IDENTIFICATION

The trees have undergone rigorous assessment at the hands of Stoiculescu (Reference Stoiculescu1985), who matched the leaves from the frieze to those on two groups of secular trees (i.e. groups of trees which have remained a consistent feature of the region) in the Southern Carpathians of Romania. The main division drawn by Stoiculescu is between resinous and deciduous trees. The resinous trees are typically evergreen and identifiable by their straight trunks, growing from a single point. These trunks are almost invariably more than half the height of the tree (some allowance must be made for the two trees in scene III, given that they are at a significant distance from the scene's action), and are topped with a narrow excurrent crown, which typically only has a few branches, often with the tree's needles directly attached to the trunk. It is through these needles that Stoiculescu (Reference Stoiculescu1985: 85) divided the resinous trees into subspecies.

The trunks of the deciduous trees are twisted, and ‘full of imperfections’, which are particularly noticeable given that they usually lack branches on the trunk's bottom half. Further, these trunks are fluted, pitted and with visible loose knots, an indication of the dedication of the sculptors to projecting an appearance of reality. The expectation would be for branching, particularly on deciduous trees, to start earlier on the trunk, as opposed to the column's depictions, in which the spreading branches of the deciduous trees are restricted to the top of the relief. The knots on the trunks suggest that the lower branches expected on these trees have been removed. The crown of these deciduous trees was described by Stoiculescu (Reference Stoiculescu1985: 85) as either umbelliform (with laterally spreading branches) or obovate (with branches spreading from a central point).

Following his division into resinous and deciduous trees, Stoiculescu further identified seven distinct groups of subspecies, which are further split into a total of 37 different leaf variations (Fig. 1). Stoiculescu contended that the first subspecies group, of resinous trees, is impossible to diagnose with more precision. Despite this, the clear differentiation between the types within group A indicates the importance of variety and specificity within the resinous tree types, as well as potential authorial intent to differentiate between the species.

Fig. 1. Leaf types on Trajan's Column, as identified by Stoiculescu (Reference Stoiculescu1985: figs 4 and 5). Reprinted by permission of Vasile Pȃrvan Institute of Archaeology, Bucharest. Stoiculescu describes the resinous leaf types (A and B) as follows: ‘Resinous species types included in the A and B species subgroup according to the crown and needle forms. Columnary crown: type 1 — long and pendent needles; type 2 — revolute, semilanceolate needles; type 3 — short and scaly needles; type 5 — needles grouped in pendent fascicles; type 6 — slightly revolute, lanceolate, big needles; type 7 — scaly and oblong needles; Pyramidal crown: type 10 (suggesting a primitive form of the actual pyramidal black poplar — Populus nigra L. cv. Italica — or the cypress — Cupressus sempervirens L.) Semipyramidal crown: type 8 — needles grouped in short fascicles; Rhomboidal crown: type 4 — long and scaly needles, catenary disposed; Ovate crown: type 9 — erect, fanned, exclusively disposed at the tree top.’ The leaf types of the deciduous trees (C–G) are listed thus: ‘Foliaceous species types divided in five subgroups according to their form of leaves: C — pentapalmary lobate leaves (suggesting Acer genus), D — broadly ovate leaves, with 3–5 triangular lobes slowly decreasing to the top (suggesting Sorbus torminalis L.), E — unregularly obovate and pennate lobate leaves (suggesting Quercus genus), F — oblong leaves (suggesting Prunus genus), G — elliptic leaves (suggesting Fagus genus).’

Depictions of the beech, or fagus, are both rare and unusual, since the tree only appears in one form (Type G.37), and is only seen once on the column. The motivation for this, Stoiculescu argued, is that ‘the beech was much too common’ (Reference Stoiculescu1985: 89), and he went on to suggest that Trajan aimed instead to impress his audience with a conquest over an exotic and foreign land, populated with strange-leafed trees. While the level of control that Trajan exerted over the frieze's arboreal contents is unclear, it is widely accepted that his account of the wars, the non-extant Dacica, would have had some influence on the overall intentions of the column, if not its individual components.

