Sonderdrucke aus der Albert-Ludwigs-Universität Freiburg HEIKO STEUER Warrior bands, war lords and the birth of tribes and states in the first millennium AD in Middle Europe Originalbeitrag erschienen in: Ton Otto (Hrsg.): Warfare and society : archaeological and social anthropological perspectives. Aarhus: Aarhus Univ. Pr., 2006, S. 227 - 236 Warrior Bands, War Lords, and the Birth of Tribes and States in the First Millennium AD in Middle Europe HEIKO STEUER 16The subject: the background It is my intention to formulate a model on war. But it is not warfare itself that will be considered in all its aspects (Steuer 2001; Jorgensen and Clausen 1997), but rather the causes and effects of wars conducted by war lords and their warrior bands. Wars from the 4th century BC to the 10th/11th century AD will be considered in devising the model - wars spanning a period of more than a thousand years, from the ancient Celts to the Normans. Based upon the reports in the written sources, characteristic phenomena (Erscheinungen) will be singled out as criteria for defining this type of warfare. Not all criteria will be obtainable for every epoch, but in the overall view a varied number of criteria can confirm the comparability of the armed conflicts and therefore also their socio-political backgrounds. (This model is not a newly formulated thesis. Most of it can be found in Wells 1999). The question of the expression of these events in the archaeological sources will only be raised in the second place, and it will be shown that, contrary to general opinion, warrior bands can never, or rarely, be recognised using archaeological methods. On the contrary, following the conclusion of the socio-political process - the birth of tribes and states - the later occupation of land as a result of these wars would leave marks on the archaeological remains. The Celtic and Germanic societies of central Europe of the first millennium were permanently changing. These variations appeared regularly, to some extent in waves or phases. We can abstract the rules; but history does not repeat itself, that is why the various epochs are not completely identical. The different reports in the written sources describe this change in an indirect manner. The recorded names of the active groups - we will call them tribes following Caesar, Tacitus and the later Ammianus Marcellinus - emerge and disappear again. The names of tribes during the time of Caesar (100-44 BC) (De Bello Gallico, 58-52 BC) differ from those of the time of Tacitus (55-116/120 AD) (Germania, 98 AD) or those of Ammianus Marcellinus (second half of the 4th century), Gregory of Tours (ca. 540-594) (Historia Francorum), or the time of the Carolingian and Ottonian historians. 150 to 250 years lie between each of these reports, the equivalent of five to eight generations. Various ethnogeneses (tribalisations) are reflected in the change of the names, but even in the preservation of an old name the old tribe does not continue. A new ethnogeneses creates other political entities. There are Ariovistusus' Suebi, the Suebi of the age of Tacitus, the Suebi (Suevi) of the migration period in Spain, etc. The same applies to the Marcomanni WARRIOR BANDS, WAR LORDS, AND THE BIRTH OF TRIBES 227 during the times of Ariovistusus or Maroboduus and those of the Marcomannic Wars 166-180 AD, which are not the same tribe. The Alamanni, mentioned for the first time around 300 AD, were made up of various groups of Germanic people, then constituted themselves in a second ethnogenesis (tribalisation) in the Merovingian Empire around 500 and then again for a third ethnogenesis (tribalisation) in the dukedom/duchy of the 8th/9th century. What are these tribes? Are they the population of territorial units, the unification of several settled landscapes with large numbers of villages (Siedlungskammern)? It is a matter of discussion whether the tribes which are mentioned by name in the ancient sources emerged only as territorially bound units in reaction to the overly powerful threat of neighbouring empires or states; as the open ranked society based on clan-like organisation with its numerous inner dependencies had become obsolete. The Celtic tribes with their central oppida mentioned by Caesar are a reaction to the pressure of the Mediterranean states and only appear in the Middle and Late La Tene period with the aid of returning mercenaries. The Germanic tribes may only have constituted themselves due to the pressure of the Celtic oppidum-civilisation on the Germanic clans, to which Germanic mercenaries in the service (army) of Celtic nobility may have contributed on their return home. Ethnology has pointed out this development for Africa during the colonial age of the 19th century: before the intrusion of Europeans there would appear to have been none of the tribal structures that later on became apparent and which up to today provide a continuous cause for conflict and wars (Vail 1989; Sigrist 1994; Lentz 1998). Today the political and social organisation of rule under Shaka (Marx 1998), the founder of the Zulu tribe or realm, is under discussion in this way, and likewise the deadly conflict between Hutu and Tutsi, today two different tribes, but before colonisation two different economic structures: farming and cattle breeding. The idea of tribes (gentes), of a people (Volk), and of a nation developed as part of the nation-state thinking of the 19th century. Events in the present (in the Balkans, in Africa etc.) show that ethnic clashes may also be conflicts among rival bands of thieves (warrior bands). But it is impossible to speak about these terms (termini) in all their complexity in this paper. In recent decades there has been a very controversial and many-sided discussion concerning the terms 'tribe/Stamm' or 'ethnogenesis/tribalisation/Stammesbildung' (compare developments for instance from Wenskus 1977 to Pohl 1998 and 2002, or also Bowlus 2002; Barth 1969). In numerous lemmata the Hoops Reallexikon der Germanischen Altertumskunde (Pohl 1998) tried to specify this discussion in its many aspects. For some other terms like Gefolgschafr (comitatus), i.e. followers or retinue, problems in the translation from Latin into German or the English language exist as well. For the structuralist attempt - an old one, not a poststructuralist or contextual attempt - made in this paper it may be enough to formulate a model as an idea. I VOlkerwanderungen' , translated as migrations of whole peoples, were in fact not migrations of peoples but rather military campaigns. These campaigns only had temporary camps as stations, not permanent settlements. Campaigns of warrior bands could be undertaken without difficulty, as the landscape was relatively empty. Although the landscapes were populated completely and systematically with settlements like a network in distances not more than 5 kilometres as a rule, there was still enough space to allow larger military units with their horses to pass through, either unnoticed or officially sanctioned. The groups called in the sources by names such as Suebi or Langobardi were warrior bands and not neighbouring societies living in villages with families or kinship groups based upon relationships among relatives, and with a clan-system. Perhaps only the leading clans went by these names. Through their concepts of order, ancient writers equated the names of the mobile and multi-ethnic warrior bands with inhabitants of the territories from which some of the warriors were drawn. The authors, with their background in the Roman Empire, did not recognise the principal difference between warrior bands and rural populations with the same names. The names randomly appear next to one another, not because parts of tribes had migrated to other countries, but because a military unit had been divided and deployed at various places. The VOlkerwanderungen of that period therefore are not the migrations of tribes with the whole family, their mobile belongings, cattle and all, as was believed in earlier research, but rather campaigns of warrior bands whose wars only much later led to the occupation of land. Their emigration did not leave the native lands depleted of their population. 228 • WARFARE AND THE STATE Depopulation had completely different reasons and backgrounds. For instance: - The Germanii disappear from the territories east of the river Elbe long after the end of the so-called migration-period in the 6th century, and Slavic tribes immigrate. - The landscape of Anglia (Angeln) in modern Jutland, and other areas along the coastline of the North Sea, only appears to become depopulated more than a century after the emigration to England around the middle of the 5th century; it would therefore seem that only latecomers were concerned. - People, i.e. warrior bands, left the densely populated areas. Even when these warrior bands settled in new countries, no major changes came into effect in the native areas, a fact that can be attested in the archaeological features - the structures of the villages show no change. At the very least no decline in the density of the population can be registered. Wars are an immanent part of social change. Ethnogenesis or tribal constitutions (tribalisations) and the development of states were not possible without wars. One therefore has to consider the effect of the feedback. Because higher organised societies (empires) influenced their surroundings, tribes developed, taking the place of clan-societies, in order to defend themselves. Warrior bands split off after the constitution of these tribes. Because warrior bands appear and influence the higher organised societies (kingdoms or empires) through mercenary services or as organised bands of thieves (warrior bands) threatening them militarily and through raids they in turn attempt to influence the tribal communities through paying mercenaries or drawing up contracts with these groups in order to use the tribes and warrior bands for their own ends. War War is the armed conflict between groups of men, i.e. armies of various sizes (Steuer 2001: 347ff). a) The armies or warrior bands or 'Gefolgschaften' as military units can be levied from the villages of a territory in order to protect them. Out of the clanbased society - due to the necessity of organisation - a tribe will arise, possessing its own territory. This tribe can then even erect special fortifications to defend the settlement area as a whole, as can be observed during the later period of the Roman Iron Age when constructions consisting of ramparts, palisades and ditches were erected along the borders on land or in the sea. The territories were up to 30 to 50 kilometres wide. These areas of roughly 2500 square kilometres would incorporate about 100 villages with areas of 25 square kilometres per village, each village consisting of 10 households with 10 inhabitants per household which in turn leads to a total population of 10,000 people. Up to 20% of this total number could go to war, which means that such an area could raise an army of up to 2000 warriors. Another method of reckoning is based on the number of weapons in the find from Illerup: based on the ratio of excavated shield-bosses made of silver, bronze and iron which are 6 to 30 to 350, a