AssoCiAZioNe NAZioNAle Per AquileiA ANNo lXXXVi 2015 COPIA ELETTRONICA IN FORMATO PDF RISERVATA AD USO CONCORSUALE E/O PERSONALE DELL’AUTORE CONFORME AL DEPOSITO LEGALE DELL’ORIGINALE CARTACEO 1 «AquileiA NostrA» © 2015 AssoCiAZioNe NAZioNAle Per AquileiA Autorizzazione del tribunale di udine n. 294 del 19-2-1972 Via Patriarca Popone 6 - 33051 Aquileia (uD) tel./fax ++39/431/91113 - e-mail: assaquileia@libero.it issN: 0391-7304 Direttore responsabile: Monika Verzár Comitato scientifico: Jean-Pierre Caillet (université Paris ouest), Giovannella Cresci Marrone (università Ca’ Foscari Venezia), reinhard Härtel (Karl-Franzens-universität Graz), Francesca Ghedini (università degli studi di Padova), robert Matijašić (Sveučilište Jurja Dobrile u Puli / Università “Juraj Dobrila”, Pola), Andrea Saccocci (Università degli Studi di udine), Marjeta Šašel Kos (ZrC sAZu- research Centre of the slovenian Academy of sciences and Arts - institute of Archaeology), Monika Verzár (università degli studi di trieste), Paul Zanker (scuola Normale superiore di Pisa) Comitato editoriale: Maurizio Buora, Paola Càssola Guida, Giuseppe Cuscito, Marta Novello, Monica salvadori, Paola Ventura, luca Villa Segreteria: elena Menon Peer-review: gli articoli e le note inviati per la pubblicazione ad «Aquileia Nostra» vengono sottoposti, nella forma del doppio anonimato, a peer-review di due esperti, di cui uno esterno al Comitato scientifico e alla Direzione. in copertina: cartolina che mostra la sala ii dell’Imperal-regio Museo dello Stato in Aquileja (con in evidenza le statue di Claudio e di Augusto, trovate nel 1879) e uno scorcio della precedente; non viaggiata e stampata a cura dello stesso Museo, è databile agli ultimi anni del XiX secolo (collezione privata, trieste). il presente volume è stato pubblicato in collaborazione con Ministero dei Beni e delle Attività Culturali e del turismo soprintendenza Archeologia del Friuli Venezia Giulia Polo Museale del Friuli Venezia Giulia la rivista non assume responsabilità di alcun tipo circa le affermazioni e i giudizi espressi dagli autori. le riproduzioni dei beni di proprietà statale sono state effettuate su concessione del Ministero dei Beni e delle Attività Culturali e del 2 turismo. riproduzione vietata. iNDiCe Giuliana Cavalieri Manasse, Marisa riGoni, Ricordo di Maria Josè Strazzulla ............................. p. 7 CeleBrAZioNe Del BiMilleNArio Di AuGusto e Dei 2000 ANNi Di EMonA (lJuBlJANA) Božidar Slapšak, emona augustea fra emporio e colonia ................................................................ Bernarda Županek, The 2000-year Anniversary of the City of Emona: Urbanism, Ideology and Everyday Life ...................................................................................................................... katharina Zanier, L’Istria nord-occidentale in epoca augustea ...................................................... palma karković takalić, iader e l’imperatore Augusto .................................................................. monika verZár, Augusto ad Aquileia e tergeste .............................................................................. » » 13 21 » » » 35 49 67 » 81 » 99 » » 111 141 » 153 MisCellANeA andrea raffaele Ghiotto, Giulia fioratto, Sul tasso di urbanizzazione della colonia latina di Aquileia ............................................................................................................................... paola maGGi, renata merlatti, Piatti da portata verniciati di nero da Aquileia. Una classe ceramica di origine efesina in uso tra tarda repubblica e inizio impero .................................. diana doBreva, anna riccato, Aquileia e il Vicino oriente. Il commercio di prodotti levantini in area adriatica ...................................................................................................................... claudio Zaccaria, Poblicio Marcello: un senatore aquileiese al servizio dell’impero .................... deniS Sami, A Passage of Troops? Late Roman Small Finds from Ad Novas-Cesenatico (north-East Italy) .............................................................................................................................. NotiZiArio Serena di tonto, dario Gaddi, marta novello, L’attività di tutela nel territorio di Aquileia (biennio 2014-2015) ............................................................................................................ » annalisa Giovannini, Un tassello di storia della Grande Guerra ad Aquileia. Museo Archeologico nazionale: la “riscoperta” dell’incartamento Egger ........................................................ » annalisa Giovannini, L’inaugurazione dell’imperial-regio Museo dello stato in Aquileja. Il ritrovamento dell’epigrafe celebrativa ...................................................................................... » NorMe reDAZioNAli .................................................................................................................. » 5 169 175 179 182 AquileiA NostrA ANNo lXXXVi 2015 deniS Sami A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) introduction What constitutes roman military equipment and who used it? For specialists in roman military material culture these are central and much debated questions 1. simplistically it may be argued that objects used by the army represent military equipment, but given the social complexity of the roman army how do we distinguish between artefacts used by soldiers and civilians 2? Furthermore, material culture can intrinsically assume multiple and different functions as well as meanings. For example, a spear-head may have been used in battle by a veles, as well as in hunting by a non-combatant person; similarly, a gladius may have been used in battle but could have also been used for personal defence by a civilian. Consequently, the object per se carries a potential ambiguous identity that can be difficult to pin down. But when we add to the single object the context where it was used and lost and the reason for its entry into the archaeological record, we can nonetheless narrow down our speculation and define more precisely its identity. Below is discussed a representative sample of small copper-alloy, iron and glass finds collected from the site of Cà Bufalini in Cesenatico during the excavations undertaken by the university of leicester between 2008-14, all of which are generally identified in current literature as “military equipment” (fig. 1). the assemblage was collected in part through metal-detector surveys and in part from stratigraphical excavation in contexts closely related to the 1 2 presence of a road and the buildings alongside the artefacts are discussed in consideration of the context of their discovery and the potential reason for their loss, giving added value to their archaeological and object status. the way these artefacts entered the archaeological record is in fact of central importance to understanding their use and their social, as well as symbolic meaning. it is clear that some roman military equipment was deliberately deposited in votive contexts and less frequently in burials, but what can we learn from the artefacts that were simply lost during their use? if we want to untangle the complex social lives of military objects, their use and role in the performance of certain identities and their final deposition, it is necessary to first understand the wider spatial, temporal and historical contexts within which these occurred. Although limited, the case study of Ad novas provides a snap-shot into these processes as the site was an interregional crossroad, pivotal to the movement of people on the Adriatic starboard and in the eastern Po Valley, as well as a focus for the local economy. the Site the Tabula Peutingeriana indicates that the roman settlement of Ad novas lay immediately north of the rubicon river and along an interregional road flanking the Adriatic littoral and connecting the cities of rimini to ravenna (fig. 2). Although we lack textual allason-Jones 1999. allison 2013. JaMes 2004. 153 DeNis sAMi Fig. 1. the North Adriatic region showing the location of Ad novas-Cesenatico and the site of Cà Bufalini. Fig. 2. Ad novas-Cesenatico, the roman road system of coastal romagna. 154 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) Fig. 3. Cesenatico, Cà Bufalini, left, the results of the geophysical survey. right, interpretation plan of the settlement. information from the roman period, the same map seems to suggest that Ad novas was a mansio/statio serving the cursus publicus. the centre must have had a certain symbolic and strategic importance since in 568-69 bishop Petrus was met here by the citizens of Classe on his way back to ravenna. We also know that later, during the revolt of George – son of John, the secretary to the Byzantine exarch who was deported and executed by emperor Justinian ii – in 719, the citizens of the nearby centre of Cervia were ordered to patrol the area of Ad novas to prevent any potential landing of imperial troops. the ninth century Liber pontificalis ecclesiae ravennatis also informs us that in 814 the settlement was described as a civitas dirupta – an abandoned town 3. With only limited information, archaeologists have sought to identify the precise location of Ad novas and the road of the cursus publicus. Although it had been largely accepted that the roman settlement of Ad novas lies in the area of modern Cesenatico and the road was linked to the modern Via del Confine, their positions remained uncertain until our recent excavation campaigns 4. the investigations at Cà Bufalini documented a large and articulated roadside settlement dating from the early first century AD to the late sixth century and comprising a substantial paved road, a circa 40 x 50 m large rectangular building with courtyard, a small thermal bath and a possible dock on the bank of the nearby Fossatone canal 5 (fig. 3). Further remains of buildings, possibly part of the residential area, were documented in 2014, making the settlement larger than previously thought. the coast of Cesenatico is today a fairly dry and flat cultivated plateau ending in a sandy shore, but until the eighteenth century this landscape was dominated by standing water characterised by reed beds and difficult to access unless moving along the roads built