As to how accurate this strange-leafed land was to the contemporary Dacian landscape, Stoiculescu argued that the trees were reasonably representative of the forest at the time. He excluded from this assertion both the absent beech, and the two non-Dacian trees (Type B.10) found on the Roman side of the Danube at the start of the relief, in a walled city. Stoiculescu's assertion is part of a broader narrative, attempting to fit the trees into an argument that the relief is an accurate depiction of Dacia itself. Meiggs concurred with this assessment, in a brief observation regarding Trajan's Column. Focusing on the deforestation scenes, Meiggs (Reference Meiggs1982: 186–7) commented that the majority of the trees felled were oak, and that ‘this was not because they were more decorative; oak was the dominant species in these central European forests’. This comment appears to have been picked up in subsequent studies, and Lepper and Frere identified trees as oaks, firs or poplars, dependent on whether they are resinous or deciduous.Footnote ¹¹ This association of the oak has also been influenced by a remark made by Pliny the Elder, in Naturalis Historia (3.147), who identified the neighbouring region of Pannonia as ‘acorn-bearing’.Footnote ¹² As a result, the impressive variety of foliage which Stoiculescu identified on the column has often been overlooked, and presented as a homogeneous block of trees, instead of the impressive range on offer.

A key difficulty in the depiction of such a wide variety of trees on the column lies in both their visibility and their recognizability. Visibility has been an issue with the column's frieze since study of it first began, with numerous solutions proposed, from the vertical axis viewing first suggested by Lehmann-Hartleben (Reference Lehmann-Hartleben1926) and later by Gauer (Reference Gauer1977), to the neighbouring viewing platforms put forward by Coarelli (Reference Coarelli and Rockwell2000).Footnote ¹³ With regard to the trees, however, there remains a difficulty. In spite of the deliberately exaggerated leaves, clearly intended for the differentiation and identification of the tree species, there is an assumption that a Roman observer of the column would be familiar enough with trees to be able to identify them, and to place them as Dacian trees.

The first difficulty with this assumption is that the trees depicted on the column's frieze are not exclusively Dacian. As already briefly mentioned, the oak, one of the more common species on the frieze, was found across Europe at this time, and the same can be said of the other two dominant foliaceous trees: the service tree (a type of wide-ranging sorbus genus, which includes mountain ash/rowan in its variations) and the sycamore. Similarly, Romans would have been familiar with a variety of resinous trees, evidenced through a poetic record stretching as far back as Ennius’ Annals (Sk. 6.177); and the agricultural handbooks of the mid- to late Republic and early Principate, written by Cato the Elder, Varro and Columella, discuss a large number of resinous trees. While these books would only have been read by a few individuals, the implication of their coverage is that there would have been some degree of familiarity with the trees in Rome and its surrounding territories among the wider community without knowledge of specific species. To what extent it matters, however, that these trees may already have been familiar by sight to the Roman people will be dealt with shortly, in the context of tree-felling on the column and the triumphal procession.

Through identifying the different varieties of trees, a study of the column reveals an almost complete depiction of the (expected) Dacian landscape of the time. The absence of the beech is particularly important, since it reveals that the sculptors were selective over what trees they carved, and the range of leaf types shows a concern for individualized depictions, as in the human figures on the relief. Clearly, they were carved with this much care, even in the unenthusiastic style of the second sculptor (as highlighted by Wootton and Russell, Reference Wootton and Russell2013), for a purpose, and that purpose may become clear as we explore the different functions of the column's trees.

DEFORESTATION

Forty-eight trees appear in deforestation scenes on the column, making it one of the most common contexts in which trees appear on the relief, and these scenes are found from the column's opening. The first example of deforestation is in scene XV (Fig. 2), and the activity continues up to the final tree-felling scene in CXXXVII. The usual content of the deforestation scenes is exceptionally revealing: beyond three instances of Dacian tree-felling, all within the same scene, Roman soldiers are felling Dacian trees, in line with the general theme of industry on the column's relief. Davies (Reference Davies1997: 63) commented that the column ‘downplay[s] the gruesome realities of war’ and instead depicts more peaceful themes, of travel, construction, adlocutio, submissio and sacrifice.Footnote ¹⁴ Coulston (Reference Coulston and Blois2003: 403) has added to Davies’s earlier comments, noting that the battles are commonly fought by ‘non-citizen auxiliary troops’, while the Roman citizen troops are lauded for their ‘building construction, field-craft and siege warfare’.Footnote ¹⁵ The absence of warfare and the prominence of construction scenes is, for Davies (Reference Davies1997: 63), part of an effort to portray a mechanized Roman army, in complete control of both its surroundings and the war. In this reading of the column, the tree-felling scenes are key, and when this is put in the context of the Romans’ relationship with the environment in times of war, a fresh understanding of the usage of trees on Trajan's Column begins to emerge. However, challenges arise when tree-felling on the column is considered within the broader tradition of tree-felling in Greek and Roman myth and history, and there has typically been an assumption made by modern scholars that any deforestation is a sacrilegious act.Footnote ¹⁶

