had not a regular staff (officium); all that the discharge of his functions required was an assistant (adiutor) selected from the schola notariorum and a number of clerks taken from that body.5 In his bureau the codicil of installation was prepared. It contained the title of the functionary, an enumeration of the provinces, or military units or functionaries, as the case might be, which were under his control, the constitution of his staff if he had one, and also, painted in colours, the insignia of his office (for instance, in the case of a Master of Soldiers, the shields of all the military units under his command). For this purpose it was necessary for the primicerius to have a list of the offices he had to deal with, the details as to their functions, and model copies of the insignia which his clerks had to copy. All this was contained in the notitia dignitatum et amministrationum. The political insigne of the primicerius himself was the Laterculum Maius. It is not, I think, identical with the notitia, but rather a register containing the names of the occupants of the various posts. It was called Maius to distinguish it from the Laterculum Minus, which contained the list of the lesser military officers — tribunes, praepositi, prefects — and was managed by the scrinium memoriae under the control of the quaestor.6

The duplication of the central ministries, in consequence of the partition of the empire, involved two primicerii, one at Rome and one at Constantinople, as independent of each other as the two quaestors or the two masters of offices. The primicerius at Rome had no regular concern with the notitia at Constantinople, and interest he might take in it would be occasional or platonic. Changes might be made in arrangements as to dignities or offices in the east without his knowing anything about them. Hence, although originally the two notitiae were framed in unison and were, so far as possible, exactly identical in arrangement and phrasing (we do not know p133 whether the common form goes back to Constantine), changes independently made from year to year produced a number of divergencies.

§ 3. Our Notitia Dignitatum consists of two such notitiae, one of the east and one of the west. The circumstance that both these books were preserved together in the west raises a strong possibility that the lost Speier MS. which was still extant in the sixteenth century and was the parent of the existing MSS., was derived ultimately from originals at Rome (or Ravenna) belonging to the bureau of the primicerius of the west. The further conclusion suggests itself that the Notitia of the west is an actual working copy, used by the primicerius and his clerks, while that of the east is a clean copy kept only for reference; and this conclusion is borne out, as we shall see, by an examination of the documents.

So far as the text is concerned, the admirable edition of Mr. Seeck which appeared forty-five years ago is probably final unless such a very improbable thing were to occur as the discovery either of a MS. independent of the Spirensis or of an ancestor of the Spirensis. The recovery of the lost Spirensis itself would probably not throw much new light. Of the four apographs, which are independent of one another, Mr. Seeck remarks: omnes tam diligenter descripti sunt, ut plerumque etiam in minimis rebus conspirent, et ubi dissentiunt, consensus partis maioris, id quod raro alias evenit, fere pro tradita lectione habenda est. But we may hope that it may some day be found possible to issue an edition reproducing the coloured insignia. No one who has not seen the pictures can form an idea of what the original Notitia used by the primicerius notariorum was like, and any English student who has occasion to use these documents should look at the beautiful book in the Bodleian, one of the Canonici, which came from North Italy and was probably written by Pietro Donato (A.D. 1436).7

§ 4. It will be convenient to consider each Notitia separately before comparing them. Each is prefaced by an index of officials. The first thing to be noticed is that there are a number of inconsistencies between the Index and the sections of the book, and some among the sections themselves, chiefly regarding the order of precedence.

(1) In the Index the primicerius notariorum precedes the castrensis; in sections XVII, XVIII the order is the reverse.

(2) In the Index the provinces of Egypt are enumerated before those of Oriens; in II and XXII, XXIII the order is the reverse.

(3) In the Index the dux Moesiae II precedes, in XXXIX, XL he comes after, the dux Scythiae.

(4) There is the same difference in order between the duces of Dacia ripensis and Moesia I.

p134 (5) Galatia precedes Bithynia in Index and II, but in XXV Bithynia comes first.

(6) The order of the praesidial provinces of Pontica in Index differs from that in II, and both differ from that in XXV.

(8) The order of the praesidial provinces of Asia in Index agrees with that of II, but differs from that of XXIV.

(9) In the Index the Dacian provinces are six, in this order: Dacia mediterranea, Dacia ripensis, Moesia I, Praevalitana, Dardania, Macedonia salutaris. In III they are five, in this order; Dacia m., Dacia r., Moesia I, Dardania, Praeval. et pars Maced. sal. The other portion of Maced. sal. is noted III.13, as added to Epirus Nova.

(10) At the end of the Index, after the list of praesides are entered the two correctores, of Augustamnica and Paphlagonia. One would expect them to come before the praesides immediately after the consulares, because they were superior in rank to the praesides; and so they do in Not. Occ. In XXV.17, Paphlagonia has its proper place between consular and praesidial provinces; but on the other hand in XXIII.14, Augustamnica appears after the praesidial provinces. Further, in II both these provinces are entirely omitted. Further, no example is given of the insignia and officium of a corrector; one would have expected to find a section on the corrector Augustamnicae, for instance, between XLIII consularis Palaestinae and XLIV praeses Thebaidos (corresponding to XLIV in Not. Occ.).

§ 5. These discrepancies cannot all be explained as errors due to the negligence of copyists. Macedonia Salutaris (9) is a clear case of an original discrepancy, and so is the treatment of the correctors (10).

There had been correctores in the west since early in the fourth century; the introduction of the title in the east was much later. We first meet the corrector of Augustamnica in A.D. 393 (Cod. Th. 1.9.2; Jan. 12) and the corrector of Paphlagonia in A.D. 395 (ib. 2.8.22; July 3). Now it seems probable that the change in the status of the governor of Augustamnica was made at the time when the Egyptian provinces were reorganised and Arcadia was created. This happened later than Feb. 17, A.D. 396, as is shown by Cod. Th. 1.14.1, where there are still only three provinces in Egypt proper.8 We may therefore date the correctorial status of Augustamnica between A.D. 386 and 392, and conjecture that the governor of Paphlagonia received the same status about the same time.

When these changes were made, what the clerks of the primicerius seem to have done was this. They inserted the two new correctors at the end of the Index, intending that when a new Index should p135 come to be written out, this addition should be transferred to its proper place between the consulares and the praesides. They inserted Arcadia among the Egyptian provinces subject to the Praet. Praef. Or. in II, and also in the provinces under the Praefectus Augustalis in XXIII. They ought to have prepared a new leaf showing the insignia and officium of a corrector, between XLIII and XLIV. But they omitted to do so, as it was not absolutely necessary because the insignia of a corrector were practically,9 and the officium absolutely, identical with those of a praeses.

Here then the Index reveals that a change had been made since it was originally drafted. In the same way it preserves the order of the six Dacian provinces as they were before they were reduced to five by the division of Macedonia Salutaris. Now in the Laterculus of Polemius Silvius, which (though compiled in A.D. 449) represents, so far as the Orient is concerned, conditions prevailing in the reign of Theodosius I,10 there is only one Macedonia. It may therefore be conjectured that Macedonia Salutaris was formed when Eutropius was guiding the eastern government (A.D. 396‑9). For Claudian holds up to reprobation his policy of dividing provinces: provincia quaeque superstes dividitur.a The Index preserves the transient arrangement introduced by Eutropius. Some time after his death Macedonia Salutaris was abolished, and the necessary change was made in the body of the Notitia, but the Index was left unaltered.

