Corippus’ Iohannis is fundamental to understanding the historical ethnography of ‘Berber’ or ‘Moorish’ society in the pre-Islamic North Africa.Footnote 3 Alongside Procopius’ Vandal War, this long Latin poem is one of our only substantial literary sources from the Justinianic period in North Africa and provides invaluable illumination on Byzantine, African and Moorish society in the 530s and 540s. At the start of the second book of the Iohannis, the poet succinctly presents his ‘catalogue’ of Berber tribes – a list of more than thirty groups, which is both the longest list of named African peoples in this period, and the sole testimony of many of these communities (Corippus, Iohannis, II.28–161; Diggle and Goodyear Reference Diggle and Goodyear1970, 28–33; ed. and tr. in Modéran Reference Modéran2003, 44–50 and Zarini Reference Zarini1997, 100–105, 121–4. Tr. Shea Reference Shea1998, 82–6).Footnote 4 This is undoubtedly the richest single source for Moorish ethnography from late antiquity, and consequently has formed the basis for all serious studies of North African society in the period. In the late-nineteenth and early-twentieth centuries, Joseph Partsch, (Reference Partsch1879; Reference Partsch1896), Charles Diehl (Reference Diehl1896), Martin Riedmüller (Reference Riedmüller1919) and Stéphane Gsell (Reference Gsell1929) all used the passage as the basic framework for their discussions of Moorish late antiquity, supplementing it (in approximate order of importance) with material from elsewhere in Corippus, Ammianus Marcellinus, Procopius, epigraphy, Arabic historiography and (most rarely of all) archaeology. Since then, the passage has remained an essential point of reference for studies of the African periphery in a variety of disciplines (cf.; Courtois Reference Courtois1955, 348, n.7; Felici et al. Reference Felici, Munzi and Tantillo2006, 657f; Mattingly Reference Mattingly1994, 173–6; Zarini Reference Zarini1997, 39). Most recently, the same passage provides the starting point for Yves Modéran's seminal Les Maures et l'Afrique romaine, and is locus classicus for his elegant explanatory model regarding the behaviour of Moorish groups of the ‘first’ and ‘second circles’ in the fifth and sixth centuries (respectively, those communities living inside and outside the Roman frontiers) (Modéran Reference Modéran2003, 43–61; Reference Modéran2003b). Prior to the Arab conquest, there is no more important textual source for the study of Moorish or Berber society than Corippus’ Iohannis, and the catalogue in Book II is undoubtedly the crucial part of that text in that regard.
My intention in the present paper is to return to Corippus’ catalogue of tribes, and specifically to examine the poet's likely source for the information that he included within it. In spite of the attention granted to the passage, the origins of this information and, crucially, the form which the poet's source material took, has never previously been considered in detail. I will argue that there is good reason to conclude that this ethnographic material derived ultimately from the triumphal procession held in Carthage in late 548 to commemorate the final victory of the general John Troglita, a climactic moment which is anticipated throughout Corippus’ poem. This proposal is supported by the internal evidence from the passage itself (chiefly the nature of the information presented and the form in which it appears), but also from the wider literary structure of the poem and its political function. Viewed in these terms, the ‘catalogue’ becomes immensely valuable as a source on changing political and social structures of the Moorish world in the 530s and 540s. The passage is not an uncritical snapshot of African ‘tribes’ in this period, as early commentators fondly thought, or simply a reflection of Carthaginian hostility towards the peoples of the periphery, as others have argued, but instead provides a specific view of the defeated Moorish groups from an imperialist perspective.
The Iohannis is a poem (certainly incomplete) of eight books and a little under 5000 lines, written in Latin hexameters probably in the later 540s or early 550s. The poet, conventionally known as Flavius Cresconius Corippus,Footnote 5 was North African in origin, and composed his work to celebrate the victories of the Byzantine magister militum John Troglita over rebellious elements in the army of occupation, as well as multiple Moorish confederations between 546 and 548 (Gärtner Reference Gärtner2008, 58–65). The narrative of the poem recounts three principal campaigns: first a dramatic Byzantine victory over a Moorish alliance under Antalas and Ierna at an unknown battlefield, perhaps in the vicinity of Hadrumetum, which probably took place in late 546 (Dodi Reference Dodi1986b, 592–3; Partsch Reference Partsch1879, xxvi, n.132). Preparations for this battle occupy books II and IV, and the conflict itself is dramatically resolved in Book V. Byzantine celebrations for this success were cut short by a second Moorish offensive, under the leadership of Carcasan which approached the walls of Carthage, before retreating first to Byzacium and then to the regions beyond the frontiers. The disastrous imperial pursuit of these barbarians in Spring 547, and the Romans’ subsequent defeat at Marta in the Tripolitanian pre-desert is recounted in Book VI. The poem closes with a narrative of John's regrouping, and the final Byzantine victory over Carcasan's Moors at the Plains of Cato, an encounter which took place in 548 (Riedlberger Reference Riedlberger2013, 40). The Iohannis ends before this conflict is fully resolved, but the epic is framed as a celebration of ultimate Roman victory throughout.