leader with a silver shield commanded a warrior band with the strength of a Roman centuria' (60 to 80 soldiers). The entire force could easily have reached the size of a Roman'auxilia' unit with 1500 to 2000 men, corresponding to the number of warriors from the area of one tribe in the Germanic territories. The warrior in the burial of Gommern in Saxony-Anhalt with his precious grave-goods and silver fitted shield corresponds to the leaders in the military unit from Illerup in Jutland, and he would have been one of the leaders of the warrior-contingent from the 30 kilometre wide settlement landscape around Gommern. b) The military units (army) can be made up of warriors and a leader (king, rex or dux) who completely separate themselves from the structures of the clan or tribe and move about in an 'unattached' manner in order to plunder and pillage (gaining the spoils of war-booty 'Kriegsbeute'). Warriors from various tribes came together and the old ethnic affiliations lost their meaning. c) These military units can be recruited and offer their services to a higher order of state (an empire) as auxiliary units, or as groups of mercenaries. Warfare is a lifestyle, combined with a certain mentality of the warrior (Bodmer 1957). This is a behaviour which accepts war as a way of life and struggle of existence (besides the rural life), from which a warrior or noble caste emerges. War was waged almost every year in the early historic societies of the first millennium. The function of a military unit varied, depending on its role in the formation of a tribe or the constitution of a state. The organised states or empires, as in the cases of the Roman Empire or the Merovingian Empire, waged war annually along various borders, sometimes in defence, but mostly in order to expand their political WARRIOR BANDS, WAR LORDS, AND THE BIRTH OF TRIBES • 229 power and to obtain war booty. The warrior bands in turn waged war every year in order to `earn' their living, either against other armies or, more often, against the settled communities of tribal societies, or they waged war against the more organised states in the form of raids. The historical phases Early and Middle La Tene Celtic mercenaries appear during the 4th century in the area of the Mediterranean, for example in the wars of the tyrant Dionysius (c. 430-367 BC) of Syracuse against Carthage. The invasions of Celtic warrior bands into northern Italy commenced soon after 400 BC, culminating in the plundering of Rome under Brennus and the battle at the river Allia in 387. The withdrawal from Rome was achieved with the payment of 1000 pounds of gold. The Celtic occupation of land began in the ager Gallicus along the Adriatic coast. The Celtic groups named in this migration were the Senones and the Boii, as well as the Insubres. None of this can be proved archaeologically; only solitary assemblages with weapons and belt ornaments are to be found among the cemeteries of local character from the 5th to the 3rd century BC; settlements remain unknown. The Celtic immigrants would appear to have adapted immediately to the local culture, to have acculturated themselves. But there is a cultural feedback to the areas from the north of the Alps to the river Marne, mirrored in the decorative style of the jewellery. A general feature of importance to this discussion is that mobile warrior bands allow themselves, as a first step, to be recruited as mercenaries. They return to whence they came with their pay. The early Celtic minting of coins in gold and silver imitates the coins of Phillip (*382, king 359-336) and Alexander the Great (*356, king 336-323). These mercenaries come from the cultures on the fringes (Randkulturen - cultures in peripheral or border regions) of the more state-like higher organised societies. In the next phase, they resort to wars of aggression under the leadership of warrior-kings (Heerkonige - war leaders, commanders). The name Brennus is an occupational name and relates to the Breton word brennin - which means king. The names of the `tribes' in the written sources are not valid for whole peoples, but rather for the warrior bands. The occupation of land only begins later and is not reflected in the archaeological evidence. Late La Tene Around 72/71 BC Ariovistus, rex Sueborum (Pliny Historia Naturalis 2,170) or rex Germanorum (Caesar de Bello Gallici 1,31,10), appears among the Sequani as a leader of mercenaries. They had employed him and his 15,000 warriors for the war against the Haedui. The number grew to 120,000 as time passed on. In the decisive battle against Caesar 58 BC 24,000 Harudes, descendants of the warriors that had once roamed Central Europe with the Cimbri, as well as Marcomanni, Triboki, Vangiones, Nemeti, Eudusii and Suebi - totalling at least 7 different groups could be found amongst the armies. That this coalition was made up from various individual warrior bands is a remarkable fact. Where Ariovistus came from is still unknown - he may have been a Tribokian. But these names again do not describe peoples, but rather warrior bands that may indeed mainly have been recruited from just one tribe. Ariovistus remained as a mercenary leader in Gaul and controlled one third of the territories of the Sequani. The campaign ended in the year 58 BC with his defeat, after which the troops moved about for 14 years 'without a roof over their heads' (qui inter annos XIIII non subissent I Caesar de Bello Gallici 1,36,7). There are no archaeological traces of Ariovistus' campaigns. But archaeologists think they may be able to prove an occupation of land through the `Germanic' finds in the