Galatia is probably another instance of the policy of Eutropius; only in this case it was not reversed. There is only one Galatia in Polemius Silvius,11 whereas the Notitia has two, consular Galatia and Galatia Salutaris. Perhaps the discrepancy, noted above under (8), in the order of Galatia (cons.) and Bithynia may be due to an actual change in favour of Bithynia introduced at the same time.

These discrepancies not only throw light on the history of administrative changes, but also serve to indicate how the Notitia was kept by the laterculenses of the primicerius. Changes as they occurred were noted marginally or interlineally in the book. But a correction which ought to have been made in two or in three places was sometimes through carelessness made only in one or two. The Index was presumably of little importance for practical use and therefore the clerks would be rather negligent in keeping it up to date. In some sections so many changes might be made in the course of a few years (for instance, in the troops under the military commanders) that the p136 accumulation of corrections necessitated the preparation of a clean copy, but this would not imply the re-copying of the whole book but only of those particular sections. Therefore a great part of the Notitia used by the bureau, say in A.D. 410, might have been actually written twenty or thirty years before and remained unaltered.

§ 6. A first prior limit for the date of the Not. Or. is given by the fact that the Prefecture of Illyricum has been transferred from the west to the east. This transference occurred just after the accession of Theodosius I in A.D. 378. This limit is brought down further by the appearance of the provinces Arcadia and Honorias, and of the military units Arcadiaci, Honoriaci, etc., named after that emperor's sons. There could be no Honoriaci before A.D. 393, when Honorius was created Augustus. But the numerous Theodosiaci and Theodosiani among the troops which are mentioned in Not. Or. seem to point to a still later time. For the fact that no units of Theodosian name appear in Not. Occ. suggests that these troops were not all formed and named by Theodosius I,12 but that some of them are named after Theodosius II, having been formed either by Arcadius and named in his honour between A.D. 402 and 408, or more probably during his own reign after A.D. 408. But there is another entry which carries the date definitely further and has escaped the notice of commentators. In XVII.8, we find in the officium of the castrensis tabularium dominarum Augustarum. Flaccilla the first wife of Theodosius I was an Augusta, Eudoxia the wife of Arcadius was an Augustus; but there were not two Augustae together till the days of Pulcheria and Eudocia. Eudocia was created Augusta in Jan. A.D. 423, so that this year is a prior limit for the composition of our Notitia Orientis. We shall find other confirmation of that limit presently. Mr. Seeck's statement that 'lässt sich im Orient keine Notiz nachweisen welche nach dem Jahre 397 fallen müsste, ja selbst was sich als Randsatz erkennen lässt,' is erroneous.13

§ 7. The Notitia Occidentis also exhibits a series of discrepancies between Index and sections, similar to that in the Not. Or.

(1) The primicerius notariorum precedes the castrensis in Index, and follows him in the section-arrangement, similar to that in the Not. Or.

(2) The vicarius Italiae who appears in Index is entirely omitted in the body of the book.

(3) The comes Britanniarum precedes the comes litoris Saxonici in Index and in V; but the order is reversed in XXVIII, XXIX.

(4) The dux Pannoniae I precedes the duces of Pannonia II and Valeria in Index, but comes after them in XXXII‑XXXIV.

(5) The consularis Campaniae comes third among the consulares subject to the vicarius urbis Romae in Index and II, but first in XX.

p137 (6) The order of the praesides under the same vicarius is Samnium, Valeria, Sardinia, Corsica in Index and II, but Samnium, Sardinia, Corsica, Valeria in XIX.

The fact that the same discrepancy between the Index and the order of sections occurs in regard to the castrensis both in Not. Or. and in Not. Occ. will call for remark hereafter, but it may be pointed out that the improbability that two clerks, one at Rome and one at Constantinople, should have independently made the same slip in copying confirms the conclusion that these discrepancies cannot be put down to mere accidental error. This Index, too, undesignedly records some changes in the provincial administration, particularly in the relative importance of the Danubian provinces. On the other hand, the absence of a section on the vicar of Italy can only be explained as an error in transmission. There is no place for it in Mr. Seeck's probable reconstruction of the quaternions of the Codex Spirensis, so that it must have been lost in an ancestor of that manuscript.

§ 8. A prior limit of date for Not. Occ. is given at once by the Gildoniacum patrimonium in XII.5, which cannot be earlier than A.D. 398‑9, after the defeat of Gildo's rebellion. We have a later limit in XV.9.14, tabularium dominae Augustae, which cannot be earlier than A.D. 421, when Galla Placidia became Augusta; and yet a later one in VII.36, where a military unit, whether legionary or auxiliary, Placidi Valentinianici felices, is named which cannot have been formed before A.D. 425‑6 when Placidus Valentinianus was restored to his inheritance by the armies of Theodosius II and created Augustus (Oct. A.D. 425).14 To this time (A.D. 425‑6) the Notitia was assigned by Mommsen. Strictly, however, it is only a prior limit (see below, § 11).

§ 9. From the most cursory comparison of the two Notitiae it is evident that they have a command form and were drawn up on the same model; so evident that it has been a common and pardonable mistake that they were two parts of one document, a Notitia of the whole empire. They correspond in the order of the contents, in the arrangement of the indexes, in the arrangement of the entries under each ministry or command or governorship, and for the most part verbally. This correspondence is so close that it could not have been merely the reflexion of the similarity of the organisations of the two divisions of the empire. It can only be explained by supposing that when notitiae were originally drawn up for the use of the primicerii the same scheme was deliberately adopted for both.

p138 (1) In Not. Or. the number of evectiones to which each official had a right is noted at the end of each section by a numeral, but not in Not. Occ.

(2) Not. Occ. has (a) a section (VII) on the distribution of the numeri of the field army, and (b) a section (XLII) on the praepositurae and praefecturae of fleets, laeti, and gentiles subject to the Master of Soldiers. There are no corresponding sections in Not. Or.

(3) The details as to the officials (rationales, procuratores, praepositi) subordinate to the comes s. largitionum and to the comes rerum privatarum are much fuller in the Not. Occ. than in Not. Or.

(4) It was observed above that no section for a corrector is given in Not. Or.; but in Not. Occ. XLIV the corrector Apuliae et Calabriae is given as an example.

