SOME PICTURES IN AMMIANUS MARCELLINUS RAMSAY MAcMULLEN R EADERSo f Ammianus are struck by the pattern which certain scenes of greeting follow (2 . Io, i, etc.), in an author who tries so hard to avoid pattern and repetition: a personage approaches a city; the populace streams out to receive him with torches and flowers; troops and senators cluster around; there are shouts hailing his presence and virtues; and thus accom- panied he enters the city gates. From other sources, however, we know that such occasions, through a very long history, had accumulated features which were indeed patterned, which included also details passed over by Ammianus such as chants and incense, and which were advertised to a larger audience in imperial coinage or in inscriptions set up by the local magistrates.' It would be interesting to find out how deeply into the society of Ammianus' time the impression of such events penetrated, and whether they were imposed by the rulers of the Dominate from above or whether they satisfied more widespread tastes. They are prominent enough in art, notably in relief sculpture; but most surviving art presents us with emperors, consuls, saints, great patrons. It is natural to attribute to the high position of men like these the qualities so often detected in the works they commissioned: grandiloquence, pose, theatricality, and dramatic richness. The present paper, however, pursues these qualities beyond painting and sculpture, into other forms and customs, and into lower levels of society. Its object is to add the support of some social history to the history of art; its conclusion is that, far from being isolated in the upper classes, fourth century art reflected with fair accuracy the enthusiasms and tastes of a popular audience. I. ARCHITECTURE,L ANGUAGE, COSTUME The first question to ask is whether there was much communication or movement of ideas of any sort up and down through the various layers of society. Ancient authors concentrate on the doings only of the higher classes. They rarely dip below a senatorial level. To this point, indeed, evidence is plentiful. Ammianus fixes a sour stare (I4. 6, 9-17; 28. 4, 8-I9) upon those senators who can afford armies of servants mustered in brigades under their officers; upon their carriages, out-riders, and uniforms for the day; upon their rigid receptions and the degrees of condescension offered to the different members of their circle. But in much of this, emperor and noble copied each other. The hem of the senator's mantle that swept the marble floors of Constantine's palace as that senator kissed the imperial purple had lately been kissed by some much humbler client, and if the senator had been ranged among the privileged of the emperor's First Admission, that rank derived from the formalities of Republican levees to which even one's meaner acquaintance i. On the ceremonies of the later Empire, generally studied Westreichs," Bonner Jahrbiicher, CLII, 1952; on ceremonial with reference only to the emperor himself, the literature is architecture, A. Boethius, "The Reception Halls of the Roman now very abundant. The name most often cited is A. AlfSldi, Emperors," Annual of the British School at Athens, XLVI, whose kindness in criticizing this paper deserves special thanks. 1951, among others; on the adventus, H. P. L'Orange, "The His own two classic articles, "Die Ausgestaltung des monarchi- Adventus Ceremony," Late Classical Studies in Honor of A. M. schen Zeremoniells am ro-mischen Kaiserhofe," Mitteilungen des Friend, Princeton, 1955; F. Cumont, "L'adoration des mages Deutschen archiiologischen Instituts, r6mische Abteilung, XLIX, et Part triomphal de Rome," Atti della Pontificia Accademia 1934, and "Insignien und Tracht der ramischen Kaiser," Romana di Archeologia, Memorie, III, 1932-1933; and espe- ibid., L, 1935, are the main foundation for the subject, which cially E. Kantorowicz, "The 'King's Advent' and the Enig- has since been divided and pursued by narrower specialists: matic Panels in the Doors of Santa Sabina," ART BULLETIN, R. Delbriick, for instance, on costume, in his Die Consular- XXVI, 1944. A convenient general r6sum6 is W. Ensslin, "Der diptychen, Berlin, 1929; "Der spitantike Kaiserornat," Die Kaiser in der Spitantike," Historische Zeitschrift, CLXXVII, Antike, vIII, 1932; and "Zu spitrfimischen Elfenbeinen des 1954, PP. I77ff. 436 THE ART BULLETIN were admitted. The aristocracy, at least, could enjoy an elaborate ceremony of which they were themselves the direct heirs, without feeling wholly dependent on some still higher model. They brought it somewhat closer to the common man. From the emperor down at least to a town councillor's petitioners, the same customs were in use. But of course society, however one divides it for analysis, is not really made up of grades and distinctions, nor are its customs really kept in compartments. Was ceremonious behavior, for example, generally accommodated in an appropriate arrangement of rooms? By way of answer, we find a series unbroken from the emperors' reception halls of Spalato, Piazza Armerina, or Constantinople down through the mansions of rich provincials, and so to the houses of the middle class: all indulge in a showier use of marble; the opening out of public rooms at the expense of bed-chambers and servants' quarters; apsidal recesses to give focus to a room; monumental entrances.2 The point cannot be pursued here, so far as architecture is concerned, but it can be supported in various other ways. Take the recently discovered panegyric directed apparently by Constantius to one of his pretorian prefects: Innate virtue holds this extraordinarayd vantagef or tested and faithful men, that when such a man is constantlyo n the alert to promotet he interestso f his emperora nd the republic,t he glory of the thing weighsa s much as the disadvantageosf the life itself and besidesh e is consideredto have soughtf or himself in respectt o fame this recognitiont, hat by merit in the serviceo f his emperorh e has prospereda s a result of industrya nd hard work. If anyonea mong all these men remainsf ixed in our sight and mind-and the felicityo f our age has drawn from fortunea great supplyo f them-Philip would be the outstandingm an whom I rightly proclaimo ur parent and friend.s The self-consciousness, sense of role, and importance which inflate the emperor can be found again and again throughout the period, more fully in the speeches of trained rhetoricians (a good ex- ample, keeping to the reign of Constantius, is Julian's Oration I. 50 C-D), less lengthily in the letters of Symmachus, and so too an unknown and ill-educated Egyptian monk, writing