Wetterau and from the Rhine-Main area to the northern part of the upper Rhine in the Late La Tene period. There can be no doubt that there are correspondences in the inventories of pottery and weapons between the Polish Przeworsk Culture and cultural groups in central Germany, Thuringia and the river Elbe region. There is a cultural link between the groups along the Rhine and the cultures further east. It cannot yet be decided if this is an extension or expansion of an archaeological culture group (Kultur- or Formenkreis), or if it is an immigration of people towards the Rhine, as the archaeological finds in both regions are contemporaneous. But there is no connection between the mobile and very mixed warrior bands of Ariovistus and the rural settlement areas along the Rhine and Main rivers with influences from the areas of the Przeworsk Culture, as Suebi from distant territories only made up a small part of the troops. 230 WARFARE AND THE STATE The age of Arminius Ariovistus was not only the enemy of Caesar, but was also amicus populi Romani in the year 59 BC. And the Cheruscan with the name of Arminius from the stirps regia of the tribe was a Roman citizen and belonged to the class of eques, he was a Roman knight: he had been to Rome and led a band of mercenaries, an auxiliary unit, in the Roman army. In the famous Varus-Schlacht (clades Variana), or Schlacht im Teutoburger Wald', some 'tribes', the Marsi, Bructeri, probably Chatti and later on Langobardi and Semnones, could be found fighting alongside the Cherusci under his command. The term Cherusci sociique is used. Again a coalition of various warrior bands came together, some groups may earlier have been in action as auxilia, or other groups that had not been in Roman military service. The so-called `Germanic Battle of Freedom' under the leadership of Arminius was - according to the historian Dieter Timpe (Timpe 1970) - no popular uprising, but rather the mutiny or rebellion of Roman mercenaries, of a regular or normal auxilia. One other war lord of this period was Maroboduus, genere nobilis (Gaius Velleius Paterculus 2,108,2), a Marcoman who had also been to Rome before he made himself a king. He then assembled a coalition of various warrior bands, made up of Lugii, Zumi (Zumern), Butoni, Sibinii (Sibiner), Semnones and Langobardi, a force of 70,000 foot soldiers and 4000 horsemen. The areas of settlement of the Marcomanni in Bohemia, to where, following his differences with Arminius, Maroboduus had retreated, and the emigration of the Marcomanni after the death of Maroboduus into Slovakia are said to be recognisable archaeologically. These are, however, patterns of rural settlements which cannot have had any direct connection with the mobile warrior bands. The age of the Marcomannic Wars 166-180 AD The wars of Marcus Aurelius and his son Commodus were conflicts involving the Roman army in defence against the warrior bands that were threatening the Roman provinces from the interior of the Germanic territory. In the beginning, the Victuales and Marcomanni, as well as others, are mentioned (Marcus Aurelius 14.1: Victuales et Marcomannis cuncta turbantibus aliis gentibus), who demand the allocation of land as a tribute following their raids. An army of 6000 allied Marcomanni, Langobardi, Obii and others is mentioned in the year 166-167. Later on, the Marcomannic king Ballomarius is mentioned as conducting negotiations and as speaking for 11 legations (Cassius Dio 71.3.1a), therefore as a leader of a coalition with ten other war lords. The traces of the Marcomannic Wars are said to be clearly recognisable in the archaeological record, for example in the presence of Roman weapons - swords with ring pommels and coats of ring-mail - as well as spurs (Stuhlsporen) of the special Mu§ov-type in a corridor of traffic between Jutland and Bohemia via the river Elbe. A feedback into the areas whence the Germanic armies came could thereby be recognised. The so-called VOlkerwanderungszeit, or Migration Period in the 3rd and 4th centuries The invasions of Alamans, Franks and Saxons into the Roman Empire began as raids of warrior bands who wanted to pillage. Some of the warrior bands are directly employed as mercenaries, not least as garrisons of the Roman forts along the imperial borders of the Rhine and Danube in late antiquity. The Roman commander - and later emperor Julian won the battle of Argentoratum (Strasbourg, France) over an Alamannic army in the year 357. The Alamannic leaders were king Chnodomar and his nephew Serapio/Agenarich, who had assembled a coalition of five kings (Vestralp, Urius, Ursicinus, Suomer and Hortar) as well as ten regales (on horseback) and a respectable number of nobles (optimates), in total 35,000 armed men (armatores) on the side of the Alamanni. There were 17 war lords for 35,000 men, which meant that every one of them had 2000 to 3000 warriors under his command. Other Alamannic kings are known from this period, as found in the descriptions of Ammianus Marcellinus; for example, the brothers Gundomad and Vadomar, and the son of Vadomar with the name of Vithikap, were leaders of war bands - which at times waged war upon Rome, and at other times were in the service of Rome as leaders of mercenaries. In the Roman military handbook, the Notitia Dignitatum from the time of around 400 AD, there are mentions of Germanic units, the Brisigavi iuniores and the Brisigavi seniores from the Breisgau, the Lentienses, the warriors from the Linzgau, and others. The Romans recruited warriors from regions in which Alamannic groups had begun to settle. A phase of occupation of land ensued, following