(5) At the end of Not. Or. there are two pictures showing the exteriors of books such as are also shown in the insignia of the officials in both notitiae; there are no corresponding pictures at the end of Not. Occ.15

To these might be added the circumstance that in certain sections in Not. Or., relating to the military commands of limes Aegypti, Thebais, Phoenice, Syria, Palestine, Osrhoene, Mesopotamia, Arabia, Armenia, and Moesia II, troops are enumerated quae de minore laterculo emittuntur, and that no such troops occur in the Not. Occ. nor is there any mention there of the laterculum minus. But this difference is evidently due to a difference in organisation. In the east certain military units were excepted from the numeri with which the primicerius notariorum dealt, and placed on the laterculum minus, whereas there appears to have been no such units in the west. This difference does not concern the character of the Notitiae or the method of the primicerii.

In regard to (1), as the primicerius had nothing to do with the management of the cursus publicus, the only possible purpose of noting the evectiones in the Notitia was to enter them in the codicils which he issued to the officials. Their absence in Not. Occ. therefore suggests that if the practice was an old one, it was continued in the east, but was for some reason abandoned in the west.

The second difference (2) points to a different method in the eastern and western bureaux. The primicerius of the west kept his files showing the distribution of the military forces in his notitia of dignities, whereas the primicerius of the east kept his files of p139 numeri quite separately, as belonging to another department of his work.

As to (4), enough has been already said. The last difference (5) is interesting. In each of the pictures eighteen books are shown inside a square frame with a pediment at the top and five vignettes of allegorical females, one at the top and the others at the four corners. The frames apparently represent armaria. These pictures are clearly ornamental, and illustrate the general difference between the two Notitiae. The Not. Occ. was the working copy in use at the bureau in Italy; the Not. Or. was a clean copy16 made in the bureau of Constantinople for the special purpose of transmission to the primicerius in Italy, and some care would have been taken to make it a presentable volume. This hypothesis may account for the other difference between the two documents that was noted above (3), the curtailment in Not. Or. of the details as to the subordinates of the two finance ministers. As the copy was not wanted for practical use, the copyist may have considered it permissible to summarise. That an interchange of copies of the Not. Or. and Not. Occ. is probable about A.D. 426 or soon afterwards will appear in connexion with other considerations to which I may now pass.

§ 10. The first far-reaching change in organisation that affected both Notitiae alike and required a considerable revision of their arrangement was the transference of eastern Illyricum from the government of Rome to that of Constantinople in A.D. 378‑9. For this purpose we may assume it to be probable that the copies of the two Notitiae were exchanged between the two primicerii. We cannot point to any other occasion in the second half of the fourth century where such an exchange was necessary. The two divisions of the empire did not preserve entirely unaltered the precise similarity of the original schemes, but only in two respects were there any serious divergencies. One of these was in the organisation of the higher military commands, which was due to the policy of Theodosius I; the other was in the arrangements of financial administration, in which different circumstances dictated improvements and alterations that could not be parallel.

Now in our Notitiae the order of rank of all the dignitaries is identical, although we know that certain changes had been made in that order since A.D. 379. This implies collusion; it implies that these changes were made by common agreement or that where one government led the way the other followed. Three such cases are known to us.

p140 The first is that of the praepositus sacri cubiculi, whose section in both Notitiae is lost. Under Theodosius the Great, the praepositus belonged to the first class of dignitaries, as is shown by a law of A.D. 384.17 It has been supposed that it was the influence of Eutropius, the powerful Grand Chamberlain who governed Arcadius for three and a half years (A.D. 396‑9), that raised his own office to rank (as he ranks in both Notitiae) immediately after the Prefects and Masters of Soldiers in order of precedence. There is, however, no proof of this, and if it were true, it was a lead which would certainly not have been followed in the west, but bitterly censured. On the contrary, we find in A.D. 414 that the praepositus ranks below the comes s. largitionum at Constantinople.18 It was not till A.D. 422 that Theodosius II raised the Grand Chamberlain to the higher place, and he explained that this promotion was prompted by the deserts of the praepositus Macrobius: nos ad hanc promulgationem Macrobi viri inl. merito provocarunt.19

The second case is that of the Master of Offices. In the reign of Theodosius I and before, the Quaestor had precedence of the Master of Offices,20 and the same order prevailed in A.D. 438 when the Codex Theodosianus was issued and probably for a year or two before,21 and afterwards this order was permanent.22 But in both our Notitiae the Master of Offices comes before the Quaestor. This reversal of order must have been therefore temporary, and it is easy to divine that the change was made for the sake of a particular individual, as in the case of the praepositus Macrobius. We may therefore conjecture as highly probable that the promotion of the Master of Offices to take rank above the Quaestor was a recognition of the merits of Helio and is to be referred to much the same time as the promotion of the praepositus. Helio had an exceptionally long tenure of the post. He held it before the end of A.D. 414 — how long we do not know, but his last known predecessor was Aemilianus in A.D. 405. He can be traced through the following years till the summer of A.D. 427;23 in A.D. 426 he is a Patrician. The confidence which the emperor placed in him is shown by the fact that he was deputed to take the place of Theodosius in the ceremony of creating Valentinian a Caesar at Thessalonica, to accompany Placidia and her son to Italy, and finally, as Theodosius was unable to travel to Rome in the autumn of 425, to represent him at the coronation of Valentinian as Augustus. Here, then, I have no doubt, the Notitiae contain the record of a temporary measure for the benefit of this distinguished and trusted minister.

These two cases have been commented on before. Mommsen p141 recognised the temporary character of the higher rank of the Master of Offices, but did not account for it. On the third case which has escaped attention I have already touched. It is that of the castrensis and primicerius notariorum. That a change was made in their relative positions in the scale of precedence is deducible from the discrepancy between the Indexes where the primicerius comes first, and the order of the sections where the castrensis precedes. A possible reason has occurred to me. The tyrant John, who seized power after the death of Honorius was put to death by Placidia in the summer of A.D. 425, was primicerius notariorum in Rome at the time of his elevation; and it is therefore conceivable that the slight degradation of the office which John had held was resolved on for the purpose of indicating that the office had been disgraced. This seems worth mentioning as a conjecture, but it is more probable that the change recognised the services of some unknown castrensis.

Whatever may be the date and circumstances of the third change, the fact that all three changes were adopted in the west as well as in the east shows that the two governments acted together in matters of this kind some time after A.D. 422, and therefore obviously not before A.D. 425‑6 when Theodosius II, having restored Placidia and her son, was taking an active interest in western affairs and was adopting a policy of restoring the real unity of the empire and promoting close co-operation between east and west. The principal expression of the new unity was to be the Codex Theodosianus which was already being taken in hand. The assimilation so far as possible between the two civil services was to be maintained, and the higher rank which had been bestowed on the Praepositus sacri cubiculi and the Magister officiorum of the east was imposed on the west.

All this is quite in accordance with the conclusion that our Notitia Orientis is derived from a copy sent from Constantinople to Italy in or soon after A.D. 426, and was therefore at that time up to date.