on top of sand dunes. this environment, as we identify, most likely played a central part in the formation of the assemblage here investigated. the north/west-south/east oriented road was a particular focus in our excavations. this was a major construction of a considerable width of circa 9 m, and was built on top of a dune levelled with clay and gravel in the highest position of the site. its surface was paved with a thick layer of cobbles corresponding to the cursus publicus construction technique. A large ditch flanked the road on the west side helping the via to remain dry, while on the east side the walls of the buildings demarcated the eastern limit of the road-space. thanks to its position, size and quality building technique this road should be identified as the Via Popilia-Annia of imperial time, the road connecting rimini to ravenna and Aquileia as depicted by the tabula Peutingeriana and described by the early medieval Liber pontificalis 6. aGn. Lib Pont. Rav., c. 93. see also the translation by C. nauerth, De Sancto Petro Seniori: XXViii, 93, 10-13. farfaneti 2000. GiorGetti 1982. 5 saMi, chriStie forthcoming. 6 saMi, chriStie forthcomming. the location of the via Popilia as been the centre of a long debate. today it is whidely acepted the road changed its original position thorugh the centuries, see GiorGetti 1982. 3 4 155 DeNis sAMi interregional viae were also known as viae militares because they were often built by the army and used predominantly, although not exclusively so, for the rapid movement of troops, dispatch of messages, animals and supplies 7. From the time of septimius severus and later with the organisation of the comitatenses (movable troops) by Constantine the mansiones and stationes of the cursus publicus became focal points in the supply of the army implying a rising militarisation of these centres 8. the use of the cursus publicus was strictly regulated by law to the extent that the eighth volume of the Codex Theodosianus is largely dedicated to the administration of the network of land communications and under emperors Valens (364-78) and Valentinian (364-75) only the emperor could have released the diplomata necessary to access the cursus 9. this attention by the roman authorities implied especially in the late Western empire a high militarisation of mansiones/stationes and of the road space 10. roads were regularly patrolled by stationarii and beneficiarii, military officials appointed to the security of the road system that together with the periodical movement of army couriers and troops with their trains of wagons and supplies also contributed to the sense of a militarised environment 11. it is this movement of military personnel that may have contributed to a notable element of the small finds collected at Cà Bufalini. in particular, discussion will focus firstly on the theoretical approach to such artefacts, seeking to define their context of use and identity. Were these objects only the prerogative of military personnel or could they also be representative of civilians? Boundaries between military and civil use are, as we shall see, often complex and blurred in the late roman period. Furthermore, attempts will be made to use the textual sources to suggest a possible provenance for these artefacts, as well as draw comparisons with other archaeological contexts and similar finds. military equipment? Belt and harnesses fittings, fasteners, weapons, as well as glass and amber beads, generally interpreted as part of, or associated with the roman military equipment appear among the most common small finds collected at Cà Bufalini 12. these groups of artefacts, despite their association with the roman army display multifunctional or cross-identity characteristics that can also offer evidence of non-combat activity on site 13. Dome-headed and flat-disc fasteners (fig. 4, 1-8) are a good example of such multifunctional and versatile equipment that cannot be exclusively and precisely assigned to horse harness or allocated to combat dress 14. the different ranges in size and thickness of these artefacts highlight their versatility. Dimensions can differ considerably spanning 14 mm width of fig. 4, 1, to 7 mm of fig. 4, 1. 5. some fasteners are decorated revealing an aesthetic element to whatever they were attached to (fig. 4, 5-7). these objects became common from the third-century and their primary use was to bond together leather belts or straps 15. Considering a maximum leather thickness of 4/5 mm, belt fastener fig. 4, 3-6 may have linked together two straps, while fig. 4, 1-2 most likely joined together three or four belts. Despite their frequency along the limes, these fasteners are completely absent in the first to fourth century AD necropolis of lugone in salò (Brescia), which appears to have some important military connotations 16. they are absent from the small find assemblages from ravenna and Classe where, given the presence of the navy fleet and military personnel, we may expect perhaps higher concentrations 17. single dome-headed fasteners have been documented at sites all around italy, but high presence of such artefacts remains so far unknown 18. the fact that these objects do not occur in the well excavated the late sixth century Byzantine castrum of sant’Antonino di Perti in liguria might suggest that their production ended sometime between the fifth and the sixth century 19. tilBurG 2007, pp. 63-64. Corsi 2000, pp. 13 and bibliography in note 49 p. 18. Corsi 2000, p. 7. 