in the 330's: To the most genuine and most enlightened, most blessed, beloved and in God's keeping and filled with the Holy Ghost and most valuedi n the sight of the Lord God, Apa Paieou,g reetingi n our masterJ esus Christ.B eforea ll thingsI pray for prosperityfo r you with the Lord God. This our letter I wrote on this papyrusth at you might read it with joy and with most securep eace from the Holy Ghost and with cheer- fulnessi n God's keepinga nd with entertainmenot f long-sufferingf illed with the Holy Ghost. To you, then, I write, most genuinea nd most securei n the sight of the Lord God, Apa Paieou. . To these two lines of illustration, architecture and language, a third may be added: costume. We are fortunate enough to have an early fourth century procession fresco, showing soldiers, "magistrates, and high functionaries" parading with the emperors, and a mosaic street scene of the next century with people walking about or sitting at tables (Fig. 3)." Here, too, the styles that one can see in use among the great men of the realm prevailed at a lower level, so far as lesser riches, leisure, and sophistication allowed-many-colored clothing, and the embroidered patches and ribbons to be discussed later-giving to Ammianus' world a homogeneity of which he was himself hardly aware. 2. On the tendencies of late Roman private architecture to- was carefully written on stone. ward more grandiosee ffects, see R. Stillwell, "Houses of An- 4. Translation of H. I. Bell, Jews and Christians in Egypt, tioch," Dumbarton Oaks Papers, xv, 1961, pp. 55ff.; I. Lavin, London, 1924, p. 83. For further discussion of the style of "The House of the Lord," ART BULLETIN,X LIV, I962, pp. late Latin and Greek, and of the influence of the capitals like iff.; R. Meiggs, Roman Ostia, Oxford, 1960, pp. 26off.; and Constantinople on the language of the man in the street, see L. Crema, L'architettura romana (Enciclopedia classica, Se- MacMullen, "Roman Bureaucratese," Traditio, xvIII, 1962, zione III: Archeologia e storia dell'arte classica, xII: Archeo- pp. 372ff. logia, I), Turin, 1959, pp. 604ff., especially p. 6o8. 5. U. Monneret de Villard, "The Temple of the Imperial 3. Translation of L. J. Swift and J. H. Oliver, "Constantius Cult at Luxor," Archaeologia, xcv, 1953, pp. 85ff., o5; D. II on Flavius Philippus," American Journal of Philology, Levi, Antioch Mosaic Pavements, Princeton, 1947, pp. 329-32. LXXXIII, 1962, pp. 248ff. It is astonishing when one thinks On clothing, see further below, Section IV. of it, that all this and a great deal more in the same vein SOME PICTURES IN AMMIANUS MARCELLINUS 437 But the three features of fourth century society collected here are chosen not only because they can be traced at several levels, but also because of a characteristic that they share. They may all be said to contain a kind of theatricality. Halls and courts in imperial palaces are deliberately adapted to dramatic appearances-appearances generally of the emperor, from behind a curtain, after the observers have been lined up formally like an audience; or, less often, appearances of consuls or high officers of state, surrounded like their ruler by a bodyguard. The fronts of private villas and the internal arrangements of the better town residences betray a similar purpose, to impress, even at some expense of comfort. A particularly significant liking for that most spec- tacular achievement, the marble stage-backdrop or scaenae frons, grows upon the empire as time goes on, in the first century in frankly ornamental fountains, nymphaea, in private houses, or shown in frescoes and mosaics; then as palace fagades, by Septimius Severus' reign, and so to the palace architecture of late antiquity.' The advantages of color and drama in curtained door- ways are much more fully exploited in the same late period than earlier, especially for churches and the emperor's residence (Figs. I, 2, 4)." As for the hyperbole of panegyrists, it was of course completely conscious, and completely artificial. This we might suspect from common sense. No man can have deserved the praise that Julian offers to Constantius. Our suspicions might be aroused, too, by the very emphasis so regularly placed on protestations of sincerity. Beyond that, we have Julian himself admitting that "this kind of praise is gravely suspected because of those who misuse it, and is considered base flattery rather than trustworthy testimony of noble deeds."' But this being so, we are hard put to imagine the scene of a panegyric. All stood except the target of praise. The rhetor spoke slowly, in a singsong manner, to bring out the rhythm, and with carefully rehearsed gestures, "Uttering platitudes / In stained-glass attitudes." Possibly Julian's mind wandered while Mamertinus addressed to him a text that now spreads over forty pages. And all to what purpose? We can only suppose that the participants in the occasion, quite aware of its falsity, valued themselves for their patience, saw themselves the custodians of an ancient literature, perhaps even savored the skill of the rhetor-but above all, felt themselves part of a tableau vivant. We have many monuments to show this, tombstones and mosaics featuring the deceased among the Muses or in the company of Plato and Aristotle.! There was a pride in such poses of cultivation. Another form of address was also popular: acclamation. The custom shades off into its opposite, malediction, into magic spells common in this and earlier periods, into the army's shouts that hailed a man emperor, savior, bringer of prosperity, and the rest. What is common to these is the repetition of a set of words. The origin of the practice can be traced to Hellenistic audiences cheering their favorite actors or jockeys the custom penetrated in the earliest rhythmically;'(cid:127) 6. The influence of the scaenae frons on other edifices can des Chronographen vom Jahre 354, Berlin, 1888, pl. 35 and be traced through J.-P. Cibe, "Une Fontaine monumentale," pp. 9off.; W. F. Volbach, Friihchristliche Kunst, Munich, Melanges d'archiologie et d'histoire de l'Icole Franfaise de 1958, fig. 152; idem, in Altchristliche Mosaiken, Bern, 1947, Rome, LXIX, 1957, pp. 90off.; P. Grimal and J. Guey, "A pro- p. 14 and fig. 12. pos des 'Bains de Livie,' " ibid., LIV, 1937, p. I53; H. Stern, 8. Oration I. 4 C; cf. 2 C, the orator's art "does not forbid "Nouvelles rech'e rches sur les images des Conciles dans l'fglise flattery, nor is it generally counted a disgrace to the speaker de la Nativite Bethl6en," Cahiers archiologiques, III, 1948, to praise falsely those who do not deserve praise." Augustine pp. 83ff.; F. Wirth, R6mische Wandmalerei vom Untergang recollects with more shame the occasion in Milan "when I was Pompejis bis ans Ende des dritten Jahrhunderts, Berlin, x934, preparing to recite a panegyric on the emperor, wherein I pp. 33ff., 1266; and G.