Fig. 2. Scenes XV–XVI. Deforestation can be seen in scene XV, on the left, watched by Trajan from scene XVI on the right. The trees are, in order, three times C.12, and one E.24. Source: Cichorius, Reference Cichorius1896: plates XIII and XIV.

Recently, Hunt (Reference Hunt2016) has readdressed this assumption, and has explored the possibilities of sacrality in trees beyond a wholesale attribution.Footnote ¹⁷ This leads to a challenge in identifying which trees are sacred, which prompted Thomas (Reference Thomas1988: 263) to conclude earlier that ‘any tree felling [is] potentially hazardous’, and the consequences for such a felling are usually severe.Footnote ¹⁸ The felling of the trees on the column operates alongside this rich literary tradition of negative deforestation, but also engages with a cultural tradition, with its roots in the late Republic and the conquest of foreign nations.

As should be expected with the relief, all the felled trees are in Dacian territory, and many felled trees are oak, a tree associated with Dacia by modern scholars, as noted above. In scene XV, the trees were identified by Lepper and Frere (Reference Lepper and Frere1988: 64) as being exclusively oaks, although Stoiculescu's analysis of the foliage has proven otherwise. Regardless of their error in this particular scene, oaks are prominent on the column, and the argument of Lepper and Frere that certain trees can be considered characteristic of Dacia is useful. While the trees in scene XV are not exclusively oak, they are all trees which Stoiculescu identified as being typical of a Romanian forest, and he added (Reference Stoiculescu1985: 84) that ‘the priority cutting of the deciduous trees suggests the preponderant waging of the two Dacian Wars just in this vegetation-zone type, where resinous trees existed only sporadically’. These authors direct us to the conclusion that the forest depicted on the column is both authentically and symbolically Dacian, being composed of trees that were both theoretically common to the area, and traditionally associated with it.Footnote ¹⁹

The need to display trees symbolically associated with the conquered Dacian territory is comparable to the triumphal importation of trees, a practice already discussed in some detail by Östenberg (Reference Östenberg2009: 184–8). The imported trees, Östenberg argued, stood for the race to which they originally belonged, and the land that they originated from. She took her lead from two brief accounts found in Naturalis Historia, the first of which indicates the importation of ebony by Pompey (12.20), and the second relating the display of the balsam by Vespasian and Titus (12.111–12). There is a clear differentiation between trees and timber, indicated by Pliny's specific use of arbores when referring to the practice in general, and the presence of the timber in Rome previously (12.17–18). Additionally, Östenberg noted (Reference Östenberg2009: 188) the usage of the verb ducere (to lead) in the context of displaying the trees in the triumph, a verb only used for the living spoils (e.g. prisoners and animals). Östenberg pointed to the balsam in particular, which pays tribute to Rome alongside its gens (12.112). In applying the attitudes which led to this practice to the trees on Trajan's Column, and to trees in conflict more broadly, a new tradition emerges: war on the landscape, which is expressed through the prominent tree-felling scenes.Footnote ²⁰

These scenes show the Roman forces waging war on the land of Dacia itself. Unlike the negative deforestation tradition, the deforestation of the trees on the relief has no adverse consequences for the Romans, and the focus is on industry, a focus clearly expressed in the first tree-felling scene on the column, scene XV. This scene runs almost seamlessly into scene XVI. Cichorius’ division between the two was contested by Lepper and Frere (Reference Lepper and Frere1988: 64–5), who identified the trunk of the oak tree at the close of XV as a ‘stage-prop rather than a divider’. The continuity between the two scenes, highlighted by the figure of Trajan, who is looking back at XV from XVI, implies that the timber felled in the earlier scene was used to construct the Roman fort of XVI. In contrast to the tradition of negative deforestation, this is not presented as a violation of a potentially sacred tree. Rather, the industry of the Romans, ratified by the emperor's approving gaze, is displayed in a positive light. The Roman soldiers are seen taking control of the foreign landscape, moulding it to their purpose, and in doing so urbanize their rural surroundings, in contrast to their Dacian enemies.Footnote ²¹

