The iconographic features of individual figures allow their categorization into ‘Figure Types’. This is not merely a modern convenience but played a key role in visually presenting imperial victory in a contemporary guise, rather than using allegory and/or traditional stylization in the representation of Romans and their adversaries. Representation of contemporary dress and equipment also fed on, and into, metropolitan expectations for imperial iconography. It was not sufficient merely to depict the actions of emperor and army beyond the frontiers, this had to be done in a manner credible to the metropolitan viewing audience. The Figure Types were also assigned different roles in the theatre of war which further defined their social status and relationship to the emperor. The sculptors working on the frieze brought their own Hellenising training to the project, and, while they exercised a degree of empirical observation, they were not themselves knowledgeable soldiers.
Thus the following Figure Types may be defined and presented in the figure catalogue:
‘Human’ figure identifiable by dress, attributes and pose as a deity. Five gods and goddesses appear on the frieze: IV Danubius, XXIV Jupiter, XXXVIII Luna(?), LXXVIII Victoria and CL Luna(?). If the town of Ancona is correctly identified (LXXIX), then its prominent temple may be that of Venus. Dolphins in the sea in the same scene might suggest the benevolent presence of Neptune.
Five deities support the Roman advance into barbaricum, participate in battle on the Roman side, divide the two Dacian Wars, or set the scene of night(?) operations.
Trajan wears a short-sleeved tunic, muscled cuirass with upper arm and upper leg lappets (pteryges), closed high boots (calcei), and cloak (sagum). The pteryges often bear zig-zag chiseled edging. On journeys he wears a tunic and cloak (sagum, sometimes a paenula); during sacrifices a toga. He is invariably bare-headed, except for a toga fold during sacrifices. He carries a scroll, sword or a spear.
In 58 places Trajan is identifiable by his portrait (variable quality of likeness), stance, dress, attendants, grouping of military standards, and the composition of crowds around him. Trajan’s facial features and hairstyle are well known from his other portraiture.
Trajan appears to supervise marches and construction activities; to travel at the head of marching columns of troops and to embark on journeys by foot, horseback or on ship; to officiate at sacrifices; to deliver speeches to audiences in adlocutio scenes; to oversee battles; to meet barbarian ambassadors; and to receive barbarian submissions.
Man usually wearing short-sleeved tunic, muscled cuirass with pteryges, leggings, a sagum, and bare-headed, the same as Trajan. Short hair and often bearded. A scroll or a sword is sometimes carried.
The 106 officers appear in Trajan’s immediate command group, in the front of adlocutio audiences, or at the head of marching columns. They represent the emperor’s harmony with the Roman elites, and his reception to good counsel. The suggestion that many officers’ represent known historical figures (Hadrian, Sura etc.) is visually unconvincing.
Man carrying a Roman standard, wearing short-sleeved tunic, leggings, military boots (caligae), ring-mail armour (lorica hamata), and an animal skin over head and shoulders, with or without a helmet underneath. A short sword (gladius Hispaniensis) may or may not appear on the wearer’s left side suspended from a baldric. Often a small round shield is carried under the man’s left arm.
Five types of Roman standard (129 instances) may be identified through comparison with literary, epigraphic and other iconographic sources: praetorian centurial signum (60 instances); legionary aquila (15, plus two empty pedestals); legionary imago (1); legionary centurial signum (32); praetorian, legionary or cavalry vexillum (16). Some standards are stuck in the ground in camps, nevertheless there are fewer attributable standard-bearers than standards in some figure groups.
The 106 standard-bearers and their standards serve to indicate movement of army columns, and to locate the emperor in sacrifice, adlocutio and submission scenes. The mix of standard types in the same scenes makes it clear that they were not intended to identify different types of troops.
Man carrying a curved horn (cornu), wearing short-sleeved tunic, leggings, military boots (caligae), ring-mail armour (lorica hamata), and an animal skin over head and shoulders, with or without a helmet underneath. A short sword may or may not appear on the wearer’s left side, suspended from a baldric. Often a small round shield is carried under the man’s left arm.
Unarmoured musicians blowing a cornu or straight bucina provide music at major sacrifices, wearing tunic, sagum and wreath.
The 25 musicians fulfill similar roles to standard-bearers in march, sacrifice and adlocutio scenes.
Sometimes standard-bearers and musicians are indistinguishable when a standard or a musical instrument is not depicted or was provided by a lost metal insert.