-C. Picard, "Le Septizonium de Cincari," was to deliver many a lie and, lying, was to be applauded Monuments Piot, LII, 2, 1962, pp. 77-92. by those who knew I lied" (Confessions 6. 6). Add some more 7. Curtains in architecture: C. Daremberg and E. Saglio, references to similar occasions in MacMullen, "Roman Bureau- Dictionnaire des antiquitis, s.v. "Velum"; on their use to close cratese," p. 375 n. 43. stage doors, see E. Bethe, "Die antike Terenz-Illustrationen," 9. H. I. Marrou, MOYCIKOC ANHP, Paris, 1938, pp. 8, Jahrbuch des Deutschen archiiologischen Instituts, xvIII, I 903, x38ff., and passim. p. o7 and figs. 5-7; E. R. Fiechter, Die baugeschichtliche Ent- io. Alf61ldi, "Ausgestaltung" (cit. above, note I), pp. 79ff.~ wicklung des antiken Theaters, Munich, 9 4, P. I zo. Curtains further, F. Staehelin, "Felicior Augusto, melior Traiano," Mu- on later doorways can be seen in Stern, op.cit., pp. 94ff.; seum Helveticum, I, 1944, pp. 179ff., and H. U. Instinsky, G.-C. Picard, "Une Schola de collge," Karthago, III, 195x- "Kaiser und Ewigkeit," Hermes, LXXVII, 1942, p. 350, point- 1952, pp. i77ff. and fig. 5; J. Strzygowski, Die Calenderbilder ing to the reflections of acclamation style in the dedications 438 THE ART BULLETIN Empire into the meetingso f the Roman senate" and turns up once more in a session of the town council of Oxyrhynchus."A famous example is the shouts of senators hailing the Theodosian Code: "Augustuseso f Augustuses,t he greatest of Augustuses [repeated eight times]. God gave You to us, God gave You to us [twenty-sevent imes]. As Roman emperors,p ious and felicitous, may You rule for many years [twenty-two times], etc., etc." We have at a still higher level the words dictated to an emperor by an angel, so he said, and distributedi n copies like an actor's script to his soldiers, who obedientlyc alled to the Deity, "GreatestG od, we beseechy ou, Holy God, we beseechy ou, To you we commit all care of justice, To you we commit our salvation, Through you we conquer,e tc." [three times repeated]." But the mob shouted like this too, in Rome, Egypt, or Africa: "May the gods keep you!"") "Rejoice! May you prosper!"t hey wrote on their drinkingv essels, for their toasts," and "May we conquer"w as the cry of the slaves as they entered the master'sh ouse in the evening." Pieces of jewelry had acclamationsin scribedo n them: "God save the wearer!,,"' All this testifies to the very wide diffusiono f a social practice,i n every class and activity.T he emphaticc heers, short phraseb y short phrase,t hat greeted a triumphatoru pon his entrancei nto Rome greeted the eyes of more ordinary folk as they looked above the lintels of their homes. Language itself had become static and stagy. II. THE POWER OF POMP AND GLITTER Yet "static and stagy" fits nothing so well as the imperial entrance,i nto whatever city. No ceremony was more carefully managed, more contrived and impressive, than the adventus. Ammianus'f requentr eferencest o adventush ave been mentioneda bove, but there is one instance (16. to, 2-Io) quite well known and worth quoting fully. Constantius approached Rome from Ocriculum accompanied by a formidable armed retinue, led in battle formation, . . . in order to show an inordinately long parade, banners stiff with gold, and the splendor of his retainers. [He entered] with the standards pre- ceding him on both sides, he himself seated alone upon a golden car shining with the beauty of precious stones of various kinds, which seemed to hold a light of mingled luster. And behind those that preceded him, many other dragons surrounded him, inwoven with purple, fastened to the gold and jeweled spear tips, with their wide mouths open to the breeze and as if hissing thus aroused with rage, trailing their twisted tails in the wind. And on both sides marched a double line of armed men with shields and crests, shining with a shifting light, clad in gleaming scale mail, and here and there the cavalry in full mail whom they call clibanarii, masked, breast-plated, girt with iron belts, whom you might take not for men but for statues of the later Empire, e.g., in H. Dessau, Inscriptiones Latinae over its doors, in his praediis NN . . . et filii . . .. vivant, sene- selectae, znd ed., Berlin, 1954, No. 597. In the Byzantine scant, et meliora perficiant (Corpus inscriptionum Latinarum amphitheater, the claques also waved colored handkerchiefs. [henceforth CIL] 8. 22774), or spes in deo (CIL 8. 21533). ii. Alf61di, "Ausgestaltung," pp. 83-86. The practice goes back at least to the Salve of a Pompeian I2. P. Oxy. (B. P. Grenfell, A. S. Hunt, et al., The Oxy- mosaic, and the greetings of slaves to the same period. Cf. rhynchus Papyri, London, 1898- ) No. 41, late 3rd-early Trimalchio's first and second "shifts," illi quidem exclamavere, 4th century. Vale Gai, hi autem, Ave Gai (Petronius Satyricon 74). 13. Lactantius De mortibus persecutorum 46. 6 and i , the 17. R. Zahn, "Zur Sammlung Friedrich L. von Gans," moment being Licinius' joining of battle with Maximin in Amtlicher Bericht aus den K6niglichen Kunstsammlungen, 313. xxxvIII, 1916, pp. 42-46, the inscription on a decorative disc 14. Scriptores Historiae Augustae (SHA), Severus Alexan- attached to a necklace5 cf. "Good luck to the Tungri," on a der 57. 5; Firmus 9. Ij Gordiani 8. 4. The SHA are particu- gold torque, A. Roes, "Some Gold Torques Found in Hol- larly rich in the texts of acclamations, or alleged ones, a few land," Acta archaeologica, xvIII, I947, P. I79; "Long life to running to several pages, but they occur in other authors Julian," on a gold fibula, CIL 3. 1639; and similar phrases too, e.g., Codex Theodosianus (Cod. Theod.) 7. 20, 3; He- on largesse-dishes of glass or silver, coins or medallions, or rodian I. 4, 8; 2. 2, 4. items of military equipment. On the last, with utere felix or 15. The common words are feliciter or vivas in the early optime conserva, see P. Wuilleumier, "Information de la XVe Empire, gaudeas or floreas later (H. Stern, Le Calendrier de circonscription, Rh8ne," Gall'i a, VIII, 1950, pp-. 46ff., and A. 354, Paris, 1953, p. Ii9). But variants are many: utere felix, Ruhlmann, "Communication propos d'une plaquette de carac- zeses, ave, bibe, etc. tire militaire," Comptes rendus de l'Acadkmie des Inscrip- 16. Amm. 16. 