Types of fortified sites

Roy General Highlands Maps

This map of the promontory fort at Burghead, Moray was drawn up by the 18th-century military surveyor General William Roy. Although the fort was associated with the Picts, it was almost certainly built earlier, during the Late Iron Age. Much of the fort was destroyed during the expansion of the town soon after Roy produced his drawing. (Society of Antiquaries, London)

Although hill-forts and brochs are the most commonly found type of Celtic fortification in Britain, other types of fortified sites existed in tandem with them. The scope of this book precludes a study of the less defensible of these, such as lake villages and crannogs, largely because these structures lacked any obvious means of defence save their encirclement by water. Although the lake village at Glastonbury, Somerset was surrounded by a wooden palisade, it was probably not designed to keep intruders out but rather as a protective barrier to prevent children and animals falling into the surrounding lake. Similarly while buildings built over the water, such as Oakbank Crannog in Scotland's Loch Tay, may well have been defensible for a short time, it lacked the protection to keep determined attackers at bay. For the purposes of this study we shall concentrate on the Celtic sites that appear to have been designed with defence in mind.

Celtic fortifications come in a variety of types, although with the possible exception of brochs they all share certain characteristics. Over the years archaeologists have developed terminology that helps them classify the intricate systems of ditches and banks they encounter. These are often augmented with more widely understood fortification terms to help explain how these features were supposed to work. For example, many archaeologists use the words 'bank' and 'rampart' interchangeably, but to be more accurate the fortification term 'rampart' should really only be applied to the innermost bank surrounding the fort's enclosure. If a fort is enclosed by a single circuit of bank and ditch, it is described as a 'univallate' fort. More complex fortifications are described as 'bivallate' (if they have two such lines of defence), 'trivallate' (if they have three circuits of bank and ditch), and 'multivallate' (if the fort is defended by more than three lines of defence). If the banks are set close together they are described as 'compact', while the opposite are regarded as 'dispersed'.

Given the range of different types of fortifications, most fall into four general categories. The first are the pure hill-forts, whose defences are placed to make best use of the terrain. The perimeter of the enclosure follows the contours of the hilltop on which the fort is built, and consequently these fortifications are rarely circular or even regular. Instead the lines bend in accordance with the shape of the hill. The result is an irregularly shaped defensive position, but one that makes best possible use of the lie of the land. A variation on this is the headland or promontory forts found on rocky spurs of coastline where the site could be made defensible with relatively little work. In sites such as Burghead in Moray, Scotland, or Rame Head in Cornwall these fortifications close off the landward side of a headland by means of a defensive bank and ditch system similar to that found in hill-forts. The only difference is that on the remaining sides of the defended position the sea itself provides a natural barrier to attackers. It is worth noting that forts of this kind are not always found on the coast. On rare occasions where the terrain provides a similar advantage far from the sea, such as at the confluence of two large rivers, a similar position could also be established. An example of this is the Iron Age settlement at Dyke Hills in Oxfordshire, where the River Thame meets the River Thames near modern-day Dorchester-on-Thames. There a bivallate defensive line was created to protect the settlement, while the two rivers protected the remaining three sides of the site.

A variant on the hill-fort is what has unsatisfactorily been described as the plateau fort, or valley fort. These are similar to the more common hill-forts, but were built on sites that possessed no defensive advantages such as slopes or even rivers. Instead they had to rely on their own man-made defences to keep attackers at bay. The only real advantage of this type of fortification seems to be that in many cases the enclosure included a natural spring, so that, unlike many hill-forts, the defenders had access to water and thus in theory could better withstand a siege. Often these appear to have been built in areas of good farmland, in valleys or on broad ridges where no more obviously defensible feature was available. An example of this type of early Iron Age fort is Rainsborough in Northamptonshire, which was excavated during the 1960s. There the fortification stood on the edge of a plateau, where the ground then fell away gently into the Cherwell Valley below. It enclosed an area of some 2.5 hectares, and although its defences were univallate, the archaeologists uncovered traces of an outer bank that had been filled in at some stage during the fort's occupation. The occupants may have come to regret the siting of their fort: archaeologists also uncovered evidence that the gateway had been destroyed by fire, and a skeleton was found amid the burned ruins of a guardhouse. The inference is of course that the fort was attacked and captured probably at some point in the 3rd century BC.