§ 11. The prior limit for the Not. Occ. can be fixed to a year later. This follows from Mr. Seeck's sagacious exploration of the sections on the Danubian provinces.24 The two 9n Pannonian provinces, Pannonia II and Valeria, had been for a long time in the power of the Huns. They were recovered in A.D. 427,25 and a reorganisation ensued which involved the building and restoring of fortresses26 and the establishment of garrisons. This work is reflected in sections XXXII º (dux Pannoniae II) and XXXIII (dux Valeriae) of Not. Occ. For the details I need merely refer to Mr. Seeck's article. The earliest year possible for our copy of Not. Occ. is thus A.D. 427‑8.27

p142 A posterior limit is given (1) by the fact that Dalmatia still belongs to the west. It was probably transferred to the east on the occasion of the imperial marriage in A.D. 437; it was the consideration Placidia had to promise her nephew Theodosius for restoring and protecting her son, but apparently she was able to bargain that the price should not be paid till the marriage arranged between their children Valentinian and Licinia Eudoxia was actually celebrated. (2) By the treaty between Gaiseric and Valentinian in A.D. 435 the Mauretanian provinces passed to the Vandals, and this meant that the dux et praeses Mauretaniae Caesariensis and the praeses Mauretaniae Sitifensis ceased to function. These provinces might, of course, though surrendered by treaty, have been retained in the list, as official optimism would have counted on their eventual recovery; so that their appearance in our Not. Occ. is not demonstrative like the appearance of Dalmatia. But (3) there is a piece of evidence which suggests that the posterior limit is very close to the prior limit at which we arrived. In the Pannonian sections (XXXIII.59‑65 and XXXIV.29, 30, 44‑46) we find a number of unnamed cohorts and one unnamed legion; the units apparently had been formed but had not yet received their titles, or else it had been decided that so many units should be sent, but they had not yet been chosen. This suggests the inference that the copy was being made just at the time when Pannonia was in the process of being organised, and that would have been in A.D. 428.

§ 12. In this paper I do not propose to discuss the insignia, connected with which there are several difficulties that have still to be cleared up. It does not seem very likely that a minute study of them will yield information bearing on the dates of the documents. But attention may be called to one feature. In the insignia of the dignitaries of illustrious rank there is a tabula or frame resting on a table with a head portrayed in the centre. This also appears in insignia of two spectabiles,28 the comes orientis and the praefectus augustalis. In Not. Or. the head is depicted in every case except that of the comes r. priv., where the central space is blank; but in the case of the first magister mil. praesentalis there are two heads. In Not. Occ. the space is blank for the quaestor, the comes s. larg., and the comites domesticorum; but the magister officiorum has two heads. The faces are all different. All are smooth-shaven, except in the case of the magister mil. per orientem, where it is bearded. Some are full-faced, others side-faced. The heads are uncovered and p143 unadorned, except in the cases of the second magister mil. praesentalis in the east, where a cap is worn, and of the magister mil. per Illyricum, where a young face with long hair is adorned with a diadem. That these heads represent the emperor was assumed by Pancirolus, and the one instance of a diadem, and the two instances of a double head, confirm the assumption. But a methodical investigation, based on a new examination of the MSS., is obviously needed.

§ 13. It was stated above that the original of our Not. Occ. was an actual working copy used in the bureau of the primicerius. The evidence for this may now be considered. We find a number of corrections as to the stations of troops. The fort to which a unit was transferred was added in the margin, and the name of the fort in which it had been stationed was not erased, and so both names appear in the text; sometimes the new station is distinguished by a nunc. These notes of change occur principally in the sections on the Danubian provinces, XXXIII‑XXXV, and have been elucidated by Mr. Seeck.

But the most important evidence consists in the differences between section VII and sections V and VI. Section VII gives the list of the field forces, both foot and horse, and shows their distribution. All these forces were under the supreme command of the two Masters of Soldiers, whose headquarters were in Italy; and therefore if the Notitia, as we have it, had undergone no changes after it was drawn up c. A.D. 428, the military units enumerated in V as under the command of the Magister peditum and in VI as under the command of the Magister equitum should correspond exactly to the units enumerated in VII. But this is not the case. A comparison is made easy by the cross-references in Mr. Seeck's edition.

A. In V there are five units (183, 207, 217, 261, 262) and in VI two (75, 85) which do not appear in VII.

The distributio numerorum, therefore, includes changes and new formations which were made after the lists in V and VI had been drawn up; these alterations were not entered under V and VI. It also records some important innovations in the military organisation.

p144 The duces and the comites (i.e. comites et duces) who defended the frontier provinces both in the east and in the west were in command of limitanei, not of comitatenses; and the units of the limitanei are given in the Notitiae with their respective stations, in the section of each dux. But in the fifth century a new class of comites appears in the west. The comes Italiae, the comes Argentoratensis, and the comes Britanniarum, each of whom has his section in Not. Occ., were not like the old duces commanders of limitanei; they commanded mobile forces which could bring relief to any threatened point. There is no sign, and no reason to suppose, that they existed in the fourth century. The geographical sphere of the comes Italiae was the tractus Italiae circa Alpes (Not. Occ. XXIV); that is, he had to look to the defence of the passes in northern Italy,31 in case the advanced posts in Raetia and Noricum were carried by an enemy. Since the invasions of Alaric and Radagaisus the protection of the peninsula was a pressing problem, and the institution of the comes Italiae may have been a part of the government's scheme for solving it. The comes Argentoratensis had his field of operations in the tractus Argentoratensis (Alsace). The defence of Gaul was also a problem of which the terms had been changed since the great barbarian invasion of A.D. 407. After the suppression of the Gallic tyrants, first Constantine, and then Jovinus, the general Constantius had to address himself to the work of organising the military defences of the empire. It appears to me that the creation of these new permanent commanders of field forces, including the comes Britanniarum (who will be more particularly considered later), may probably be ascribed to him.32 We know that part of that statesman's work, in the second decade of the fifth century, was the restoration of order in northwestern Gaul. We have no knowledge when the dux tractus Armoricani was first created, but we may conjecture that at this time his sphere was extended along the Aquitanian coast in order to protect it against Saxon pirates. The extent of the command is specially noticed in XXXVII.24‑29. At all events the three comites in question were in being before A.D. 428 (Not. Occ. XXIV, XXVII, XXIX).

At the same time other commanders of field forces had been appointed during the reign of Honorius to meet the emergencies of the time, but without the intention of making these commands permanent institutions. In A.D. 408 Generidus was entrusted with a command including the diocese of Illyricum (Dalmatia and Pannonia) and the transalpine provinces of the diocese of Italy (Noricum, Raetia), with the title apparently of Master of Soldiers.33 And in A.D. 419 we find Asterius commanding in Spain, as Hispaniarum p145 comes. Both the commands may have continued (perhaps intermittently) till A.D. 428, but it was not contemplated to make them permanent parts of the military organisation (and the persons appointed might have either the title of comes or of magister militum, just as in the fourth century when we find Aequitius as comes given the command in Illyricum, and afterwards receiving the higher title of Master of Soldiers),34 for in that year, unlike the comes Britanniarum etc., they have no separate sections, with insignia and officia, in the Not. Occ. and are mentioned only in VII.