War on landscape is a common theme in descriptions of the Dacian Wars, particularly in the accounts of Pliny the Younger. Control of the landscape is a key component of Trajan's portrayal, in contrast to Decebalus in the Panegyricus, Pliny's only surviving speech, written and delivered prior to Trajan's first Dacian War. In this panegyric, written in the expected hyperbolic fashion,Footnote ²² Pliny described the land of the ‘barbarian king’, and its betrayal of the native ruler. The river and the mountains prefer to part for Trajan to advance, before turning and fighting for the emperor (Plin. Pan. 16.5). Here, a good emperor is seen in complete mastery of the forces of nature. Hutchinson compared this (Reference Hutchinson and Roche2011: 128, citing Plin. Pan. 50.1) to the negative portrayal of Domitian later in the speech. He added that Domitian's mastery of nature is artificial and incomplete compared with the willing submission of the landscape to Trajan. Moving genre, to Pliny the Younger's letter to Cannius Rufus, the theme continues. Here, Pliny related Trajan's command of the landscape, his creation of new rivers, and the subsequent bridging of these rivers, an echo of the bridging of the Danube at the base of the column's frieze (Plin. Ep. 8.4).Footnote ²³

It can and should be assumed that Trajan invoked the tradition of arboreal triumphal imports in his triumphs over Dacia, and that he displayed the trees of the land alongside the river deities and other natural features, particularly in the absence of a foreign ruler. It is therefore not unlikely that the emperor followed similar precedent and incorporated these trees, Dacian by their association with the triumph, into his forum complex, in the pits excavated by Packer in 1982, which form double colonnades in the Area Fori, identical to the known avenues of plane trees in the Porticus Pompeiana (Packer, Sarring and Sheldon, Reference Packer, Sarring and Sheldon1983; Packer, Reference Packer1997: 418–19).Footnote ²⁴

While tree-felling is prominent on the column's relief, it is not portrayed within the expected negative deforestation tradition. Instead, it taps into a broader cultural tradition that incorporates the importation of the trees in the triumphal procession, the urban habitat of the Romans as opposed to the rural character of the Dacians, and the industry required. I have characterized this earlier as a war on the landscape — an attitude which acknowledges the Roman war on the country, represented here by the trees, and by the similarly tamed Danube — as well as a war on the people. As a result, the victory depicted on the column is not only a victory over the Dacians, but also over Dacia itself, which has now come under Roman control.

TRAJAN AND DECEBALUS

The preceding section showed how the deforestation scenes on the column could be characterized as a part of war on the landscape. The use of trees as a narrative device can also be found in the depictions of the two leaders, who appear among trees several times on the column: Decebalus is shown in a forest setting four times, and Trajan three times. On the face of it, these numbers seem unremarkable until compared with the total number of appearances of each leader: Decebalus is depicted six times on the column, compared with Trajan's 57–60 appearances. In other words, the Dacian king is over ten times more likely to be found in a forest setting than his Roman counterpart. This might be expected, given the conflict between culture and nature inferred by Ferris (Reference Ferris2000: 61–85), which he extended to being between Roman and Dacian, and then between barbarism and civilization. The portraits of Trajan on the column, which may have been coloured purple and thus be extremely visible to the viewer, are in line with his general portraiture. Tuck (Reference Tuck2015: 226) has argued that the expected Augustan qualities of virtus, clementia and pietas are visible throughout this portraiture, as opposed to being restricted to depictions on this monument. The common forest setting of the Dacian king is often in direct contrast to the expected virtues from the clipeus virtutis, as will become apparent.

