Roman Chester (Deva)

Chester was one of three legionary fortresses — alongside York and Caerleon — in Britain, located on the River Dee. It was constructed in the 70s a.d. by Legio II Adiutrix and garrisoned by Legio XX Valeria Victrix from the 80s until the late fourth century, although it was thinly manned when many cohorts of Legio XX were seconded for the building of Hadrian’s Wall and the Antonine Wall.Footnote ⁷ The fortress was quite extensive with an amphitheatre, a number of bathhouses both inside and outside the walls,Footnote ⁸ a port and the foundations of a perplexing ‘elliptical building’.Footnote ⁹ Chester was a cultural melting pot: soldiers, merchants, veterans, natives, women and children from a variety of backgrounds mixed together in the fortress and its surrounding settlements: the canabae legionis, the vicus of Heronbridge, two smaller settlements at Saltney and Boughton, and a pottery facility at Holt.Footnote ¹⁰ There are 18 religious inscriptions from the fortress and surrounding area, excluding three dedications to the emperor — consistent with other fortresses in Britain but insignificant compared to the legionary bases on the Rhine such as Bonn or Xanten.

Jupiter Optimus Maximus Tanarus

The altar dedicated to Jupiter Optimus Maximus Tanarus was inscribed as follows (RIB 452):

The princeps was the second highest-ranking centurion in a legion and the consular dating allows us to place the altar in 154 a.d.Footnote ¹¹ Just like Collingwood nearly a century ago, most scholars today see IOMT as an interpretatio.Footnote ¹² Tanarus most likely derives from the Celtic taran, meaning ‘thunder’, and should be identified with (and as a misspelling of) the Celtic god Taranis, who appears in Lucan as well as several other inscriptions with slightly different spelling:Footnote ¹³ for example, at Bockingen, Godramstein, Orgon, Scardona, Bribir and Thauron.Footnote ¹⁴ Versions of Taranis appear alone on the altars in Bockingen, Godramstein and Orgon, and Lucan treats him as a single deity, suggesting that Taranis is not an epithet meaning ‘thunderer’ as some scholars have suggested.Footnote ¹⁵ He was a god who was paired with IOM or Jupiter in a select few local areas with specific circumstances that might have encouraged such a link.Footnote ¹⁶ The substantial iconographic variation on the Jupiter Columns in the Rhineland — Jupiter sitting on a throne, mounted and attired for battle, or fighting Giants — exemplifies the various roles of Jupiter in different places.Footnote ¹⁷ The association with Taranis at Chester drew out a particular aspect of IOM to the exclusion of other aspects of his multifaceted role as an all-powerful sky god.Footnote ¹⁸

Now Lucius Elufrius Praesens was from Clunia in Hispania Tarraconensis and like many of the soldiers at ChesterFootnote ¹⁹ — originating from Syria, Lusitania, Gallia Lugdunensis, Noricum, Gallia Narbonensis, northern Italy, Germania Inferior, Lower Moesia and TripoliFootnote ²⁰ —would not have known Taranis from his hometown. Instead, it is likely that there was already a cult to Jupiter Optimus Maximus Tanarus in Chester. Perhaps Taranis was worshipped by a contingent of Gallic cavalrymen in Chester and was slowly syncretised with Jupiter, who would have been worshipped by the Romans. Indeed, an earlier IOM altar at Chester had a goose — a symbol of Juno — engraved into the side-panel, possibly indicating that this Jupiter was originally understood to be connected to the Capitoline Triad.Footnote ²¹ This is made manifest in an altar at Maryport (IOM Capitolinus) and the Capitoline, Mars and Victory altar from Carlisle which must be linked to the 3 January festival prescribed in the Feriale Duranum.Footnote ²² The later IOMT altar is shed of such links with the Triad and the January festival. Therefore, IOMT was the product of the local conditions at Chester, which might not have been present elsewhere in Britain, and became the fortress’s premier deity by the mid-second century. In this sense, Lucius Elufrius Praesens was not Romanised or Celticised in making a dedication to IOMT, but rather localised.

Aldhouse-Green suggests that Taranis was a god of thunder, weather and storms, which might have overtones of battle.Footnote ²³ Some have also suggested that the cult was represented by the wheel-symbol and was therefore common across Britain, despite the name only appearing at Chester.Footnote ²⁴ But there is serious doubt about the association between the wheel-symbol and Jupiter or Taranis,Footnote ²⁵ and there is no wheel on the IOMT altar, but a six-petalled flower and a wreath enclosing a rosette. It is possible that the concept of ‘thunder’ was linked to the din of battle and the parade-ground — the clashing of weapons on shields, the stamping of feet, the shouts of soldiers —Footnote ²⁶ and it was in the parade-ground at Chester that the IOMT altar was discovered.Footnote ²⁷ But it seems Taranis had deeper links with a particularly Celtic aspect of warfare: headhunting.