The Iohannis is certainly epic in scope and structure, and Corippus prominently displays his debts to earlier writers in the tradition. As might be expected, Virgilian aspects are particularly marked, but Corippus was also directly familiar with Lucan, Statius and Silius Italicus, as well as late imperial poets, and also drew upon long-standing Homeric motifs (Zarini Reference Zarini2003). Several commentators have argued that the focus of the poem on contemporary events and the formulaic celebration of its protagonist represent such strong panegyric elements that the work is better read as a generic hybrid, something like an epic-panegyric, not unlike the later Roman praise poems of Claudian, whose work Corippus knew well (Compare Gärtner Reference Gärtner2008, 26–32; Gärtner Reference Gärtner and Goldlust2015; Riedlberger Reference Riedlberger2013, 90–6; Schindler Reference Schindler2009; Zarini Reference Zarini and Goldlust2015). That the Iohannis was intended to evoke the formal ceremonials of victory is clear enough. In his opening lines, Corippus frames his work as a contribution to the celebrations of victory and peace in the heady aftermath of the war:
My Lords: I have undertaken the task to speak of the victor's laurels; I will sing festive verses of peaceful times. (Corippus, Iohannis, Proem 1–2; Diggle and Goodyear Reference Diggle and Goodyear1970, 1.)
Corippus explicitly positions the Iohannis in the atmosphere of Justinian's Carthage in multis … triumphis, but the precise relationship between the poem and the victory ceremonies of 548 is not clear. Jósef Mantke (Reference Mantke1990) argued that the last book of the epic may originally have included a description of the triumph – a conclusion that was lost in the imperfect transmission of Book VIII. More recently, Peter Riedlberger (Reference Riedlberger2013, 83–8) has suggested that the epic was composed for performance at the ceremony itself, and that particular passages may have connected the narrated events to the immediate audience. Most commentators, however, have been content to assume that the Iohannis was shaped by the spirit of the recent celebrations, but may not have taken a formal role within them, and this seems to be the most persuasive interpretation in the absence of further evidence to the contrary (cf. Dodi Reference Dodi1988a, 111).
The theme of triumphal celebration recurs throughout the Iohannis. Throughout the poem, the belligerents on both sides anticipate victory in precisely these terms. For the unfortunate Moors, this is effected through the prophecies of their oracles, in which the leader Carcasan is imagined repeatedly as the focus of a triumphal procession. The narrator is at pains to stress the dramatic irony here: the decapitated Moorish commander will be a trophy in John's triumph, and not the victor himself. (Corippus, Iohannis, VI.166–87; VII.513–542. Cf. Castronuovo Reference Castronuovo, Silvestre and Squillante1997; Mantke Reference Mantke1990, 326–9). For the Romans, of course, these hopes are better founded. Victory and triumph are assured to John through prayer and appeals to divine dispensation, a trust in eventual success that remains with him (and the audience of the poem) even in the darkest days of defeat. (Corippus, Iohannis, VII.314–5; VII.491–4; VIII.294–9; Mantke Reference Mantke1990, 328–30). Amidst the fighting in Book VIII, for example, Putzintulus, a Roman officer, anticipates the victory celebrations with his dying breath. Cf. Mantke Reference Mantke1990, 331; Riedlberger Reference Riedlberger2013, 395:
The gates of Carthage will receive you in the greatest triumph, your strong ranks only missing one. (Corippus, Iohannis, VIII.502–3; Diggle and Goodyear Reference Diggle and Goodyear1970, 184.)
There is even a foretaste of these celebrations in Book VI, when the Roman troops return to Carthage after their victories in 546, and are received warmly by the inhabitants of the city (Corippus, Iohannis, VI.56–103). Although this passage includes many of the constitutive features of a military triumph – including the homecoming of the victorious troops, and a parade of captive Moorish women – this does not mark a definitive conclusion to the fighting. No defeated soldiers are included in the procession, and indeed the fighting starts up again soon afterwards. This is a reminder of the greater victory to come, rather than an endpoint in itself.
Corippus’ celebration of Byzantine victory represented a clear statement of support for the imperial power during a period of prolonged uncertainty (Cameron Reference Cameron and Cairns1984; Dodi Reference Dodi1986a; Tommasi Moreschini Reference Tommasi Moreschini2002; Zarini Reference Zarini and Briand-Ponsart2005). The Justinianic occupation of North Africa had not been straightforward in the 530s and 540s: even after the suppression of successive pretenders, military rebels and hostile Moors, political and religious difficulties continued to blight the administration of the province into the 550s (Cameron Reference Cameron and Cairns1984; cf. Blaudeau Reference Blaudeau and Goldlust2015). In this context, the composition (and perhaps performance) of a Latin epic-panegyric must be regarded both as an assertion of Africans’ loyalty to the Justinianic regime and a contribution to its acceptance among the Latin-speaking elite. The very close ties that bound the Iohannis into the wider discourse of imperial victory deserve closer investigation, especially in understanding the presentation of the Berber or Moorish groups within the poem. In describing his belligerent and malevolent barbarians, Corippus was not simply rehearsing the traditional chauvinisms towards the inhabitants of the African periphery, but adopted the very specific rhetoric of imperial victory. In understanding this, we can look anew at some of the specific details of his account.