9 Corsi 2000, p. 8 (see also chapter 3 source 130). 10 see šašel 1977. 11 fuhrmann 2012, pp. 197-98. tilBurG 2007, pp. 66-68. petraccia 2001. Militarised space: tilBurG 2007. Corsi 2000, p. 8. 12 allason-Jones 1999. 13 allison 2013, p. 3. 14 allason-Jones 1985. 15 BiShop, coulSton 2006, pp. 182-83. 16 redžić et alii 2014; redžić 2013, pp. 98; fiScher 1988. 17 the excavation at Classe focused on the commercial area and the Basilica severiana where the presence of military personnel may have been limited (elena Baldi personal communication). 18 in emilia-romagna dome-headed fasteners have been documented in Mirandola near Modena (corti 2012, ig. 1.5) and Borello near Cesena (maraldi 2014, ig. 4.2-3), but it is possible that lak of published exemplars prevent a full understanding of the distribuition of such objects. 19 Mannoni, murialdo 2001. 7 8 156 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) Fig. 4. Cesenatico, Cà Bufalini, Copper-alloy fasteners (1-8) and ring junctions (9-11). 157 DeNis sAMi Fig. 5.Cesenatico, Cà Bufalini, Copper-alloy mount (1), scabbard slide (2), lorica squamata scale (3), lorica segmentata hinge (4) and buckle (5). iron buckle (6). 158 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) one mount, fig. 5, 1 was most likely a belt or harness decoration as its two pins are too narrow to function as a fastener. it is decorated in blue and green glass, a production technique paralleled in roman Britain and on the Danubian limes where brooches and mounts have enamelled embellishment, notably such mounts are unknown at, for example, the third century military assemblage of Dura Europos in syria or that of tamusida in Morocco 20. the mount, fig. 5, 2 is more likely to belong to a scabbard slide. the upper end is disc-shaped with an integral rivet to the reverse for attachment, while the lower section is long and narrow, culminating in a small hook 21. Although such simple items may have had multiple functions, the iron D-shaped buckle, fig. 5, 6 and plain copper alloy rings, fig. 4, 10-11 are perhaps horse harness fittings and were used to secure saddles. the two rings present signs of heavy wear on one side suggesting they were subject to intense mechanical stress 22. Weaponry is probably the most functional category of objects with the strongest military connotations. the iron spearhead, fig. 6, 1 is a simple sub-pyramidal point possibly from a catapult or arciballista, but a second head spearhead, fig. 6, 2 is the terminal part of a pilum, a pyramidal-head javelin unlikely to have been used outside a military context 23. Although other functions and interpretations cannot be excluded – for example, a fitting from a casket or small item of furniture – our hinge fragment, fig. 5, 4 seems in fact most likely to be part of a lorica segmentata. Very similar parallels are documented at the castrum of Colchester in england 24. to my knowledge, the only elements belonging to a lorica segmentata in north italy comes from the necropolis of lugone where a well preserved hinge was recovered from a burial 25. A small copper alloy buckle, fig. 5, 5 may also have been part of lorica segmentata, although, again, its simple form means that other equine-related and domestic functions cannot be ruled out. the fragment of a copper alloy armour scale documents the use of lorica squamata on site. this armour was often employed in the protection of horses and may connect with the other horse related objects documented on site. As seen, the group of objects investigated here, despite their potential military connection could also Fig. 6. Cesenatico, Cà Bufalini iron subpyramidal bolt (1), pilum (2). be assigned to different users and so, automatically assuming a military connection can lead to the obvious risk of an overrepresentation of the presence and role of army personnel in passing through Ad novas. lindsay Allason-Jones, in analysing the ambiguity of the small finds from turrets on the Hadrian’s Wall showed the importance of the contexts in which finds are recovered to assign them a function 26. to the 20 similar mount has been documented in lincolnshire (Portable Antiquities scheme: liN-B798F1) and North yorkshire (PAs: NCl-309B11). JaMes 2004. MuGnai et alii 2013. 21 A slightly similar scabbard slide is documented in Darlington (PAs: BM-7CC415). 22 BiShop 1988, p. 106. JaMes 2004, pp. 60-62. 23 this sub-pyramidal point has similarity with artillery bolt-heads in BiShop, coulSton 2006, p. 169, n 2-7. About the pilum see d’amato, suMner 2009, pp. 6-7. BiShop, coulSton 2006, pp. 73-76, 130, 151. connolly 2001. 24 crummy 1992, p. 189, n 161, 234, n. 362-64. coulSton 1990. 25 Massa 1997; appelS, laycock 2007, pp. 62-3. 26 allason-Jones 1988. 