8, 9, exclamasse ex usu 'vincamus', verbum tions et Belles-Lettres, 1935, pp. 69ff. sollemne; similar Gliickwiinsche to protect the house, written SOME PICTURES IN AMMIANUS MARCELLINUS 439 polishedb y the hand of Praxiteles. .. [As for Constantiush imself] he showed himself immobile. . . . . . keepingh is gaze straighta head,n or turnedh is face to left or right, and like a statue of a man was never seen eithern oddingw hen the wheel jolted, nor spitting,n or wiping or rubbingh is face or nose, nor moving his hands. This long passage contains a number of points that deserve discussion. We may begin at the end, with the immobility of the emperors, which they assumed to make more perfect their resemblance to a god, and which was upon need completed by painting their eyes or by adding a wig lest their hair blow.'" Makeup went back to Oriental monarchy, the motionlessness of the emperor to the rules of the stage: "not to sit down if weary, not to wipe away the sweat save with the robe [the performer] wore, to allow no discharge from nose or mouth to be seen" (Tacitus Annals 16.4). Such traditions ended in a curious rapprochement: at the same time that imperial statues were coming to resemble their subjects by being borne about in processions, carried in chariots, wreathed and hailed and addressed as witnesses to oaths, the emperors themselves copied their own statues. They were increasingly forced into an ideal impersonal mold, encompassing the whole list of virtues necessary in a ruler, expressed outwardly in strength of body, in the splendor of their eyes, in their gait and voice, in the serenity of their behavior. Individual differences tended to recede in official representations. Once the commissioned artists had caught the proper character, it was repeated again and again: "The flattened head, low forehead, the enormous muscles of the jaw, great eyes glassy and placed somewhat too far apart, the bull neck, short snub nose, beard with stiff short hair like bristles," recur from one portrait to another, and emperors with a change only of the title re-used the portraits of their predecessors, even of pretenders against them, in coins and busts." In all of these, however, nothing is so striking as their rigidity. Metal or stone seems only to delineate a man made of the same substance, whose expression is fixed in a show of imperturbable omnipotence, and who is addressed as Your Serenity. He could be com- plimented on "the gravity of his visage, the tranquillity of his eyes and countenance."20 And if he responded to the artistic and philosophic impulses of his time, he transformed himself into a sort of icon, for display and adoration. He did not even scratch his nose in public. It is hardly necessary to mention here the quality of frontality, which is spread so broadly across the arts of late antiquity. Theodora and her party (Fig. 2) typify it. They stand in repose, their gaze set on nothing, their motions frozen, their faces full toward the viewer. So also Stilicho, in Ammianus' day (Fig. 5). What the description in Ammianus allows us to imagine, however, is that men who wished to emphasize their own importance did so through gestures and poses exactly resembling those of art. In this period, the fourth century, a more heavily armed kind of elite cavalry troops was com- ing to the fore. They wore scale-mail, less often chain-mail, covering both man and mount. In parades they frequently added masks, gilded, silvered, or bronze. These are what Ammianus i8. M. P. Charlesworth, "Imperial Deportment," Journal 91, and p. 118, on both coins and portrait busts. This feature of Roman Studies, xxxvii, 1947, p. 36. Yet wigs may rather of art begins suddenly at the turn of the 3rd-4th centuries, have served the purpose of giving a very full head of hair, and though there are some later developments, the ideal image which was thought to express a plenitude of power. See H. P. remains remarkably stable throughout the next 250 years. L'Orange, Apotheosis in Ancient Portraiture, Oslo, I947, pp. 20. Panegyrici veteres, Galletier, ed., 7 [6]. 4, 4, referring 33ff., 68ff., 94. Themistius had proclaimed that the good to Constantine. Serenitas is a frequent title, e.g., Cod. Theod. ruler should indeed resemble an &yaXjLa, or rv5aXlza.T he I. 12, 5; 6. 29, 3; cf. L'Orange, Apotheosis, p. 95, on the references and earlier sources for this exeltKr(cid:127)'aV,o rdinary view are rpooadrovy aX71ry1br9 Oeos, the calm that bespoke divine power collected by L. Delatte, Les traites de la royauti d'Ecphante, resident in the philosopher or ruler, recalled in Synesius' address etc., Liege-Paris, 1942, pp. 157 and 216. Some aspects go back on kingship (J. P. Migne, Patrologia graeca, LxvI, col. 1o69). to Plato. Compare his Republic 420 C and Politicus 277 A Tranquillitas is more often ascribed to the empire than to the with Synesius, De Regno 5. io68. emperor, but coins of the 2nd and 3rd centuries carry the 19. P. Castelfranco, "L'arte nella moneta nel tardo impero," legend Tranquillitas Augusti or Augustorum, and Beata tran- Critica d'arte, II, 1937, pp. I3ff., and C. C. Vermeule, "Ro- quillitas is familiar in the 4th century. Gravitas harks back to man Numismatic Art, A.D. 200-400," Numismatic Circular, Lxv, Republican ideals of weighty calm, but it too is ascribed to 1957, pp. 2-6, on the coins; L'Orange, Apotheosis, fig. 66, p. the emperor (Panegyrici veteres o[4]. 9, 5) 440 THE ART BULLETIN calls, in the description just given, clibanarii. The word comes from Persia21 as did this partic- ular style of armor." Clibanarii received much attention in the sources, being something really fantastic and terrify- ing. Ammianus returns to them in another passage (25. I, 12); a Latin panegyrist (Panegyrici vet., ed. Galletier, Io[41]. 22, 4), like Claudian (In Rufinum 2. 357f), dwells on them; Julian (Oration I. 37 C-D; 2. 