British Archaeology, no 6, July 1995: Features

The Roman historians Dio and Herodian were dismissive of the campaigns waged by the emperor Septimius Severus and his son Caracalla in northern Britain between AD208 and 211. No battles were fought, and following the death of Severus at York all the territories that had been campaigned over were abandoned.

Yet Dio and Herodian seem to have missed the point. Recent archaeological research in Scotland suggests that Severus had no intention of bringing the Caledonians to battle, but instead attempted to wipe them out by systematic devastation of the landscape. His policy, moreover, seems to have been successful, as peace beyond the northern frontier lasted for most of the following century.

The principal evidence consists of military bases associated with Severan activity. At South Shields, overlooking the Tyne estuary, a Hadrianic fort was reconstructed as a gigantic provisions depot. Up-river the great base at Corbridge, always a nexus for projected campaigns in the north, underwent major refurbishment which included the building of granaries. Far to the north, on the south banks of the Forth and Tay estuaries, forts were established at Cramond and Carpow. The purpose of these appears to have been to sustain by sea large armies campaigning north of the Forth, so avoiding the long and manpower-consuming lines of communication through southern Scotland which had characterised earlier Roman incursions.

In addition, a number of temporary camps have been convincingly identified as Severan. At Ardoch in Perthshire, for instance, two large camps (covering 25ha and 55ha) - which post-date the annexe of a fort in commission until the mid-2nd century - seem to represent successive seasons of activity, and the most likely recorded historical context is Severan campaigning in 209 and 210.

Fourteen other camps, similar in size, proportion and general layout to the 25ha camp at Ardoch, are known in eastern Scotland beyond the Forth. They trace lines north-eastwards through Strathmore towards Aberdeen, along the Angus coastlands, and into Fife. A similar pattern is followed by a series of 55ha camps, which thrusts inexorably from the Forth to the head of Strathmore. The camps are set on average 10 or 12 miles apart - a comfortable day's march for a big army.

What were these camps for? Contrary to general belief the progressive movement of a single large force through the landscape was not the normal method of Roman campaigning; and such substantial and coherent groups of camps as these are without parallel in the Roman world. Hostile territory was most readily dominated by ensnaring it in a web of strongpoints and roads, which allowed Rome's most powerful weapons - literacy and communication - to prevail. It seems therefore that Severus and his generals had something completely different in mind than the control and exploitation of a subjugated landscape.

When plotted against a modern map of agricultural potential the putative Severan camps run unerringly through the most productive land of eastern Scotland, and if a radius of 10 miles is drawn round each, virtually no hectare of prime agricultural ground remains uncovered. From the secure base represented by each camp, determined troops would have had little difficulty in systematically destroying the productive capacity of such an area - burning the standing or stored crops and killing the livestock. If the business was conducted around harvest time, the crops would have been at their most vulnerable and the army itself could live off the countryside it was laying waste. The American Civil War Unionist General Sherman pursued just such a policy during his infamous march through Georgia in 1864.

No direct contact with the enemy, whom Dio and Herodian describe as elusive, would have been necessary. Few would have survived winter in the devastated landscape; and in the following spring, competition for what little remained, combined with a chronic lack of seed and breeding livestock, would have made the catastrophe self-perpetuating. Severus's policy, in other words, seems to have been nothing short of an attempt at genocide of the Caledonian population.

Dr Colin Martin is a Reader in Maritime Studies at the University of St Andrews

In the flurry of excitement that greeted the discovery of a new set of Upper Palaeolithic cave paintings in southern France six months ago, the most interesting aspect of the paintings was rarely mentioned, let alone discussed at length. It is that the Grotte Chauvet, unlike any other known prehistoric painted cave in the world, is full of paintings of carnivorous animals.

Palaeolithic paintings 20-30,000 years old are notoriously hard to decipher, but on one matter nearly all scholars are agreed. In depicting a range of large herbivores such as ibex and reindeer, horses and bison, cave paintings in France and Spain reflect the main types of meat eaten by people in the period. Herbivores are depicted in the paintings at Grotte Chauvet too - recently dated to 30-33,000BP - but in addition there are numerous paintings of lions and bears, and perhaps the only Palaeolithic paintings in Europe of spotted hyenas and a panther. In one panel, several lions form a single composition, possibly stalking a herd of bison facing them. So what does it all mean?

Caves commonly formed dens for large carnivores in the Palaeolithic period, and in the Grotte Chauvet the bones and paw-prints of carnivores have been found associated with - and dating from the same period as - traces of human activity, such as hearths, footprints, worked stone, and places where pigments had been grubbed up and rolled into pellets. Some cave bear bones lie intact in hibernation nests, suggesting that the animals died in hibernation; but in many instances, carnivore bones have been disturbed, and placed in what may be `privileged' positions - as with a bear skull positioned on a large rock slab in the centre of a large chamber. Cave bears have sharpened their claws on the cave walls, and the claw marks often deface the paintings.

However, the evidence of animal-bone assemblages from caves throughout Europe shows that by the Upper Palaeolithic carnivores were using caves considerably less than in earlier periods. Whereas Neanderthal bones from the Middle Palaeolithic are often found in caves with gnaw-marks, suggesting the Neanderthals had been eaten by animals, no known Upper Palaeolithic human bones have been disturbed by animals. Taken together, the evidence suggests that, over time, humans overcame - or at least contained - the threat of carnivorous animals; and by fire, projectiles and other weapons, generally kept them away from caves they were using.

It may well be that man's broadly successful struggle against carnivorous animals partly explains their presence in the Grotte Chauvet paintings. Most Palaeolithic paintings were in some sense `religious' - executed as they often were in dangerous and inaccessible places, where the experience of seeing them would have been heightened - and perhaps in overcoming the major threat posed by bears, hyenas and lions, humans invested these animals with a religious or mythological significance. Such Palaeolithic `carnivore myths' may have been told, enacted and reinterpreted in ritual ceremonies held in the mysterious confines of the cave.

Any such rituals would have involved `reading' the paintings and interacting with the animal bones; and the Grotte Chauvet raises many new questions about the mythological thinking of Palaeolithic people. For instance, did they read the claw marks, paw prints and bones of bears as they read their own art? What did they make of the claw marks scratched through paintings? Did people see themselves as carnivores; or did they rather project human attributes onto carnivores; or did they do both?

The Grotte Chauvet paintings suggest that Palaeolithic people had a complex imaginative relationship with animals, including carnivores, that went far beyond the need to eat meat, and the need to compete for food and shelter. The precise character of Palaeolithic carnivore myths - if they existed at all - will probably always remain a matter of speculation. However, further excavations are planned in the cave, and it may prove to have more secrets yet to reveal.

Paul Pettitt works in the Research Laboratory for Archaeology and the History of Art, at the University of Oxford