159 DeNis sAMi importance of the context and the concentration of these finds, we should add the reasons for why these artefacts entered the archaeological record. Was their disposal accidental for example being broken, replaced during repair or discharged through age? or, perhaps was their removal or loss intentional? depoSition Among the many reasons for the disposal or loss or burial of military equipment was deliberate deposition 27. Military equipment – and in particular weaponry – has in fact been documented extensively within ritual contexts, suggesting that it held an important symbolic meaning for military communities 28. unintentional loss was also a major cause of disposal in particular for smaller sized, everyday artefacts. the objects from Cà Bufalini, as we have seen, are functional, rather cheap, ordinary items, intensively used and exposed to heavy wear and rupture, causing different degree of damage. losing a small mount may in fact have been unnoticed to the owner, causing only minor inconvenience. But the loss of a dome-headed fastener securing a waist belt, a horse harness or a scabbard slide may have caused functional faults that could hardly have been unnoticed. the loss of a horse harness ring junction or that of a waist buckle, similarly most certainly caused major problems that needed fast repairs. Beyond the potential damage cause by modern ploughing, if we look at the assemblage it appears that the most common reason for loss or disposal is heavy wear due to mechanical stress. For example, rings fig. 4, 11-12, despite concretions, clearly presents indications of intense use on one side. the mount decorated with inlay fig. 5, 1, features broken rivets. leather belts and straps no doubt frequently broke; this may explain why almost all the domeheaded and the flat-circular fasteners, as well as junction rings and buckles were found complete. However, considering the high quantity of such at Ad novas does this make a place of heavy traffic and potential stress and rupture of securing and fastening artefacts? Although the road crossing the site was of considerable width this was still a narrow space which hundreds, if not thousands of cavalrymen, soldiers and wagons loaded with supplies had to transit. even if the movement of the roman army was extremely well planned and strictly regulated by law to ensure fast movement of soldiers and supplies to minimise the trains and to reduce multi-user conflict 29. Nonetheless, jam, delays, jostling must have been common place in the empire. According to modern statistics a horse averages 210.31 cm in height, and 237.74 cm in length and when a rider sits in the saddle the widest point is 121 cm 30. roman horses were of smaller size, closer to the size of a modern pony, spanning between 111.76 cm and 121.92 cm of the exmoor pony to the 160.02 cm of the Friesian horse 31. in determining the space occupied by a cavalryman we need to consider that he had to travel at an acceptable speed, but he also had to move fast, changing direction or setting a formation in moments of necessity. it is reasonable therefore to think that enough space was left between horses – a minimum of 1 m on each side and 2 m from the front and rear horse would have been enough. A schola of the late fourth century consisted of circa 500 cavalrymen, which, considering the above measurements, implied a marching column divided in two rows circa 750 m long. to this we have to add wagons or other horses packed with supplies and possibly non-combative personnel. Accordingly to these measurements the road at Cà Bufalini was large enough to accommodate two Cavalrymen comfortably (fig. 7). on the related ground, the mansio/statio of Ad novas may have represented a critical point in the movement of an army required to transit in formation through a restricted area delimited by buildings and ditches. the restriction caused by the landscape possibly intensified form of mechanical stress to the equipment – particularly to horse harness – that was at the origin for the high concentration of these artefacts in the archaeological records of the site. provenance and chronoloGy the distributions of such finds within forts and fortifications or from scattered funerary contexts has mostly been understood only within the regions of the limes 32. As a consequence, these artefacts are not very well known outside the ‘militarised’ regions of the empire. the standardisation of the roman military equipment means that the provenance of these finds is generally difficult to establish. the dome-headed fasteners, ring junctions and D-shaped buckles col- BiShop, coulSton 2006. BiShop. coulSton 2006, p. 116. 29 Cth Vii 4, 22 (May 396); roth 1999; tilBurG 2007, p. 127. For the movement of the Army at the time of emperor Justinian see coSentino 2011. 30 hopper 2007, p. 165. 31 Speidel 1994, p. 108, JohnStone 2004. 32 Swift 2000. 