the initial period of raids and service as mercenaries, constituting the basis for the first ethnogenesis (tribalisation) of the Alamanni. WARRIOR BANDS, WAR LORDS, AND THE BIRTH OF TRIBES • 231 Archaeological traces from the beginnings are very rare. There are only a few burials, the total number being out of all proportion to the strength and numbers of the armies of which we know. It may therefore be assumed that at first these war bands returned to their homelands on both banks of the river Elbe some hundreds of kilometres from the Limes. The discoveries of enormous lost booty from the Rhine, as in the cases of Neupotz and Hagenbach, are proofs of this thesis. In the areas of origin beyond the rivers Elbe and Saale, in Thuringia, the so-called princely graves of the Hassleben-Leuna group are interpreted as the burials of leaders of mercenaries and their families. A number of years ago Joachim Werner (1973) pointed out the rich furnishings with vessels made of bronze and glass, with Aurei as Charon's-pennies and with knobbed bow fibulae (Bagelknopfribeln) — Roman officers' fibulae — to mark mercenaries from the'Gallisches Sonderreich' , the separated Empire of Gaul under emperor Postumus. It appears that from the 3rd to the 5th centuries warrior bands moved to the south from beyond the river Elbe and later made way for the ethnogenesis (tribalisation) of the Alamanni. Warrior bands from northern Germany headed into northern Gaul and led to the ethnogenesis of the Franks, and the warrior bands of the Angles and Saxons from the coastal area of Schleswig-Holstein contributed to an ethnogenesis in England. The density of settlement after the acquisition/occupation of land only began to increase during the 5th century in the Alamannic area. Military hillforts and rural settlements belonged to this pattern. Only during this phase does an increase in the population become archaeologically evident, which maintains close cultural relations to the areas of oilgin beyond the Elbe. A steady flow of people towards the south is also recognisable, but so is a cultural feedback which influenced material culture, from fibulae to ceramic inventories. It remains unanswered how far the Roman military-belts of the 4th/5th century which can be found in grave furnishings in Germany — should to be seen as indicators of returned mercenaries. Germanic craftsmen themselves began to produce belt mounts of that type even quite far from the border to the Roman Empire as a result of a fashion trend. It is rather the deposits of gold coins, ring-jewellery and fibulae which were probably the possessions of former high-ranking leaders of mercenaries. The Vikings in the west during the 9th and 10th centuries The number of raids of the Danish and Norwegian Vikings into the Carolingian Empire was on the increase from 840. War lords with war bands came in order to plunder with their ships that could transport 30 to 60 oarsmen and warriors. In the beginning they returned to the north, but later they began to spend the winters in enemy territory. With 100 ships easily 3000 to 6000 men were involved. The Danish king and war lord Harald Klak offered mercenary services, and Louis I ('The Pious' [died 840]) granted him a fief in Friesia, so that he would fight other Vikings. The Danish king Horik I offered Louis his support for the same reasons. He also demanded Friesia, and furthermore the lands of the Abodrites as fiefs. In order to give his demands more weight, he sent several hundred ships against Hamburg. As the Germanic kings and war lords had offered mercenary services and 'foeratii contracts' after their raids in Roman times, so the Viking kings were now attempting the same. Under the reign of Karl II (`The Bald', dem Kahlen [838-877]) the payment of `Danegeld' began. It was the time of the Viking irregulars under independent chiefs, who in turn were linked in alliances. A chief or war lord by the name of Weland let himself and his followers be recruited by Karl for payment, and fought other Viking irregulars/bands. He had access to 200 ships and wanted 5000 pounds of silver as well as provisions for driving other Vikings away. Such alliances between Vikings and local rulers were also to be found in England and Ireland before the Vikings openly resorted to raiding. A great band or army of Vikings gathered in the year 865 in Kent and remained together until 879; the army conquered several Anglo-Saxon kingdoms. Parts of this army began to settle in England from 876 onwards; others moved on against France in 879 and besieged Paris in 885 and 887, but returned to England in 892 and were defeated by Alfred the Great. This socalled Great Army, a coalition of numerous war lords and their armies, was continuously strengthened through fresh units. Attempts at colonisation only began after 900 (see Coupland 1998). The Normans and the Mediterranean during the 11th century The Viking leader Rollo spent the first ten years of the 10th century fighting for booty in the area of 232 WARFARE AND THE STATE the river Seine. He was granted a fiefdom in the year 911 after converting to Christianity - land at the estuary of the Seine, later known as Normandy. He thereby became the first duke of Normandy. The descendants of Rollo tried to strengthen the fiefdom and to incorporate the 'Viking' nobility in a tighter fashion, but parts of this group rebelled and moved on to new raids in the Mediterranean. The first group of Norman knights appeared in Salerno in southern Italy in the year 999: They were a group of pilgrims returning from Jerusalem and had stopped over in Salerno when the city was attacked by Saracens. They procured weapons and horses and chased the Saracens away. The lord of Salerno, Gaimar IV (999-1027), wanted to employ them as mercenaries, but they returned to Normandy accompanied by envoys who in turn were to employ others as mercenaries. Other versions mention that warriors from Normandy appeared in southern Italy at the latest in 1015-16. They were in the military service of the local nobles and therefore in the service of the emperor against the Muslims or Saracens and Byzantine enemies. 