The Online Library of Liberty

A project of Liberty Fund, Inc.

• The Works of Edward Gibbon
• The History of the Decline and Fall of the Roman Empire
• Chapter Xxvii
• Chapter Xxviii
• Chapter Xxix
• Chapter Xxx
• Chapter Xxxi
• Chapter Xxxii
• Chapter Xxxiii
• Appendix Additional Notes By the Editor
• 1.: Theology In the Market-places of Constantinople — ( P. 13 )
• 2.: Did Theodosius Visit Rome In 394? — ( P. 66 )
• 3.: The Libraries of Alexandria — ( P. 85 , 87 ) 1
• 4.: Worship of Relics — ( P. 98 , 99 )
• 5.: Stilicho In Inscriptions — ( P. 119 , 134 , 159 )
• 6.: The Two Eastern Expeditions of Stilicho and His Illyric Policy — ( P. 122 , 144 )
• 7.: Alaric In Greece — ( P. 140-143 )
• 8.: Penetrabis Ad Urbem — ( P. 148 )
• 9.: Alaric’s First Invasion of Italy — ( P. 148 , 151 Sqq. )
• 10.: Radagaisus — ( P. 167 )
• 11.: The Second Carausius — ( P. 178 )
• 12.: The Tyrant Constantine — ( P. 178 )
• 13.: “the Statue of a Poet Far Superior to Claudian” — ( P. 191 )
• 14.: Death of Maximus — ( P. 266 )
• 15.: Septimania — ( P. 286 )
• 16.: Rate of Travelling By Sea — ( P. 289 )
• 17.: Armenian Affairs — ( P. 331 , 333 )
• 18.: Procopian Legends — ( P. 354 , and Vol. Vi . P. 80 )
• 19.: The “egyptian” of Synesius — ( P. 304 )

APPENDIX

ADDITIONAL NOTES BY THE EDITOR

1.

THEOLOGY IN THE MARKET-PLACES OF CONSTANTINOPLE — (P. 13)

The humorous description of the interest taken in theological subtelties by the mechanics and slaves of Constantinople is quoted by Gibbon on the authority of Jortin, but Gibbon acknowledges that he does not know where it comes from, and implies that Jortin does not state his source.

A striking instance of the slumbers of Homer. Jortin indeed omits to give the reference, but he expressly ascribes the passage to “Gregory,” that is, Gregory of Nyssa, with whom he is dealing in the context. It would seem from Gibbon’s note that he took Gregory to be the Nazianzen.

The passage occurs in Gregory Nyssen’s Oratio de deitate Filii et Spiritus Sancti (Migne, Patr. Gr. 46, p. 557) and runs as follows: —

ἐὰν περὶ τω̂ν ὀβολω̂ν ἐρωτήσῃς ὁ δέ σοι περὶ γεννητον̂ καὶ ἀγεννήτου ἐϕιλοσόϕησε κἂν περὶ τιμήματος ἄρτου πύθοιο, Μείζων ὁ πατὴρ, ἀποκρίνεται, καὶ ὁ υἰὸς ὑποχείριος. εἱ δὲ, Τὸ λουτρὸν ἐπιτήδειόν ἐστιν, εἴποις, ὁ δὲ ἐξ οὐκ τὸν υἰὸν εɩ̂̓ναι διωρίσατο.

2.

DID THEODOSIUS VISIT ROME IN 394? — (P. 66)

According to Zosimus (iv. 59 and v. 30), Theodosius went to Rome after the battle of the Frigidus. This is likewise attested by Prudentius (against Symm., i.), and is implied in Theodoret’s statement, in reference to the visit of 389, χρόνου δὲ συχνον̂ διελθόντος εἰς τὴν Ῥώμην ἀϕικόμενος πάλιν ὁ βασιλεύς. This evidence has been accepted by Jeep; but the objections urged by Tillemont against it seem quite decisive, and it is rejected by Clinton and most authorities. It is a case of a confusion between the suppression of Maximus and the suppression of Eugenius; the visit to Rome after the second war is merely a duplicate of the visit after the first war. Guldenpenning thinks that Theodosius sent a message to the senate signifying his will that pagan worship should cease (Der Kaiser Theodosios, p. 229-30).

3.

THE LIBRARIES OF ALEXANDRIA — (P. 85,

87

)1

“The valuable library of Alexandria was pillaged or destroyed.” That is, the lesser library in the Serapeum, which was situated in the Rhacôtis quarter of the city. Gibbon has failed to distinguish it from the great Library of the Brucheum, of which Zenodotus, Callimachus, and other famous scholars were librarians. This Library is said to have been burnt down when Cæsar was in Alexandria (but see Mahaffy, Egypt under the Ptolemies, p. 454).

For the distinction of the two libraries see Epiphanius, de mensuris et ponderibus, 168 (Migne, Patr. Gr. vol. 43, p. 256): ἔτι δὲ ὕστερον καὶ ἑτέρα ἐγένετο βιβλιοθήκη ἐν τῷ Σεραπίῳ [sic] μικροτέρα τη̂ς πρώτης, ἥτις καὶ θυγατὴρ ὠνομάσθη αὐτη̂ς. For the first or mother library, see ib. 166 (Migne, p. 249). For other references see Susemihl, Geschichte der alexandrinischen Litteratur, i. p. 336.

But is it an attested fact that the lesser or daughter library was destroyed in 391? The sanctuary of Serapis was demolished, but does that imply the demolition of all the buildings connected with the Serapeum?2 The only evidence on which Gibbon’s statement rests is the sentence which he quotes from Orosius (p. 87, n. 53). But Orosius does not mention the Serapeum or speak of a large library. He merely says that he had seen bookcases in temples (which he does not name); and that, since then, he had been informed that the temples had been pillaged and the bookcases emptied. It seems to me highly improbable that Orosius is thinking either of the Alexandrian library or of the Serapeum. There is no reason to suppose that the library was in the temple. I conclude then that there is no evidence that the library of the Serapeum did not survive till the Saracen conquest, notwithstanding the verdict of Susemihl (ib. 344): “Omar fand 642 schwerlich noch Bücher in Alexandreia zu verbrennen.”

4.