27 28 160 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) Fig. 7. Cesenatico, Cà Bufalini, plan of trench 10 showing part of the north/east-south/west road and the flanking structure. lected in Ad novas are, for example, identical to the equipment documented at Dura-Europos in syria, as well as in Colchester in england. if the object per se cannot help in defining its origin, other finds associated with them, combined with more careful consideration of the local and regional historical context may help to shed light on the potential provenance of the assemblage. Glass beads of cubic, cylindrical, globular and hexagonal faceted forms dating from the second to the fifth century – with increasing concentrations from the fourth century – are recurrent finds along the north and the east limes 33. the dark blue hexagonal faceted bead from Cà Bufalini, fig. 8, 3 is one of the most emblematic late roman personal decorations documented in the Danube region and are generally associated with the presence of the army 34. turquoise melon beads made of glass or more frequently of faience (fig. 8, 2) are also distinctive and widely – although not exclusively – occur within military contexts. Probably manufactured in the Danube area they are known from the first to the fifth century AD and a connection with military horse decoration has been suggested 35. in addition, Swift 2000. Swift 2000, pp. 57-58. 35 Penelope Allison categorise melon beads as potential part of dress or horse equipment as well as possible female and child decoration. allison 2003, pp. 83-85. 33 34 161 DeNis sAMi Fig. 8. Cesenatico, Cà Bufalini, beads: amber (1), faience (2), glass (3-4). amber beads (fig. 8, 1) occur widely in roman funerary contexts and with a particularly high incidence in burials of the Danube region 36. As small portable objects, glass and amber beads may have travelled extensively through the empire in different modes: traded as necklaces or bracelets, as well as passed from one person to another as gifts. A comprehensive study and distribution maps of blue hexagonal beads and melon beads may change our view of the production, use and distribution of these artefacts, currently; however, they are associated with the presence of the army and the north-east limes. textual sources indicate an intensive presence of the Army particularly in the fourth century. the network of land communications of the Po valley was essential in the movement of troops from the key province of Gaul to illiricum at the time of Constantine 37. As consequence, most likely a season of renovation and investment in the system of roads and mansiones/ stationes occurred 38. in 312, preparing for war, Massentius concentrated his army in the north east of italy, between Verona and Aquileia. Constantine moving from treviri and taking Massentius by sur- 36 37 38 2012. 39 40 41 42 prise, defeated Massentius’s army at susa, occupied Verona and in the late summer he besieged Aquileia. the fall of Verona and Aquileia opened Constantine the way to the Po Valley finally completed with the fell of Modena and rimini. At this point Constantine had secured his access to the via Flaminia and reached rome 39. From rome, in the early months of 313, he moved to Milan and then Gaul. Further presence of Constantine and his army in north east italy is documented in 315 and 316 when he prepared for the battle against licinus. Finally, in 318 Constantine resided between Aquileia and Milan and moved at the end of the year to sirmium in Pannonia 40. Further consistent movement of troops are documented in 361, when emperor Julian besieged Costantius ii in Aquileia 41. in addition, the battle of the river Frigidus fought between theodosius and eugenius in 394 – possibly around Vipava (slovenia) not far from Aquileia – may have meant a regular flow of Pannonian troops along the Via PopiliaAnnia both before and after of the battle 42. in the early fifth century, stilico and Alaric transited with their army through the Po valley possibly passing through Ad novas and the garrison of the near fifth Swift 2003. porena 2013, p. 338. troops from southern Pannonia for example were deployed in other contexts and have a high degree of mobility. radman-livaJa roBerto 2014. porena 2013, pp. 335-36. roBerto 2014, p. 137. štekar 2013. 162 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) century capital ravenna may have frequently moved along the via Popilia-Annia too. Germanic mercenary troops led by general Aetius were defeated by magister militum Boniface in the battle of rimini in 432 and both the armies must have passed through Ad novas 43. Given the relative proximity of the Pannonian limes and the frequent and massive transit of troops along the cursus publicus of the Po Valley, it is not surprising that the objects here discussed had a north-eastern provenance. taking into consideration the historical background, a chronology spanning from the third to the fourth/fifth century is most likely. the high concentration of fourth to fifth century coins, ceramics and glass also support a late roman date for the assemblage here investigated 44. ConClusions When investigating every day and multi-functional objects we need to be cautious in drawing links with specific identities. However, supported by the textual evidence, a military origin for many of the finds from Ad novas remains plausible, whilst simultaneously bearing in mind their ambiguous nature. As we have seen, the militarisation of the road-space and the administration of the mansio/statio in which army personnel were widely and