In his Pharsalia, Lucan wrote: et quibus inmitis placatur sanguine diro Teutates horrensque feris altaribus Esus et Taranis Scythicae non mitior ara Dianae (and those who propitiate with horrid victims ruthless Teutates, and Esus whose savage shrine makes men shudder, and Taranis, whose altar is no more benign than that of Scythian Diana).Footnote ²⁸ The implication here is that human victims were sacrificed to these three Gallic gods: Teutates, Esus and Taranis. The comparison with the Scythian Diana is relevant here because she, too, was linked with human sacrifice.Footnote ²⁹ There is further evidence of Taranis’ connection to headhunting specifically in the Bern scholiast: praesidem bellorum et caelestium deorum maximum Taranin Iovem, adsuetum olim humanis placari capitibus (the chief of wars and greatest of the celestial gods, Jupiter Taranis was once accustomed to being placated by human heads).Footnote ³⁰ Duval argues that, given what we know of the Gallic practice of headhunting (see below), ‘on offrait à l’arbitre céleste des guerres soit des têtes prélevées sur des ennemis morts, soit des têtes de prisonniers décapités ou d’autres victimes spécialement désignées pour ce supplice.’Footnote ³¹ Is there any further evidence for this link between Taranis and headhunting? Interestingly, Aldhouse-Green suggested there might be some evidence for this in Britain based on a find in Staffordshire: ‘this is a stone from a temple-site at Wall, Staffordshire, carved on which is a naked figure holding a circular object (thought by the excavator to be a wheel) over an altar. Beneath this figure is a human severed head.’Footnote ³² There is another possibility found at Chester itself: the tombstone of Aurelius Lucius.

Aurelius the Headhunter

Discovered along the North Wall in 1887, the tombstone reads (RIB 522):Footnote ³³

More important than the text itself is the image on the tombstone (Fig. 1). It shows a male figure, obviously Aurelius Lucius, reclining on a couch with armour on. Hope characterises this as a ‘totenmahl or funerary banquet’, but it deviates from other such scenes by having a very explicitly military overtone.Footnote ³⁴ He holds a cup in his raised right hand and a scroll in his left hand. Above him, fixed to the wall, is a full-faced cavalry helmet with plume and a sword with large pommel, presumably a spatha. Aurelius Lucius was probably a member of the 120 equites legionis along with other cavalrymen who appear on tombstones at Chester.Footnote ³⁵ The decision to mark his status in this way is surprising, given the usual iconography on the tombstone of a cavalryman was a warrior on horseback over a defeated enemy, normally with a spear.Footnote ³⁶ Indeed, four are from Chester (RIB 499, 541, 550 and 551) and another (RIB 538) shows a slight variation of this scene. But most interesting of all, a small figure is displayed next to the couch holding a severed head by the hair. Now it has been known for a long time that headhunting by Celtic auxiliaries was practised in the Roman army of the imperial period, especially amongst the cavalry.Footnote ³⁷ Headhunting had been an important feature of Gallic warfare, culture and religion centuries before the Roman conquest and was noted by numerous Roman sources.Footnote ³⁸ Archaeological discoveries of decapitated heads in Roman Britain have been used to support this view, though the best evidence comes from southern France, such as the warrior statues from Entremont.Footnote ³⁹ For Voisin, its practice in the Roman army of the Imperial period was simply a continuation of a Roman practice dating back to the regal period and early Republic;Footnote ⁴⁰ but against this Petitjean has argued that by the Second Punic War such practice had fallen out of use. It was reintroduced by Celtic auxiliaries who continued their long-standing traditions which became accepted by imperial authorities and adapted to represent the domination of the empire.Footnote ⁴¹ Similarly, Fields has argued that Celts in the Roman army continued to practise headhunting as ‘an organised, coherent form of violence in which the severed head retained its specific ritual meaning’.Footnote ⁴² For Fields, Aurelius Lucius’ tombstone is thus, and one of the first pieces of evidence of, ‘a conscious exhibition of the traditional Celtic practice of headhunting’.Footnote ⁴³

Fig. 1.

Tombstone of Aurelius Lucius, CHEGM 1999.6.65 (H: 137 cm; W: 71 cm) (Courtesy of West Cheshire Museums).