The catalogue of tribes
The catalogue of tribes in II.28–161 provides the richest concentration of ethnographic and geographical material in the whole of Corippus’ poem. Ostensibly an order of battle for the first great engagement of the poem in Book V, the catalogue is a wide-ranging survey of the people, places and physical features of the African periphery. This is not always straightforward, however. As the poet himself recognized, the unusual names of the African tribes sometimes fitted poorly into Latin hexameters (Corippus, Iohannis, II.25. Cf. Skutsch Reference Skutsch1900; Zarini Reference Zarini1997, 142). For the modern historian, this problem is compounded by possible errors of transmission, particularly given the survival of the text in only a single, late manuscript. (Modéran Reference Modéran1986; Zarini Reference Zarini1997, 41) Nevertheless, the basic outline of the passage is clear enough. In all, somewhere between thirty and thirty-five groups are listed, each of which is defined either by the collective ‘tribal’ name, the name of their leader, or a nearby geographical landmark. Stripping out Corippus’ various digressive asides, these are:
1. Antalas (leader), who is followed by the Frexes (named group) (II.28–46)
2. Sidifan (leader from the same part of Africa), followed by unnamed horsemen (II.47–50)
3. Sinusdisae (named group) (II.51)
4. Silvaecae (named group) (II.52)
5. Naffur (named group) (II.52)
6. Silcadinet (named group) from tall forests (II.53–5)
7. Unnamed inhabitants of Gurubi montana (II.56)
8. Unnamed inhabitants of the Mercurios colles (‘Hills of Mercury’) (II.57)
9. Unnamed inhabitants of Ifera. Led by Autiliten, son of an unnamed father (II.57–61) (cf. Modéran Reference Modéran2003, 67–70; Zarini Reference Zarini1997, 154–5)
10. Silvaian (named group) (Nomads) (II.62)
11. Macares (named group) who live in huts on the tops of high mountains (II.62–64)
12. Silzactae (named group) (II.66)
13. Caunes (named group) (the River Vadara flows between them and the Silzactae) (II.66–8)
14. Unnamed inhabitants from near Agalumnus mountain (II.69–73)
15. Reference to Macubius mountain (apparently nearby) (II.72)
16. Unnamed inhabitants of the region around Sascar (or a group called the Sascar) (II.74) (cf. Zarini Reference Zarini1997, 159)
17. Astrices (named group) (II.75)
18. Anacutasur (named group) (II.75)
19. Celianus (named group) (II.75)
20. Imaclas (named group) (these last three together) (II.75) (cf. Partsch Reference Partsch1896, 24; Zarini Reference Zarini1997, 160).
21. Unnamed inhabitants from Zersilis (and Gallica) (II.76–7)
22. Unnamed inhabitants from Tillibaris and the Talalatean fields (II.78–84) (cf. Zarini Reference Zarini1997, 160–1).
23. Ilaguas (named group) (from the farthest regions of Libya) (II.85–88)
24. Austur (named group) closely associated with Ilaguas. Ierna is their leader (II.89–112)
25. Ifuraces (named group) (II.113–115)
26. Muctinian gens (named group) from the wastes of Tripoli (II.116–7)
27. Unnamed group from Gadabis (II.117–8)
28. Unnamed group from Digdiga (II.119)
29. Unnamed populi who use ‘Velanidean boats’ and fish in the lakes (II.120–2) (Partsch 1897, 28; Diggle and Goodyear Reference Diggle and Goodyear1970, 31; Zarini Reference Zarini1997, 173–4).
30. Barcaei (named group) (II.123–137)
31. innumerae gentes from Gemini Petra and Zerquilis. Apparently following an unnamed dux (from the other parts of Africa) (II.144–148)
32. Unnamed inhabitants of the Navusum wastes (II.146–7)
33. Unnamed inhabitants of abominable Arzugis (II.147–8)
34. Unnamed group of cavalry from Aurasitania. (II.149–155)
These groups are not all treated equally – the Frexes (group 1 in the list above) and Ilaguas/Austur (23/4) are each granted digressions of eighteen and twenty-seven lines respectively. Corippus also devotes some space to the dress or weapons of the Ifuraces (25), Barcaei (30) and Aurasitanians (34). Other groups are passed over very rapidly, and are included only as a name, or by passing reference to a nearby mountain or river.
In literary terms, the passage is conventional (Zarini Reference Zarini1997, 60; cf. Ehlers Reference Ehlers1980; Gärtner Reference Gärtner2008, 33-40). The most direct textual reference was to the catalogue of Latin tribes who resisted Aeneas and the Trojans in Aeneid, VII.641–817, and this would certainly have been recognized by Corippus’ immediate audience. They may also have appreciated the echo of later poetic catalogues of African gentes which themselves took their inspiration from Virgil (and from one another). The most obvious examples here at Lucan, Pharsalia, IV.666–701 (which describes Rex Juba and his African allies), Silius Italicus, Punica, III.265–320 (on Hannibal), Claudian, De Consulatu Stilichonis, I.248–58 and 351–7 (on Gildo's Moorish allies) and Sidonius Apollinaris, Carmen, V.332–8 (on the Vandal King Geiseric's African followers in his raids on Italy). Corippus himself was aware of all of these authors (with the possible exception of Sidonius), and placed his own work firmly within this tradition. Viewed in these terms, the catalogue of Book II was a straightforward addition to a Virgilian-inflected epic on North Africa.