WORSHIP OF RELICS — (P. 98,

99

)

In Gregory Nyssen’s Encomium of St. Theodore (Migne, vol. 46, 736 sqq.) there are passages, which, coming from such an eminent and learned ecclesiastic, are an important illustration of the growth of the veneration of relics. For example, he says: — εἰ δὲ καὶ κόνιν τις δοίη ϕέρειν τὴν ἐπικειμένην τῃ̑ ἐπιϕανείᾳ τη̂ς ἀναπαύσεως, δω̂ρον ὸ χον̂ς λαμβάνεται, καὶ ὼς κειμήλιον ἡ γη̂ θησαυρἰζεται. τὸ γὰρ αὐτον̂ τον̂ λειψάνου προσάψασθαι, εἴ ποτέ τις ἐπιτυχία τοιαύτη παράσχοι τὴν ἐξουσίαν, ὄπως ἐστὶ πολυπόθητον, καὶ εὐχη̂ς τη̂ς ἀνωτάτω τὸ δω̂ρον ἴσασιν οἱ πεπειραμένοι καὶ τη̂ς τοιαύτης ἐπιθυμίας ἐμϕορηθέντες . . . τὸ μὲν ἁπλω̂ς ἀποθανὸν ῤίπτεται ὡς τὸ τυχόν· τὸ δὲ τῷ πάθει τον̂ μαρτυρίου χαριτωθὲν, οὕτως ἐστὶν ἐράσμιον καὶ ἀμϕισβητήσιμον, ὼς ὸ προλαβω̂ν λόγος ἐδίδαξεν (p. 740).

5.

STILICHO IN INSCRIPTIONS — (P. 119,

134

,

159

)

The inscription celebrating the rescue of Africa by Stilicho, referred to by Gibbon, p. 119 (note) and p. 134 (note), will be found in C.I.L. vi. 1730. It runs as follows: —

• Flavio Stilichoni inlustrissimo viro, magistro equitum peditumque
• comiti domesticorum, tribuno prætoriano, et ab ineunte ætate
• per gradus clarissimæ militiæ ad columen sempiternæ
• et regiæ adfinitatis evecto, progenero Divi Theodosi, comiti
• Divi Theodosi in omnibus bellis adque victoriis et ab eo in
• adfinitatem regiam cooptato itemque socero D. N. Honori
• Augusti Africa consiliis suis et provisione liberata.

For inscriptions referring to the restoration of the “walls, gates, and towers” of Rome, undertaken through Stilicho’s influence before Alaric’s first invasion of Italy, see C.I.L. vi. 1188-1190.

Another inscription records Stilicho’s victory over Radagaisus: C.I.L. 6, 1196 (p. 249). Gibbon (after Mascou) refers it to the Gothic was of 402-3, and expresses surprise at the description of Alaric’s defeat as the total extinction of the Gothic nation (p. 159). Pallman took the same view (Volkerwand, p. 243); but the title is rightly referred in the Corpus (loc. cit.) to the events of 405.

Imppp. clementissimis felicissimis toto orbe victoribus DDD NNn Arcadio Honorio Theodosio Auggg. ad perenne indicium triumphorum quod Getarum nationem in omne ævum docuere extingui arcum simulacris eorum tropæisq decoratum S.P.Q.R. totius operis splendore.

6.

THE TWO EASTERN EXPEDITIONS OF STILICHO AND HIS ILLYRIC POLICY — (P. 122, 144)

An unwary reader of Gibbon might fail to realise that on two separate occasions Stilicho came, an unwelcome helper, to the assistance of Arcadius in the Illyric peninsula. As there has been a difficulty about the dates, and as Zosimus inverts the order of events, it is important to grasp this clearly. On the first occasion ( 395) Stilicho started from Italy in spring (Claudian, in Rufin. 2, 101), came up with Alaric in Thessaly, and was then commanded to return, before he had accomplished anything, by an order of Arcadius. Gainas and the Eastern troops went to Constantinople, and Rufinus met his fate; while Stilicho returned to Italy. In the following year ( 396), when Alaric was in southern Greece, Stilicho again, came to help the realm of Arcadius, landed at Corinth, blockaded Alaric in Pholoe, and allowed him to escape. (Zosimus, v. 7, places the blockade of Pholoe before the death of Rufinus. The charge of Zosimus that Stilicho indulged in debauchery in Elis cannot safely be pressed; for the phrase he uses is borrowed from Julian’s Misopogon. See Mendelssohn, ad. loc.)

395. Claudian represents Alaric as shutting himself up in a fortified camp on the news of Stilicho’s approach (in Ruf. 2, 124-9). Stilicho arrives in Thessaly (implet Thessaliam ferri nitor, l. 179) and prepares to attack the enemy. If he had been permitted to do so, the invasion of Greece would have been averted (186 sqq.), but alas! regia mandata arrive from Arcadius, and he has to sacrifice the “publica commoda” to the duty of obedience. This must have been about the beginning of November, if Rufinus was slain on 27th November (as Socrates states, vi. 1; cp. Chron. Pasch. ad ann.). Thus the advance of Stilicho from Italy to Thessaly would have occupied more than six months. What was the cause of this delay? It is significant that the charge brought against Rufinus by Claudian of having incited the Visigoths to the invasion of Greece is uttered only as a suspicion by Socrates (loc. cit., δόξαν ε[Editor: illegible character]χεν ὡς κ.τ.λ. “was supposed to have,” &c.); in the following century the suspicion has developed into a positive statement in the chronicle of Count Marcellinus ad ann. (Alaricum . . . infestum reipublicae fecit et in Graeciam misit).

396. (Gibbon wrongly places the events of this year in 397. It is not clear why he deserts the guidance of Tillemont.) Stilicho landed at the Isthmus (Zosimus, 5, 7), and is said to have had Alaric at his mercy at Pholoe. Three views have been held as to the escape of Alaric: (1) he out-witted Stilicho, who was culpably negligent (cp. Zosimus); (2) the suggestion of Claudian (B.G. 516) that Arcadius and his ministers, jealous of Stilicho’s intervention, treated with Alaric and secured his retreat, might be supported by the circumstance that Arcadius created him Master of Soldiers in Illyricum soon afterwards; (3) Stilicho is supposed to have made a secret treaty with Alaric, and permitted his retreat, for purposes of his own.