Man wearing a short-sleeved tunic, caligae, and no leggings. Banded plate armour is worn on the torso from the waist up and on the shoulders (‘lorica segmentata‘) and appears in thirty variants as defined by types of fastenings. Bare or helmeted head. Short hair and often bearded. Thirty helmet variants, based on bowl features, falling into three super-types: brow-plate, frontal peak, all-round peak. On the march the helmet is sometimes not worn but slung over the chest (IV, XXXIII, XLVIII-IX, CI). A waist belt is present or absent, with or without a short frontal ‘apron’. A short gladius on the soldier’s right side may or may not be suspended from a baldric. A curved, rectangular shield is carries bearing either a wreath or a thunderbolts (fulmina) with eagle wings blazon. A shafted-weapon, sword, building block, pickaxe (dolabrum), or other tools are held, dependant on scene context.
Form of armour and shield clearly identify the figure as a member of the citizen army formations, with reference to numerous literary, artefactual and iconographic comparanda (the fulmina blazon references Rome’s chief divine patron, Jupiter). No distinction between praetorians and legionaries was intended by the sculptors, as is clear from the mixing of standards together, thus the generic term ‘citizen soldiers’ is employed here.
The 622 citizen soldiers form the core of advancing armies and undertook all construction work. In battle they generally hold back while the auxiliary figures dominate the action. The exception is the prominence of citizen troops in sieges where the technical aspects of siege warfare were presented as a ‘civilised’ function. Thus only citizen troops operate artillery.
Man wearing a short-sleeved tunic with varying skirt treatment (plain, fringe, tufts, other), leggings and caligae. A mail shirt (lorica hamata) covers torso, upper arms and upper legs, also with various edgings (plain, fringe, tongues, zig-zag). Finely chiseled zig-zags were applied to indicate the mail armour structure on all auxiliaries in some Scenes, and none in others. A sagum is sometimes also worn. A helmet is generally worn, exhibiting many of the thirty variants seen with citizen troops, but with proportionally more all-round peaks and apex rings. A gladius Hispaniensis on the soldier’s right side may or may not appear, suspended from a baldric. A flat, oval shield carries a decorative blazon on the front in several variants: axial, triangular ‘piles’; vegetal tendrils; palmettes; wreath; concentric rings; ‘Roman’ (thunderbolts-and-wings, eagle, lupercal, scorpion); plain/unfinished. Seen from behind the shield may be carried in ‘Roman’ fashion (hand on central, horizontal grip) or by Hellenising carriage (handle and arm-sleeve). Shafted-weapons and swords were depicted in stone, or defined by pose with open hand for metal insert.
The 461 auxiliary infantry march in columns and participate in adlocutio audiences. They scout and destroy enemy buildings, and they take the brunt of the fighting in open battle. They also assault walls in sieges. They do not do any construction work or handle artillery.
Man with exactly the same dress and equipment as the auxiliary infantryman, except that he is directly associated with a horse or group of horsemen.
The 82 auxiliary cavalrymen participate in march, adlocutio and open battle scenes. They also pursue fleeing enemies.
Man in belted, short-sleeved tunic, paenula or sagum, and caligae. Short hair and often bearded. Without cuirass or worn helmet. A variant man wears a belted, knotted, sleeveless tunic and caligae alone. Only in Trajan’s travels between the Wars, a variant appears with belt, apron and sword, a curved oval shield and a helmet slung over the chest (LXXXVI-VIII).
The 229 unarmoured soldiers attend sacrifices, travel with Trajan and greet him on his journeys. Those with knotted tunics man ships (XXXIII-IV, LXXIX, LXXXII), and occasionally do construction work (XCII, XCVII). They represent naval personnel (classiarii). The soldiers with oval shields appear exclusively with praetorian signa and this is the only time a distinction is drawn between different types of citizen soldiers. These depict the praetoriani who were the emperor’s bodyguard in Rome and on journeys.
This Figure Type is the most varied and includes all bowmen within the Roman forces. The first occurrence is a standard auxiliary figure with shield and other weapons replaced by a bow (XXIV.38). Four archers in scene LXVI are all clad in loricae hamatae and ribbed, flat‑topped, conical helmets. The next group of men (LXX) exhibit loricae hamatae and cylindrical quivers with conical caps suspended on their backs from baldrics. One archer’s lower body is clad in an ankle-length skirt. Helmets are conical with many ribs, a narrow peak, cheek‑pieces and a widely curving neck‑piece. Each man wears a multi‑strapped bracer on his left wrist and shoots a segmental bow with curled‑over ears. In Scene CVIII archers wear loricae hamatae and ankle‑length skirts. On the right hip they have a short gladius suspended from a baldric and a quiver is strapped to the back. The archers’ helmets are conical with ribs quartering the bowl, a narrow peak, cheek‑pieces and a small, square neck‑guard. One bow is curved and short with little detail, whilst a second has a set‑back handle, and recurved limbs. Lastly, three archers have loricae squamatae, with fringed undergarments appearing below them, and short gladii on baldrics (CXV). One man clearly wears a long skirt. The helmets have more ribs than those on the last archers, and cheek‑pieces, prominent bowl finials and widely curving, solid neck-flanges. Their bows have narrow staves, recurving limbs and set‑back handles. All of the archers described have short sleeves and the few visible feet have closed shoes (CVIII).