If these were old bottles, however, the wine was new: while earlier writers may have inspired the form of Corippus’ catalogue, the content seems to have been the poet's own. Among more than thirty groups listed, very few are known from other sources, and none are familiar figures who recur in classical geographical accounts of Africa. Imperial accounts of the African periphery recycle the same small handful of conventionalized tribal names – Aethiopes, Nasamones, Garamantes and so on – with little concern for the precise identities of the groups concerned (Modéran Reference Modéran2003, 51). This is the pattern that we see across the Latin poetic tradition, and which recurs even in much historiographical and geographical writing. Indeed even Corippus himself adopts the same approach in a shorter catalogue later in his poem (cf. Riedlberger Reference Riedlberger2013, 46):
other gentes joined them, and not only those who had already waged war: the cruel Nasamon who ploughs the Syrtic country; he who cultivates lands near the Garamantes, and those who drink the waters on the banks of the Nile. Who could list these peoples, or number them? (Corippus, Iohannis, VI.196–201; Diggle and Goodyear Reference Diggle and Goodyear1970, 120.)
By contrast, the list in book II is much longer, eschews such familiar figures and includes a bewildering variety of ethnographic and toponymic novelties, including many proper names that are not otherwise known.
The few clear correspondences between Corippus’ list and our other sources suggest that Corippus’ material was based on authentic knowledge of the frontier regions of imperial North Africa. His references to the belligerent inhabitants of Gemini Petra and Zerquilis, for example, are certainly to be identified with the Geminianou Petra and (perhaps) Zerboulé which Procopius identifies as Numidian strongpoints, besieged by Solomon during his campaigns in the Aurès in 539. (Procopius, Vandal War, II.2–23; II.19.19–20. cf. Zarini Reference Zarini1997, 179). Similarly, Corippus’ allusion to the ‘horrid land of abominable Arzugis’ (Iohannis, II.147–8: horrida tellus| Arzugis infandae) evidently refers to a region around the Chott El-Djerid that is also known from the letters of Aurelian of Carthage, Augustine of Hippo and the geographical sketch of Orosius, but was scarcely a commonplace among the learned geographers of the Mediterranean world (Modéran Reference Modéran2003, 364–74). The allusion to Gurubi montana (‘Mount Gurubis’) has plausibly been connected to the mons Gurbessa or the Curbissenses gens which the late Roman geographer Julius Honorius places to the south of the Roman territories, probably in the frontier regions of Byzacena (Julius Honorius, Cosmographia, 42, 48. Partsch Reference Partsch1896, 22; Modéran Reference Modéran2003, 77–112). Corippus’ Astrices may be identical to Ptolemy's Astakoures, and his Imaclas (rendered as Mecales in Iohannis, III.410) have parallels in the pages of Herodotus and Ptolemy. (Herodotus IV.178; Ptolemy, Geographika. IV.13. Partsch Reference Partsch1896, 24; Riedmüller Reference Riedmüller1919, 20). Other names are unknown in their present form, but plausibly adhere to the conventions of Libyan (or Berber) onomastics, particularly once problems of transmission have been considered. The Silvacae, Silvaizan, Silzactae and Silcadenit are a vivid case in point: we may not encounter these groups elsewhere in the literary or epigraphic record, but they sound like plausible Berber tribal names, especially when presented together (Modéran Reference Modéran2003, 107–9; Partsch Reference Partsch1896, 23; Zarini Reference Zarini1997, 153). The mountains Agalumnus and Macubius and the River Vadura, similarly sound convincing as local toponyms, even if they have not been definitively identified (Modéran Reference Modéran2003, 70–4; Zarini Reference Zarini1997, 157–9).
Corippus’ catalogue is clearly divided into three sections, separated by authorial lamentations or apostrophes to God. Scholars have worked from the small handful of identifiable groups or places in each section of the list, and have extrapolated wider patterns from them, with predictably mixed results. Early commentators argued that these divisions were essentially geographical: the first section (lines 28–84, or groups 1–21 in the list above) were thought to relate to Antalas and the Moors of Byzacium, the second (lines 85–138, groups 22–30), to the inhabitants of Tripolitania and the pre-desert beyond, and the third (lines 140–161, groups 31–34) to Numidia and the Aures (Courtois Reference Courtois1955, 348, n.7; Partsch Reference Partsch1879, viii–ix; Partsch Reference Partsch1896, 21–8; Riedmüller Reference Riedmüller1919, 15). Following this reasoning, Pierre Morizot suggested that the relative prominence of the Libyan barbarians in Corippus’ catalogue may betray a substantial demographic shift on the southern frontier, a position that continues to gain adherents (Morizot Reference Morizot1991: cf. Fentress and Wilson Reference Fentress, Wilson, Stevens and Conant2016; Merrills Reference Merrills2018; Modéran Reference Modéran2003, 55–6). More recently, Yves Modéran proposed that the divisions in the passage were determined by political affiliation, rather than location, and reflected the relative relations of different Moorish groups towards the empire. In this reading, Antalas and his immediate followers were Moors of the ‘first circle’, who lived within the frontiers of the empire in southern Byzacium and western Tripolitania, and who consequently enjoyed long-standing political relations with the imperial authorities in Carthage (Modéran Reference Modéran2003, 63–7). Those listed in lines 85–138 and 140–161 were ‘second circle’ Moors who lived beyond the frontiers, in eastern Tripolitania or southern Numidia and who were much less closely integrated into imperial systems of government. Their contact with the empire was more occasional, or occurred through seasonal transhumance, and they were thus a distinct category of outsider (Modéran Reference Modéran1986, 211; Modéran Reference Modéran2003, 58–61 and 109; Zarini Reference Zarini1997, 143ff, 163ff; Zarini Reference Zarini and Briand-Ponsart2005, 413–5). Modéran's interpretation of the passage has much to recommend it, not least as a catalyst for his own magnificent interpretation of the wider patterns of Moorish history, but it does not fully explain whether Corippus’ audience is likely to have understood this structure in quite the same terms as the brilliant French historian.