It appears to be generally assumed that the post of Magister equitum per Gallias was a regular and standing institution before the end of the fourth century. There is no good evidence for this assumption. The Jacobus magister equitum to whom Claudian addressed a poem35 in the late months of A.D. 401 was evidently the magister equitum praesentalis,36 who was left to protect north Italy while Stilicho was absent in Raetia fighting with Radagaisus. Nor is there any reason to suppose that it was not the same post which was held by Gaudentius the father of Aetius,37 although he happened to have been killed in Gaul,38 and in A.D. 423 by Crispinus.39

The first Master of Soldiers mentioned in literary sources who was almost certainly mag. eq. per Gall. was Aetius, who was raised to this rank in A.D. 429,40 and the post was evidently created to satisfy that ambitious commander, who had been defending Gaul during the past few years and was feared by Placidia and Felix the magister utriusque militiae whom he succeeded not without violence in the following year.

Now this evidence happens to agree remarkably with the indications of Not. Occ. If the magisterium equitum per Gallias was an institution of long standing, it is impossible to explain the fact that the sections of the two magistri praesentales are not followed by a section for the vir inlustris magister equitum per Gallias with his insignia and officium, and the troops under his command. What we do find is the troops of the Gallic mag. eq. enumerated in section VII, like those of the comes Hispaniarum and the comes Illyrici. But there is a difference. In the middle of this section, where it has no business to be, we find the officium of the mag. eq. p146 per G. It is easy to conjecture what happened. It was decided at this time (A.D. 428‑9) to establish a permanent commander of the field forces in Gaul, with the highest rank. His officium was duly drawn up in the bureau of the primicerius and was provisionally inserted after the troops under his command, but the preparation of the insignia was postponed and never executed so long as this particular book was in use. Thus the negligence of the clerks of the primicerius enables us to detect the date of the creation of this post.

§ 14. If we examine the distribution of the field forces, as given in VII, we find that it corresponds to the situation of the empire in the years 428‑437. The numbers are as follows:41

The growth of the power of the Huns, which was a constant menace on the middle Danube frontier, is sufficient to account for the presence of a considerable field army in Illyricum under the newly-appointed commander, the comes Illyrici. In Gaul a large army was now needed, not only to reinforce the limitanei of Germania I and Belgica II (where the Salian Franks under Clodio were seeking to extend their territory), and Armorica, at any point, but also in the interior where continuous vigilance was required, and fighting was almost incessant, to keep in check the encroachments of Visigoths and Burgundians. Ever since its invasion by the Vandals and other barbarians in A.D. 409 we might expect to find armies in Spain which till then had been an un-military diocese. From A.D. 428 onwards, when the Vandals were threatening Tingitania, then conquering and occupying the Mauretanias, field armies were naturally despatched to Africa, to serve under the comes Tingitaniae and the comes Africae.

This army list has been generally interpreted by historians as supplying figures for the military strength of the empire at the end of the fourth century. What it really supplies is data for estimating the size of the armies of Valentinian III at the end of the third decade of the fifth century. The total number appears to be just 132,000.

§ 15. The Notitia of A.D. 428 represents Britain as still a diocese of the empire, under the civil government of the vicarius, with its five provinces under two consulars and three praesides, and still defended by Roman troops, (1) limitanei under (a) the count of the p147 Saxon shore, in the south-east, (b) the duke of the Britains, in the north, and (2) a field army under the count of the Britains. It has been usual in recent years to reject this evidence on the ground that it is incompatible with the literary evidence which has been held to prove that Britain had been entirely abandoned by the Roman government twenty years earlier. Mommsen maintained that the picture of the Britannic armies presented in Not. Occ. corresponds to the pre-Constantinian organisation and would have been actual in a Notitia of A.D. 300, but was quite obsolete in a Notitia of A.D. 400 or later;42 and concluded that whoever compiled Not. Occ., having no information relating to contemporary Britain which had passed out of the control of Rome but was still theoretically recognised as a part of the empire, had recourse to an ancient list more than a hundred years old in order to fill in the military units. This view has been very widely accepted, and is certainly erroneous.

The argument on which it is based does not apply to all the Britannic sections. It does not apply to XXVIII, the section on the Saxon shore. The Saxons, who had begun to raid the British coasts before the end of the third century, did not become a serious menace before the middle of the fourth, and of the forces which are enumerated here only two units, the old legio II Augusta, which had been in Britain since its conquest, and the cohors I Baetasiorum, are found in inscriptions of the second or third century. Nor does it apply to the field forces of the comes Britanniarum, under whose command we find a vexillation of equites Honoriani (VII.202), which cannot have existed before the end of the fourth century. Nor yet to the limitanei in Yorkshire, all of which, with the exception of a detachment of legio VI victrix (XL.18), are so far as we know later than the third century. In fact, Mommsen's argument, if it were valid, would apply solely to the troops per lineam valli in XL.

There, 23 stations and 23 units are named, and all the units with three exceptions can be shown to have been in Britain, by the evidence of inscriptions, since the second or third century.43 The three exceptions for which there is no epigraphic evidence are 34 cohors I Cornoviorum at Newcastle, which seems to have been a cohort of native British; 53 ala I Herculea at Olenacum, whose name testifies that it was formed in the reign of Diocletian and Maximian, and 47 Mauri Aureliani, whose name points to their creation by Aurelian. The general military policy had been to keep the same troop formations there from the time of Hadrian onwards, and there is no evidence to show that this policy was radically changed by p148 Constantine or that the old formations which defended the Wall were broken up so long as the Wall continued to be defended at all.

That a primicerius notariorum of the fifth century unearthed documents of the beginning of the fourth century, lists which had long ago ceased to have any actuality, and inserted them in a register which was intended purely for use in his office, is a theory which is not very plausible. In fact it is incredible. It is a procedure which might have been adopted by Hierocles or George of Cyprus; but the interest of the primicerius in his Notitia was not antiquarian. And it is quite from the purpose to suggest, as has been suggested, that the object was to disguise the loss of Britain. Disguise it from whom? From the notarii?