Decebalus’ appearances on the relief are framed by two in which the trees have a central role in the scene's events. His first appearance, in scene XXIV, is one of surveillance, as he watches from the back as the armies clash in the opening battle of the relief, likely the second Battle of Tapae, which was won by the Roman forces (Fig. 3). The trees that dominated the previous scene, one of Roman deforestation, reappear at the close of this one, as the Dacian troops attack from the forest that shelters their watching king. He hides underneath a resinous tree and behind an unusually small deciduous tree with broad palmate leaves, identified by Stoiculescu as a service tree, and is the final figure in the top half of the scene. This first appearance of Decebalus is clearly not a positive one, and attention is drawn to him by the projected line of the thunderbolt wielded by Jupiter Tonans, seen here to the upper left of Decebalus, slightly to the right of the centre of the battle. This is simply conjecture, since the thunderbolt itself is missing, but by extending Jupiter's line of sight, and combining it with the direction of his throwing arm, we find that he appears to be aiming at the Dacian king (Settis et al. Reference Settis, La Regina, Agosti and Farinella1988: 146). Lepper and Frere (Reference Lepper and Frere1988: 68) wondered whether this was the sole purpose of the deity's inclusion in the relief, or whether it was instead to indicate the thrust of the Roman forces, who advance into the forest under divine direction, or whether Jupiter Tonans’ function is perhaps to indicate that the battle occurred during a thunderstorm (as Cichorius had earlier suggested), and to further imply the ‘general support of Olympus’ for the Roman cause. The ultimate conclusion, of course, is that the figure of Jupiter Tonans can have all the above functions, but the most obvious one to a viewer is the literal interpretation, that the god is aiming his wrath at the hiding Decebalus.

Fig. 3. Scene XXIV. Decebalus’ head can be seen in the trees on the right, while Jupiter Tonans aims to hurl his thunderbolt from the left. The trees are identified as a D.18, A.5, A.5, D.16 (underneath the window) and a final A.5. Source: Cichorius Reference Cichorius1896: plate XIX.

The trees themselves are mentioned in the commentary of Lepper and Frere (Reference Lepper and Frere1988: 71), who identified them as reappearing ‘to form the right-hand boundary of the scene’, which underplays their significance in the scene. The trees at the close of scene XXIV do not function solely as a scene divider (the final resinous tree does this for the top half of the relief, but is obscured by the Dacians carrying their wounded comrade in the bottom half). Rather, they have multiple functions within the scene. The first function of the trees is that which was highlighted by Lepper and Frere, and is a structural one, closing a scene, and interrupting the narrative of the column. However, they directed their readers to the previous scene, XXIII, and the trees there. This is a deforestation scene, following a Roman advance through the forest (scene XXII), and the lack of ostentatious construction leads to the conclusion that the Roman soldiers are clearing a path for the advancing army (Coarelli, Reference Coarelli and Rockwell2000: 64). Although Lepper and Frere (Reference Lepper and Frere1988: 69) linked these trees to those which Decebalus reportedly dressed in Dacian armour at Tapae (Cass. Dio 67.10.2–3), it is easier to link these specific trees, and the episode as a whole, to the broader tradition of a war on the landscape. Nature is tamed by Roman forces in the scene leading up to the battle, and in the battle itself the trees are found only on the Dacian side of the scene, attempting to hide the king from the wrath of Jupiter Tonans.

When compared with the imperious figure of Trajan, both in scene XXIV with Decebalus and immediately behind him in scene XXV, the contrast between the rustic king and the urban emperor can be seen clearly. In scene XXIV, Trajan is dominant, giving the order to advance, as the heads of Dacians are shown to him. Meanwhile, Decebalus hides in the forest, distanced and uninvolved in the action of the scene, and identifiable at first viewing only by the direction of Jupiter's thunderbolt. This comparison is strengthened by the direct contrast between the figure of Decebalus in this scene, and Trajan, only two figures after him, in the following scene. In this appearance, Decebalus’ depiction is a complete mirror image of that of the Roman emperor: he hides where Trajan leads,Footnote ²⁵ and, unlike Trajan's backdrop earlier in scene XXIV, of advisers and a stone wall, Decebalus’ surroundings are rustic and arboreal. Thus, the opening depiction of Decebalus is not a flattering one on the relief, and the trees are integral in establishing the tone of this depiction.

Decebalus’ final appearance on the relief is in the moment of his suicide underneath a tree identified as belonging to the quercus (oak) genus, in scene CXLV (Fig. 4). In this scene, the oversized figure of Decebalus is seen taking a knife to his throat at the foot of the tree as the Roman forces arrive. The tree is passed over by Lepper and Frere, and by Settis, the latter of whom only referred to it as a place identifier for Decebalus’ prone form, although without an appreciation of the identifier's redundancy, given the gigantic proportions of Decebalus compared with the other figures on the column, and the direction of the other figures’ gazes, focused on the fallen king.