It is certain that Stilicho’s assertion of the unity of the Empire by appearing with armed forces in the Prefecture of Illyricum was viewed with suspicion and distrust at Constantinople. The feeling at the court of Arcadius is aptly expressed in words which Claudian has put into the mouth of Rufinus (in Ruf. 2, 161): —

• Deserat (sc. Stilicho) Illyrici fines, Eoa remittat
• agmina, fraternas ex aequo dividat hastas.

It is certain too that Stilicho afterwards, if not in 396, made it the aim of his policy to detach Illyricum from Arcadius and add it to the realm of Honorius. This is stated in so many words by Zosimus (v. 26), and it was doubtless Stilicho’s object from the beginning. This is the view of Jung (Römer and Romanen, p. 188: ich sehe darin vielmehr die consequente Verfolgung der durch Stilicho von Anfang an beabsichtigten Politik), who has some good remarks on the geographical importance of Illyricum; the unsatisfactoriness of the line of division of 395 which cut off Dalmatia from the rest of the Balkan peninsula (p. 186); and the circumstance that all northern Illyricum belonged to the Latin-speaking part of the Empire.

After the first invasion of Italy, Stilicho intended to use the help of Alaric for this purpose, and established him on the borders of the territory on which he had designs; but the execution of the plan was continually deferred, on account of other events which claimed the care of Stilicho. Alaric during this time was playing his own game, between the courts of Ravenna and Constantinople. His object was to obtain permanently Dalmatia, Noricum, Istria, and Venetia, with a regular grant of money from the Empire. This was what he asked in 410 (Zos. v. 48), and his aim throughout was doubtless a settlement of this kind.

The certainty that from 402 forward Stilicho made use of Alaric for his Illyric designs rouses the suspicion that he was playing with Alaric, with the same intent, in 395 and 396. The famous words of Orosius (vii. 37): Alarico rege cum Gothis suis saepe victo saepe concluso semperque dimisso, are strikingly true of Pollentia, of Verona, and of Pholoe; I suspect that they are also true of the campaign of 395, and that the unaccountable delay between Stilicho’s start in the spring and his return to Italy in Oct.-Nov. was due to diplomatic dallyings with Alaric. Of course nothing would be said of that by Claudian.

While Stilicho aimed at annexing eastern Illyricum, the court of Constantinople aimed at the acquisition of Dalmatia. Olympiodorus says that Stilicho employed Alaric to defend it (fr. 3). The object was pursued in the reign of Theodosius ii. and was finally attained at the marriage of Eudoxia with Valentinian iii., when the boundary was changed to the advantage of the East. Compare Cassiodorius, Var. ep. 1, Güldenpenning, das oström. Reich, p. 310. But even as early as 414-15 there is epigraphic evidence suggesting the conclusion that at that time Salonae was under the government of Constantinople. See Jung, op. cit. p. 187 note.

It is possible to regard (with Keller; Stilicho, p. 27) Stilicho’s special Illyric policy and his relations with Alaric as part of a larger policy which had two chief aims: to maintain the unity of the Empire, under two emperors, and to infuse new blood into it by absorbing barbarians. Stilicho’s policy has been generally misunderstood. A monograph appeared in the year 1805 with the curious title: Flavius Stilicho, ein Wallenstein der Vorwelt (by C. F. Schulz).

7.

ALARIC IN GREECE — (P. 140-143)

Though no record tells that Alaric burnt down the Temple of Eleusis, it is certain that the invasion of the Goths was coincident with the end of the Eleusinian mysteries. The sanctuary of the two goddesses must have already suffered much under Jovian and Theodosius. The cult, restored by Julian, was suppressed by Jovian, but renewed again under Valentinian through the intervention of Praetextatus, proconsul of Achaia. It must have been affected by the intolerant edicts of Theodosius; certainly the demonstration of the Christian section of the Athenian community forced the last Eumolpid high priest to resign. Subsequently — probably on the death of Theodosius — the pagan party felt themselves strong enough to appoint, as hierophant, a priest of Mithras from Thespiae, and he persided at Eleusis at the time of Alaric’s invasion.

See Gregorovius, Hat Alarich die Nationalgötter Griechenlands zerstört? (Kleine Schriften, vol. i.), and Geschichte der Stadt Athen im Mittelalter, i. p. 35 sqq.

As for Athens, there is no doubt that it capitulated and was spared by Alaric, and that the Goths did not destroy or rob its art treasures. Athens suffered, as Gregorovius remarks, less in the invasion of Alaric than in the invasion in the time of Dexippus. There were of course acts of cruelty; some are recorded in the Vita Prisci of Eunapius. But we must not press the words of Claudian (in Rufin. ii. 189): nec fera Cecropiae traxissent vincula matres, further than at the most to interpret it of the rural inhabitants of Attica. Gregorovius observes that in the other passages where the devastation of Greece is mentioned (iv. Cons. Hon. 471, Eutrop. 2, 199, cons. Stil. i. 180), there is not a word about Athens.

As to the Zeus-temple of Olympia, it is supposed that the Phidiac statue of Zeus had been removed about two years before the Gothic invasion (in 394, when Theodosius suppressed the Olympic games) to Constantinople and was afterwards burned in the Palace of Lausus. Cp. Cedrenus, i. p. 364 (Gregorovius i. p. 43). The temple of Olympia was burnt down in the reign of Theodosius ii.

The general conclusion of Gregorovius is that it is a gross exaggeration to ascribe to the Goths the deliberate destruction of the temples and sanctuaries of Greece.

8.

PENETRABIS AD URBEM — (P. 148)

The clear voice which Alaric heard in the grove uttered an acrostich with the help of Claudian’s art. It has been pointed out that the first and last letters of the two verses (B.G. 546-7) spell ROMA.

• R umpe omnes, Alarice moras; hoc impiger annO
• A lipibus Italiae ruptis penetrabis ad urbeM.

So it is printed in Koch’s edition.

9.

ALARIC’S FIRST INVASION OF ITALY — (P. 148, 151sqq.)

That the battle of Pollentia was fought in 402 is now universally agreed by all competent historians; there is no conflict of evidence on the matter, and there is nothing to be said for 403.1 But there is still room for difference of opinion as to the date of Alaric’s entry into Italy, and possibly as to the date of the battle of Verona.