Archaeological evidence — that is, inscriptions and coins — does not enable us to determine the date at which the Wall was abandoned as a military frontier. Towards the end of the fourth century, inscriptions in all parts of Britain (and not only in Britain) are rarer, and at the same time the issue of copper from western mints is very scanty. We have to take this into account in considering the import of the fact that, so far, the camps and mile-castles of the Wall furnish no evidence of occupation after the rebellion of Maximus.44 The legitimate inference is merely negative, namely, that the abandonment of the Wall at that time is compatible with the archaeological facts, but a late date is not excluded. In other words, while archaeology does not support the view that the Wall was held after A.D. 395, 'it gives equally little support,' as Mr. Haverfield observed in reviewing an able article of Mr. Carter, 'to the theory that it was not held after 395.'45

§ 16. Leaving aside for the moment the linea valli, the evidence in the Notitia is quite clear that Britain as a whole was still held by the empire in A.D. 428. Apart from the Wall, systems of military defence had been organised in the south-east from the Wash to Portsmouth under the comes litoris Saxonici, and on the east from the Tees to the Humber or the Wash under the dux Britanniarum; and the defence was further provided for by mobile forces under the comes Britanniarum. These forces consisted soon after A.D. 428 of three infantry units, the Victores iuniores Britanniciani (probably auxilia), the Primani iuniores and the Secundani iuniores46 (both legions), which seem to have been new formations, and six vexillations of cavalry of which one at least, the equites Honoriani seniores, was in existence before A.D. 424, and at least three (eq. Catafractarii iun., eq. scutarii Aureliaci, and eq. Syri) were later than A.D. 428, while it is uncertain whether the eq. stablesiani and the eq. Taifali were old or new.

p149 Such being the military establishment in Britain at the time when the patrician Felix was the head of the armies of the west, when Aetius was defending Gaul, and Gaiseric was conquering Africa, the existence of the civil government is a matter of course — the vicar at the head of the diocese, the two consulares governing Valentia and Flavia, and the three praesides governing Britannia I and II and Maxima.

§ 17. Inferences, however, of this kind from the Notitia are held to be quite inconsistent with the literary evidence. Even Mr. Haverfield, who is more cautious in his statements than other writers as to the end of Roman rule in Britain, says, for instance: 'It is agreed that these chapters (Not. Occ. XXVIII and XL) do not exhibit the garrison of Britain at the moment when the Notitia was substantially completed, about A.D. 425, for the good reason that there was then no garrison left in the island.'47 The literary evidence is notoriously scanty and fragmentary, and it does not appear to me to lead to the conclusion which is generally accepted as unquestionable, that Britain was finally abandoned in the first decade of the fifth century.

When Italy was menaced by the invasion of the Visigoths in A.D. 401‑2 troops were moved from Gaul and Britain to reinforce the army in Italy. The Britannic contingent is described by Claudian as legio praetenta Britannis quae Scotto dat frena truci.48 Mr. Haverfield observed that this legio 'is not necessarily legionary,'49 as legio might be used poetically for any unit of foot soldiers; but, in the absence of positive evidence to the contrary, the presumption is that a legion, not a cohort, was meant. It was evidently stationed somewhere in the west, as it guarded against invaders from Ireland. We are not told whether Stilicho sent it back to Britain after the battle of Verona in A.D. 403. Probably not; since the menace from Alaric was not over, and, if it was not sent back in A.D. 404, it is not likely that it was sent back by Stilicho, for then came the invasion of Radagaisus, A.D. 405‑6, and in A.D. 406 the Britannic armies revolted. It would have been bad policy to restore to Britain at that moment troops which were associated with it and would have been likely to make common cause with the rebels. In A.D. 407 the tyrant Constantine, following the example of Maximus, crossed over to Gaul. As to the size of the army which accompanied him we have no information. Modern historians50 have generally assumed that all the Roman troops p150 in Britain crossed the channel and never recrossed it. The assumption that he entirely denuded the island of military forces and left it defenceless against Saxons, Picts, and Scots, is merely an assumption and an improbable one. There is no likelihood that he proposed to make an empire for himself which did not include Britain or that he was ready to let the island slip from his grasp. Before he left its shores he must have sent agents to sound some of the regiments in Gaul and prepare the path for his occupation of the Gallic provinces; the invasion of the barbarians from beyond the Rhine who in A.D. 407 were devastating north-eastern Gaul facilitated and appeared to justify his enterprise; he might have gone without taking all the Britannic troops. The probability would seem to be that, leaving some troops in Britain on the main lines of defence, he was at first able to maintain his authority in the island from his headquarters in Gaul.

But it is easy to understand that as his difficulties in Gaul increased and his plans extended into Spain he could spare little attention to Britain, and in A.D. 408‑9 the power of his government there was shaken by a rebellion. The notice we possess of this movement comes probably from the contemporary Greek historian Olympiodorus, who was unusually well-informed about affairs in the west, and is preserved by Zosimus,51 who states that invasions of the barbarians beyond the Rhine forced the inhabitants of Britain and some of the Celts in Gaul to revolt:

τῆς Ῥωμαίων ἀρχῆς ἀποστῆναι καὶ καθ’ ἑαυτὰ βιοτεύειν οὐκέτι τοῖς τούτων ὑπακούοντα νόμοις, οἵ τε οὖν ἐκ τῆς Βρεταννίας ὅπλα ἐνδύντες καὶ σφῶν αὐτῶν προκινδυνεύσαντες ἠλευθέρωσαν τῶν έπικειμένων βαρβάρων τὰς πόλεις.

and that Armorica and other provinces of Gaul imitated the Britons expelling the Roman governors. The barbarians in this context must refer to the Saxons, and the inference evidently is that as the troops in Britain were insufficient against the Saxon raiders, the people, in some parts of the island at all events, disregarding the Roman authorities, took the defence and administration into their own hands. But the movement was a revolt against Constantine, and the notice is important as showing that Imperial authority, that is, his authority, was still represented in Britain after he had left the island.

News of what had happened reached Ravenna, and the government there took advantage of it in the interest of the legitimate emperor. In the course of A.D. 409 Honorius despatched an emissary to Britain authorising the cities to take measures for their own defence,52 and p151 thus legalising their rebellion against the authorities which represented Constantine. It was an interim measure evidently, designed to further the eventual re-establishment of his own rule in Britain when he had overcome the usurper.

In view of this concern of Honorius for Britain, there is no a priori reason for supposing that, when the tyrants had been suppressed in Gaul, and it was possible for the energetic Master of Soldiers, Constantius, in whose hands the conduct of the empire rested, to reorganise the defence of western Europe, Britain was not included in the reorganisation. The evidence of the Notitia supplies the proof that it was. I hazarded the conjecture (above, § 13) that Constantius introduced the new policy of permanent commanders of field forces in some regions hitherto defended only by limitanei, the comes Italiae, the comes Argentoratensis, and the comes Britanniarum.

Whether it was at this time, or earlier in the time of Stilicho as Mr. Craster believes, that the defences of Yorkshire under the dux Britanniarum were organised, cannot, I think, be determined with certainty. For the whole reign of Honorius there are only two inscriptions that give any information as to Roman activity. Hon. Aug. on tiles used to repair the fort of Anderida may be early or late in the reign, but the inscription recording that the signal station on the summit of Peak on the Yorkshire coast was built by Justinian is presumably to be placed in the first years of the century, inasmuch as Justinian was the name of an officer who accompanied Constantine to Gaul.53 This, so far as it goes, is in favour of Mr. Craster's view.