Fig. 4. Scene CXLV. The trees are identified as C.11, E.22 (under which can be seen Decebalus), C.13 and E.23. Source: Cichorius, Reference Cichorius1896: plate CVI.

Unlike the first appearance of Decebalus, with which this scene is directly comparable, the king is not hiding in the forest, but is clearly visible to the audience of the column. Here, the king's presence is the focal point, and his visibility is paramount. In addition to the oversized figure, and the concentrated gaze of all the surrounding figures and spears, the tree acts as a framing device for the king. In an extremely rare appearance on the relief, and contradicting the expected structure of the tree, there is a branch halfway up the trunk. This branch, which is short and leafy, shades Decebalus, framing him as an individual within the forest, and forcing the figure into a prone position. The tree, the trunk of which is behind the king, further isolates Decebalus, not from the Roman forces, but from his own sons, as Cichorius, followed by Lepper and Frere (Reference Lepper and Frere1988: 177), identified the two figures. Here, the tree acts as an internal scene divider, providing compositional emphasis, similarly to the way that the rich drapes that surround Jupiter Tonans do in scene XXIV.

The choice of the oak tree can be construed as a continuation of Jupiter's ire, echoing the thunderbolt of scene XXIV, which is further down the same side of the column, on the well-attested northwest, or ‘Victory’, axis. The column could never be read as a continuous spiral, and if we are to assume that viewers would have been standing at a gallery, or unable to circle the column to read the continuous relief, they would have read the column as a vertical axis (Bianchi Bandinelli, Reference Bianchi Bandinelli1978: 139; Settis et al. Reference Settis, La Regina, Agosti and Farinella1988: 202–3; Coarelli, Reference Coarelli and Rockwell2000: 19).Footnote ²⁶ As a result, the viewer will have been able to compare the first appearance of Decebalus with the final one, directly above it, and would see corresponding symbolism. First, the thunderbolt of Jupiter Tonans isolates the figure of Decebalus, who is only watched by the deity. And at his death, the oak tree, sacred to Jupiter, again isolates the Dacian king from his sons, forcing him from the shadows of his first appearance, and thrusting him into the foreground at the moment of his death. At this, the climactic moment of the Dacian Wars, and of the column's relief, the landscape of Dacia, represented by the tree most commonly associated with Dacia, betrays the Dacian king, and works for the Romans, thus paralleling the rivers and mountains in Pliny the Younger's writing. Thus, the scene not only shows the defeat of the Dacians, but also the subjugation of Dacia itself, and the Romans’ war on the landscape is complete.

There are two other appearances of Decebalus on the relief in which trees play a part. First, Decebalus observes a battle from a forest, and is framed by trees, in scene CXXXV (Fig. 5). Lepper and Frere commented (Reference Lepper and Frere1988: 173) that ‘the tree at the junction of Casts 360 and 361 (and the rocks beside it) detach Decebalus from the thick of the fight’, and added that ‘a second tree firmly separates Decebalus and his bodyguard from the next scene’. Settis (Settis et al. Reference Settis, La Regina, Agosti and Farinella1988: 144) further drew attention to the isolation of the Dacian king, although he focused on the eyelines of the king's entourage, which are directed away from the king, focusing instead on the action. This is, for Settis, a passive scene, unlike an earlier, comparable scene (XCIII), in which Decebalus’ entourage are focused on him. The trees function as the most obvious isolating tool here, and are combined with a low wall to separate Decebalus further from his surroundings. With a similar emphasis on framing to that in scene CXLV, the trees arch over the scene, containing Decebalus and his entourage as separate from the action, and only one forearm is allowed to intrude on Decebalus’ contained entourage, thus connecting the two scenes, together with the gaze of the figures, who focus beyond the bounds of their frame, much like Decebalus’ first appearance on the column. The trees have a dual purpose in this scene: on the one hand they act as a haven for Decebalus and his entourage, much like the trees do in Decebalus’ first appearance, and distance him from the battle happening around him, to which he is only linked by eyeline and an intruding arm. This separation is not necessarily a positive one, and it is noticeable that there is no comparable appearance of Trajan, who is elevated from his people, but never separated from them. This is the second function of the trees within the scene, which focus attention on the Dacian king, elevating him and highlighting his importance, and his regality as the enemy forest king. In this aspect, perhaps, Decebalus is compared with Trajan, but the comparison ends with the elevation. The trees separate the Dacian king, highlighting both his isolation in the scene, and his rusticity.