None of the explanations for the detail and structure of Corippus’ catalogue seem persuasive. It has been suggested that the poet drew upon his personal knowledge of North Africa in composing this section of the Iohannis (Baldwin Reference Baldwin1978, 372; Riedlberger Reference Riedlberger2013, 36; Riedmüller Reference Riedmüller1919, 6–8). While Corippus was certainly brought up in Africa, and may well have come from a rural region, his precise origins are far from clear, and it would be a stretch to infer that detailed ethnographic knowledge of this kind was commonplace in the African hinterland (cf. Corippus, Iohannis, Proem, 25; Diggle and Goodyear Reference Diggle and Goodyear1970, 2: quondam per rura locutus. cf. Riedlberger Reference Riedlberger2013, 36). Others have argued that Corippus had access to the campaign records of John Troglita, and may well have derived his ethnographic information from that source. (Cameron Reference Cameron and Cairns1984, 169; Riedlberger Reference Riedlberger2013, 88–9; Zarini Reference Zarini1997, 43–4). That Corippus had some knowledge of the principal events of John's campaigns, and a confident mastery of a range of Berber group names and toponyms is clear from the Iohannis as a whole, but these are only partial solutions to the peculiar form of the catalogue in Book II, not least because it is hard to see why military reports would have been structured in this way (Cameron Reference Cameron and Cairns1984, 169; Gärtner Reference Gärtner2008, 66–127; Zarini Reference Zarini1997, 43–4). We also need to consider the extent to which the details of this passage would have been meaningful to Corippus’ audience. While the overt display of detailed ethnographic and geographical information had both political and literary valence – a demonstration both of Corippus’ mastery over his material, and the resurgent empire's domination over the peripheral regions of Africa – such rhetoric depended in some sense upon the resonance of this information with the intended audience (Modéran Reference Modéran1986, 211).
Far from being a straightforward Moorish order of battle, or a systematic survey of the local tribes of North Africa, Corippus’ catalogue is a puzzling amalgam. Particularly remarkable are the obscurity of many of the names, (in the sixth century as well as the twenty-first), the intermingling of different types of material – leaders, peoples, rivers, mountains and so on – and a somewhat eccentric system of ordering, which has some geographical coherence, but does not seem to have been pursued consistently. None of these features has clear antecedents in extant late antique geographical texts, (which tend to group geographical features by type), historiographical or poetic catalogues (which typically fall back on familiar ethnographic topoi, and are rarely as full as the list presented here), and none of them seems entirely appropriate to a poetic order of battle in the traditional sense.
If we set the details of Corippus’ ethnography to one side, we can start to detect a more convincing explanation for the origins and function of this passage. In all, Corippus identifies only eighteen gentes by name. Five more are defined in part by their leaders (named or unnamed), although two of these groups are also named in their own right (Antalas’ Frexes and Ierna's Ilaguas). Most remarkably, Corippus lists fifteen further groups who are not named but who are defined instead by a geographical feature (primarily specific mountains, forests and rivers, but also wider regions). What remains, then, is not an ordered and systematic list of named gentes, but an extraordinary jumble of places, people and groups. While the catalogue in Aeneid VII provides some precedent for this, Virgil's treatment of named commanders is much more systematic, and his use of geographical referents is largely to lend drama to his account, rather than to stand in for ethnographic detail. In Corippus’ account, mountains, streams and lakes are included alongside named groups and leaders in seemingly ad hoc fashion, transformed into an order of battle only by the dubious association of these features with anonymous inhabitants living nearby. It is this that provides us with the most convincing explanation for the origin of this material.
A very close parallel to Corippus’ jumble is to be found in the Elder Pliny's account of the trophies carried in the triumph of Cornelius Balbus in 19 BCE, preserved in the fifth book of his Historia Naturalis. As it happens, this relates to the same part of the world as Corippus’ catalogue – the territories that would later become southern Numidia, Byzacena and the Tripolitanian interior – but the more significant point of correspondence is structural:
… It is wonderful that our authors had recorded the above-mentioned towns captured by [Balbus], and in addition to Cidamum and Garama, the names and representations of the other peoples and towns were led forth and in his triumph, and came in this order: the town of Tabudium, the natio of Niteris, the town of Miglis Gemella, the natio or town of Bubeium, the natio of the Enipi, the town of Tuben, the mountain called Niger, Nitibrum, the town of Rapsa, the natio of Viscera, the town of Decri, the river Nathabur, the town of Thapsagum, the natio of Tamiagi, the town of Boin, the town of Pege, the river Dasibari; then following towns: Baracum, Bulba, Halasit, Galsa, Balla, Maxalla, Cizania; and mons Giri, and a titulus stating that precious stones came from there. (Pliny, Historia Naturalis, V.36–7; Desanges Reference Desanges1980, 63.)