57 C) twice describes them at length, each time, like Ammianus, empha- sizing their similarity to statues. They contributed to parades and martial ceremonies exactly the quality desired in the emperor himself: dramatic immobility. Beside clibanarii moved troops with resplendent equipment, shields, crests, mail (Amm. 16. Io, 8). These too entered Rome with Constantius. While soldiers like the praetorian guards in the earlier Empire might well have matched the pageantry of these fourth century ones, the latter attracted a growing attention by their greater number, their exotic splendor, and the contrast they offered to the very shabby soldiers of the regular army. They are prominent in the mind's eye of contemporaries. The commander's helmet, arms, his shield, even his horse, are decorated with gold and precious stones, his bodyguards are agleam "with the splendor of gold and colors,"" and wear an increasing weight of jewelry: torques, pendant medallions, rings, bracelets, engraved or stamped with the emperor's image, inscribed "Luck to the bearer," or a gift from the throne."' Such ornament answered an obviously barbarian demand for a wealth more intelligible than coins, especially during the third century when the increasing use of troops recruited from beyond the Empire was matched by the increasing adulteration of the coin- age to pay them. In the fourth century, sculptured representations of soldiers with elaborate decoration are more often barbarians,a nd men (not only women) in excavated graves are buried with increasingly rich, and wholly un-Roman, jewelry. The habit of showy costume spread rapidly throughout society, along with Germanic motifs such as the opposed swans' heads on a belt buckle from Rome (Fig. 8 and note 37 below). Even a private citizen's lowest servants wore torques, fancy fibulae, and embroidered tunics.25 Court fashions were higher. Theodora's attendant wears a big brooch on his left shoulder (Fig. 2), of a type that can be studied more closely in a mid- fourth century fresco from Bulgaria (Fig. 6 and note 25 above), and on Stilicho (Fig. 5). Its exact counterparts are recovered by the archaeologist in Africa, Switzerland, or elsewhere round 21. Meaning "oven," which gives the wearer's point of I am inclined to believe that they are accurate statements of view. On these troops see my article, "Inscriptionso n Roman conditions in the period when they were written, though of armor," American Journal of Archaeology, LXIV,I 96o, pp. little value for the period which they pretend to describe. 30ff.; ibid., passim, for many referenceso n elaborate armor, They are confirmed by the soldiers' uniforms, some with seg- adding H. Klumbach, "Ein r5mischer Legionarshelm aus menta, shown in frescoes and mosaics5 by the gold-and-silver Mainz," Jahrbuchd es R6misch-GermanischeZne ntralmuseums chased helmets of Cod. Theod. 10. 22, 1; and by the evidence Mainz, viii, 1961, pp. 96ff., on helmets mostly of the early referred to in notes 21 above and 28 below. Empire. 24. Gold torques went back to the Republic, borrowed from 22. The precise origin is disputed,b ut it is generally East- Germanic and Gallic usages (Alf6ldi, "Insignien," [art. cit. ern. See H. Seyrig, "A Helmet from Emesa," Archaeology, above, note I] p. 54; A. Biittner, "Untersuchungen iiber Ur- v, 1952, p. 69, and B. Rubin, "Die Entstehung der Kata- sprung und Entwicklung von Auszeichnungen im r6mischen phraktenreiterei,"H istoria, IV, 1955, pp. 264-283. Another Heer," Bonner Jahrbiicher, CLVII, 1957, pp. 133ff., 152ff.) piece of equipmentw ith an Easternh ome was the ring-pommel but are far more frequently noticed in the late Empire, be- used on the hilts of longer swords, introducedi nto the Roman coming by the mid-4th century a standard instrument of coro- army from Severant imes on. See H.-J. Hundt, "Ein tauschiertes nation ceremonies (Amm. 20. 4, 18; 29. 5, 20). Several au- r6misches Ringknaufschwerta us Straubing," Festschrift des thors speak of gifts by emperors of torques, rings, fibulae, and R6misch-Germanischen Zentralmuseums Mainz, III, 1953, pp. armillae to troops (MacMullen, "The Emperor's Largesses," o09-I 18, and on the overall Orientalization of the later army, Latomus, xxI, 1962, pp. i59ff.). One branch of the Roman R. Grosse, "Bewaffnung und Artillerie des spiitrbmischen army got its name from its ornaments, Bracchiati (Alf51ldi, Heeres," Archiiologischer Anzeiger, I-II, 1917, p. 43. "Cornuti: A Teutonic Contingent in the Service of Constan- 23. On the commander's extravagant costume, see Pane- tine the Great," Dumbarton Oaks Papers, xIII, 1959, PP. gyrici veteres Io[4]. 29, 5; Claudian De Consulatu Stilichonis I74ff.), and Roman civilians put on jewelry that bore Ger- 2. 88ff.5 Amm. 23. 3, 6. The quotation on the guards is from manic motifs in decoration (idem, "Eine spiitr6mische Helm- Amm. 31. io, 14. The SHA often mention gifts by the throne form," Acta archaeologica, v, 1934-1935, p. Ix 4; J. Heurgon, to favored officers, upon advancement, and stress the elegant Le tresor de Tenes, Paris, 1958, PP. 4off.). results of this generosity: "fine bright uniforms," "nobly 25. D. P. Dimitrov, "Le systeme decoratif et la date des armed" (Severus Alexander 33. 3ff.5 50. 3), belts with gold peintures murales du tombeau antique de Silistra," Cahiers and gems (Gallieni 20. 3) or gilded (Claudius 14. 5), etc. archiologiques, xII, 1962, pp. 38ff. SOME PICTURES IN AMMIANUS MARCELLINUS 441 the empire."2L aws that restricted the use of jeweled fibulae to the emperor suggest some similar implications of rank in slightly less sumptuous examples. At least they were found only on soldiers and officials entitled to wear the heavy cloak (chlamys), and chlamydati certainly did not include the majority of the emperor's servants. People who commissioned their portraits naturally wanted their importance to be made evi- dent, just as Rembrandt's sitters wanted the lace to show. Thus fourth century reliefs, paintings, and mosaics emphasize details of costume and equipment at the expense of form, a fact often noted by critics and sometimes attributed to a contemporary theory that the reality of detail should prevail over the demands of perspective and proportion." But the true explanation is probably more social and psychological than aesthetic. So a passion for pure display took hold throughout the empire. But there is a curious inter- pretation laid upon this display, wholly typical of the period and echoed in a Latin rhetor (Paneg. vet. Io[4]. 14, 3), the Scriptores Historiae Augustae, Julian, and Ammianus. The first writer, describing an army, mentions how "their shields flashed forth something terrifying, and the awful splendor of their heavenly arms gleamed." This is slightly expanded (Scriptores Historiae Augustae, Sev. Alex. 5o. 3) by the statement that in an army well equipped at all points, well mounted, and so on, an enemy "might recognize the Roman state itself." Julian says more ex- plicitly (Oration I. 23 C), "the enemy were discountenanced [E'KTTTXwrr-LEVObLy] ]t he close order, splendor, and calm" of the Roman troops. When we come to Ammianus, however, we find no less than seven passages ascribing the alarm of the enemy solely to the brilliant appearance of the Roman army.28 Vegetius adds his support in the view that "the radiance of arms carries the great- est terror to the foe."29 This is to attribute to something purely external a force that it cannot have possessed. The belief is puzzling. One must interpret it, I think, in terms of the psychology implied in what has already been discussed-in terms of the value placed