directly employed contributed to generate a militarised environment. the proximity of important political centres such as rimini, ravenna and Aquileia, as well as the battles fought between the late third and fifth century in the north Adriatic region may have further strengthened such atmosphere through the regular movement of army personnel in the eastern Po Valley 45. its location along an important interregional road and next to a waterway, made the mansio/statio of Ad novas a strategic point in the support and possibly the supply network for moving troops. the discovery of a consistent military equipment assemblage in such a context is therefore not surprising. the small finds from Ad novas represent a unique corpus of information that offers scope to understand more about the people who frequented the settlement and travelled along the road. At the same time, they shed light on the transient presence of the army in a region far from the limes. these finds open new theoretical insight into the possible social and emotional impact of militarisation, moving from the materiality of finds to a more intangible understanding of the site. Did militarisation bring a sense of security or insecurity? Did the transit of troops cause material and economic damage or represent a point of structural development and a new market? the questions remain, but are important to raise and introduce more. aBStract Keywords: this paper describes a collection of small finds generally identify as part of the roman military equipment collected during the excavation organised by the university of leicester at Cà Bufalini in the north-east italian town of Ad Novas-Cesenatico. Archaeological investigations carried out from 2006 to 2014 documented a first to sixth centuries AD road settlement organised alongside a large glareata road most likely to be identified with the Via Popilia-Annia of imperial time. the military origin of the artefacts here discussed is defined not only by their functional aspect, but also by the context of provenience and disposal. Keywords: Cesenatico; Militaria; late Antiquity; Popilia; Adriatic. riaSSunto un paSSaGGio di truppe? Militaria tardo romana da ad Novas-ceSenatico (italia nord-orientale) questa ricerca analizza un gruppo di reperti - genericamente identificati come militaria - raccolti durante la campagna di scavo organizzata dall’università di leicester a Cà Bufalini di Cesenatico tra il 2008 e il 2014. lo scavo ha portato alla luce i resti della mansio/statio di Ad novas costruita lungo una via glareata identificabile come la via Popilia-Annia di epoca imperiale. il rapporto tra lo spazio della strada, il passaggio di truppe e la perdita di militaria vengono utilizzati per definire l’uso militare degli oggetti presentati. Parole chiave: Cesenatico; militaria; tardaantichità; Popilia; Alto Adriatico. Gwatkin, whitney 1911, pp. 410-11. the ceramic, coins and glass assemblages are currently under study. For a preliminary idea of the material evidence from Cà Bufalini see saMi et alii 2014, saMi, chriStie forthcoming. 45 chriSite 2007. Buora 2002. 43 44 163 DeNis sAMi BiBlioGrAFiA Agnellus = Agnellus of ravenna The Book of Pontiffs of the Church of Ravenna, trans. by D. Mauskopf Deliyannis, Washington DC 2004. Agnellus = Agnellus of ravenna, Liber Pontificalis Bischofsbuch i-ii, trans. by C. Nauerth Auerth, Freiburg 1996. alison 2013 = P. alison, People and Spaces in Roman Military Bases, Cambridge. allason-Jones 1999 = l. allason-Jones, What is a Military Assemblage?, in «JromMilst», 10, pp. 1-4. allason-Jones 1988 = l. allason-Jones, Small Finds from turrets on Hadrian’s Wall, in Military Equipment 1988, pp. 197-233. apleS, laycock 2007 = A. appleS, S. laycock, Roman Buckles and Military Fittings, Witham. BiShop 1988 = M. C. BiShop, Cavalry Equipment of the Roman Army in the First Century A.D., in Military Equipment 1988, pp. 67-195. BiShop 1998 = M. C. BiShop, Military Equipment, in Roman Castleford. Excavation 1974-85, Vol. i. The Small Finds, H.e.M. Cool and C. philo (eds), Wakefield, pp. 61-82. BiShop, coulSton 2006 = M. C. BiShop, J. C. coulSton, Roman Military Equipment. From the Punic Wars to the Fall of Rome, oxford. Buora 2002 = M. Buora (ed.), Miles romanus dal Po al Danubio nel Tardoantico, Atti del Convegno internazionale (Pordenone-Concordia sagittaria, 17-19 marzo 2000), Pordenone. chriStie 2007 = N. chriStie, From the Danube to the Po: The Defence of Pannonia and Italy in the Fourth and Fifth Centuries AD, in «Proceedings of the British Academy», 141, pp. 547-578. connolly 2001 = P. connolly, The pilum from Marius to nero: a reconsideration of its development and function, in «JromMilst», 12/13, pp. 1-8. Corsi 2000 = C. Corsi, Le Strutture di Servizio del Cursus Publicus in Italia. Ricerche Topografiche ed Evidenze Archaeologiche, BAr, international serie, 875, oxford. corti 2012 = C. corti, Miscellanea. instrumentum e oggetti di uso personale in metallo e pasta vitrea, in L’insediamento romano della Tesa di Mirandola (Mo). Ricognizioni e scavi 1930-2011, M. calZolari and F. foroni (eds), in «quaderni di Archeologia dell’emilia romagna», 30, pp.187-90. coSentino 2011 = s. coSentino, Sul