Pliny states explicitly that this information was taken from contemporary records of the triumphal parade, and so we can be confident that the list presented here corresponded to the order in which these trophies were presented in procession. No defeated leaders are named in this list, but such information would scarcely have fitted with Pliny's immediate concern with the physical geography of the region. The remainder of the information otherwise anticipates Corippus’ quite closely in both form and content. Here we see nationes (Niteris, the Enipi, Viscera and so on), towns (Tabudium, Miglis Gemella), mountains (Niger, Giri) and a river (Nathabur). Careful analysis of the text reveals a jumble of ethnonyms and toponyms that may be traced to the Tripolitanian coast, the pre-Sahara and southern Numidia, and hence represented the trophies of two separate campaign seasons in 20 and 19 BCE (Desanges Reference Desanges1980, 394–410; Merrills Reference Merrills, Mugnai, Nikolaus and Ray2016). Yet if these places were evidently not organized geographically, there was still an order of sorts in the parade. The particular prominence of Cidamum (Ghadames) and Garama at the head of the list suggests that they were also focal in the procession itself, and indeed this emphasis on ‘headline’ victories is common in other accounts of classical triumphs (Merrills Reference Merrills2017, 78–86). Typically, the most impressive (or best known) geographical trophies would be paraded first (and would draw the attention of spectators), and the remaining peoples, places and things follow behind without clear geographical order, but in sufficient profusion to impress the audience, as was evidently the case here.
Pliny, of course, wrote almost five hundred years before Corippus, and was describing a ceremony that was itself almost a century old at the time of his composition, but a scattering of sources illuminate the likely form that this triumphal display took, and suggest that similar patterns of commemoration survived into the later Roman and Byzantine empire (Börm Reference Börm2013; McCormick Reference McCormick1986). In early imperial triumphs, ethnic groups and geographical features were represented in a variety of ways, from elaborate models, tableaux or personifications to written placards (tituli), all of which might be further glossed with the aid of claquers in the crowd (Merrills Reference Merrills2017, 69–105; Östenberg Reference Östenberg2009). In later centuries, victory ceremonies were celebrated in provincial capitals, as well as the imperial centres, but some of these conventions seem to have survived (Ammianus Marcellinus, 29.5.56; cf. Symmachus, Epistula 37 and McCormick Reference McCormick1986, 231–59). The parade of victorious soldiers remained a focus, as Corippus’ account of John Troglita's first triumph in Carthage suggests:
All marvelled when gazing at the battle standards, at the clothing made stiff with dust, and at the men made fierce by the grim slaughter. They gazed at breastplates, helmets, shields and threatening swords, sword-belts, bridles, resounding bows and quivers, and spears made purple with Massylian blood. (Corippus, Iohannis, VI.76–82; Diggle and Goodyear Reference Diggle and Goodyear1970, 116.)
Prisoners were the other mainstay of Roman victory celebrations. While the triumph described in Iohannis VI represents the defeated barbarians only through the abject representation of captive Moorish women (and thereby underscores the incomplete victory and the premature nature of this triumph), other sources of the period show the defeated enemy were a common feature of parades (Corippus, Iohannis, VI.56–103; contra Cameron Reference Cameron and Cairns1984, 172; cf. McCormick Reference McCormick1986, 87–99; Riedlberger Reference Riedlberger2013, 86; Sannicandro Reference Sannicandro2012, 471–5) There are occasional hints that other forms of display also survived into late antiquity, although here our evidence remains slight. Ambrose hints at such in sermons written in the aftermath of Theodosius’ triumphs of the 388, where he alludes freely to placards which identified the defeated enemies as well as models of cities. (Ambrose, Expositio in Lucam, 10.109–11; McCormick Reference McCormick1986, 87). Our fullest late antique source for this is perhaps Claudian's evocative account of the triumphs that Stilicho might have celebrated had he so wished:
He himself would be drawn in a laurel-decked chariot by white horses, and be followed by his soldiers chanting their festive songs. Some would lead captive kings, others carry conquered towns wrought in bronze or mountains or rivers. Here would go in sad procession the river gods of Libya, their horns broken, there Germany and the Rhine God in chains. (Claudian, De consulatu Stilichonis, III.20–25, tr. Platnauer Reference Platnauer1972, 44–5.)
Here we do read of captive kings, alongside towns, rivers and places, along with some hint of how these representations might look.
We know much less about the choreography of John Troglita's triumph in 548. It seems likely that his final celebrations would have marked the conclusion of a long series of campaigns, (as was the case with Cornelius Balbus), rather than the fighting of a single season. Given this, we might expect that the parade would have afforded a prominent role to the major Moorish antagonists – Antalas and the Frexes, the Laguatan and Austuriani – but might also have incorporated earlier Byzantine victories (such as those of the 530s over the Moors of Numidia). The procession would be filled out with captives, trophies and tituli commemorating other peoples, towns, territories rivers and mountains encountered in these campaigns, and intended to reflect the magnitude of the imperial victories, rather than provide precise geographical information.