The detailed description demonstrates that this is the most volatile of all Figure Types. Archers never appear twice with the same cuirass, form of helmet, or bow. As sculpting progressed up the Column shaft the Type was first a simple auxiliary infantryman with an unrealistic bow, then armour was added in common with the Sarmatian cavalry Figure Type (LXVI drawing on XXXI and XXXVII). Real composite bows first appear in CVIII, but with a reversion to mail armour. Finally in CXV proper bows are combined with helmets of a realistic type seen on the Column pedestal reliefs. Presentation of this Figure Type was clearly a dynamic process, and it seems that spoliated barbarian equipment was only employed as models when work was far advanced.
The 17 archers appear in march, battle and siege scenes, never in adlocutio groups. Scholars have generally identified the archers as Syrians, Palmyrenes of ‘Levantines’, on the basis of literary notices. While this is reasonable, in the context of the Column’s iconography the actual features of the Figure Type are more Sarmatian than Syrian. The long-skirted tunic, so beloved of modern reconstruction artists, has no ready parallels in eastern clothing, but might be linked most readily to nomad kaftans (see below). This indicates that the sculptors were creating a wholly artificial Figure Type as they progressed. Apart from providing missile support to other troops, they are present to demonstrate the range of ethnic troops under Trajan’s commanding hand.
Man with a short-sleeved, belted tunic and sagum. Short hair and beard; bare feet. One man has a short sword on his right side. Less than half the slingers carry a flat, oval shield, bearing in one case a wreath blazon, and indistinguishable from auxiliary infantry boards. A sling is wielded in the right hand, or a stone is thrown by hand. Satsuma-sized missiles are sometimes carried in the folds of the sagum.
Quite what ethnic type the slingers represent is unclear. They may be have been intended as Iberian irregulars, for which there is some epigraphic evidence, but the sources are silent for Balearic slingers after the republican period. The Figure Type is very simply formulated and seems an artificial construction.
The 10 slingers appear in battle, siege and march scenes (LXVI, LXX, LXXII, CVIII, CXIII). Apart from providing missile support to other troops, they are present to demonstrate the range of ethnic troops under Trajan’s commanding hand.
Man wearing trousers or leggings with a thick roll-over of textile round the waist; caligae or closed shoes; but no tunic; short hair, with or without beard. Three men (CVIII.22, 24, 26) may wear their hair combed to the side but it is not entirely certain that it is tied in the Germanic Suebian knot (nodus). Flat, oval shield with piles, wreath, tendrils, small motifs or ‘Roman’ blazon. Club or short sword are wielded in combat, and a scabbarded short gladius appears on either the right or left hip, twice suspended from a baldric.
The rather awkward term ‘bare-chested irregulars’ has been adopted as a descriptive and neutral usage. Generally these men have been identified as Germans, and this would be strengthened if the men in CXVIII do indeed wear the Suebian nodus.
The 25 bare-chested irregulars appear mainly in battle and siege contexts (XXIV, XXXVIII, XL, LXVI, LXX, LXXII, CXV), but also in marching columns (XXXVI, CVIII). The one bare-chested irregular in an adlocutio, the only irregular figure of any kind to appear in this scene genre, is made even more unusual by the leggings he wears in common with the auxiliary infantry around him, and by the thunderbolt-and-wing blazon on his shield (XLII.10). Apart from providing additional combat troops, the Figure Type demonstrates the range of ethnic troops under Trajan’s commanding hand.
Man with a corkscrew-locks hairstyle, wearing only a sleeveless tunic and carrying a round shield, mounted on a saddleless horse with plaited cord neck band. Open handed poses suggest that shafted-weapons were inserted. Comparison with literary texts and the ancient iconography of North African peoples allows identification with the Moorish cavalry generally included in Roman imperial armies, and attested specifically as participating in the Dacian Wars (Dio 68.18).
The 15 riders appear in only one scene (LXIV) riding down Dacian infantry. Apart from providing additional combat troops, the Figure Type demonstrates the range of ethnic troops under Trajan’s commanding hand.