(1) We have to set the statements of two chronicles against each other. On one hand Prosper, sub ann. 400: Gothi Italiam . . . ingressi (see next Appendix). On the other, the Fasti Vindobonenses (Chronica Italica; see above, vol. iv. Appendix 5, p. 353) have, sub anno 401, the more precise notice: et intravit Alaricus in Italiam, xiv. kl. December.2

Pallmann (followed by Hodgkin) accepts the date of Prosper. Tillemont, also accepting Prosper, but putting (in spite of Prosper) the battle of Pollentia in 403, found himself driven to assume that Alaric having invaded Italy in 400 was driven out of it in 401 and returned in 402 — in fact a double invasion.

As there is little or nothing to choose between Prosper and the Fasti Vindobonenses — both being equally prone to error — we may be disposed to allow the argument of Seeck3 (approved by Birt) to determine us in preferring the date of the Fasti Vindobonenses. In describing the entry of the Goths Claudian speaks of constant eclipses of the moon among the terrors which preyed upon men’s minds: —

• territat adsiduus lunæ labor atraque Phœbe
• noctibus aerisonas crebris ululata per urbes.
• nec credunt vetito fraudatam Sole sororem
• telluris subeunte globo sed castra secutas
• barbara Thessalidas patriis lunare venenis
• incestare iubar.

• (B.G. 233 sqq.)

These data (cp. adsiduus) are satisfied by the two lunar eclipses which took place on June 21 and December 6, 401.

After Pollentia, there must have been another engagement at Asta (vi. cons. Hon. 203). Keller thinks that this took place before that of Pollentia. In any case Gibbon is wrong in supposing that Asta was the town in which Honorius was shut up, till delivered by Stilicho. Honorius was in Milan, as is clear from Claudian’s description (ib. 456 sqq.). To reach Asta Stilicho would have had to cross not only the Addua (488), but the Padus (which is not mentioned).

(2) That the battle of Verona did not take place later than 403 is proved by the fact that it is celebrated in the Panegyric composed by Claudian before the end of that year for the sixth consulate of Honorius, which began on Jan. 1, 405. That it took place in summer is proved by a line of that poem (our only source for the battle): —

Those therefore who like Tillemont and Gibbon set Pollentia in spring 403 were obliged to set Verona in the summer of the same year. The question therefore arises whether, when we have moved Pollentia a year back, we are to move Verona along with it. Pallmann leaves Verona where it was in 403, and he is followed hesitatingly by Mr. Hodgkin. That the victory of Verona was won in 403, and that more than a year elapsed between the two battles, has, I think, been proved convincingly by Birt (Preface to ed. of Claudian, liv.-v.). The argument is that, if Verona had been fought in 402, the long interval of sixteen months would have stultified the whole tone of Claudian’s poem, which breathes the triumph of a recent victory. Such a line as

is inconceivable on any save the first First of January following the victory. Cp. also lines 406, 580, 653. The transition in l. 201 is suggestive of a considerable interval between the two battles —

• te quoque non parvum Getico, Verona, triumpho
• adiungis cumulum nec plus Pollentia rebus
• contulit Ausoniis aut moenia vindicis Hastae.

The resulting chronology is: —

• 401. Alaric enters Italy (Venetia) in November; at the same time Radagaisus (see next Appendix) invades Raetia. Stilicho advances against Radagaisus.
• 402. Battle of Pollentia on Easter Day.
• 402-403. Alaric in Istria.
• 403, Summer. Alaric again moves westward; Battle of Verona.

10.

RADAGAISUS — (P. 167)

Radagaisus invaded Italy in 405 , at the head of an army of barbarians. He was defeated by Stilicho on the hills of Faesulae. There is no doubt about these facts, in which our Western authorities agree, Orosius (vii. 37), Prosper, ad ann. 405, and Paulinus (Vita Ambrosii, c. 50). Prosper’s notice is: Radagaisus in Tuscia multis Gothorum milibus cæsis, ducente exercitum Stilichone, superatus et captus est. But Zosimus (v. 26) places the defeat of Radagaisus on the Ister. “A strange error,” Gibbon remarks, “which is awkwardly and imperfectly cured by reading Ἄρνον for Ἴστρον.” Awkwardly and contrariwise to every principle of criticism. It is an emendation of Leunclavius, and Reitemeier’s Ἠριδανὸν is no better. But Zosimus knew where the Danube was, and the critic has to explain his mistake.

From Gibbon’s narrative one would draw the conclusion that this invasion of Italy in 405 (406 Gibbon incorrectly; see Clinton, ad ann.) was the first occasion on which Radagaisus appeared on the stage of Imperial events. But he appeared before. A notice of Prosper, which there is not the smallest cause to question, represents him as co-operating with Alaric, when Alaric invaded Italy. Under the year 400 (there may be reason for questioning the year; see last Appendix) in his Chronicle we find the record: Gothi Italiam Alarico et Radagaiso ducibus ingressi. It is perfectly arbitary to assume that the notice of the action of Radagaisus on this occasion is a mere erroneous duplication of his action, which is separately and distinctly recorded under the year 405. Pallmann emphasised the importance of the earlier notice of Prosper, and made a suggestion which has been adopted and developed by Mr. Hodgkin (i. p. 711, 716, 736), that Alaric and Radagaisus combined to attack Italia, Alaric operating in Venetia and his confederate in Raetia in 400-1, and that the winter campaign of Stilicho in Raetia in 401-2, of which Claudian speaks, was directed against Radagaisus. This combination has everything to recommend it. The passages in Claudian are as follows: —

Leaving aside the question whether (as Birt thinks) the barbarians whom Radagaisus headed in Raetia were the Vandals and Alans who invaded Gaul in 406, we may without hesitation accept the conclusion that in 401 Radagaisus was at the head of Vandals and other barbarians in Raetia. Birt points out the statement that Radagaisus had intended to cross into Italy (εἰς τὴν Ἰταλιαν ὥρμητο διαβη̂ναι), with which Zosimus introduces his account of the overthrow of Radagaisus by Stilicho; and proposes to refer that statement not to the campaign of 405 but to that of 401.