If such a procession were rendered in verse, it would look more or less like the catalogue of tribes in Iohannis II. Rivers, towns and mountains, which would have been common features in a triumph, but would have little obvious place in an order of battle, are rendered fit for purpose by the simple process of alluding to unnamed tribal groups who lived nearby. Corippus presents this information as preparation for war, of course, rather than as a straightforward description of the victory ceremony, but in so doing he lends a profound irony to the hubristic gathering of the Moorish tribes which is entirely in keeping with the tone of his epic. As the barbarians of the poem purport to gather in order to fight, they take the recognizable form of the warriors whom Corippus and his audience had seen pass, defeated through the streets of Carthage. In many ways, the passage can be read as a poetic and imaginative response to the triumphal procession itself: this was precisely the sort of reflection that the ceremony was intended to encourage.
The Moorish leaders included in the catalogue also support this interpretation, and may again have been inspired by the Carthaginian triumph of 548. The Iohannis as a whole is well-populated with heroic figures, including more than thirty named Moors. Many of these individuals perish in the stylized combats described in Books V and VIII, and aristeia of this kind typify Corippus’ battle narratives. A true dramatis personae for this fighting would introduce these figures to the audience, and indeed elsewhere the poet does precisely this. At the close of Book IV, for example, Corippus list sixteen Moorish leaders, five of whom are killed in the battle that follows and eight more are depicted in combat (Corippus, Iohannis, IV.619–44). Yet the catalogue in Book II bears little clear relation to the conflict that follows. Only four leaders are named there – Antalas, Sidifan, Ierna and Autiliten – and while all are later shown in combat, three of them may have survived the campaigns that followed.Footnote 6 This truncated account thus reads less as a prolegomena of the bloodshed to come, and more as a list of the most prominent survivors of this conflict – precisely the individuals, in other words, who would have featured in the Carthaginian triumph.
There are two exceptions. Ierna was certainly killed during the campaign: his death forms the emotional climax of Book V, and he could not have appeared in person in the triumph (Corippus, Iohannis V.519–21). It is conspicuous, however, that Ierna is not presented in the same way as his Moorish allies within the catalogue. Rather than place Ierna at the head of a group (as he does Antalas or Sidifan), Corippus instead includes him only as a brief aside in line twenty two of his twenty-five line account of the Ilaguas and Austur. Here, it is the peoples themselves who are the focus of attention, and not their (lost) leader:
Ferocious Ierna was their leader, and was the priest of Gurzil. (Corippus, Iohannis, II.109; Diggle and Goodyear Reference Diggle and Goodyear1970, 30.)
The more striking case is the unnamed dux who appears in the third section of the catalogue:
Now an audacious leader rose up from the other side of Africa, seething at the ruin that Roman arms had earlier given him through carnage: so great was the anger that the savage now poured forth! (Corippus, Iohannis, II.140–43; Diggle and Goodyear Reference Diggle and Goodyear1970, 32.)
Corippus’ reluctance to identify this dux seems peculiar, given his careful use of proper names elsewhere in the catalogue. The references to Gemini Petra and Zerquilis that follow have led commentators to assume that the leader was from Southern Numidia – ‘the other side’ (aduersa de parte) of Africa from the Tripolitanian groups described in the preceding lines (Corippus, Iohannis, II.141; Diggle and Goodyear Reference Diggle and Goodyear1970, 32; Partsch Reference Partsch1896, 28; Riedmüller Reference Riedmüller1919, 26–7; cf. Zarini Reference Zarini1997, 141). It is possible, however, that Corippus was alluding here to Carcasan, a member of the alliance in 546, the leader of the Moorish coalition in 547 and 548 and the principal antagonist of the latter stages of the Iohannis. According to Corippus, Carcasan was among the Moorish leaders in 546, but the motivations ascribed to the dux here fit more closely with the second rising in the Spring of 547 – at least as they are presented later in the poem. If this dux was intended to evoke Carcasan, the catalogue could be read as a stylized list of the major belligerents faced in three years of campaigning, along with some acknowledgement of the causes of the war. Such a list would serve poorly as an order of battle, but would correspond closely to the likely form of a triumphal ceremony.
Like Ierna, Carcasan was killed in combat, which may explain his euphemistic presentation here, but he was certainly represented in the triumph that marked the end of the campaign (Corippus, Iohannis, VIII.627–34). Carcasan's misunderstanding of a Moorish prophecy that he would appear in a triumphal parade is a recurring motif throughout the Iohannis (Corippus, Iohannis, VI.166–87; VII.513–542. But cf. VII.506–7; Mantke Reference Mattingly1990, 329–30). Corippus’ statement that the leader's severed head was borne as a trophy as part of these celebrations seems to allude to a specific event, and doubtless one that would have been chillingly memorable to the inhabitants of Carthage. It is far from certain that the unnamed dux was Carcasan, but it is certainly possible. It seems likely that John's final triumph in 548 afforded comparable prominence to Antalas, Ierna and Carcasan, even if only the first of these had survived to take part in the procession. As a celebration of the final victory over multiple campaigns, the triumph could not be translated directly into an order of battle dated to 546, but Corippus made the minimal changes necessary to make his catalogue both consistent from a narrative point of view, and comprehensible to an audience who well remembered the dramatic celebrations within the city.