Man with long trousers; belted, long-sleeved tunic, slit up the sides of the skirt; fringed or plain-edged sagum; and closed shoes. Medium length hair, and often a beard are worn. Individuals in groups randomly wear a ‘Phrygian’ cap (pilleus) without any prominent rank function (despite Dio 68.9). An oval shield is carried, indistinguishable in form and carriage from auxiliary shields. Blazon types are ‘piles’; vegetal tendrils, palmette, concentric rings, thunderbolts-and-wings, and plain/unfinished. Stone weapons and poses with open hands indicate use of short, straight sword, curved Dacian sword (falx), shafted weapon, bow, and stone block missile. In 18 instances Dacians are associated with horses, but are otherwise indistinguishable from other Dacian foot warriors (XXVII, XXXI, C, CXXXIX, CXLIII-IV, CXLVI).
Wolf-headed windsock standards (21 ‘dracones‘) and vexilla (10) appear with Dacian figures. Some standards appear in association with buildings, nevertheless there are fewer attributable standard-bearers than standards in some figure groups, and bearers do not always wear pillei.
The Dacian king, Decebalus, known from the literary sources and other Roman iconography, appears at key points, and is recognizable more by context and pose than by portrait (LXXV.67, XCIII.26, CXX.21(?), CXXXV.2, CXXXIX.4(?), CXLV.11, CXLVII head only). He always wears a pilleus.
The 660 barbarians of the Figure Type form the massive preponderance of barbarian adversaries, and may be securely identified as Dacians by the context of the monument and by comparison with other Roman iconography. Dacians congregate, fight, die, submit or flee up the helix in large numbers. They attack Roman installations (XXXII, XCIV-VI, CXXXIV) but do not engage in sophisticated siege warfare. Dacian horsemen are most prominent in a disastrous river-crossing (XXXI) and during Decebalus’ flight at the end of the Second War (CXLIII-IV).
Man wearing trousers, cloak, closed shoes and no tunic. His hair is swept up into a ‘Suebian’ knot (nodus, cf. Tacitus, Germania 38) at the side or front of the head. No tunic is worn.
They may be identified by comparison of hairstyle and lack of tunic with literary sources and iconography as ‘Suebian’ Germans. The seven figures only appear as emissaries in two Scenes (XXVII, C) and their function is to emphasise the range of peoples coming to treat with Trajan.
Man in scale armour tightly fitting body and limbs, closed shoes, and a truncated conical, ribbed helmet, riding a scale armoured horse. The latter’s armour fits tightly to body, neck, head and limbs down to the hooves, with perforated eye-guards. Poses and stone details show the men armed with shafted-weapon or bow, and some wear a short sword on their right side. An unarmoured man wears a long-skirted kaftan, slit up the front, and closed shoes(?). The head is bare, has a hair band, or a pilleus.
The nine armoured barbarians appear in just two scenes, contiguously, one above the other (XXXI, XXXVII). Literary sources and iconography allow their identified as Sarmatian steppe nomads, and these are almost the only armoured barbarians in the entire canon of Roman art. They impart an exotic variety to the forces opposing Trajan. The three unarmoured figures may also be Sarmatians in steppe horseman’s kaftans, seen elsewhere in Roman iconography. They only appear as emissaries in one Scene (C) and their function is to emphasise the range of peoples coming to treat with Trajan.
Sacrificial attendant or lictor
This Figure Type includes youths and boys attendant at sacrifices, wearing a loose tunic and specific hairstyle with locks. Youths present an incense box (acerra) and boys play pipes. Muscular men with short hair and wearing only a wrap-around skirt and a wreath, have sacrificial knives at their waist, hold a pole-axe, and attend sacrificial animals. Men in tunics and cloaks with axe and rods (fasces) are lictores escorting Trajan.
These 42 figures help to define ritual context (VIII, LIII, LXXXV-VI, XCI, XCIX, CII-III, and the lictores denote the emperor’s status as a magistrate (LXXXIV, CIV).
Man in tunic and toga, or tunic and sagum; closed shoes; short hair, wreathed at sacrifices. Male child dressed the same. Woman in floor-length, short-sleeved tunic and wrap-around stole; closed shoes; hair brushed back and tied up in a bun, wreathed at sacrifices. Female child dressed the same. Infant in swaddling clothes, gender indeterminate.
The 77 Roman civilians appear to greet and attend Trajan during his travels, and they witness sacrifices (LXXX-XXXI, LXXXIII-VI). Occasional togate figures appear elsewhere in military groups (LXXXIV, XCI). The latter may be high-ranking officers, the unarmoured equivalent of cuirassed officers. The groups of civilians serve to contextualise Trajan’s travels through the provinces.
