[ELH] Anglo-Saxon England. Settlement - rural and town life

Summary of lecture:

In our own day we have a very clear idea of the difference between town and country; indeed we are heirs of a whole bundle of intellectual concepts and prejudiced snobbery about one or the other.. Townies and bumpkins both go back a long way - to Roman times, in fact. Then too, town and country were easily distinguished, but not so in Anglo-Saxon England. With the collapse of the Roman empire - or rather, since yesterday I was proposing that the empire itself never died, with the collapse of the imperial economy, towns and cities as the Romans knew them disappeared from Britain. The empire had required a huge standing army and a huge bureaucracy to service it, and these were paid for by taxation; taxation paid in coin, not kind, just as the army was paid in coin, not kind. A monetary economy encouraged the growth of markets, indeed required them, and markets led to towns. Within the imperial economy, Britain appears to have been what we would call today a net exporter of produce. Coin flowed in to Britain, and the economy encouraged what academics call conspicuous consumption: the building of lavish villas, for example. When the imperial army was withdrawn from Britain, taxation and the export of supplies for the army ceased, and the flow of imperial coin dried up. Towns and cities as the Romano-British knew them ceased to exist.

Towns in the Dark Ages: an invisible phenomemon?

There then followed a long period of several centuries before the emergence of the medieval town as we see it in our mental picture. What was happening in the interim? Did towns and cities really vanish entirely from the landscape, until a time when something happened called, in the words of an important collection of conference papers, 'the rebirth of towns in the West'? What was the role of Alfred's foundation of 'burhs' in that process? - for burhs are generally thought to have originated as strong points of defence against Danish marauders. Did burhs lead to towns, or were there towns already in existence? Did Aethelflaed, Lady of the Mercians, found burhs in the Midlands on green-field sites, or did she rather strengthen the fortification of towns founded earlier, perhaps under Offa? Did Pope Gregory the Great know what he was talking about when he instructed Augustine to found bishoprics in the twenty-four cities of Britain, or was he blindly copying some long-out-of-date Roman official list, such as we know from Nennius, and therefore writing in ignorance of the true state of affairs? We used to think that archaeology had confirmed the death of the Roman city by the discovery of layers of what is called 'black earth', taken to be evidence that the sites of cities were turned over to agriculture. Now we are not so sure. To what extent are such layers the residue of generation after generation of wooden buildings, decayed and replaced? Archaeologists were puzzled by the lack within the Roman city of London of evidence for the trading settlement that was documented by at least the eighth century - until they discovered it outside the Roman walls, in the area of what is called Aldwych, literally, 'the old wic', or market place.

The problem is compounded by the use by ecclesiastics of words for 'city' when describing monasteries: Cogitosus on Kildare in the life of St Brigid, for example. Nennius described Worcester as the 'metropolis of the Hwicce'. Major monasteries of the seventh and eighth centuries may indeed have been the largest concentrations of people in the Anglo-Saxon landscape. Yet monastic 'cities' - given that the writers were intending to use words like 'urbs' and 'polis' literally, rather than in the sense of 'the city of God' - do not seem to qualify as urban either in our terms or in Roman terms. To paraphrase Richard Hodges, the rebirth of towns in the West, when it comes, seems like a revolution: the emergence of 'a fully-commercialised society' involving all levels of the population, with a network of 'interlocking central places'.

What is a town?

Does this mean then that urban functions were absent in Anglo-Saxon England? Is that the question we should be asking? Probably, yes it is. Arguing about the presence of absence of towns in Anglo-Saxon England misses a crucial point. It fails to address the question, 'What is a town?' A town is defined by the activities of the people who live in it. A town is a coming-together of functions. I spent the weekend at Spilsby in Lincolnshire. Spilsby has one short main street and less than 2,000 inhabitants, a big village, you might say. But Spilsby calls itself a town. So what is found there? A town-sized church, for one thing, but then churches follow population. More instructive are the doctor's surgery, the bank, the agricultural supplies depot, the post office, the newsagents, the White Hart inn with a courtyard entrance wide enough for horse-drawn coaches, the Spilsby Theatre housed in what was a law courts building, disused local government offices, the statue of Spilsby's famous son, the explorer John Franklin, discoverer of the North West Passage, a sure sign of municipal pride, and, most important of all, a market street so wide that its centre came to be occupied in part by a market hall, now extended and operating as a garage and filling station. Thus a town is not defined by its size but by its functions: administrative, judicial, social, but above all, commercial. It is a service centre and a market centre.