upon the almost super- natural power supposed to lie behind a pose, a costume, an expression, a set of words. What arms and armor only represent or imply is thought to reach out against the enemy in some effectual fashion. The point I am making will be supported further in the discussion of uniforms, Section IV. III. ANIMAL DECORATION AND METAPHOR Roman soldiers throughout the Empire put their names on their shields, writing them, or punching the letters on some metal part. By this means, according to Dio, they identified them- selves and their exploits in the very midst of battle."0 If that was indeed their purpose, it was better served by a custom that developed, at least by the later Empire, of painting shields in some heraldic way-diversa signa in scutis ping ebant, ut ipsi nominant, digmata-and perhaps these de- vices can be detected even in the second century as well as later, some like a fleur-de-lis, some harder to describe."3I t seems probable, however, that most of these designs were adopted for their own 26. The so-called Z~qiebelknopffibel, in, for example, R. Preliminary Reports, vi, New Haven, 1936, p. 466, we have Laur-Belart, "SpitrSmische und friihmittelalterliche Griiber," the written mark of a maker or owner; stamped or punched on Ur-Schweiz, xxI, 1957, PP. 7ff. and fig. 6; Heurgon, Trisor tabulae ansatae or umbones in E. Hiibner, "Ramische Schild- de Tnkis, pp. 23ff. buckel," Archiiologisch-epigraphische Mitteilungen aus Oester- 27. The views of Plotinus are discussed by A. Grabar, reich, xI, 1878, pp. ioSff.; and a late example (sth century) "Plotin et les origines de l'esth6thique mddi6vale," Cahiers in J. Werner, "Kriegergriber aus der ersten Hilfte des 5. archeologiques, I, 1945, pp. i8ff. Jahrhunderts," Bonner Jahrbiicher, CLVIII, 1958, pp. 4o6ff.; 28. Amm. I8. 2, 17; 21. 13, 15i 27. 2, 6; 27. 5, 3; 28. 5, on the general practice, consult Grosse (op.cit. above, note 22) 35 29. 5, 15; 31. Io, 9. The enemy pavore torpescent, territi p. 42. Against Dio, probability favors the use of shield in- stetere, etc., because of the Romans' flashing eyes, splendid scriptions simply to protect the articles against loss. equipment, shining standards, etc. Ammianus is not saying 31. The quotation is from Vegetius 2. I8. A neglected text that splendor terrifies by implying military efficiency, but that is clearer (Amm. 16. I2, 6): Alamanni enim scutorum insig- it terrifies by itself alone. nia contuentes, norant eos milites . .. quorum metu aliquo- 29. Vegetius 2. 14, cited by Alf6ldi, "Eine spitr6mische tiens . . . abiere dispersis. Cf. corps identification implied ear- Helmform," p. I 17. lier in Tacitus Hist. 3. 23. As to shield insignia, the earliest 3o. Dio Cassius 67. Io, i. In M. Rostovtzeff et al., Dura examples that I find are on Trajan's column, and then on 442 THE ART BULLETIN sake, for they are hard to fit into a pattern of actual legionary organization even as early as the Trajanic and Antonine columns, and are afterwards exploited for purely ornamental ends. The taste for such bright decoration was not Roman, though whether imported from western auxiliary or from the East is not Shields of the third and fourth centuries unfold scenes regiments clear.32 of the sack of Troy or of battles with Amazons (no heraldry here, at least) or are colored a vivid red, rose, or green, or blue-green with patterns in yellow and red, or borders of alternating red, white, and orange, black and gold, etc."3S ome are shown in the mosaic from Piazza Armerina (Fig. 7). They contribute richly to that splendor of armament thought to terrify the enemy. On some shields were painted opposed goats' heads, giving their name to the soldiers of these units, Cornuti. They were especially favored by the emperors from Constantine on, promi- nent and envied. Another unit which appears first in the fourth century bore the name, "The Lions," clipeoque . . . teste Leones."3 The two together declare the popularity of animal sym- bolism among the western barbarian troops, both being called after their shield device. A. von Domaszewski pointed out that animal insignia in the earlier imperial army are almost all derived from the signs of the zodiac, chosen through horoscopy to represent important dates in the legions' history."3F avorite types are the Bull, Ram, Capricorn, and Lion; the ram and the griffin (sacred to Apollo, hence leg. XV Apollinaris?) appear on helmets in the Antonine column. The griffin is the more easily explained by the eastern home of the legion, and so also, in the third century, is the raising of units' animal signs from a lower place on the legionary standard to the very tip, a development "reflecting the spread of forms of worship from the Orient, in which animal cults had their true home.""3 Later, however, apotropaic animal symbolism comes in through Celtic and Germanic influences." In writers from Homer on, animals are used in comparisons, for the sake of vividness. Warriors are likened to lions, for instance; mythology affords a fairly complete bestiary from which further comparisons can be drawn. In the largely derivative literature of Ammianus' time, such classical comparisons are fairly frequent and fairly easy to detect, along with vaguer references to men who are "beasts" or "monsters"-beluae, O'rppa,o r the like."8B eyond these, late writers draw in the Antonine column. All the others are abstract. Later pic- animal cults and horned helmets of later centuries, see Al- tures can be seen in G. V. Gentili, La Villa Erculia di Piazza faldi, "Cornuti," pp. i72ff., adding the shield shown in E. AB.r mPearcien, a.I Im mosoasiaciic id fi igPuiraaztzia, MArilmane,r in19a5, 9,R opmlse., 2149 5a5n,d p3l0. -3117x 5 SRcihviasftfar adni, a"rcEhienoel oVgiialk ecrrwisatinadnear, uxnxxgIsIz, eit1li9c5h6e, pBp.r o2n4z5efsft. aItut edtatet,e"s W. F. Volbach, Altchristliche Mosaiken, pl. 2; Levi, Antioch ca. 6oo. For lions on various objects in metal such as 4th cen- Mosaic Pavements, p. 73; P. Muratoff, La peinture byzantine, tury belt-buckles, see J. Lafaurie, Le Tresor de Chicy (Suppl. Paris, 1935, pl. 7i H. Stern, "Les Peintures du mausol6e 'de to Gallia, xII, Paris, 1958) PP. 3oIff.; for a dolphin- and l'Exode' a El-Bagaouat," Cahiers archeologiques, xx, 196o, horse-motif on a 4th century buckle, J. M. C. Toynbee, Art pp. i and Volbach, Early Christian Art, New York, in Roman Britain, London, 1962, p. 178. On the general sub- 1961, 1p2flf. .; 89. ject of the revival of animal forms in Western provincial art 32. I presume that the shield-devices on the Trajanic and of this time I have gathered some references in "The Celtic Antonine columns represent painted ones, but the earliest sur- Renaissance," to appear in Historia, xiv, 1965, fasc. 1. viving example in color that I know is on the mural in the 38. "Beasts" or the like as a term of vituperation: SHA Temple of Gadd6 at Dura, a scale design on a gold and red- Maximini 17. Zosimus 2. 