Male child wearing Dacian male dress. Woman wearing belted, floor-length, long-sleeved tunic and wrap-around stole (sometimes knotted at the front or caught up in the belt); closed shoes; hair up in a back bun with a twisted coil around the crown, or in a bun and covered by a cloth(?) scarf. Earrings are sometimes depicted. Female child dressed the same. Infant in banded garment or swaddling clothes, gender indeterminate.
The 48 Dacian civilians occur in compact groups in Scenes of capture or submission (XXIX-XXX, XXXIX, LXXVI, XC, CXLVI, CLV), once forming a cultural hierarchy in a crowd with Roman civilians at the front, progressing through to barbarian onlookers (XCI). They also occur in one Scene of torture (XLV).
Figures represented by a head alone where group identity does not allow identification, or which are substantially damaged beyond reconstruction, are defined as ‘unidentifiable figures’. There are 54 individual instances.
Discussion: Figure Type contraventions, documentary accuracy and stylisation
Formulation of the Figure Types was an ongoing process carried through by the sculptors as they worked at the Column face. Some of the major Figure Types settled down in their general features early on. Thus the citizen soldiers in Scene IV appear with two and even four waist-belts, where one was the proper provision. They also carry their belongings on poles above their heads, an impractical method as has been demonstrated by re-enactors. Scene V is the only one in which musicians’ cornua have an extended cross-bars with peltaform terminal, the standards present are tall and highly detailed, and the leading cavalry have extraordinarily long spears, unusually carved in stone. Here the sculptors were still in the early stages, but they were also using equipment to fill the triangular space above the human figures which was created by the second winding of the frieze mounting up on the ‘tail’ of the first.
Further up the helix sculptors sometimes confused the features and roles of Figure Types in what is termed here ‘Figure Type contraventions’. The animal skin head-dress of standard-bearers and musicians twice appears on other Figure Types. In a column of citizen soldiers approaching Trajan in Scene L, three men wear ‘loricae segmentatae‘ and animal skins (L.1, 2, 4; clear on the cast, lost detail on the original stone). Moreover, while Trajan has the normal zig-zag edging to his pteryges, this is extended across his muscled cuirass in the mail armour convention (L.8). These details can only be interpreted as mistakes on the sculptor’s part and are clearly Figure Type contraventions, not reflections of what the Roman army was really wearing in the field.
Similarly, a group of auxiliary infantry wear animal skins (XXXVI.12-4, 16-20), mixed with auxiliaries with unparalleled openwork ‘helmets’ which expose the hair underneath (XXXVI.2, 4-6, 8-11). Much has been made in the modern literature of these supposedly ‘elite’ and ‘Germanic’ soldiers, but they are again a product of sculptural processes, not actual military practice. A host of other detailed departures from Figure Type conventions might be mentioned here because they are statistically so rare, such as an officer wearing mail instead of a muscled cuirass; Roman blazons on auxiliary, irregular and Dacian shields; additional waist-belts on citizen soldiers; waist belts on auxiliaries; slipped tunics and absent cloaks for Dacians; a slipped tunic on a bare-chested irregular; and the bare heads of auxiliaries in battle scenes.
Very occasionally, the wrong form of shield is assigned to men of a Figure Type, such as the single auxiliary infantryman carrying a curved, rectangular shield with wreath blazon (XXXVIII.5), and one auxiliary infantryman has a small, round, standard-bearer’s shield under his left arm (XVI.7-8). Two citizen soldiers bear flat, oval, auxiliary shields (LXXII.14-5). These instances might have been based on sculptors’ observations, and scholars have not hesitated to over-interpret such features, but more likely they are sculptors’ confusions between Figure Types.
The roles of Figure Types are also confused in a number of Scenes, thus only in two do auxiliary infantry engage in construction work. The first is one of the earliest appearances of the genre (XII.10). In the second the awkward poses of figures suggest a change in composition during sculpting (CXXIX.5-6). In Scene LXXVII an adlocutio audience made up entirely of auxiliary infantry was given legionary signa. Citizen troops sometimes participate in open battle (as distinct from sieges), but one much-discussed instance involves the men with flat, oval auxiliary shields, who also have short mail armour sleeves under their ‘loricae segmentatae‘ (LXXII.14-5). Equipment features and figure poses are very confused in this scene. Similarly, only one bare-chested irregular ever stands in an adlocutio audience (XLII.10), but his leggings are a regular auxiliary feature and he bears a Roman shield blazon.