It was satisfactory to find that Birt had already taken a step in a direction in which I had been led before I studied his Preface to Claudian. The fact is that Zosimus really recounts the campaign of 401, as if it were the campaign of 405. His story is that Radagaisus prepared to invade Italy. The news created great terror, and Stilicho broke up with the army from Ticinum, and with as many Alans and Huns as he could muster, without waiting for the attack, crossed the Ister, and assailing the barbarians unexpectedly utterly destroyed their host. This is the campaign of the winter of 401-2, of which we know from Claudian’s Gothic War; only that (1) Zosimus, placing it in 405, has added one feature of the actual campaign in 405, namely the all but total annihilation of the army of Radagaisus, and that (2) Zosimus, in placing the final action beyond the Danube, differs from Claudian, who places it in Noricum or Vindelicia (l. 365, cited above) and does not mention that Stilicho crossed the river. But the winter campaign was in Danubian regions; and the main difficulty, the appearance of the Danube in the narrative of Zosimus, seems to be satisfactorily accounted for by the assumption of this confusion between the two Radagaisus episodes, a confusion which must be ascribed to Zosimus himself rather than to his source Olympiodorus.1

11.

THE SECOND CARAUSIUS — (P. 178)

A new tyrant in Britain at the beginning of the fifth century was discovered by Mr. Arthur Evans through a coin found at Richborough (Rutupiae). See Numismatic Chronicle, 3rd ser. vol. vii. p. 191 sqq., 1887. The obverse of this bronze coin “presents a head modelled in a somewhat barbarous fashion on that of a fourth century Emperor, diademed and with the bust draped in the paludamentum.” The legend is: DOMINO CARAVS IO CES. “The reverse presents a familiar bronze type of Constans or Constantius ii. The Emperor holding phoenix and labarum standard stands at the prow of the vessel, the rudder of which is held by Victory. In the present case, however, in place of the usual legend that accompanies this reverse — FEL. TEMP. REPARATIO — appears the strange and unparalelled inscription: —

This coin cannot be ascribed to the well-known Carausius of Diocletian’s reign; for the type of the reverse is never found before the middle of the fourth century. The DOMINO (without a pronoun — nostro) on the obverse is quite unexampled on a Roman coin. Mr. Evans conjectures that CONSTANTINO is to be read on the reverse and makes it probable that this obscure Carausius was colleague of Constantine iii., left behind by him, with the title of Caesar, to hold the island while he was himself absent in Gaul; and would refer the issue of the coin to 409. “The memory of the brave Carausius, who first raised Britain to a position of maritime supremacy, may have influenced the choice of this obscure Caesar, at a moment when the Romano-British population was about to assert as it had never done before its independence of Continental Empire.” Whether chosen by Constantine or not the coin “may at least be taken as evidence that the new Caesar stood forth as the representative of the interests of the Constantinian dynasty in the island as against the faction of the rebel Gerontius and his barbarian allies.”

12.

THE TYRANT CONSTANTINE — (P. 178)

The best account of the rise, reign, and fall of the tyrant Constantine, ruler of Britain, Gaul, and Spain, will be found in Mr. Freeman’s article, “Tyrants of Britain, Gaul and Spain,” in English Historical Review, vol. i. (1886) p. 53 sqq.

At first, in 407, Constantine’s Gallic dominions “must have consisted of a long and narrow strip of eastern Gaul, from the Channel to the Mediterranean, which could not have differed very widely from the earliest and most extended of the many uses of the word Lotharingia.” That he was acknowledged in Trier is proved by the evidence of coins (Eckhel, 8, 176). Then he moves down to the land between Rhone and Alps, which becomes the chief theatre of operations, and Arelate becomes his capital. His son Constans he creates Caesar, and a younger son Julian nobilissimus. Early in 408 Sarus is sent against him by Stilicho. Sarus gains a victory over Constantine’s officer (Justinian); and lays siege to Valentia, in which Constantine secured himself. But he raises the siege on the seventh day, on account of the approach of Constantine’s able general Gerontius, from whom he with difficulty escapes (by coming to an understanding with the Bagaudae, who appear to act as a sort of national militia) into Italy.

Constantine’s next step is to extend his rule over the rest of the Gallic prefecture, — Spain. We are left quite in the dark as to his relations with the Barbarians who in these years (407-9) were ravaging Gaul. Spain at first submitted to those whom Constantine sent; but very soon the influential Theodosian family organised a revolt against it. The main part of the resistance came from Lusitania, where the four Theodosian brothers had most influence. The rustic army that was collected was set to guard the Pyrenees. To put down the rising, Constantine sent troops a second time into Spain — this time under the Caesar Constans, who was accompanied by Gerontius and by Appollinaris (grandfather of the poet Sidonius), who accepted the office of Praetorian Prefect from Constantine. The Theodosian revolt was suppressed; Constans set up his court in Caesar-augusta (Zaragoza), but soon returned to Gaul, leaving Gerontius to defend Spain.

The sources for this story are Orosius, Sozomen, and Zosimus. For the Spanish events we have no fragments of Olympiodorus. “On the other hand the local knowledge of Orosius goes for something, and Sozomen seems to have gained, from some quarter or other, a singular knowledge of detail of some parts of the story” (Freeman, p. 65). It is practically certain that Sozomen’s source (as well as that of Zosimus) was Olympiodorus (cp. above, vol. ii. Appendix 10, p. 365).

Thus master of the West, Constantine forces Honorius, then ( 409) too weak to resist, to acknowledge him as his colleague and legitimate Augustus. Later in the year he enters Italy with an army, avowedly to help Honorius against Alaric (so Olympiodorus), his real motive being to annex Italy to his own realm (Soz. ix. 12). At this time he probably raised Constans to the rank of Augustus. It appears that Constantine was in league with Allobich, the general of Honorius, to compass his treasonable designs. They were discovered, Allobich was cut down, and then Constantine, who had not yet reached Ravenna, turned back.

Meanwhile the revolt of Gerontius in Spain had broken out, and Constans went to put it down. Gibbon’s account of the revolt is inadequate, in so far as he does not point out its connection with the invasion of Spain by the Vandals, Sueves, and Alans. There is no doubt that Gerontius and Maximus invited them to cross the Pyrenees. (Cp. Olymp.; Oros. 7, 28; Sozom. ix. 113; Zos. 6, 5; Renatus, in Gregory of Tours, 2, 9; Freeman, p. 74: “The evidence seems to go for direct dealings between Gerontius and the invaders, and his treaty with them is more likely to have followed the proclamation of Maximus than to have gone before it.”) The dominion of Maximus was practically confined to the northwestern corner; the seat of his rule was Tarraco. As for the relation of Maximus to Gerontius, it is very doubtful whether παɩ̂δα in Olympiodorus is to be interpreted son and not rather servant or retainer.