Another distinct group of forts comprises those that were clearly built as non-defensible enclosures, probably to house livestock or to provide a seasonal home for a farming community. Sites of this kind were often built on the sides of a hill or in a flat area, and often involved multiple enclosures encircling a central area. While not primarily designed as a defensive position, these could serve as an emergency refuge in time of danger. Examples of forts of this type include Lordenshaws in Northumberland and Clovelly Dykes in Devon, and they are generally located in either the north of England, the south-west, or in the western

Types Fortifications

The Iron Age hill-fort at Woden Law in the Scottish Borders was built in three phases, and in its final form consisted of a double rampart and ditch.The curving double bank in the foreground has been identified as a Roman siegework, suggesting that the small hill-fort might have been besieged during Agricola's campaign in southern Scotland around AD 80-81. (RCAHMS)

Types Fortifications

Major areas of broch concentration 50 miles



The major areas of broch concentration in Scotland; the sites mentioned in the text are also indicated by name.

Shetland Islands

Clickhimin b


Old Scatness %jarlshof

Old Scatness %jarlshof

Dun Vulan


Dun Vulan

Old Scatness MapBarry Hill Angus Fort

The Iron Age hill-fort on Barry Hill, Angus on the edge of the Scottish Highlands was built over two phases: the large oval bank of rubble that represents the later phase was built on top of an earlier earthwork. A lesser outwork protected the southern and eastern approaches to the fort, while a small area on the north side may have served as an animal enclosure. (RCAHMS)

The Iron Age hill-fort on Barry Hill, Angus on the edge of the Scottish Highlands was built over two phases: the large oval bank of rubble that represents the later phase was built on top of an earlier earthwork. A lesser outwork protected the southern and eastern approaches to the fort, while a small area on the north side may have served as an animal enclosure. (RCAHMS)

coastal fringe of Wales. Clovelly Dykes is a prime example, with a central compound of 1.2 hectares surrounded by an oval bank and ditch, which in turn was ringed by another full bank and ditch, then two more low-banked enclosures, some of which had associated ditches in front of them. The site also boasted a large annex enclosure on its western side, protected by another ditch. In all the fortification encompassed an area of some 9.6 hectares. This complex site was expanded during its period of occupancy, as it seems the outer enclosures were added after the innermost enclosure was built. Given the minor nature of the outer works it is probable that these served as livestock enclosures, while the inner enclosure housed a small farming community or even a large farmstead.

Even smaller fortified enclosures have been found, particularly in Scotland where these structures are known as ring forts or raths. These small enclosures were almost always univallate, protected by a circular or near-circular bank and ditch system. In some cases the enclosure encompassed a natural or artificial mound, but they could also be built on level ground, such as the ring forts found in Glen Lyon, which may have been built as boundary markers or outposts, guarding the developed Iron Age community of Loch Tay and the summer pastures in the hills between the glen and the loch. Other ring works seem to have served to protect farmhouses, a little like the fortified farmhouses found in medieval Europe. Examples of these include Dan-y-Coed and Woodside, both in Dyfed, Wales. What is unusual about these last two sites is that they were strongly fortified, with a bivallate defensive system, and both were approached along a pathway lined by more banks. It has been suggested that this banked approach served as a means of herding cattle to and from the main enclosure. The impression that these were farming settlements rather than villages is enhanced by archaeological evidence of roundhouses lacking central hearths - suggesting the presence of farm buildings rather than dwelling houses.

Finally there are the brochs, the majestic stone towers found in northern Scotland. There may well have been a crossover between these buildings and the stone-clad ring forts of central Scotland or the stone-built hill-forts found in Wales. However, there is virtually no evidence that the people who built these structures had any association with the builders of Celtic sites in the rest of Britain. Given the limited scope of this book we can only touch on their basic features and suggest how they might have served the people who built them. Armit (2003) covers the subject in considerable detail, and is thoroughly recommended for those who want to explore this topic further.

+3 0

Post a comment