Another notable occurrence is the representation of two Dacians manning an artillery piece (LXVI.43-4). Scholars have been quick to link this weapon with artillery and Roman technicians acquired by Decebalus through wars with Domitian (Dio 67.7, 4; 68.9, 3, 5). This formed an important part of the ‘shameful legend’ of Domitianic defeat to be avenged by Trajan. While this (and Dio) might have picked up on a note in Trajan’s commentarii, the actual execution on the frieze is curious. The weapon is pointing back at the other Dacians, as the slider indicates, not at the Romans. Reflection of historical circumstance or not, this looks as though the sculptor originally meant the weapon to be shot against the barbarians and Roman figures finished up as Dacians because the lateral space of the Scene was re-planned during carving.
Thus in many cases confusion of Figure Type features and roles may reflect individual figures being originally intended as one Figure Type, but completed as another, incorporating elements of both.
Strictly speaking there is also another class of Figure Type contravention whereby figures are represented in variant dress and equipment, but demonstrably based on empirical observation of actual military practices. Thus standard-bearers not wearing animal skins, but adding belted tunics over their armour, may be paralleled by other iconography, principally bearers’ figural gravestones. In only one scene mounted auxiliaries carry the correct form of sword, the long cavalry spatha, with additional features which accurately match the artefactual record (XXXVII). The saddlery and tack of their horses is also rendered in an unusually detailed and correct manner.
The praetorians which form part of the unarmoured soldier Figure Type are another case in point (LXXXVI-VIII). There is ample evidence for curved oval shields continuing in use with the Praetorian Guard in Rome, and the order of dress, with belts, weapons and shields, but no other armour, is well attested throughout metropolitan and provincial iconography. Moreover, the slinging of a helmet over the owner’s chest on the march, seen in several Scenes, is also a practice attested in provincial sculpture.
With regard to the barbarians, there are several Scenes in which they have additional or variant detail which rings true. Scene C stands out as a selection of ethnically diverse peoples coming to pay court to Trajan. Not only are kaftan-wearing Sarmatians and Germans present, but also the Dacians wear flat-topped hats, tunics much longer than usual, and, in one case, an entirely anomalous scabbarded long sword with a scabbard slide, best paralleled by weapons on the Column pedestal reliefs. Similarly, right at the top of the helix, barbarians also wear long tunics and hats, perhaps denoting new northern adversaries contacted at the end of the Second Dacian War (CLI). However, the long tunic actually conforms much better with depictions of Dacians on other monuments, such as the bound Dacian prisoner statues which adorned the Forum of Trajan. One might ascribe the usual tendency to ‘understate’ Dacian tunics to the sculptors’ desire to reveal more, rather than less, of the human form within.
The overall perspective of sculptural execution of the Figure Types suggests a remarkable degree of applied detail, and a willingness to incorporate elements of empirical observations.
Thus the rendition of ‘loricae segmentatae‘ includes a great variety of fastening features. Generations of attempts to reconstruct this armour form which started with the Column, rather than recovered artefacts, were doomed to failure precisely because the Column sculptors did not understand the practicalities. Yet the Column renditions were so fatally seductive precisely because they did incorporate real detail. The ‘fleur-de-lys‘ hinge fittings of actual armours appear as buckle plates on chest and back sections of Column loricae (e.g. XI.8). This detail can only have come from observation of real armour, albeit misunderstood. Similarly technically specific, and similarly garbled, are the sculptors’ renditions of artillery (XL, LXVI). The bolt-shooting pieces are impractically rendered and differ at every appearance, lacking arms or shown with multiple long arrows, yet the arched cross-piece (kamarion), the projecting ‘slider’ (diostra), and the enclosed sinew-housings denote visual experience of actual weapons. The same might be observed of the design of warships in travel scenes. The knowledge that biremes and triremes had two or three banks of oars on each side translated into three lines of oars projecting out from under and impractically through simple, cross-braced outriggers (XXXIII-IV, LXXIX, LXXXII, LXXXVI).
Less misleading are the acutely and accurately rendered military standards, which almost always have varied detail each time they appear. It is important to note that equal numbers of praetorian and legionary standards types are depicted (62 and 62), far more of the former than would be proportionally present in a Danubian expeditionary army, reflecting sculptors’ greater familiarity with the metropolitan troop formations. Tents which occupy camps on the frieze also exhibit details of construction which accord well with leather tent sections recovered along the imperial frontiers. Even the marching equipment of Scene IV, however impractically slung to fill the overhead space, does include a well-observed assemblage of practical items.
The term employed here for this application of specific features based on empirical observation, but in a somewhat vague manner, is ‘detailed approximation’. Real features caught the sculptor’s eye in the streets of Rome and he made a mental note. They were then applied without clear technical understanding of their practical functions.