47, 5i Pacatus in Panegyrici vete- dish-brown shield, dated to the second half of the Ist century res, Galletier, e(cid:127);d ., 12[2]. 24, 6 Julian Oration I. 38 C; (Dura Preliminary Reports, VII-VIII, 1939, PP. 232, 269ff.). 2. 62 C, the monster of Lerna5 Panegyrici veteres 2 [io]. 4, 3, The Monza diptych of Stilicho (Fig. 5) has a similar scale monstra biformia like the ones Hercules faced; Eusebius Mar- pattern. tyrs in Palestine 2. 16; idem, Vita Constantini 1. 49; 2. I 33. Homeric scenes on shields painted in the year 256, ibid., 3. 66; Athanasius Historia Arianorum 3; 20; and 25 idem, pp. 326ff., 349ff., and plates 41-45; vivid colors and borders, Vita Sancti Antonii 95 Amm. 16. 5, 17; 28. 1, 12. 28. 3, 43 Gentili, op.cit., pls. 24, 30, and 527 Pace, op.cit., pl. 17; Vol- 28. 6, 4; 31. 15, 12. Examples of Homeric metaphors are bach, Mosaiken, pl. 2. There is one mention, SHA Alexander Amm. 19. 3, 3 and 29. 4, 7, leo magnitudine corporis et tor- Severus 50. 5, of special corps with silver and gold shields. vitate terribilis5 Julian Or. 1. 48 C, bS6reprl u& X'KoV; 2. 84 34. Claudian De Bello Gildonico 1. 423. D, cf. Iliad 17. 20; 2. 87 A; 2. 98 C, cf. Iliad 22. 262; Sy- 35. Domaszewski, "Die Tierbilder der Signa," Abhandlun- nesius De Regno 6. 1o69, comparison with the Hydra. Ear- gen zur r6mischen Religion, Leipzig-Berlin, 1909, p. 3, and lier authors exercised a tyranny over the imaginations of their again in E. Petersen et al., Die Marcus-Siiule, Munich, 1896, successors in Latin and Greek, and readers tire of this bor- pp. rowed and mechanical savagery as of other stock types of 361.1 2Dffo. maszewski, Abhandlungen, pp. 12ff. metaphor, nautical (pilots, storms, rudders, favoring winds), 37. On these, on the Cornuti, and on their connection with medical (doctors, wounds, diseases), military (bastions, walls, SOME PICTURES IN AMMIANUS MARCELLINUS 443 a variety not found earlier, and use them with increasing freshness and effectiveness. In Claudian we find, beside the ordinary classicizing metaphors, others that involve an assortment of monkeys, boars, ostriches, and dogs." Ammianus has a still larger collection: bulls, vultures, kites, lions, dogs, which he uses sometimes in extended similes, always with vigor.'0 If we look for the source of the images that come to his mind, we may assume a certain amount of borrowing, from the elder Pliny or the Georgics; a certain amount of ordinary observation of country life; but beyond these, a third source as well, belonging especially to the Roman world, that is, the amphitheater. Often he compares men to "wild beasts with tusks, in their cages" under the stage (19. 6, 4), to "the beasts of the amphitheater" (15. 5, 23; 28. I, 10; 29. I, 27; 31. 8, 9), "like a beast in a hunting show" (venatio, 30. I, 15)-very much as Claudian (In Rufinum 2. 394f) refers to an enemy, in about the same period, "as a beast who has lately left his native mountains and, torn from the high forests, is doomed to the shows of the arena." Scholars are sometimes so scholarly that they forget the obvious, the things that came instantly to the common mind when in search of a phrase to convey excitement, ferocity, color, or danger. The Romans not only spent more public money on their amphitheaters than on anything else-it is usual to find a provincial city putting one up even before its public baths, and sometimes before it could really afford either-but when they could not actually attend a performance, they gawked at the placards in the street advertising the next day's show" or solaced themselves in their own homes with scenes of venationes and combats between animals, extremely common in frescoes and mosaics. They had a robust taste for blood. Valentinian kept two pet bears in cages by his bedroom, which he playfully called Goldflake and Innocence (Amm. 29. 3, 9), and a private citizen specified his favorites, Cruel and Mankiller, written over portraits of bears.'2 These theatrical scenes and gaudy paintings certainly must have made a profound impression on Ammianus and his contemporaries, and help especially to account for comparisons of people with rare animals like monkeys and ostriches. The largest number of Ammianus' animal metaphors concern snakes, serpents, and dragons: an enemy is "like an underground serpent lurking below the hidden entrance of its hole," or "like a viper swelling with its store of poison."'3 It is tempting to find the inspiration for this in, or somehow connected with, the dragon standards which we have already seen accompanying Con- stantius on his tour of Rome. A closely similar passage, to which we will return a second time below, is Themistius' Oration I. 1-2: Most admirers see [not the emperor's soul but see] rather, and sing in their discourses,t hings such as the expanse of the realm, the number of subjects, the invincible regiments of infantry and the troops of cavalry and the great wealth of their equipment and the insuperables creens of weapons and the dragons on the delicate banners, raised high on gilded shafts, filled and shaken by the breeze. The more elegant of those towers, armor, weapons), and miscellaneous old favorites such public and Empire assembled by L. Friedlander, Darstellungen as shepherds, watch-dogs, seasons of the year, rivers, sun, stars, aus der Sittengeschichte Roms, 9th ed., Leipzig, 1920, III, pp. etc., etc., to all of which men or events are likened. The 50ff.); but there are three direct references. Herodian (i. 15, animal and other metaphors in Ammianus to which I draw 4) speaks of animals collected for Commodus' games which attention are, however, much further elaborated than in Taci- "we then saw that we had [earlier] wondered at in pictures." tus, Cicero, or other writers. Such is my own impression, con- SHA (Carus 19. i) mentions games "which we can see painted firmed per litteris by Prof. Geo. Kennedy, whose kind response in the Portico of the Stables, on the Palatine," and Gordiani I must acknowledge. (3. 