250 Normans are listed as defeated mercenaries in a battle against Byzantine forces; various later sources mention contingents of 300 knights under the command of leaders who were granted Sicilian towns as fiefdoms. The Norman Robert Guiscard followed his brothers to southern Italy in 1046-47, conquered Calabria and became duke of Apulia and Calabria as well as Sicily. He sent his younger brother Roger (26 years old) with just 60 knights to hold the area of Calabria. The conquest of Sicily followed in 1061 through a first wave of attack with 13 ships and 270 knights which was then joined by a group of another 170 knights in the second wave of attack in order to conquer Messina. The battle and victory of 700 knights over 15,000 Muslims followed. Sicily, which had become Muslim in the 9th century, had been reconquered for the empire in the later part of the 11th century; Palermo was taken in 1072, the last Muslim outpost fell in 1091. The rules of the war lords and warrior) band warfare Criteria for this special kind of warfare can be given with the help of the following scheme of develop- ment: 1 In the beginning it was mercenaries from the central European clan and tribal communities who offered their services to the armies of the more advanced organised states. They returned home with gold and silver which they had received as pay, as well as with concepts and ideas of the more advanced forms of societies. 2 The formation of warrior bands under the leadership of war lords - who gathered entourages of warriors around them - followed next; the war lords and their warriors emerged from the old tribal society, separated from it and became individual entities with their own names that had less and less involvement with their communities and lands of origin. 3 In order to keep these warrior bands together, the war lords had to see to it that a steady income was ensured, which in turn was gained through warfare, although sometimes the mere threat of war was sufficient to trigger payments of tributes. These raids were aimed at areas outside the home territories, often enough (in terms of civilisation) more organised - and therefore promising more booty - political entities, empires or states. 4 Successful leaders attempted to strengthen their armies, for which they needed more income, which in turn meant that they had to wage wars more frequently, in the end almost continuously. This phase is characterised by a lack of fortifications, as territory was of no importance to warrior bands that were constantly on the move. 5 In order to be successful against growing military resistance from the affected areas a number of leaders joined together and formed military coali- tions. 6 After a series of raids the warrior bands began to occupy and at the same time to settle in the territories they had fought in, instead of returning to their lands of origin. Or the most powerful leader attempted to build his own empire in enemy territory. He thereby had to assert himself against his opponents. The threatened countries solved the problem through the inclusion of war-band leaders into their own system of government (foederati or fiefdom contracts). They then tried to use these forces against further threats from without. In the case of a lack of a local government organisation, the leaders formed their own empires on the territories of the state which they had occupied, or where they could settle as mercenaries. WARRIOR BANDS, WAR LORDS, AND THE BIRTH OF TRIBES • 233 All these proceedings refer to Central or temperate Europe — from this part of the world I have provided examples, but it may also be possible to apply this sequence to other districts and cultures in the world — with individual features dependent on the special character of the epoch. Generalisation is only possible in a limited manner, but will be attempted here for model-forming purposes. Comparisons are possible with other parts of the world and other epochs, for instance in the special circumstances of neighbouring traditional and nomadic empires, as in China (Barfield 1989). The movement from warrior band to empire or territorial government always follows a pattern of which the individual stages cannot — or at best in some stages only — be archaeologically recognised. It is only the final goal, the completed occupation of land and new settlements complete with an ethnogenesis (tribalisation), which leaves its traces in the archaeological material. The signs of the constitution of a tribe are (1) the filling of the area of settlement with latecoming immigrants and (2) the connection to the various lands of origin (the Saxons in England and lower Saxony; Franks and northern Germany; Alamanni and the area between Mecklenburg and Bohemia beyond the river Elbe; Normans in Italy and in Normandy). It is an unanswered question, then, what the furnishing of male burials with weapons in Early La Tene, in the Roman Iron Age, in Merovingian times and during the Viking Ages may imply. Are weapons symbols of social status and rank, or indeed signs of actual participation in combat, or even in wars? In