Models for the Figure Types need not have been numerous. One citizen soldier in ‘lorica segmentata‘, and one auxiliary in lorica hamata would have sufficed for the majority of Roman figures (622 + 543 = 1,165 out of 1,876, or 62% of the total). Unarmoured soldiers, sacrificial attendants, lictores and Roman civilians were common sights in the metropolis. Deities and cuirassed officers were part of the usual canon of Roman sculpture. The range of barbarian prisoners of war, slaves, diplomatic visitors and ‘tourists’, visible in Trajanic triumphs, elite entourages, in the streets, and in the auditorium audiences (cf. Martial, de spectaculis 3), would have provided elements for the Dacians, Germans, kaftan-wearing Sarmatians and Moors. More difficult were the armoured Sarmatians, perhaps based purely on a short passage in Trajan’s lost write-up (commentarii) of the wars, or a verbal description from an informed source. The Roman archers and slingers, whatever their intended ethnic identities, were entirely fabricated Figure Types, realized in part by drawing on war booty equipment left over from triumphs in Rome. Even the bare-chested irregulars, German or otherwise, were reduced to the bare minimum of features.
The overall accuracy of the distinctive dress and equipment used to define Figure Types is another issue. The majority of evidence points to the use of the ‘lorica segmentata‘ and curved shield by close-order praetorian and legionary troops; flat shields by other soldiers. However, mail and scale armours were also used by citizen formations, as were additional limb defences, greaves (ocreae) for the shins and a segmental defence (manicae) for the sword arm. This is clearly demonstrated by the contemporary metope reliefs of the Tropaeum Traiani (Adamclisi, Romania), and by artefacts recovered from legionary installations. Thus, while the Column’s restriction of segmental armour to citizen soldiers reflected actual practice, it was employed as a clear visual simplification, reserving mail and scale for auxiliaries and irregulars.
It has also been supposed by some scholars that the shield blazons on the Column were employed to identify specific legions and auxiliary formations. Examination of the application of these motifs does show that thunderbolt-and-wings designs were part of the citizen identity. Citizen wreath blazons were distributed vicariously without any reference to specific ‘events’ or formations. The auxiliary blazons were random overall, dependant on the penchants of individual sculptors, and this especially goes for the ‘Roman’ forms. The majority of oval shield blazons are indistinguishable between auxiliaries and barbarians, sharing as they do the odd, mistakenly applied ‘Roman’ design (e.g. XXXII.10, 13; XLII.10; CXVI.1). In one instance four auxiliary infantry and one bare-chested irregular, lined up in one rank, all bear the same ‘small motifs’ blazon (LXX.6, 9, 12, 14).
Similarly the distinctive roles assigned to Figure Types may be evaluated. The overwhelming evidence of dedicatory inscriptions and tile stamps suggests that the majority of building activities along and beyond the imperial frontiers were carried out by legionary troops. Substantial building was not undertaken regularly by auxiliaries until after the Trajanic period. Legionary troops also provided the skills and manpower for siege operations, and artillery was almost exclusively the preserve of citizen formations. On the other hand, the legions did still form the core of armies and the battle-winning heavy infantry (reading Josephus, Arrian and other authors), while the function of the auxilia was to provide the support roles of light infantry, missile troops and flanking cavalry. Presentation on the Column of citizen troops usually standing back ‘in reserve’ from battle, and the auxiliaries and irregulars surging forward, is exactly the image drawn by Tacitus of his father-in-law’s victory at Mons Graupius in Caledonia (Tacitus, Agricola 35). Victory was made all the more glorious if won without the loss of Roman, i.e. citizen, blood. This was a long-standing elite ideal, also expressed on the Arch of Claudius inscription in Rome (CIL VI.920), and Tacitus was writing contemporaneously with the creation of Trajan’s Column. Thus, in this respect, ideal victory was being presented as a not entirely accurate version of conflict reality.
Another somewhat stylized element was the deployment of irregular Figure Types on the Column frieze. In battle scenes archers, slingers and bare-chested irregulars appear almost formulaically, inserted behind or amongst the auxiliaries. In march Scene XXXVI figure groups of eight auxiliaries with heterodox head-wear and of eight bare-chested irregulars are in compositional balance. In CVIII ‘triplet’ groups of slingers, bare-chested irregulars and archers follow each other in marching column. This looks very much like an iconographic translation of an order of march of the kind rehearsed by Josephus, Pseudo-Hyginus and Arrian. Here perhaps there is another reflection of Trajan’s commentariiwhich very likely listed formations advancing into Dacia ordered by status and type. Here the upper line of figures is exclusively made up of citizen troops, while the lower consists only of auxiliaries and irregulars. It might even be suggested that the irregular Figure Types represented the four quarters of the empire: North (Germans), East (Syrians), South (Moors) and West (Iberians). Together they contributed to the ethos of firm Trajanic control over heterodox ethnic groups.
