History can be made anywhere and in any way but the history we remember and are generally taught in schools is very often that of battles - Hastings (1066), Agincourt (1415), Waterloo (1815), Normandy (1944), and the rest. A revival of interest in military history has created an interest in ancient and modern battles as events, and historic battles have become a vital part of our national heritage.

While the places where those events took place - the battlefields themselves - are usually only considered in passing by students of warfare, the creation of the category of `historic battlefield' (through English Heritage's 1995 Battlefields Register) invites us to examine battlefields in their own right. They have a great deal to tell us.

The type of landscape where battles were fought has changed greatly over time. Three phases can be recognised, from the earliest periods of organised warfare, through early modern battles, and on into our own age. At Maldon (991) the battlefield is flat and featureless, generally typical of warfare from Megiddo (1469BC) - the earliest battle of which we have record - to the end of the Middle Ages. At Maldon, two forces of heavily armed and armoured foot warriors slugged it out hand to hand in bloody m�l�e, Anglo-Saxons unsuccessfully defending their land and homes from marauding Norsemen. The flat ground was integral to this style of fighting and type of war. The Saxons allowed the Norsemen space and time to deploy, giving up the advantages of both surprise and manoeuvring space to engage in combat.

The same idea can be seen in other places at earlier times. At Marathon (490BC) the Athenian hoplites waited eight days before charging into the more numerous Persians, whose strength had grown over that time; and at Plataeia (479BC) eight days also passed before the Greek army abandoned its strong positions in hills to deploy onto the empty plain to meet its outnumbering foe. For the ancient Greeks, it seems, battle with a foreign enemy could perhaps only take place after eight days had passed; and in both these cases strategic advantages were given up in order to use the battlefield in the `correct' way.

The Greeks' internecine wars were also governed by rules, with a fixed sequence of events - invasion of territory; the pointless infliction of limited damage to agricultural production; the mustering of an army; and on the day of battle the shock of a sudden charge followed by an intense but very short period of hand to hand butchery.

Flat featureless space was the landscape of macho fighting men and highly ritualised warfare where the architecture of battle was created by bodies of armed fighters.

The landscape of battle began to change by the end of the Middle Ages, albeit slowly and gradually. At each of the three great English victories of the Hundred Years' War against France - Cr�cy (1346), Poitiers (1356) and Agincourt (1415) - apparently flat space was sought to fight on, neatly bounded by woods or obviously wet ground, an ideal theatre for the headon clash of armoured horsemen and foot soldiers. In all three, the French mounted men at arms encased in heavy armour recklessly charged home, foundered in mud and confusion, and died not only under accurate fire from longbows, but also of drowning, heat-stroke, dehydration and exhaustion. They never learned - charging home was how war was done.

By the 17th century, however, gunpowder ruled the battlefield and landscape features were coming into use although the majority of battles of the English Civil War were fought on relatively featureless ground. At Edgehill (1642) an attempt to use the advantage of a steep slope as an obstacle failed since the enemy refused to attack; the Royalist forces occupying the hill had to deploy in the valley below to meet the enemy. They need not have - strategy simply required blocking the Parliamentarian advance, an objective met by the placement of the Royalist army on its high position. By moving downslope, battle was met but without military necessity.

The use of landscape features was more prominent at Naseby (1644) where dis-mounted dragoons were stationed along a fence to fire into the flank of enemy cavalry. From this period on, features which provide obstacles to movement, cover and places to deploy are highly prominent in the battlefield landscape. At Waterloo (1815), hillslopes, buildings and sunken roads were essential components of the battlefield space.

It is no accident that the 18th century is also the period of the Enclosure Acts, the division of common land into private plots, of Model Farms, of Capability Brown and the landscape garden - where landscape features themselves take on significance and meaning. In early modern history, the landscape of battle is made up of bodies of troops and landforms in conjunction.

The third period of battlefield landscapes emerged during the course of the 19th century, together with the increasing industrialisation of the modern world. In the United States, the American Civil War (1860-1865) saw the first battlefield use of barbed wire and extensive entrenchment. Sherman's `March to the Sea' (Atlanta to Savannah, 1864) saw the deliberate devastation of territory. The fighting around Port Arthur between Russia and Japan (1904-1905) demonstrated the effects of the machine gun on infantry. By the time of the Battle of the Somme (1916), military technology could encompass the destruction of everything within the battle area - soldiers, trees, buildings, whole hillsides. The featureless landscape had returned, not chosen this time but made by destruction on a truly industrial scale. The power of the nuclear blast is the ultimate expression of this mode of warmaking.

Battles are essentially material phenomena because they comprise the coming-together of three material elements - the landscape of the place of battle; the technology of a particular style of warfare; and the people who fight. The type of landscape chosen or constructed for battle reflects the attitude of its age towards how war should be conducted.

The relationship between landscape form and the technology brought to it determines the precise tactics of the battle. The relationship of technology and soldiers determines systems of movement and control for that battle. The relationship of landscape with the soldiers converts that landscape to a place with meaning.

Where landscape as a category meets the idea of place, the location begins to evolve a cultural significance. The battlefield becomes a receptacle of memory of the event that took place there, and the people who were present become associated with the place; it becomes part of their identity. Where memory meets identity, we enter the field of heritage, where the category of `historic battlefield' finds its home.

The form of battlefield landscapes gives expression to the ideology of a particular period. As temporal and spatial lenses focusing the concerns of an age, they are the places where issues of political, moral and legal legitimacy were contested and decided. They express attitudes towards life, death, place and landscape. They are places where questions of identity - local, regional, national, professional, individual - were resolved. They formed and disrupted social bonds. Battlefields have hitherto suffered a lack of interest as landscapes for study, but their historical importance gives them the right to claim further investigation.

Dr John Carman is a Research Fellow in Archaeology at Clare Hall, Cambridge