6), a picture set up in the House of Beaks showing zoo 39. P. Fargues (Claudien, Paris, 1933, PP. 322ff.), gathers stags, 30 wild horses, Ioo wild sheep, 300 red ostriches, etc. the references and distinguishes between the old (lions, horses, The number of animals was probably written on a representa- bees, etc.) and the new animals. The latter he derives without tive of each species, as is done in a surviving mosaic (L. Poins- discussion from images in vulgar speech. M. P. Brown, A,4u- sot and P. Quoniam, "Bates d'amphith6atre sur trois mosaiques thentic Writings of Ignatius, Durham, N.C., 1963, p. ix 2, du Bardo," Karthago, III, 1951-1952, p. 14o), and on gladia- even fixes on a fondness for extended and exotic animal meta- tors' figures on cups and mosaics to show the total of their phors as among the differentia of the mid-4th century Pseudo- victories (J. Gricourt, "Les 'Marques chiffrbes intra-dicora- Ignatius, as opposed to the more limited range of metaphor tives' de la Graufesenque," Hommages a Albert Grenier, in the genuine Ignatian corpus. Brussels, 1962, II, pp. 763-770). 40. Amm. I4. 4, 1ix 14 9, 9; 15. 3, 55 28. 4, Io5 28. 6, I3. 42. Poinssot and Quoniam, op.cit., pp. 144ff. (later 4th cen- 41. It is a safe assumption that these existed, given the use tury). of placards in triumphs and in the light of other public paint- 43. Amm. I4. 5, 65 I4. 7, 13; 15. 2, 4 x5. 7, 4i 18. 4, 4 5 ings described below (see the ample material from the Re- 28. I, 7i 28. I, 33. 444 THE ART BULLETIN speakersc ome a little nearert o yourselfa nd lay hold of your crown and your gleamingr obe, your strong girdle and tunic. Here we meet again the same parade device that struck Ammianus, the dragon banner. By his day it was a familiar sight in the army. Introduced originally from the East," it appears in many Eastern writers: Theodoret, Claudian, Eusebius, and others (but in the Spaniard Prudentius as well)." Eusebius in a rather well-known passage describes how Constantine causedt o be paintedo n a lofty tableta nd set up in the front of the porticoo f his palace,s o as to be visible to all eyes, a representatioonf the salutarys ign [the cross] placedi n the paintinga boveh is head, and below it [Licinius] . . . undert he form of a dragonf allingh eadlongi nto the abyss.F or the sacredo raclesi n the bookso f God's prophetsh ave describedh im as a dragon and a crookeds erpent,a nd for this reason the emperort hus publiclyd isplayeda paintingo f the dragon,s trickent hroughw ith a dart, beneathh is own and his children'sf eet.4" The image recalls one closely similar but far more ancient and widely diffused, that in which some hero on horseback spears a wild beast. In such a posture Claudian imagines Stilicho's son (De cons. Stilichonis 2. 35of) portrayed in gold in a panel of an embroidered cloak (cf. the togae pictae, below, p. 449). Scenes of exactly this kind survive in Coptic textiles, as in ivory reliefs, frescoes,a nd mosaics.T he Coptic examples lead us to the suggestion that there was a symbolic aspect to the hunter's triumph, clear not only in the Babyloniana nd Assyriank ings' trampling upon their enemies, but, more abstractly,i n the victory of Good over Evil. Strzygowskib rings forwarda work of Egyptian art, Horus depictedo n his horse spearinga crocodile."A' n Egyptian already quoted more than once, Claudian( In Rufinump r. 15f), comparest he murderedR ufinus to a poisonous,c oiling serpent." Now that second Python has been killed by the weapons of our master.., .who preservedt he world unshakenf or the brothere mperors[ Arcadiusa nd Honorius] and held sway in peace with justice, in war with energy." But Good victoriouso ver Evil, the one mounted, the other wriggling below, meets us in crudely carved magical amulets from Asia Minor, from Syria, and from Egypt, demonstratingt he prevalenceo f the image in the common mind.'8S o do other animals,n ot only dragonsb ut lions, boars,s tags, representingv ices or enemies, these being interpretedb y both pagan and Christiani n an abstractw ay, in an age devoted to symbolism. Their appeal especially to a popular audience has been emphasized.49 Finally, the serpenti n particulara s the embodimento f Evil enters the literatureo f the fourth centuryt hrough a sourceu niquelyi nfluential,t he Bible, which offersa host of texts and images that fill the authors of East and West, not only with such hunt-motifsa s have been mentioned,b ut with other situa- aux4i4l.i arIite si s ofnilrys,t sbeuetn ino nV eTgreatjiauns'' s dacyo luitm nis, fboournnde ablyso Eaamstoenrng 2I.s ai4a6h. 2T7h:e I sc(rcifp. tuLruakl ea lolu1s:i o1n9s aanrde tPos aRlmev e9l1a:t ioIn1s3 ff2.0):, 2w haincdh the legions. It was known among Germans, Dacians, and Sar- Constantine's allegory closely follows. The Eusebius passage matians (J. Dobial, "Roman Imperial Coins as a Source for is discussed by G. Rodenwaldt, "Eine spitantike Kunststr6imung Germanic Antiquities," Transactions of the International Nu- in Rom," Mitteilungen des Deutschen archiiologischen Instituts, mismatic Congress, r936, London, 1938, pp. I7off.) but its r6mische Abteilung, xxxvI-xxxvII, 1921-1922, pp. 85ff., and more obvious origins were recognized: Persici dracones (SHA by Alf61ldi, The Conversion of Constantine and Pagan Rome, A4urelianus2 8. 5). On its history, see R. Grosse, R6mische Mili- Oxford, I948, pp. 84ff. The picture is repeated in Con- tirgeschichte, Berlin, 1920, p. 23 1. There is only one picture stantine's coinage for wider circulation and again (ibid., p. of it, dating to the 4th century. See H. Stern, "Peintures" 134 n. 28) in Firmicus Maternus De errore profanarum re- (art.cit. above, note 31), pp. 112ff. ligionum 21. 2. 45. Athanasius Vita Sancti Antonii 6; idem, Historia Ariano- 47. J. Strzygowski, Hellenistische und koptische Kunst, rum -8o; Theodoret Historia ecclesiastica I. 14, a political Vienna, 1902, p. 27 fig. 16; p. 35 fig.9g; and p. 37. enemy is "that dragon"; Claudian (an Egyptian) De III 48. G. Schlumberger, "Amulettes byzantins anciens destinbs Consulatu Honorii 138ff., idem, In Rufjnum II i77ff., 346ff., a combattre les malifices et maladies," Revue des itudes Epithalamium 193, De IV Consulatu Honorii 545, De VI grecques, v, 1892, pp. 73ff., especially p. 91; N. Thierry, Consulatu Honorii 566ff.i Eusebius Vita Constantini 2. 1; and "Notes sur l'un des bas-reliefs d'Alahan Manastiri," Cahiers add the two Eastern provincials just cited, Ammianus and archAologiques, xIII, 1962, p. 43. Themistius (but also Prudentius Peristeph. i. 35, Contra Orat. 49. G. Downey, "The Pilgrim's Progress of the Byzantine Symm. 2. 713, Cathemerinon 5.56). The earliest reference to Emperor," Church History, Ix, 1940, pp. 207-2x17 idem, dracones that I know is in the African poet Nemesianus (Cy- "Ethical Themes in the Antioch Mosaics," ibid., x, 1941, pp. negetica 85), of the later 3rd century. 367-376. He finds vices presented as animals, and their con- 46. Vita Const. 3. 3; cf. Licinius again a 8pKc(v, ibid., quest as a venatio, in a 2nd century moral tract, and in later art.