The sculptors of the frieze seem to have been working to a set of broad parameters. The commentarii may have provided the structure of wars and campaigns, and some prose descriptions of army composition (and armoured Sarmatians). Above all the remit was to depict Roman soldiers and barbarians in contemporary dress and equipment. Working in a Hellenising koine, relief sculptors had a repertoire of genres, especially combat compositions, to draw upon going back in Greco-Roman art to the Archaic period. Indeed, from the Aegean Bronze Age onwards the convention was dominant that winners moved towards the viewer’s right and the vanquished proceeded from right to left. For the Column frieze this worked harmoniously because the helical frieze followed the anticlockwise direction of the internal spiral staircase. Thus the Roman forces always proceeded inexorably up the helix from left to right, and the barbarians constantly fall back to their inevitable defeat.
However, there were also tensions between the contemporary depictional remit on one hand, and the artistic training and inclinations of the sculptors on the other. The ‘barbaromachia‘ of the Column drew heavily on Hellenistic period conventions of combat pose and pathos. This is most prominently seen in recurrent Pergamene style figures of bested barbarians falling to one knee, one of whom is the dying Decebalus (XXXVIII.11, LXVI.36, XCIV.20, CXII.3, CXLV.11, CLI.7). One unfortunate bi-product of this was the reluctance to depict fighting stances which were alien to, or which fitted uncomfortably with, convention. The signature Dacian weapon in the Trajanic period was the curved, single edged sword in two variants: the short, single-handed sickle-sword, and the long, double-handed scythe-sword. Both are covered by the Latin word ‘falx‘, and both are attested in the artefactual record. The former is repeatedly carved in stone on the frieze (e.g. LXVI, LXXII, LXXV, XCV-VI, CXLV, CLI). The latter type appears on the Column pedestal reliefs and on the trophies of Scene LXXVIII, but never elsewhere on the frieze. Instead, a hybrid was created, essentially a scaled down version of the scythe-sword with a proportionally long haft which could be wielded in one hand alongside use of a shield (LXVI.39-40, LXXII.33, 35). This was done precisely to avoid having large numbers of Dacian figures in unconventional, double-handed stances. Such poses were acceptable in small numbers for men throwing blocks off besieged fortification (CXIII.11, 27, CXV.21, CXXXIV.13, 22), but not for multiple figures in Hellenising barbaromachia.
Battles between Greeks and Persians, Greeks and Amazons, Lapiths and Centaurs, and Gods and Giants often involved the use of Greek hoplite equipment, especially the large, round shield (hoplon or aspis) carried with a forearm-sleeve (porpax) and a rim handle (antilabe). The Column sculptors could not stop themselves from often reproducing this carriage for Roman and barbarian oval shields. In reality the horizontal, central handgrip was the real method for all boards in the northern European tradition which dominated the Roman equipment from the 4th century BC onwards. Moreover, in traditional combat art the ‘civilised’ Greeks were often depicted in heroic nudity, while the barbarian Persians and Amazons wore much more feminizing clothing coverage. Even when the complete Greek panoply was worn, artists managed to reduce the coverage of pteryges to reveal as much as possible of the limbs, and even the genitalia, of hoplites.
Thus for the Column project the barbarians were conveniently covered up, although, again, the sculptors could not resist exposing more by shortening tunic skirts. For the Roman Figure Types they reduced the size of helmet cheek-pieces and of shield boards. Tunic skirts rode right up over skin-tight leggings for auxiliaries, and some skirts were omitted altogether, while lorica hamataskirts were also unrealistically brief (e.g. I, XI, XVIII, XXXIII, XXXV-VI, XXXVIII, XL, XLII, XLIV, LIV, LIX, CXXXVI-VII, CXLVIII-CLIV) exposing buttocks and groin area. Buttocks and ribs show though the armour in a ‘wet T-shirt’ manner, whereas mail armour was invariably worn over a thickly padded support garment. Such clinging armour led observers from the 16th century to the present to mistake the garment for cloth or leather, whereas the zig-zag chiseling proves that all such loricae were made of metal ring-mail. This also extended to the reduction in size of horses, vehicles, ships and other scenery elements, so as not to dominate and obscure the human form. Indeed, it would not go too far to admit an element of homo-eroticism in the sculptors’ treatment of the male body.