Fig. 5. Scene CXXXV. The trees are identified as E.26 and D.16. Source: Cichorius, Reference Cichorius1896: plate XCIX.

The separation of Decebalus from his men by trees is again seen in his penultimate appearance on the relief, in scene CXXXIX (Fig. 6). Here, Decebalus holds council with his fellow pileati in a woodland setting after the looting of his treasure by the Roman forces. The trees in this scene are intermittent, two acting in a rough framing capacity, and a third, background tree, between the two, to the right of Decebalus. As in scenes XXXV, CXXXV and CXLV, this tree could be seen to be isolating Decebalus from his men, although it does so in so subtle a manner that it serves as an indication of the continued forest throughout the scene. The choice of tree as a framing device is consistent with the usage of other deciduous trees as framing devices elsewhere on the relief. The shaping of the tree, with overhanging branches rather than a single straight trunk, allows for the scene to be contained both at its edges and at the top of the scene. By containing Decebalus here with a background tree, as opposed to the foreground tree which isolates him from his troops in scene CXLV, the relief identifies him within the scene. This is directly comparable to Trajan's appearance in scene XXXVI, in which two background trees frame him as the key figure of the scene. Decebalus’ appearance in the forest is similar to Trajan's appearances elsewhere on the relief, since this is the only scene in which a figure other than the emperor appears in an adlocutio stance. This is one of the most flattering depictions of Decebalus on the relief, since he begins to conform to the expected role of a statesman on the column, which is typified by Trajan's behaviour. As a result, the sculptor here is highlighting rather than isolating the figure of Decebalus, bringing him to the fore in a defining moment of the war, as the Dacian king organizes a retreat. It is not necessarily a good moment for the king to be highlighted, and all the elements of the Dacian defeat come to the fore in this scene: their king, in their native forest, orders their flight, abandoning their country to the advancing Romans, including the trees which will later separate the king from his men (scene CXLV).

Fig. 6. Scene CXXXIX. The trees are all identified as D.21. Source: Cichorius, Reference Cichorius1896: plate CII.

Unlike Decebalus, Trajan is found in proximity to trees only a handful of times, and each with a similar purpose: to reiterate the ‘war on landscape’ tradition, and to frame parts of the action. In the first of three appearances, Trajan is seen leading auxiliary forces into a forest, where he receives a report from a group of scouts (scenes XXXVI–XXXVII). He is next seen receiving a Dacian embassy, which emerges from the trees of a forest that is being cut down and tamed by the Romans (scene LII: Fig. 7). The final appearance of Trajan in the context of a forest is in scene LXVIII, in which a prisoner is brought in front of the emperor, again from the forest. Each of these scenes is one of two sides, of presentation to Trajan, either the scouts presenting a report of the Dacian territory to Trajan (scenes XXXVI–XXXVII), or the prisoner being presented in scene LXVIII. In these scenes, the trees additionally function in a framing context, dividing the two sides, and in the case of the battle, the two scenes.Footnote ²⁷ However, in the case of the Dacian embassy (scene LII: Fig. 7), the scene dynamic has shifted noticeably. Here, Trajan stands facing away from the deforestation behind him, as his forces tame the Dacian landscape, and he receives a Dacian envoy, who is seen extending his hands in greeting and supplication across the division of the tree, and the safe haven of the Dacian forest. The emotion of this scene is clear. What was once Dacian territory is now being controlled and manipulated by the Roman forces, and the Dacian embassy must now enter a foreign, newly urbanized landscape in order to negotiate with their Roman enemy. This is common throughout Trajan's arboreal appearances on the column, and in every instance the forest is associated with the opposing forces, and the urban sophistication of the emperor is contrasted with the rustic simplicity of the Dacians, epitomized by their almost exclusively arboreal king, who is eventually betrayed by the tree which fundamentally represents the land of Dacia.

Fig. 7. Scene LII, showing Trajan greeting a Dacian embassy. The trees are both identified as E.22. Source: Cichorius, Reference Cichorius1896: plates XXXVII; XXXVIII.