The rest of the episode of Constantine’s reign — the sieges of Vienna (which, some have suspected, is a mistake for Narbo) and Arelate — have been well told by Gibbon. These events must be placed in the year 411; for Constantine’s head arrived at Ravenna on 18th September (Idatius ad ann.), and it was in the fourth month of the siege of Arelate that Edobich’s troops came on the scene (Renatus ap. Greg. Tur. ii. 9).

Mr. Freeman thus contrasts the position of Constantine with that of contemporary tyrants: —

“Constantine and Maximus clearly leagued themselves with the barbarians, but they were not mere puppets of the barbarians; they were not even set up by barbarian help. Each was set up by a movement in an army which passed for Roman. But the tyrants who appear in Gaul in the following year, Jovinus, Sebastian, and Attalus — Attalus, already known in Italy, is fresh in Gaul — are far more closely connected with the invaders of the provinces. Attalus was a mere puppet of the Goths, set up and put down at pleasure; his story is merely a part of the marches of Ataulf in Gaul and Spain. Jovinus was set up by Burgundian and Alan help; his elevation to the Empire and the earliest Burgundian settlement in Gaul are simply two sides of one event. Even Maximus was not in this way the mere creature of the invaders of Spain, though he found it convenient at least to connive at their invasion.”

13.

“THE STATUE OF A POET FAR SUPERIOR TO CLAUDIAN” — (P. 191)

Other readers may, like myself, have been puzzled by this reference of Gibbon. Professor Dowden has supplied me with what must, I believe, be the true explanation. The statue of Voltaire by Pigalle (now in the Institut) was executed in 1770. The actress Mlle. Clairon opened a subscription for it. See Desnoiresterres, Voltaire et la Société au xviii. Siècle, vii. p. 312 sqq.

14.

DEATH OF MAXIMUS — (P. 266)

The chronicle of Count Marcellinus states that the tyrants Maximus and Jovinus were brought in chains from Spain (to Ravenna) and executed in the year 422, on the occasion of the tricennalia of Honorius (sub ann. 422, p. 75, ed. Mommsen, Chron. Min. vol. ii.). This, like some other unique notices in Marcellinus, was doubtless taken by him from the Chronica Italica (see above, vol. iv. Appendix 5, p. 353), which have come down in a mutilated condition (cp. Mommsen, ib. p. 46). It is borne out by Orosius, who, writing in 417, says (vii. 425): Maximus exutus purpura destititusque a militibus Gallicanis — nunc inter barbaros in Hispania egens exulat; which alone is of sufficient authority to refute the statements of the Eastern writers followed by Gibbon.

15.

SEPTIMANIA — (P. 286)

An error prevails in regard to the name Septimania. It first occurs in Sidonius Apollinaris, Ep. iii. 1, 4, where it is said of the Goths of the kingdom of Tolosa: Septimaniam suam fastidiunt vel refundunt, modo invidiosi huius anguli (that is, Arverni) etiam desolata proprietate potiantur. In his Index Locorum to Luetjohann’s ed. of Sidonius, Mommsen points out that Septimania is not derived from septem (the etymon is septimus) and therefore did not signify either the Seven Provinces of the Viennese Diocese, or seven cities granted to the Goths (Greg. Tur. 2, 20). It means the coast-line from the Pyrenees to the Rhone, in Sidonius as well as in Gregory of Tours and later writers; Sidonius means that the Goths declared themselves ready to exchange this coast district (including towns of Narbo, Tolosa, Bæterræ, Nemausus, Luteva) for Arverni. Bæterræ was a town of the Septimani; hence Septimania.

16.

RATE OF TRAVELLING BY SEA — (P. 289)

In connection with Gibbon’s note on the length of journeys by sea in the reign of Arcadius, I have found some contemporary data in the Life of Porphyry of Gaza by the deacon Marcus. (1) From Ascalon, in Palestine, to Thessalonica: 13 days, p. 6, ed. Teubner. (2) Back from Thessalonica to Ascalon: 12 days, p. 7. (3) From Gaza to Constantinople: 20 days, p. 24. (4) Back from Constantinople to Gaza: 10 days, p. 25. (5) From Cæsarea (Palæst.) to Rhodes: 10 days in winter, p. 30. (6) From Rhodes to Constantinople: 10 days, winter, p. 33. (7) From Constantinople (starting 18th April) to Rhodes: 5 days, p. 47. It must be remembered that we are not informed about intermediate stoppages. These references may be added to those in Friedläander’s Sittengeschichte, ii. 13-17. With a good wind one could sail 11 or 12 hundred stadia in 24 hours.

17.

ARMENIAN AFFAIRS — (P. 331, 333)

Gibbon wrongly places the division of the Armenian kingdom into Roman and Persian Armenia in the fifth century. This division was arranged between Theodosius the Great and the Persian King. See Saint Martin, Mémoires, p. 316. Persarmenia was at least two-thirds of the whole kingdom. Arsaces, who had already reigned 5 years over all Armenia, continued after the division to rule over Roman Armenia for 2½ years; while Chosrov (a Christian) was appointed by Persia as king of Persian Armenia. On the death of Arsaces, Theodosius committed the rule of the Roman part to a native general, who was induced to recognise the authority of Chosrov; while Chosrov, in order to secure his position in Roman Armenia, acknowledged the suzerainty of the Roman Empire. This did not please Persia, and Jezdegird, son of the Persian king, overthrew him, after he had reigned 5 years. Jezdegird then gave Armenia to Chosrov’s brother; but Chosrov was subsequently restored through the influence of the archbishop Isaac, and reigned about a year. He was succeeded by Sapor, a royal prince of Persia, who made himself hated and attempted to proselytise the Armenians. On his father’s death he returned to Persia, endeavoured to win the crown, failed, and perished. After an interval Ardeshir (Gibbon’s Artasires) was appointed — the last of the Armenian kings. His deposition is described by Gibbon. The government was then placed in the hands of Persian marzbans.

18.

PROCOPIAN LEGENDS — (P. 354, and vol. vi. p. 80)

(1) Boniface and Aetius; (2) Valentinian and Maximus

In his Italy and her Invaders, vol. ii. (p. 206 sqq., ed. 2) Mr. Hodgkin has discussed and rejected the romantic story connected with the death of Valentinian, the elevation of Maximus and his marriage with Eudoxia. The story is told by Procopius (de B. V. i. 4); and, in accordance with Gibbon’s criticism that “Procopius is a fabulous writer for the events which precede his own memory,” Mr. Hodgkin relegates it to “the fables of Procopius.”