'Dispersed proto-urban functions'

What seems to have happened in the earlier centuries of Anglo-Saxon England is that these functions were carried out, but rather than being brought together as in the towns of Roman Britain and of the later Middle Ages, in many cases they were dispersed. Administration was carried out at one place, justice at another, trade at somewhere else again. Mick Aston has suggested that so-called 'dispersed proto-urban functions' were dispersed among places within a single territorial network, and has given as an example three places in north-east Wiltshire: Bedwyn, the site of a royal hall, Marlborough, a market town, and Ramsbury, the location of a minster church. This is a useful model. John Blair has taken further the location of royally-founded minster churches in relation to royal halls, often some distance apart, and James Campbell's discussion of royal tuns and dependent settlements leads in the same direction. Thus we can see in Leicestershire, for example, the possibility that Loughborough, Barrow and Rothley represent three functions within a single landed unit: the royal hall at Barrow (simplex place name), the minster at Rothley (high status church, AS cross, cult of St Wistan), and the king's tun or collecting point for tribute at Loughborough (enclosure pn, earl's vill with large soc, markets and church). A similar trio of settlements may explain the apparent relationships between Higham-on-the-Hill, Hinckley, and Market Bosworth.

The king's tun

Let us concentrate for a moment on the king's tun, since it was such an important focal point in the Anglo-Saxon landscape. Its importance rests in part, an emphatic part, on the nature of Anglo-Saxon kingship. You are already aware than Anglo-Saxon kings, whether over-kings or under-kings, were peripatetic, that is, they travelled around their kingdoms, dispensing justice and gifts, resolving disputes, and entertaining their nobles and visiting dignatires. Where they went, their court went also, and these mouths were fed locally by the produce of the king's estates and collected tribute, that is, rent in kind and other dues in kind required from those with lands in the district. We find lists of such tribute in Anglo-Saxon charters and description of surviving customary dues in entries in Domesday Book. In Domesday Book also appear descriptions of manors owing what the editors of the Phillimore edition of the survey describe as 'one night's revenue', the feorm, or 'farm' of 'one night'. 'Farm' is here in its oldest sense, that is, meaning land leased out in order to receive in return payments of rent - specifically the amount of rent sufficient to lodge and board the king and his entourage for one night. The feorm is always mentioned in connection with royal manors, and in the same way, royal tribute and dues were collected at settlements known as king's tuns, not necessarily, or indeed often, identical with the location of the royal hall where the king stayed when in the district. What king would choose to stay within sound and smell of a market?, for as time went there was increasing taxation, and taxes had to be paid in coin by sale of surplus produce, while at the same time, dues were increasingly commuted to payment in coin rather than in kind. The word 'tun' may indeed give away the fact that such places were secondary or subsidiary, for James Campbell some time ago and Ros Faith more recently, have both put forward cogent reasons why we should see tuns as locations of demesne, parcels of demesne, within a single landed unit, demesne being land directly in lords' hands rather than land rented out to members of the lords; entourages, or to colonising tenants.

Places of assembly

One characteristic that the king's tun shared with some other groups of places is that it was a point of public assembly. People came together not only to give the king and his entourage board and lodging; they 'attended upon the king' in order to have disputes settled, cases appealed, lands granted, gifts given, appointments made, laws promulgated, policy debated, and ambassadors heard and replied to. People also assembled for other reasons. To hold fairs was one reason. Fairs had been held probably since prehistoric times, and often on tribal boundaries, for communities needed to exchange their surpluses and specialisms. Some fairs were concerned with a specific commodity: horse fairs appear to have been particularly ancient. A number of places are called long ports, 'port' being Old English for market (thus showing the original nature and purpose of seaports). One such place is Lamport in Northamptonshire. I wonder whether such places were the locations of fairs strung out along a road or street and how many were on ancient boundaries. Fairs were one reason for coming together, another was the holding of annual sports or games, and sometimes the two went together. Brigg Fair in Lincolnshire was held at a bridge, as its name implies, but the place was also known as Glanford, from the Old English gleam, which as well as being the root of our word 'glamour', also meant 'revels', 'festivities, 'games'. Close by Glanford Brigg is Hibaldstow, and stow in Old English means a place of assembly - in this case to visit the resting place and shrine of St Higbald, a bishop of Lindsey. Stow frequently has a religious connotation, but sometimes it is difficult to decide on a religious or secular explanation for the gathering. Bristol means 'bridge stow' and could equally refer to assembly for buying and selling, and assembly to visit a holy place. Again, both could co-exist. Indeed, within a mile or so of each other in Gloucestershire there were in the late medieval and early modern period, annual games at Coaley, a fair at Nympsfield, and a shrine at Nympsfield also. Though the provenance of all three is dangerously late for our period, Nympsfield means in Old English 'the cleared land of the sacred place' - nemet, a British borrowing - and this can only refer to the Romano-British healing shrine of Mercury at adjacent Uley, whose temple was replaced by a Christian church, probably in the fifth century. A final category of assembly point is the place where representatives gather to make laws or decide difficult cases. The Hundred meeting place is one such, the shire moot another, but larger areas also had assemblies. There are places near Peterborough and Moreton-in-Marsh called Guildenbeorg - gild being an Old English word for 'association'. The latter was the meeting place of the 'Four Sheriffdoms' - that is, the counties of Worcester, Warwick, Gloucester and Oxford.