Within Britain there are a few other attestations of headhunting in the iconographic record, yet the evidence is limited.Footnote ⁴⁴ One of the best pieces of evidence is the Lancaster horseman. The tombstone of Insus, son of Vodullus, shows the Treverian cavalryman with helmet, shield and sword atop his mount. Below him is the decapitated body of an enemy, still holding shield and sword, whilst his head is grasped by the hair in the right hand of Insus.Footnote ⁴⁵ Without many other examples from Britain of headhunting it is hard to know how prominent this aspect of martial culture was. Yet it is probably fair to assume that in Chester there was a sizeable enough group engaging in the culture of headhunting for Aurelius (or rather more precisely his heir) to include a severed head on his tombstone which would have been visible to all. It could be countered that other cavalry tombstones from Chester (RIB 499, 541, 550 and 551) do not include references to headhunting. This is true; but, given three are highly fragmentary, it is dangerous to conclude that they would not have contained headhunting iconography. Furthermore, while the ‘rider-over-enemy’ was a stock iconography that ‘simply served to distinguish the skills of the cavalryman and to contrast his military role with that of the foot soldier’ or even presented ‘the deceased as the heroes of Rome, trampling the vanquished enemy’,Footnote ⁴⁶ in the eyes of certain Roman soldiers in Britain, the iconography might well have been associated with headhunting, marking the moment before the act of decapitation of a defeated enemy just as Insus’ tombstone showed the moment after the act. Indeed, the Bridgeness Slab from the Antonine Wall depicts a rider scene on the left with four defeated enemies, one of whom has been decapitated.Footnote ⁴⁷

Although it is impossible to establish any sure connection between the IOMT altar and the tombstone, it is striking that there was a cult to a god associated in Gallic culture with headhunting and a soldier wishing to commemorate his headhunting credentials on his tombstone in the same fort. Aurelius Lucius likely lived during the second century, though it is hard to be precise. Based on the depiction on the tombstone, Stewart argued that ‘his beard and vigorous hair, as well as his name, probably give an early Antonine terminus post quem … the use of the nomen Aurelius, and the very fact that this eques is a citizen might point beyond a.d. 211’.Footnote ⁴⁸ However, earlier dating based on the lettering of the inscriptions on all the tombstones found in the lower portion of the North Wall suggests that they cannot be dated beyond c. 200 a.d.Footnote ⁴⁹ Therefore, it seems that Aurelius Lucius could have lived during the century in the middle of which an altar was dedicated to IOMT at Chester and was a central aspect of the local pantheon at the fortress. It thus seems reasonable to connect the two together and see them as manifestations of a particular martial and religious culture within Chester that valued the practice of headhunting and the god Taranis connected with it.

Therefore, the cult of IOMT was quite different to other Jupiter cults in Britain, and perhaps across the empire, through its association with Taranis and headhunting. Equally, this Taranis cult was quite different from other manifestations, for it had been linked with a central god within the Roman pantheon, giving it a centrality within a Roman fortress unachievable elsewhere. The cult of IOMT was thus local and unique. It is possible that originally the cult of Taranis was brought by Gallic cavalrymen who served at Chester and engaged in the culture of headhunting, within which Taranis was an important deity. It was then combined with the cult of Jupiter Optimus Maximus and so created local variation. Soldiers who then came to Chester and engaged in the cult, such as Lucius Elufrius Praesens who erected the altar which survives today, were thus integrated into a local religious manifestation. And this was likely the case with Aurelius too. There is little sign that he was Gallic. The name Aurelius Lucius was obviously Roman, and it is likely he was serving not in an auxiliary cavalry unit, but the equites legionis. It has been suggested that the brushed-back hair and presence of a moustache and beard signifies Aurelius’ Gallic origins.Footnote ⁵⁰ Yet the hair is not spiked, nor is the facial hair any secure marker of identity. Perhaps most significantly, Aurelius chose not to have a rider scene depicted on his tombstone as was typical of Gallic cavalrymen in the Roman army. Aurelius wanted to depict ‘civilian and leisure attributes’; indeed, ‘the majority of British totenmahl stelae, however, commemorate women and/or children’.Footnote ⁵¹ And yet his identity as a cavalryman was important enough that he had a sword and helmet added to the background. Similarly, the cult of headhunting was so important to him that he wanted it explicitly included in a scene that it did not traditionally fit. It is possible that the cult of headhunting was so important an aspect of his identity not because of his ethnic origins, but because at Chester there was a particularly intense significance attached to headhunting because of the cult of Jupiter Optimus Maximus Tanarus that had developed there.