In the English Historical Review for July, 1887 (p. 417-465), Mr. Freeman published a long criticism of the historical material for the careers of Aetius and Boniface. He held the account of Procopius (B. V. i. 3) to be “legend of the sixth century and not trustworthy history of the fifth,” and tried to “recover the true story as it may be put together from the annalists, the writings of St. Augustine, and other more trustworthy authorities.” In this case Mr. Hodgkin takes a completely different view and argues (ib., vol. i. p. 889 sqq., ed. 2) that the Procopian legend “has still a reasonable claim to be accepted as history,” while admitting that in some points it has been shaken by Mr. Freeman.

Now, while the two stories need not stand on the same footing so far as historical credibility is concerned, while it may be possible to follow Mr. Hodgkin in rejecting the one and accepting the main part of the other, there is a preliminary question which must be discussed before we attempt to decide the ultimate question of historical fact. Procopius is not the only authority for these stories. They are also found in the Salmasian Excerpts, which were first printed by Cramer in his Anecdota Parisina, ii. 383 sqq., and afterwards included among the fragments of John of Antioch by C. Müller, in the Fragmenta Hist. Græc., vol. iv. p. 535 sqq. The fragments in question are 196 and 200. It was a serious flaw in Mr. Freeman’s essay that he was not aware either of the Salmasian Excerpt 196, or of the Constantinian Excerpt 201, which also bears on the question of Aetius and Boniface. Mr. Hodgkin refers to fr. 196, which (with Müller) he ascribes to Joannes Antiochenus, and says: “Though a comparatively late author (he probably lived in the seventh century) and though he certainly used Procopius freely in his compilation, he had also some good contemporary authorities before him, especially Priscus, and there seems some probability, though I would not state it more strongly than this, that he may have found the story in one of them as well as in Procopius.”

But Mr. Hodgkin, while he takes account of fr. 196 in defending one “Procopian legend,” takes no account of fr. 200 in rejecting the other “Procopian legend,” though fr. 200 bears to the latter the same relation which fr. 196 bears to the former.

Now in the first place it must be clearly understood that the author of the work from which the Salmasian Excerpts are derived cannot have been the same as the author of the work from which the Constantinian Excerpts are derived. There is no question about this, and it could be proved merely by comparing the two (Salmasian) fragments under consideration (frags. 196 and 200) with (the Constantinian) fragment 201. If then we accept the Constantinian Excerpts under the name Joannes of Antioch, we must be careful not to ascribe the Salmasian Excerpts to that writer. Which is the true Joannes, is a question still sub judice. (See below, vol. vi. Appendix 2.)

The vital question then is whether Procopius was the source of S. (as we may designate the author of these Excerpts) for these fragments or not. For if he was, S. adds no weight to the authority of Procopius and may be disregarded; it he were not, his statements have to be reckoned with too. From a careful comparison of the passages, I find myself in complete agreement with C. de Boor (who has dealt with the question in Byz. Ztsch. ii. 204 sqq.) that Procopius was not the source of S. but that the accounts of both authors were derived from a common source.1 The proof in the case of fr. 200 is very complete; because we happen to have in Suidas sub voce θλαδιάς (see Müller ad loc.) a fragment of what was evidently that common source.

The inference, for historical purposes, is important. We cannot speak with Mr. Freeman of “Procopian legend” or “legend of the sixth century.” Procopius cannot be described in these cases as setting down “the received tale that he heard.” He was using a literary source; and there is not the slightest proof that this literary source belonged to the sixth century. It seems more probable that it was a fifth century source. It may have been Priscus or it may not.

These two episodes therefore depend on the authority of a writer (who has so far not been identified) earlier than Procopius and distinct from John of Antioch. They may for all we know have very early authority, and they cannot be waived away as “Procopian legend.” Each must be judged on its own merits.

It seems to me that there was probably a certain foundation of truth in both stories, but that they have been dressed out with fictitious details (like the story of the Empress Eudocia and Paulinus). I do not feel prepared to reject the main facts implied, that Aetius intrigued against Bonifacius and that Valentinian seduced the wife of Maximus.

The story of the single combat of Aetius and Boniface is derived from Marcellinus (like Procopius, a writer of the sixth century). But rightly interpreted it contains nothing improbable. It does not imply a duel; but a single combat in a battle. It is however important to observe that “John of Antioch” (fr. 201, Müller, p. 615) says nothing of Boniface’s wound but states that he was out-generalled by Aetius, and that he died of diseases due to depression and chagrin.

τὸν δὲ Βονιϕάτιον σὺν πολλῃ̑ διαβάντα χειρὶ ἀπὸ τη̂ς Λιβύης κατεστρατήγησεν, ὥστε ἐκεɩ̂νον μὲν ὺπὸ ϕροντίδων νόσῳ τελευτη̂σαι.

It remains to be added that the essay of Mr. Freeman throws great light on the career of Boniface in Africa and the doings of Castinus, Felix, and Sigisvult.

19.

THE “EGYPTIAN” OF SYNESIUS — (P. 304)

The interpretation of the Egyptian allegory of Synesius has caused a good deal of trouble, owing to the fact that our other sources supply such meagre material as to the details of the political transactions at Constantinople in the reign of Arcadius. It had long been recognised that Egypt stood for the Empire, and Thebes for Constantinople; and the Praetorian Prefect Aurelian had been detected under the veil of Osiris. But no certainty had been attained as to the identity of Typhos, the wicked brother of Osiris. It was chiefly in consequence of this lacuna that the able attempt of Güldenpenning to reconstruct the history of the years 399 and 400 on the basis of the work of Synesius (cp. my Later Roman Empire, i. p. 79 sqq.) did not carry complete conviction. But O. Seeck has recently succeeded in proving the identity of Typhos and in interpreting the allegory more fully (Philologus, 52, p. 442 sqq., 1894). His results must be briefly noted.