Hierarchies of settlement within large landed and economic units

As well as dispersed urban functions, then, there appears to have operated in Anglo-Saxon England a hierarchy of settlements, and more than that there appears to have been a system of specialism at individual settlements. This is one of the benefits of studying place-names. We may be sure that a place named Sutterton, 'shoe-makers' tun' (in the area of the Danelaw such places are Sutterby) was so-named because local circumstances allowed the growth of a craft recognised by the people of surrounding places. Similarly with Sapperton, the 'soap-makers' tun. While Boultham, the 'meadow with burdock plants', may well have developed a specialism in the production of burrs for wool-carding, since meadows with burdock merely growing in them must have been relatively numerous, if not two-a-penny, depending on local topography and climate. From places named for their services or location within a single district, a category of which the most obvious perhaps are the Eastons and Westons, it is possible to move outwards to glimpse component settlements within larger economic units. Take, for example, places whose names betray some role within a system of seasonal pasture, either the intercommoning of some large extent of summer meadow or rough peripheral grazing land, or transhumance, the seasonal movement of livestock over long-distances to summer pastures or home to winter quarters. Winderton in Warwickshire is the winter tun, not a winding place. The various Somertons are self-explanatory. Hardwicks are dairy farms, perhaps, and Swinhopes the valleys where pigs were pastured. Maidenwell in Lincolnshire may be a well characterised by may-flies; more attractive is the etymology, or name-explanation, that interprets it as a source of water used by young women driving the livestock to the high wold summer pastures belonging to the people of Louth. A nineteenth-century description of transhumance in Ireland evokes the living conditions of maidens herding in the summer shielings that may not be far removed from those of young women tending herds in Anglo-Saxon England. For example, they were said to spend their days spinning, and watching the cattle, and at night they slept on beds of black sedge, from which they rose 'as fresh as trout'.

The 'multiple estate' debate

Dispersed urban functions, tribute collecting-points, specialist settlements, intercommoning, and movements of livestock to distant pastures all imply the ability and authority to organise and allocate units of land within larger entities. Few themes relating to Anglo-Saxon England have led to fiercer debate than the concept of what has variously been called 'the multiple estate' and 'the federative manor'. Some scholars see early- and middle-Anglo-Saxon England as composed of landed units smaller than the modern shire but thought to underlie districts confusingly known in the north of England and Scotland as shires: Howdenshire, and Richmondshire, for example. Elsewhere they have been thought to underlie the Rapes of Kent and Sussex; the Wapentakes and Shipsokes of midland and eastern England; districts composed of multiple Hundreds and appurtenant to some great church or a single lord; and so on. Lindsey, which is one of the so-called Parts, or large sub-divisions of Lincolnshire, was itself divided into Ridings (that is, thirds). Gloucestershire had its Ferdings (farthings or quarters), as well as the large Hundred of Berkeley which was also known as a 'Hernesse', literally a 'listening' or 'harking' in the sense of obedience to a lord's jurisdiction. Certainly such landed units, where they can be shown to exist, were larger than the parish. They prefigured the later Hundreds (unless these already existed) and paralleled the so-called minster parochiae. These were areas under the pastoral care of major minster churches, some of which survived until the Reformation, like that of Melton Mowbray. The most well-known proponent of such landed units is Glanville Jones, whose concept of 'multiple estates' came under attack substantially because it relied on late medieval, model law codes from Wales. These described the division of large standard areas of land, composed of fifty farms or homesteads, between the king, his reeve, and his tenants. But the concept has also had its supporters, partly because Wales was indeed divided administratively, into the early modern period, into Cantrefs, a Cantref meaning literally 'a hundred home- or farmsteads', each Cantref being composed of two Commotes, literally areas which 'come together'. The similarity with English Hundreds, the basic area of medieval local government and policing below the level of the shire, and indeed with similar groups of settlements on the Continent, is too striking to be ignored. Many English Hundreds in the west Midlands, it is worth noting, were composed at Domesday not of 100 hides, but of 50, and it is possible to see pairs of such Hundreds which considered together seem to echo the Welsh Commotes and Cantrefs. At the time of Domesday, much of England was covered by a network of sokes, that is, groups of vills owing customary dues to, or for certain purposes under the jurisdiction of, a particular manor and its lord. As with the Hundreds, whose number was fluid and continued to vary right up to the last century, some Domesday sokes look like relatively recent creations in the late eleventh century while others look distinctly more archaic. One problem is that land continually changed hands, and possessions of a single lord might wax and wane. Thus fission may have been as common a process of fusion. Entirely different landed units might underlie those thought to be recognisable in the mid-Saxon centuries.