But if Stilicho was the author of the defensive system in Yorkshire,54 that system still existed in A.D. 428. There is no reason for suspecting the evidence of the Notitia. The only possible stumbling-block is the first entry under the disposition of the duke, praefectus legionis sextae, where as Mr. Seeck pointed out, the words victricis, Eburaci have fallen out of the text. Now Mr. Craster thinks that the sixth legion was that which went to Italy in A.D. 402.55 I am inclined to agree with him, and I am also inclined to think that it never returned. But it would not follow that the Notitia is here reproducing an item from a list which was obsolete after A.D. 402. For the sixth legion p152 might be simultaneously in Italy and in Britain. Sexta victrix had been in Britain since the reign of Hadrian; it was one of the old pre-Constantinian legions, which after Constantine's reform of the army had the capacity of being in several places at once. Take Quinta Macedonica. It was at the same time in four different stations in Dacia Ripensis (Not. Or. XLII.31‑33, 39) and also at Memphis (ib. XXVIII.14). The old legions of 6,000 were broken up into detachments of 1,000, but each detachment retained the name of the legion. Sexta victrix may therefore have left a station in the west of Britain in A.D. 402, and left it for ever, and yet Sexta victrix may have been the garrison of York in A.D. 428.

The same consideration applies to Secunda augusta. It has been observed by Mr. Craster that it accompanied Constantine to Gaul, because we find it there later; it remained there and became a legio comitatensis.56 But this is quite compatible with the simultaneous existence of Secunda augusta as a legio limitanea in Britain. The legion had been in the island since the first conquest, and at many places, among others Caerleon; there was still a detachment of it at Richborough in A.D. 428, serving under the Count of the Saxon Shore, and probably another under the comes Britanniae.57

§ 18. The certainty that Britain remained under Roman rule into the thirties of the fifth century under the regency of the Augusta Placidia may make us hesitate to leap hastily to the conclusion that the sub-section of XL, per lineam valli, was obsolete in A.D. 428. Archaeological discoveries may yet make it certain or probable that the line of the Wall was abandoned in the reign of Honorius; and in that case the sub-section in question must have been copied from the immediately preceding Notitia by an error, and represent the actual organisation of the defence of the Wall at the time the Wall was abandoned. There is an internal argument in favour of this view. The castella per lineam valli do not appear in the insignia of the dux Britanniarum; his insignia consist entirely of the fourteen castella in or around Yorkshire. This consideration, however, is not decisive, because in the Danubian provinces the numbers of the castella in the insignia of the dukes is considerably smaller than the number in the text. But it may establish a presumption; for just after the Notitia was drawn up the Danubian provinces, as we saw, were in the process of reorganisation, and in the case of all the other dukes the numbers of the castella pictured in the insignia exactly correspond to the numbers recorded in the text. There is therefore, apart from archaeology, some reason for thinking that the Wall ceased to be manned by troops during the reign of Honorius, and p153 if so the date which naturally offers itself to conjecture is the time of the rebellion of Constantine (A.D. 407‑11).

§ 19. From A.D. 410 to 429 Roman literary sources tell us nothing of Britain, but in A.D. 429 Germanus, bishop of Auxerre, was sent thither by Pope Celestine to stem the Pelagian heresy, and in his biography by Constantius, written not many years after his death, there is a notice, which, though legendary in form, is clearly based on an actual victory won over combined forces of Saxons and Picts in the neighbourhood of St. Albans.58 The story makes it a bloodless victory gained by the efforts of Germanus. It illustrates the evidence of the Notitia that at this time or soon afterwards newly-formed units were being sent to the Count of the Britains. The pressure of the Saxons is further confirmed by the British tradition that in the consulship of Felix and Taurus, A.D. 428, the Saxons came to Britain.59

The next notice we have in a Roman source is a few years posterior to the chronological limits of the Notitia Occidentis. A Gallic chronicle states that in the eighteenth year of the joint reign of Theodosius II and Valentinian III (calculated from the death of Honorius)60

Britannia usque ad hoc tempus uariis cladibus euentibusque uexatae61 in dicionem Saxonum rediguntur.

It looks rather as if this, A.D. 442, were the year in which Roman rule finally came to an end in Britain. Valentinian III and Aetius had neither money nor men to continue to defend the island; all they had were urgently needed nearer home. It was the same year in which Rome was compelled formally to surrender to Gaiseric the best provinces of Africa.

§ 20. It will be convenient to summarise the principal results of this examination of the Notitia Dignitatum.

1. The Notitia Orientis comes from a clean copy, prepared at Constantinople in or soon after A.D. 426, and transmitted to the primicerius notariorum at Rome. It was copied from the notitia then in use at Constantinople.

2. The Notitia Occidentis is derived from the working copy in use in the office of the primicerius at Rome, drawn up in A.D. 427‑8, and in use during the following decade. Thus it contains a number of corrections and additions of the years 428‑37.

3. The troops enumerated in Not. Occ. VII represent the field forces of the west as they were at this period (428‑37), not, as is generally assumed, at the end of the fourth century.

p154 4. The sections of Not. Occ. appertaining to Britain represent the actual situation in 428 and following years; with the possible exception of the sub-section of XL on the troops on the Wall.

5. It is therefore possible that the Gallic Chronicle may preserve the true date of the Roman abandonment of Britain, A.D. 442.

6. The office of magister equitum per Gallias was introduced, as a permanent command, in A.D. 429.

The Author's Notes:

2 Not. Or. XVIII. omnium is Schonhov's correction of omnis; the corresponding text in Not. Occ. XVI has Notitia omnium, but omits the words scholas — tractat.

3 Carm. Min. 25.83 sqq. The father of Celerina is the primicerius referred to, and he held office in the west. Cp. Birt in the index to his ed., sub Palladius.

4 Cp. Justinian, Nov. 24 ad fin., 25 ad fin., etc. Karlowa, Gesch. d. röm. Rechts, I.991. It is not probable that mandata principis were issued to provincial governors in the fourth century, nor (if they were), is there any evidence that the primicerius had anything to do with them. We only know that this practice had fallen out of use before the age of Justinian and that he revived it (Nov. 17).

9 There was only a different lettering on the codicilli; see Not. Occ. XLIV, XLV.

11 This writer also records only one Cappadocia, but we know from Cod. Th. 13.1.11, that there were already two Cappadocias before A.D. 386. In enumerating the provinces of western Illyricum (Diocese) along with those of eastern Illyricum (Prefecture) Polemius reproduces the conditions existing before A.D. 379. Then there was only one province of Dacia; in Not. Or. there are two Dacias, but we cannot say whether this change was due to Theodosius or to Eutropius.

14 It is possible that some of the units named Valentinianenses were also formed in the early years of Valentinian III (cp. Not. Occ. VII.47, 61, 71, 165).

15 I do not include the clarissimate of the praeses in Not. Or. XLIV.4, and the perfectissimate of the praeses in Not. Occ. XLV.4. It is pretty certain that perfectissimi in the latter passage is an error. The perfectissimate had probably been entirely done away with in A.D. 412, as Hirschfeld has shown, Die Rangtitel der römischen Kaiserzeit, in S. B. of the Berlin Academy, 1891, p592. Compare Mommsen, Hist. Schr. I, 558 sqq.