The explanation proposed here is that Corippus’ ethnography was drawn from a triumphal display which would have marked the successful conclusion of John Troglita's long period of campaigning in North Africa. This interpretation explains the form and content of the catalogue as it stands, but also fits neatly into the narrative structure of the poem, and its wider social and political function. In evoking the ceremony which marked the final defeat of the Moors at the very moment of his poem when belligerent barbarians are first introduced, Corippus cast their hubristic revolt with a profound irony – a dramatic mode which he deploys extensively throughout his poem. With the imagery of the procession vivid in their minds, Corippus’ audience would certainly have appreciated this allusion. Triumphs were a mainstay of commonplace geographical knowledge in the early empire, and there is good reason to think that this continued into the early Byzantine period, but their salience was perhaps less straightforward than is sometimes assumed.
Conclusions
Corippus’ accounts of the Moors in the Iohannis drew upon a range of different influences, political as well as literary. It is too simple to dismiss his descriptions of African groups as well-worn stereotypes regarding the rural fringes, although he did deploy such topoi elsewhere in his account. Nor can we regard the Iohannis as a simple reflection of imperial ideology, contrasting Justinianic, Christian order with pagan Moorish chaos, although these oppositions too are apparent in his poetry. A full appreciation of the Iohannis demands a sensitive appreciation of the different media with which Corippus was working, and against which his poem would be interpreted. The military triumph of John Troglita in around 548 represented a particular manifestation of imperialist ideology, which Corippus was able to appropriate within the second book of the Iohannis. The triumphal ceremony almost certainly included ethnographic and geographical details, which would have been tantalisingly obscure to the poet and his audience, and which could scarcely have become common knowledge within Carthage in any other way. By repurposing this material in the form of a Moorish order of battle, Corippus both lent his narrative a verisimilitude and charged the epic as a whole with a dramatic irony. The poem that resulted doubtless contributed to the rhetoric of imperial victory in the years after 548, but it was the combination of different elements – poetic and political – which made this contribution so effective.
This interpretation of Corippus’ sources substantially complicates our use of the catalogue in Book II as a source for Moorish ethnography and history in this period. There can be little doubt that North African society in the early-sixth century underwent dramatic change (Modéran Reference Modéran2003, 510). The shrinking and subsequent collapse of the Vandal kingdom in Carthage, the Byzantine occupation of 533-4 and the prolonged conflict that followed, would all have had a catalytic effect upon social and political identities in the African provinces. If our principal literary sources provide only a distorted perspective on this world, the inconsistency of their attempts to classify Moorish society hints at the sheer variety that they encountered. Corippus, as we have seen, defines the Moorish world in terms of leaders, tribal groups and confederations, as well as falling back on archaic terms – Mazax Massylies, Mauri – to embrace the whole (Partsch Reference Partsch1896, 21–2; Riedlberger Reference Riedlberger2013, 45–6). His contemporary Procopius includes almost no ‘tribal’ names within his account of Africa, preferring the generic Mauroisioi to refer to barbarian groups, and distinguishing among them primarily through reference to their leaders or their points of origin (Modéran Reference Modéran2003, 304). Corippus and Procopius – and our other fragmentary sources from the period – all reveal the fluidity of the boundaries between ‘Moor’, ‘Roman’ and ‘African’, and describe individuals and populations who might identify with their local town, with the old empire or with their local church as much as with any essentially ‘tribal’ affiliation (compare Castronuovo Reference Castronuovo, Silvestre and Squillante1997 with Conant Reference Conant2012).
Corippus’ catalogue may seem to offer a more secure picture of Moorish tribal identities, but this needs to be treated with caution. For more than a century, scholars have displayed remarkable erudition in connecting the tribal names and toponyms of Iohannis II to other group names preserved for us in Ptolemy, the Arabic historiographers or the epigraphy. This work reveals particularly clearly the extraordinary longevity of some collective names, and perhaps group identities, from the classical to the medieval period, and this is certainly important. But these correspondences should not be taken as evidence for a stable network of ‘tribal’ identities, sublimated in the Roman period, and resurgent thereafter. Corippus’ list is not primarily a list of tribes, but rather a jumble of leaders, peoples and placenames, gathered together first by the conventions of imperial triumphal display and then adapted to the literary requirements of an epic poem. Both of these rhetorics – ceremonial and poetic – were predisposed to represent the Moors in terms of military or tribal groups bearing exotic names. This is not to suggest that these lists did not bear important correspondences to certain social and political constellations within the African world – particularly in times of war – but we cannot view Corippus’ as a timeless ethnography of the African world, or as a straightforward reflection of its dominant social organization.
Ultimately, Corippus cannot offer us the disinterested glimpse of Moorish North Africa in the sixth century that we so desperately need. But the catalogue that he does provide, shaped by the formal conventions of imperial triumph, at least provides some suggestion of how the empire represented this world, in all of its social and political complexity.