1. Taurus. — Synesius states in the Preface that the name of the father of Osiris and Typhos was Taurus. There can be no question that he is the Taurus who appears in the Consular Fasti of 361. He was quaestor in 353, and became praetorian prefect in 355. He held this office (the μεγάλη ἀρχή of Synes. c. 2, p. 1213, ed. Migne) till 361. He was appointed to decide a theological disputation (Epiphanius, de Haer. 71, 1); and presided at the Council of Ariminum (359). He was an author as well as an official. The arguments of Borghesi and Seeck establish his identity with Palladius Rutilius Taurus Aemilianus, the author of 14 Books De re rustica. Taurus had a son named Harmonius who was killed by Arbogastes 392 (John Ant., fr. 187).

2. Aurelian. — He appears first about 383 as builder of a Church (Acta Sanctorum, 6th May, p. 610). In 393 we find him (C. Th. 2, 8, 23, &c.) Prefect of Constantinople before Rufinus held that office. Then after the fall of Eutropius, he appears as Praetorian Prefect of the East (399-400). In 400 the revolt of Gainas causes his fall (see above, p. 304-305). But he was to rise again and become Prefect a third time (402-404), as Seeck has shown from two letters of Synesius (31 and 38: cp. Cod. Th. 4, 2, 1, and 5, 1, 5, where the false dates have to be amended). He is therein described as τρισέπαρχον, “thrice Prefect,” in an epigram (Anth. Plan. 4, 73) on a gilt statue dedicated to him by the senate. His son’s name was Taurus (Synes. epist. 31), which confirms the identification.

Osiris (i. c. 3, p. 1217) held a post which is described as ἐπιστάτης δορυϕόρων γενόμενος καὶ ἀκοὰς πιστευθεὶς, explained by Seeck to be that of magister officiorum; he was then Prefect of the city (πολιαρχήσας, ib.); he was consul (ii. 4, p. 1272), and he twice held the μεγάλη ἀρχή or praetorian prefecture, — the second time μετὰ συνθήματος μείζονος (ib.), which means the Patriciate. What happened to Osiris on his fall corresponds even more strikingly to that which happened to Aurelian. The leader of the foreign mercenaries is on the other side of a stream (like Gainas), Aurelian crosses it (p. 1252) and is spared. His companions in misfortune (Saturninus and Johannes) are alluded to, p. 1268.

3. Arcadius. — The insignificance of Arcadius is reflected in the myth by the fact that he is never mentioned except in one passage (p. 1268), where he appears as the High Priest. The person who through his influence over the Emperor had the real power appears in the myth as holding the kingly office — e.g. Osiris while he was in power.

4. Caesarius. — In the allegory Typhos is in close alliance with the barbarian mercenaries, and instigates their attack on Thebes in order to overthrow his brother Osiris. When Osiris surrenders himself to the barbarian leader, Typhos urges that he should be put to death. Typhos then receives the kingdom and administers it tyrannically; nor is his position shaken by the fall of the barbarian leader. Before the first rise of Osiris to power1 he had filled a post which gave him patronage in distributing offices, the power of oppressing towns (p. 1217), and the duty of regulating measures in connection with the payment of taxes in kind (p. 1219). These hints taken along with the mention (ib.) of torch-bearing attendants show that the office was no less than that of Praetorian Prefect. It follows that Typhos was Praetorian Prefect before 399, and again in 400.

Eutropius had endeavoured to reduce the power of Praetorian Prefect of the East by making it a collegial office; and Eutychianus appears as holding that office (1) along with Caesarius while Eutropius was in power; (2) along with Aurelian, 399-400; (3) along with Aurelian when he was restored 402. It may be assumed that he also held it between 400 and 402.

It follows that Caesarius, whom we find Praetorian Prefect from 396-398, and again in 400 and 401, was the prototype of Typhos, the son of Taurus and the brother of Aurelian. Some other points confirm the conclusion. The tendency to Arianism, of which Typhos is accused, is illustrated by C. Th. 16, 5, 25, and the passion of Typhos for his wife by a notice in Sozomen, 9, 2.

The great political object of Aurelian was to break the power of the Germans in the army and at the court — the policy for which Synesius pleaded in his De Regno. The question arises: What was the attitude of the Empress Eudoxia to this policy? The fall of Eutropius which she brought about (Phil. 11, 6) led to the rise of Aurelian, and when Aurelian fell, her intimate friend — scandal said, her lover — Count John, fell with him.2 Further, Seeck makes it probable that the second Praetorian Prefecture of Aurelian ended, and Anthemius succeeded to that post, about end of 404; and it was on 6th October, 404, that the Empress died. We are thus led to infer a close political union between Eudoxia and Aurelian; and, if the inference is right, it is noteworthy that the Empress of German origin, the daughter of the Frank Bauto, should have allied herself with a statesman whose policy was anti-German.

[1 ]I must note that in the Nation, July 7, 1898, Mr. Frederick I. Teggart has made a good case for Gibbon’s view that the Serapeum Library was burned in 391.

[2 ]The statement of Eunapius in the Vita Aedesii: καὶ τὸ Σαραπεɩ̂ον ἰερὸν διεσκεδάννυτο οὐχ ὴ [Editor: illegible character]εραπεία μόνον ἀλλὰ καὶ τὰ οἰκοδομήματα, cannot be pressed to mean more than that not only was the worship suppressed but the temple itself was demolished.

[1 ]“The date 403 seems to have originally obtained currency from a simple mistake on the part of Baronius, a mistake fully acknowledged by Tillemont (v. 804).” Hodgkin, i. p. 736.

[2 ]The Additamenta to Prosper in the Cod. Havn. give the date: x. kal. Sept. (Mommsen, Chron. Min. i. p. 299).

[3 ]Forschungen zur deutschen Geschichte, 24, p. 182 sqq. (1884).

[1 ]Mr. Rushforth points out (in a review of this volume in Eng. Historical Review, xiii. p. 132, 1898) that the statement of Zosimus that the threatened invasion of Radagaisus caused a panic at Rome, taken in connection with the restoration of the walls of Rome in 402 (which Gibbon omits to mention), is a confirmation of the view which I have tried to establish that Zosimus is really relating the campaign of 401.

[1 ]Cp. further E. Gleye in Byz. Ztsch. v. 460 sqq., where some other of the Excerpts (esp. fr. 12) are treated in their relation to Procopius, with the same result.

[1 ]He also held a financial post: Seeck conjectures that of a rationalis of a diocese.

[2 ]Further, Castricia, wife of Saturninus, who was banished with Aurelian, had influence with Eudoxia, as we know from Palladius, Life of Chrysostom.