transito dei soldati (in merito a Iust. nov. CXXX), in «Bulgaria Mediaevalis», 2, pp. 25-38. coulSton 1990 = J. C. N. coulSton, Later Roman Armour, 3rd–6th centuries AD, in «JromMilst», 1, pp. 139-60. crummy 1992 = N. crummy (ed.), Colcester Archaeological Report 6: Excavation at Culvert Street, the Gilberd School, and other sites in Colchester 1971-85, Colchester. d’amato, Sumner 2009 = F. D’amato, G. suMner, Arms and Armour of the imperial Roman Soldier: From Marius to Commodus, Barnsley, Frontline Books. farfaneti 2000 = B. farfaneti, Cesenatico romana. Archeologia e territorio, ravenna. fiScher 1988 = t. fiScher, Zur römischen offizierausrüstung im 3. Jahrhundert n. Ch., in «BayVgBl»,53, pp. 167-190. fuhrmann 2012 = C. J. fuhrmann, Policing the Roman Empire, oxford. Gwatkin, whitney 1911 = H. M. Gwatkin, J. P. whitney (eds), The Cambridge Medieval History, Vol. 1, Cambridge. GiorGetti 1982 = D. GiorGetti, Cesenatico: note di antichità, in Storia di Cesena, I, L’evo antico, G. susini, (ed.), rimini, pp. 221-230. hopper 2007 = l. J. hopper, Landscape Architectural Graphic Standards, Hoboken. JaMes 2004 = s. James, Excavation at Dura-Europos 1928-1937. Final Report VII: The Arms and Armour, and other Military Equipment, london. JohnStone 2004 = C. J. JohnStone, A Biometric Study of Equids in the Roman World, thesis submitted for PhD, university of york, Department of Archaeology. maraldi, miari 2012 = l. maraldi, M. Miari (eds), Borello archeologia dalla preistoria alle soglie del Medioevo, ravenna. Massa 1997 = s. Massa, Aeterna domus: il complesso funerario di età romana del Lugone, Salò, Comune di salò, Museo civico. Military Equipment 1988 = Military Equipment and the Identity of Roman Soldiers: Proceedings of the Fourth Roman Military Equipment Conference, J. C. N. coulSton (ed.), BAr, international series, 394, oxford MuGnai et alii 2013 = N. muGnai, l. BiGi, e. Bernardoni, r. fanelli, m. perotti, Metalli, in Sidi Ali ben Ahmed – Thamusida, 3. I materiali, A. akerraZ, s. camporeale and e. papi (eds), rome, pp. 369-444. mannoni, murialdo 2001 = t. mannoni, G. murialdo (eds), S. Antonino: un insediamento fortificato nella Liguria bizantina, Florence. petraccia 2001= M. F. petraccia, Gli stationarii in età imperiale, rome. radman-livaJa 2012 = i. radman-livaJa, The Roman Army, in The Archaeology of Roman Southern Pannonia. The State of Research and Selected Problems in the Croatian Part of the Roman Province of Pannonia, B. miGotti (ed.), BAr, international series, 2393, oxford, pp. 159-89. porena 2013 = P. porena, La riorganizzazione amministrativa dell’Italia. Costantino, Roma, il Senato e gli equilibri dell’Italia romana, in Costantino I. Enciclopedia Costantiniana sulla figura e l’immagine dell’imperatore del cosiddetto editto di Milano 313-2013, Vol. i, rome, pp. 329-349. redžić 2013 = s. redžić, Rimske pojasne garniture na tlu Srbije od I do IV veka, Doktorska teza u rukopisu, Belgrade. 164 A PAssAGe oF trooPs? lAte roMAN sMAll FiNDs FroM AD noVAS-CeseNAtiCo (NortH-eAst itAly) redžić 2007 = s. redžić, nalazi rimskih fibula na nekropolama Viminacijuma, Belgrade. redžić, Jovičić, pantelić 2014 = s. redžić, m. Jovičić, s. pantelić, Unpublished grave-goods of belt-sets with ring-shaped buckles from Viminacium, in «Archaeology and science», 9, pp. 37-42. roBerto 2014 = u. roBerto, Aquileia tra Massenzio e Costantino: l’assedio della tarda estate 312, in «Antichità Altoadriatiche», 78, pp. 129-143. roth 1998 = J. P. roth, The Logistics of the Roman Army at War: 264 B.C. – A.D. 235, leiden. saMi et alii 2014 = d. Sami, e. Baldi, a. Booth, t. chinni, l. toniolo, Ad Novas-Cesenatico. From Roman Road to Late Antique Wood Structures An Interim Report on the Evolution Test Pits and Excavation at Ca’ Bufalini 2006, in «Fastionline», 319 (www.fastionline.org/docs?FolDerit-20014-319.pdf) Sami, chriStie forthcoming = D. Sami, n. chriStie, Digging along the Canal. The Roman Road and the Mansio of Ad Novas at Cà Bufalini (Cesenatico, FC), in Proceedings of the conference Economia e territorio nell’Adriatico centrale tra tarda Antichità e alto Medioevo (ravenna 02-03 May 2013). šašel 1977 = J. šašel, Viae militares, in Studien zu den militärgrenzen Roms, Vorträge des 10 Internationalkongresses in der Germania Inferior, ii , D. haupt and G. heinZ (eds), Köln, pp. 235-244. Speidel 1994 = M. P. Speidel, Riding for Caesar. The Roman Emperor’s Horse Guard, london. štekar 2013 = A. štekar, Poskus lociranja bitke pri Frigidu leta 394 na območju med Sanaborjem in Colom, in «Annales: analiza istrske in mediteranske študije», 23, 1, pp. 1-14. Swift 2003 = e. Swift, Transformations in meaning: amber and glass beads across the Roman frontier, in trAC. Proceedings of the Twelfth Annual Theoretical Roman Archaeology Conference, G. carr, e. Swift and J. weekeS (eds), Canterbury 2002, oxford, pp. 48-57. Swift 2000 = e. Swift, The End of the Western Roman Empire. An Archaeological Investigation, stroud. tilBurG 2007 = C. tilBurG, Traffic and Congestion in the Roman Empire, Abingdon on thames. Denis Sami oxford Archaeology east. 15 trafalgar Way, Bar Hill, Cambridge, CB23 8sq, united Kingdom denis.sami@oxfordarch.co.uk 165