all cases only a small proportion of the total numbers of male burials from each period include weapons in their inventories; these weapon assemblages often depend on the age of the dead man. There is a difference between younger and older warriors. For the early period there is also the possibility that at first only members of warrior bands or auxiliary units of the Roman Empire were buried together with their weapons. Pattern of interpretation The function of the warriors in the constitution(s) and development(s) of tribes (tribalisation) and states (empires, territorial governments) is recognisable through the described stages. In conclusion it can be said, that (Fig. 1): Warfare forces clan societies to organise themselves in tribes. Tribes develop, which have to set up military units in order to fend off threats and to defend themselves. A part of the warrior group joins together to form warrior bands that offer services as mercenaries or undertake raids. A cyclical development creates war: The contrast between states at different levels of development and between states and clan-based societies leads to warfare, which in turn leads to the constitution of tribes. As tribes developed, warfare between tribes followed. As warrior bands, which searched for and needed 'work', developed on another level (higher than tribes and above and also independent of the tribes), the raid was invented and mercenary services offered to state-organised societies. These always had reasons for waging war against their neighbours. The question must be asked, why did the social form of organisation called the warrior band and its behaviour — the annual raid — develop? They must have had either a regular or specific function in European history; the regularity with which this behaviour can be observed cannot be overlooked. Possible explanations could be: 1 The warrior bands always existed, as they were necessary due to certain inner structures of their societies. A part of the young warriors could not live in the rural surroundings of the settlements at home because, for example, of laws regulating inheritances which stood against them. They separated themselves from the clan and tribal societies. The wars necessary for their income were mostly internal, as can be seen in the Celtic world (sacrificial areas as in the case of Gournay in France) or in the Germanic world (the great weapon deposit of Illerup in Jutland). They only appear sporadically in historic sources — mainly when the raids were conducted against more organised states. 2 The co-existence of states — with varied levels of organisation — and open ranked societies, clan and tribal societies, forces the cultures along the fringes (border lines) to adopt the new organisational form of the warrior bands which make profits by mercenary services or raids on the of more advanced political entities. The cyclic developments can thus be explained by the fact that there are always new structures of states emerging, which in turn influence their neighbourhood. That is why my repeated model 234 WARFARE AND THE STATE clan-societies 4, 1. tribalization tribes with territories internal wars from tribe to state mercenaries and warrior bands war clan-societies 1. tribalization tribes with territories internal wars from tribe to state war war 2. ethnogenesis or tribalization tribes in occupied territories 1 internal wars from tribe to state mercenaries and warrior bands war war From Tribe to State and WarfareImperial societies, kingdoms and states war war 2. ethnogenesis or tribalization tribes in occupied territories migrations internal wars from tribe to state "names of tribes" ( mercenaries and warrior bands war war 2. ethnogenesis or tribalization tribes in occupied territories internal wars from tribe to state clan-societies 4, 1. tribalization tribes with territories internal wars from tribe to state war STEUER/PLONNER 2001 war F I G . 1: Warfare and transformations of tribes and states. can be used - and not only for the times beginning with the Celts and the La Têne period, but equally for the Hallstatt period and earlier epochs of temperate Europe. - The Mediterranean world with Greece, Rome and Carthage influenced the clan-based societies of the La Têne period - the first recognisable response being the Celtic mercenaries in the area of the Mediterranean. - The secondary structuring development of tribes and areas of rule (oppidum-civilisation) in the Celtic world in turn influenced the early Germanic world of beyond the Rhine. The first recognisable response was the Suebi mercenaries of Ariovistus employed by the Celtic lords of the Sequani. - The Roman Empire in its varied stages of expansion influenced the Germanic world, after integrating and assimilating the Celtic world in the Empire as the Gallic provinces. During the Roman Iron Age far inside the Germanic countries areas of rule developed which defended themselves with ramparts and trenches. How many of these basic units merged into a tribe still remains unknown. 