16 The inconsistencies noted above, § 4, show indeed that it was not meticulously corrected. There is only one case of the inadvertent inclusion of a name which had been corrected in the margin XL.48; and in the same place the troops of the laterculum minus, 44‑49, ought to have been transferred immediately after 36, so as to precede the officium (as in other sections). Some small inconsistencies, pointed out by Mr. Seeck, op. cit. p73, have also been allowed to remain in XXXIX and XLI.

18 Ib. 11.28.9, in the note as to sending copies of the law to certain officials.

21 Because in Cod. Th. 1 the title 8 on the quaestor precedes title 9 on the mag. off.

23 See Mommsen's ed. of Cod. Th. vol. I, p. clxxxvii. Before Feb. 22, 430, Helio was replaced, Cod. Th. 7.8.15.

26 This activity is referred to by Vegetius in De re Militari. Mr. Seeck has made it probable that those were right who, like Gibbon, saw Valentinian III in the emperor of Vegetius.

27 No arrangement had yet been made for the civil government of Valeria; the province does not appear in section II, nor a praeses in the Index. Probably it was left temporarily under the military rule of the dux.

28 The other spectabiles (except the proconsuls and consulares) have, instead, volumes with various inscriptions on the covers, which present great difficulty. Eight different inscriptions can be distinguished. In the most important of these (which appears in the insignia of the castrensis, primicerius notariorum, magistri scrin., vicarii, comites and duces), fl̅|intall|comord̅|pr Böcking (Not. Dig. II, 528, n9) rejects the most obvious explanation of the last words, comes ordinis primi, on insufficient grounds. The true reason for rejecting it — on the assumption that the inscription designates the ranks of the particular official — is that it appears in the insignia of the simple duces and they were not comites ordinis primi.

29 Valentinianenses, to one of which Valentinianenses iuniores, 11, V.190, will correspond.

30 203 equites stablesiani. It might be conjectured that in VI equites stablesiani Britanniciani fell out after 82 equites st. Italiciani. It is probable enough that 204 equites Taifali may be the same as VI.59, equit. Honoriani Taifali iuniores, but these may possibly be identical with VII.172, equit. Hon. iun. who were in Gaul.

32 There is no evidence for a comes Britanniarum in the fourth century. Theodosius, the emperor's father, had the rank of comes when he was sent to Britain in 369, but he did not take the place of any one else, nor is there anything to show that he was succeeded by a comes permanently stationed there. In consequence of his success he received the post of a magister equitum.

36 Mommsen held that from the time of the supremacy of Arbogastes and throughout the fifth century, the two masterships in praesenti were combined (under the title Mag. utriusque mil.), Hist. Schr. I, 556. This view is certainly wrong. It is refuted by the Notitia, and it is quite clearly untrue of the later part of the reign of Valentinian III when we have the clearest evidence for two magistri, both of whom have the title mag. u. m.

40 Prosper, Chron. sub a., Aetius magister militum factus est, interpreted, rightly as I think, by Mommsen (ib. 535). In the previous year Aetius had only the title comes. In the Life of St. Hilary of Arles we read of an inlustris Cassius qui tunc (A.D. 429) praecerat militibus (c. 6, in Migne, P.G. 50, col. 1227) and evidently stationed at Arles. Inlustris might suggest that he was a mag. mil., but that seems highly improbable. He is otherwise unknown.

41 Taking the strength of the legion as 1,000, of the auxiliary cohort and the vexillatio as 500 each. Of course, some of the units may not have been up to full strength.

42 See Historische Schriften, III, 214, n2: 'die britannischen Abschnitte der Notitia gehören der vordiokletianischen Epoche'; ib. 117, they belong 'der vorconstantinischen Militärordnung.' Mommsen's theory has been accepted, e.g. by Mr. Sagot in La Bretagne romaine (1911), p230, and Mr. Grosse in Römische Militärgeschichte (1920), p28. It is rejected by Sir C. Oman, England before the Norman Conquest (1910), p151.

43 For the cohort I or II Asturum (XL.42) and the ala Sabiniana (XL.37) cf. Mr. Cheesman, The Auxilia of the Roman Imperial Army, p147.

44 See Mr. Craster's study of 'The Last Days of the Roman Wall' in Archaeological Journal, LXXI.25 sqq. 1914.

45 'Roman Britain in 1914' (Supp. Papers, III, of the British Academy, 1915), p40. In the same way the archaeological evidence is consistent with the destruction of Calleva about A.D. 420, but does not disprove its survival till a later year.

47 See also his Romanization of Roman Britain (ed. 3, 1915), p80: 'After 407 the Romanized area was cut off from Rome.'

50 Mr. Haverfield (ib.) explains the assumed 'departure of Romans' after 406‑7 as meaning, not a great departure of persons, but that the central government ceased to send 'the usual governors and other high officials and to organize the supply of troops.' If the Caesar Carausius of the Richborough coin published by Sir Arthur Evans (Num. Chron. VII.191, 1887) was, as he suggested, a colleague of Constantine, this would support the probability that Constantine did not propose to let Britain slip from his hand.

52 Ib. 10, Ὁνωρίου δὲ γράμμασι πρὸς τὰς ἐν Βρεττανίᾳ χρησαμένου πόλεις φυλάττεσθαι παραγγέλλουσι. Gibbon dated the end of Roman government in Britain not to the departure of Constantine but to this rebellion (chap. XXXI). Of a Saxon invasion at this period we have a record in the Chronica Gallica (Chron. Min. ed. Mommsen, I, p654): Britannia Saxonum incursione devastatae in the sixteenth year of Honorius, i.e. A.D. 410. But we cannot depend on the accuracy of the dates in this chronicle to a year. The death of Arcadius, e.g., is entered under XII instead of XIV. That of Honorius comes in the thirty-second instead of the twenty-ninth year of his reign, and some of the other dates are three years wrong. (Possibly there was a confusion between 395, and the year of Honorius's creation as Augustus in 393.) The true date of the invasion may thus be two or three years earlier, 408 or 407, or it may be later.

53 CIL VII.268. See A. J. Evans, Numismatic Chronicle, 1887, p208 and cp. Haverfield, Ephemeris epigraphica, IX, p561. For Justinian, see Olympiodorus, fr. 12 (Ἰουστῖνον for Ἰουστινανόν),º the source of Zosimus, VI.2.

58 Constantius, Vita Germani, c. 17. See Levison, Bischof G. von Auxerre, in Neues Archiv, XXIX.97 sqq. 1903.

59 Historia Brittonum (in Chron. Minora, III), c. 66, p209, cp. c. 31, p171.

61 So I read for latae; the correction seems simpler than Mommsen's la<te uexa>tae.

p131 The Notitia Dignitatum
By J. B. Bury.

The Author's Notes:

Thayer's Note:

p131 The Notitia Dignitatum By J. B. Bury.

The Author's Notes:

Thayer's Note:

p131 The Notitia Dignitatum
By J. B. Bury.