One strategy to counteract this problem is to examine the extent to which Hundred and soke boundaries approximate to natural boundaries, especially watersheds, since these can often be demonstrated to define social, cultural, linguistic and agrarian districts with distinctive communities. Another is to examine the place-names of a district and to categorise them both in terms of the functionality they reveal, and also in terms of their age. A putative caput or head vill of a multiple or federative estate might appear as a Domesday manor in the hands of the king and with an archaic place-name. Such names include those ending in -ham, an early Old English name for a high-status home- or farm-stead, those ending in -ingas, like Hastings, place of the people of a specified lord, and simplex names so-called because composed of a single element, often a topographical term, for example 'Barrow'.

The 'rebirth of towns in the west'

What, then, caused the coalescence of urban functions and the emergence of the medieval town? And when did this happen? Was it a feature of Anglo-Saxon England? Yesterday's lecture focused on trade, and discussed the emergence in the seventh century of coastal trading posts, emporia in Latin, wics in Old English- Hamwih near Southampton; Ipswich; Norwich; what became Jorvik at York; and Aldwich outside the walls of London to take four examples. All these places became centres of administration and locations of major churches. Though the growth of towns took off only in the tenth century, driven by what scholars have seen as a tenth century revolution in trade and technology, we see here how one urban function attracts others. In the same way, saints' shrines attract markets, or perhaps markets attract saints' shrines, as they may have attracted places for Christian baptism. The tenth century revolution was accompanied by population growth and what has been called 'a fully commercialised economy', and encouraged the growth of a countrywide network of district markets with concomitant growth of market towns, superseding the small number of early fairs described earlier. Scholars of urban morphology, the study of town plans, discern planned blocks of properties in town which they assign to the tenth century onwards. These speak of royal or comital involvement - they also demonstrate a practice of regularity in laying out settlements.

Rural settlements and nucleation

Such regularity is also visible in many villages and there too may be evidence of planning in Anglo-Saxon England. Settlement patterns as well as village plans in England fall into two great categories: scattered farms and homesteads in upland and woodland Britain, nucleated villages across a swathe of central England. The chronology of nucleated villages is much debated and not yet clear. Yet there is strong evidence to support the view that nucleation occurred in the tenth century or perhaps the ninth, and was a development parallel to the growth of towns.

Scholars have seen nucleation as the coming-together of scattered or individual farmsteadings as pressure built up on pasture land. Was this because of population growth? Or because of the successful foreign trading of English cloaks and other woollen goods and an explosion in the number of sheep needing pasture? Another explanation - and quite likely the two should be considered together - is that nucleation was a forced consequence of the introduction of open-field agriculture. Large, large strip fields from the Carolingian period have been identified in Saxony which are similar to strip fields found in England. It is quite likely that existing farms and hamlets had to be evacuated, in similar fashion to the abandonment of villages in order to allow the creation of medieval parks. Both processes betoken central authority. Thus is no surprise to find that nucleated villages have either a regular form, planned in one or more rows fronting streets, or are polyfocal, like frogs-spawn, possible agglomerations of farms which had moved in together from their previous locations elsewhere in the district or landed unit. That nucleation was accomplished by lordly power seems to be supported by the numbers of villages where the church is sited so as to appear as an appendage of the manor house - which indeed is how thousands of churches originated.

What of the remaining villages where the church site does not fit so neatly into the overall plan? It is possible that in some cases of nucleation at least, the choice was made to found the new central village at the site of a pre-existing church or cemetery or some other sacred site. Settlement sites had in any case been moving from one place to another within a neighbourhood from long before the arrival of the English. One challenge facing archaeologists is to relate Romano-British villa sites to nearby contemporary and perhaps associated hamlets. Another is to relate Romano-British habitations to evidence for settlement in the so-called Dark Ages and later: what is known as the Mid-Saxon Shift.