100 tribal districts of the Suebi are mentioned by Caesar (de Bello Gallici 4.1). These sent out 1000 armed men every year on raids, altogether 100,000 men - perhaps a reflex of this situation. (Caesar goes on to report that the rest of the warriors stayed at home in order to do their farming so that their families would have an income, but were sent out the following year when it was their turn). The `Suebenknoten' (knot of the Suebi), the characteristic knot of hair worn on the side of the head of the warriors, is spread across the whole of Germania, and Tacitus points out that it was an honour for a warrior to wear this knot (Germania 38). The first response outside Germany was the bands of mercenaries serving as auxilia-units in the Roman army since the times of Arminius. A later response was the groups of mercenaries WARRIOR BANDS, WAR LORDS, AND THE BIRTH OF TRIBES • 235 made up of the so-called Alamanni and Franks who served during the 3rd century. - The Frankish Empires of the Merovingian and, even more so, the Carolingian dynasty had the same stimulating effect on the Scandinavian peoples. Scandinavians can be traced back to the Merovingian or Vendel period where they had been employed as mercenaries of the Franks (socalled ring-swords may be signs of such alliances: Steuer 1987). - Raiding bands of Vikings are employed as mercenaries against other bands of Vikings, the situation later being resolved through the granting of fiefdoms. - Mercenaries from Normandy conquer the Muslim south of Italy and found a kingdom. The feedback such activities provided to the places of origin only become apparent when written sources report on it. Otherwise, the development from clan- and family-based societies to tribal societies and from that stage - with the key word 'centre of wealth' - to early areas of rule (empires, states) can only be deduced in an indirect manner (compare: Mortensen and Rasmussen 1988; 1991). BIBLIOGRAPHY • Barfield, T.J. 1989. The Perilous Frontier: Nomadic Empires and China. (Studies in Social Discontinuity). Oxford: Blackwell. • Barth, F. 1969. 'Introduction'. In: F. Barth (ed.): Ethnic Groups and Boundaries. The Social Organization of Culture Difference. Oslo: Universitetsforlaget, pp. 9-38. • Bodmer, J.P. 1957. Kriegertum als Form der menschlichen Existenz im Frahmittelalter. Zurich: Fretz und Wasmuth. • Bowlus, Ch.R. 2002. 'Ethnogenesis: The tyranny of a concept'. In: A. Gillett (ed.): On Barbarian Identity. Critical Approaches to Ethnicity in the Early Middle Ages. (Studies in the Early Middle Ages, 4). Turnhout: Brepols, pp. 241-56. • Coupland, S. 1998. 'From poachers to gamekeepers: Scandinavian war lords and Carolingian kings'. Early Medieval Europe, 7(1), pp. 85-114. • Jorgensen, A.N. and Clausen, B.L. (eds.) 1997. Military Aspects of Scandinavian Society in a European Perspective, AD 1-1300. Copenhagen: Nationalmuseet. • Lentz, C. 1998. Die Konstruktion von Ethnizitiit. Eine politische Geschichte Nord-West Ghanas 1870-1990. Cologne. • Marx, Chr. 1998. 'Die "Militdrmonarchie" des Shaka Zulu - koloniale Mythen und neue Forschungen'. Arbeitskreis Militargeschichte e.V. Newsletter 6, pp. 6-9. • Mortensen, P. and Rasmussen, B.M. (eds.) 1988. Jernalderens Stammesamfund. Fra Stamme til Stat i Danmark 1. Aarhus: Aarhus University Press. • Mortensen, P. and Rasmussen, B.M. (eds.) 1991. Hovdingesamfund og Kongemagt. Fra Stamme til Stat i Danmark 2. Aarhus: Aarhus University Press. • Pohl, W. 1998. `Gentilismus'. In: Hoops Reallexikon der Germanischen Altertumskunde. Vol. XI. Berlin and New York: Walter de Gruyter, pp. 91-101. • Pohl, W. 2002. 'Ethnicity, theory, and tradition: A response'. In: A. Gillett (ed.): On Barbarian Identity. Critical Approaches to Ethnicity in the Early Middle Ages. (Studies in the Early Middle Ages, 4). Turnhout: Brepols, pp. 221-39. • Sigrist, Chr. 1994. Regulierte Anarchie: Untersuchungen zum Fehlen und zur Entstehung politischer Herrschaft in segmnentiiren Gesellschaften Afrikas. (3rd edn.). Hamburg. • Steuer, H. 1987. 'Helm und Ringschwert Prunkbewaffnung und Rangabzeichen germanischer Krieger'. Studien zur Sachsenforschung, 6, pp. 189-236. • Steuer, H. 2001. 'Kriegswesen. III. Kulturgeschichtliches und Archaologisches'. In: Hoops Reallexikon der Germanischen Altertumskunde. Vol. XVII. Berlin and New York: Walter de Gruyter, pp. 347-73. • Vail, L. (ed.) 1989. The Invention of Tribalism. London. • Timpe, D. 1970. Arminius-Studien. Heidelberg: Carl Winter. • Wells, P.S. 1999. The Barbarians Speak. How the Conquered Peoples Shaped Roman Europe, Princeton: Princeton University Press. • Wenskus, R. 1977. Stammesbildung und Verfassung. (2nd edn.). Cologne and Vienna: BOhlau. • Werner, J. 1973. 'Bemerkungen zur mitteldeutschen Skelettgrabergruppe Hassleben-Leuna. Zur Herkunft der ingentia auxilia Germanorum des gallischen Sonderreichs in den Jahren 259-274 n. Chr.'. In: H. Beumann (ed.): Festschrift flit. Walter Schlesinger. (Mitteldeutsche Forschungen, 74/1). Cologne and Vienna: BOhlau Verlag, pp. 1-30. 236 • WARFARE AND THE STATE WARFARE AND SOCIETY Archaeological and Social Anthropological Perspectives EDITED BY TON OTTO HENRIK THRANE AND HELLE VANDKILDE AARHUS UNIVERSITY PRESS 2006 WARFARE AND SOCIETY Archaeological and Social Anthropological Perspectives © Aarhus University Press, Ton Otto, Henrik Thrane & Helle Vandkilde 2006 ISBN 87 7934 110 1 Editors: Ton Otto, Henrik Thrane & Helle Vandkilde English revision: Stacey Cozart, Nick Thorpe, Mary Waters Lund Proofreading: Steffen Dalsgaard Layout & cover: Hanne Kolding Printed by: Narayana Press Type face: Stone Serif, Stone Sans Paper: Arctic Volume Aarhus University Press Langelandsgade 177 DK-8200 Aarhus N Denmark www.unipress.dk White Cross Mills Hightown Lancaster LA1 4XS UK Box 511 Oakville, CT 06779 USA Cover: "War Magic", 1975, screenprint by the Papua New Guinean artist Timothy Akis (deceased 1984). The picture illustrates the connection between warfare and social identities. In some Melanesian societies war magic is used to transform men into warriors, so that they can kill people and thereby establish group identities and social boundaries. Published with the financial support of the Danish Research Council for the Humanities