Let us examine a case study. I have chosen Wiltshire, because, like Leicestershire, it is not easy to explain its existence by reference to its topography. Wiltshire's heart is the uplands of the Wiltshire Downs. Its northern parts lie mainly in the upper Thames basin, look eastwards to Oxford and ultimately to London, and were frequently contested between Wessex and Mercia. Its southern parts, on the other hand, are drained by south-flowing rivers and streams, and look towards the Channel. Barbara Yorke has shown that the centre of the power of Wessex shifted from the Thames valley to the region bordering the channel and that the two areas may originally have been separate. Nick Higham and Ken Dark have seen them as deriving from the cantons of different tribes of Roman Britain.

Wiltshire's Domesday boroughs were, in the south, Wilton, the tun on the Wylye (the county town, hence 'Wiltshire') and Salisbury, the burh at Sarum, and in the north Marlborough ('Maerla's burial place'), Cricklade (the lade or river crossing, here across the Thames, associated with the craig, Old Welsh for prominent hill), Bath, which fell ecclesiastically in the Somerset diocese of Bath and Wells, and Malmesbury (M's burh). Note the three burh names, and that Cricklade and Bath were burhs also. Fording or bridge places like Cricklade were normally in 'safe pairs of hands', such as the sheriff's, if not controlled directly by the king. Malmesbury may also have been a royal tun, as well as being the location of a major early monastery, since the Domesday survey stated that it was responsible for collecting the revenues of two Hundreds. Wilton and Sarum were contiguous and with Salisbury form a probable trio of places with dispersed urban functions: Sarum, as a hillfort, may have been the site of the royal hall, Salisbury was the location of a minster church, and Wilton represented the third centre, the collection point for the king's tribute.

The second class of important Domesday settlement was those six royal manors which each paid the feorm of one night. These were, in the south Amesbury (Ambre's burh), Warminster (the minster church on the river Were), and Tilshead (meaning 'Theodwulf's hide of land'); and in the north Calne and Bedwyn (both British stream names) and Chippenham (Cippa's meadow).

Only one other Wiltshire manor was in the hands of Edward the Confessor in 1066. This was Britford ('the Britons' ford'), immediately south of Salisbury and perhaps once part and parcel of Sarum. However, Edward's queen held two major manors and the smaller manor of Winterbourne Stoke. The two major manors were Wootton Rivers ('the woodland tun', possibly a settlement concerned with woodland management), and Westbury (the west burh).

A third group of what by 1086 had become royal manors had belonged to Earl Harold, that is, king Harold, and his mother Gytha. Harold's vills were Melksham ('the milk ham'), Bromham ('the broom ham or hamm'), Netheravon ('avon' is a British name for 'river'), Collingbourne Ducis ('Cola's stream'), and Compton Chamberlayne ('the valley farm'). His mother's vills were Aldbourne ('Ealda's stream'), Rushall ('Rust's nook of land'), and Coombe Bissett ('the valley').

The other manors in the king's hands had been forfeited or inherited. Thus Earl Tostig had held the major manor of Corsham ('Cossa's ham'); one Aeleva, perhaps a nun, had held another prestigious manor, East or Great Knoyle ('the knuckle-shaped hill'); one Godric had held Lydiard Millicent (Lydiard is a Welsh name, its first element obscure and the first denoting garth, a hill); and an unspecified person had previously held Ogbourne ('Occa's stream').

In addition, there were a number of royal churches elsewhere in Wiltshire in 1086, whose existence tends to indicate that the places these churches served had been themselves, at one time, in the king's hands. They were at Highworth (until the thirteenth century simply known as 'the enclosure'), Burbage ('the burh by a stream'), Pewsey ('Pefe's eg or island'), Avebury ('burh by the river Avon'), Heytesbury ('Heahthryths burh'), Hazelbury (hazel burh or beorg 'hill'), Sherston ('stone on a steep slope'), and Upavon (river name), as well as one of Wiltshire's boroughs, Marlborough ('Maerla's burial place' [beorg, but not of a hill]), and Bedwyn. Two points of interest arise from this list of royal churches. First the number of places indicating an enclosure (burh, worth) or other prominent and discrete landscape feature (hill, stone, burial place). Could these indicate early monastic sites, and/or instances of Christianisation of pre-Christian sacred sites? Second, the cluster of royal churches around Ramsbury and Marlborough: is this evidence of dispersal among constituent chapelries of the endowment of a single early church of high status?

[Leicester University] [*]
Last updated: 03 February 1999 11:39
Dr G.R. Jones

The views expressed in this document are those of the document owner.
If you are an authorised user you may edit this document through your Web browser.