Recovering the lost religious place-names of England: Gavin Smith


Why, apparently, are religious beliefs so little reflected in the earliest English place-names? Gavin Smith explores various relationships within this thought provoking essay, which "At The Edge" is sharing with the Druidic Dawn and greater internet Community.

Why, apparently, are religious beliefs so little reflected in the earliest English place-names? Openly religious Old English (OE) place-names - St Albans, Whitchurch, Axminster - are rare. Yet Wales at this period is full of llan names commemorating the activities of peripatetic ecclesiastics. Was England by comparison really so secular? Or is it that the religious content of early place-names has simply been disguised and forgotten?
Seventh-century place-names

It would be difficult to overstate the role played by religion in the England of the Dark Ages. At least two kings, Sigeberht of East Anglia (circa 630-7) and Caedwalla of Wessex (680s), abdicated to become penitents. And in this same period, the seventh century - because of the emergence of a common language, English, but also, I would argue, because of the increasing administrative centralism of the Church - the basis of English place-naming was laid down. Thus the charters and boundary lists of two religious establishments in Surrey, Chertsey Abbey and the minster at Farnham, believed to be genuine seventh century documents, contain at this early date a whole set of typical OE place-name elements including -ingas, -ham, - hyth, - brycg, burh, -eg, -ford, -leah, - feld (today’s -ing, - ham,- hythe, -bridge, bury, -ey, ford, -ley, -field) [1].

Most of our knowledge of the earliest OE place-names comes from ecclesiastical charters, so we might expect to find some religious content. Is it there? And if we find it, what is its significance? Two outcomes might be hoped for. First, do we find that a good proportion of the names of religious or cultural centres might be reinterpreted to reveal a religious meaning? The answer, I suggest, is yes. Glastonbury, Canterbury, Chertsey, Wells, Lincoln, Lichfield, Sherborne, Bristol, Windsor, all will be claimed as examples. Second, is there a chance that place-names might reveal lost archaeological sites? This is something of a holy grail for place-name students, who (rightly) have received a hard time for their troubles from sceptical archaeologists. Care is advised. A place-name fixes neither a geographical site nor a date of first occupation. But there is the prospect, to take one example, that place-names may provide the ‘other trace’ that identifies Stenton’s ‘ancient parish churches, actually early monasteries which have disappeared without other trace’ [2].

There is also the prospect that pagan religious activity underlying Christian sites may be identified, so helping us to understand the mechanics behind the Pope’s recommendation to Mellitus, first bishop at St Paul’s in London in 604, that existing pagan sites be co-opted [3]. The Pope anyway was following standard Roman Empire practice (under the Empire the god worshipped at Bath was ‘Sulis Minerva’).

So, let us look at religious activity in the seventh century, and see whether any OE place-name elements might be linked to parts of it. Monasteries and minsters are a good place to start, because they are relatively well documented. Morris, an ecclesiastical historian, lists the seventh century names of all churches documented from that period [4]. Cox, a linguist, gives us the relative frequency of occurrence of OE place-name elements in all place-names documented up to the 730s [5]. Other linguists (e.g. Gelling [6]) have looked at the possible meaning of OE elements. A geographer (like myself) would look at spatial patterns. All these approaches will be used below; they are brought into focus in a confined study area, Surrey.


burh - 'fort' or ‘monastic enclosure’.

Most promising is OE element burh, found in the names of monasteries including Glastonbury, Bury St Edmunds, Peterborough, Canterbury, Malmesbury, and significantly in Paulus byrig aet Lundaenae for St Paul’s in London. Three of these names clearly mean ‘the burh dedicated to saint Edmund/Peter/Paul’.

It does look likely that to the accepted meanings of burh as ‘Iron Age hill-fort, Roman town, fort, manor, borough’ we must add ‘monastic enclosure’. Canterbury could thus either be ‘the cathedral’ or ‘the fort’ of the people of Kent. Such would be consistent with Caradoc’s use of urbs (town) for Glastonbury, and Bede’s statement that the great Welsh monastery of Bangor Iscoed (some of these religious establishments held thousands of people) ‘is called by the English Bancornaburg’. Ekwall interprets the first part of Malmesbury Abbey’s name to be an amalgamation of the names of two of its famous abbots, Maildulf its Irish founder, and the later English Aldhelm. Equally interesting is Barker’s thesis of a series of lost Celtic monasteries (llan) in southern England, several of whose putative sites have burh names (Ramsbury, Amesbury, Westbury, Charlbury) [7].

In Surrey, Newark Priory was called Aldebury (1204) and then Novo loco de Andebir (1210), i.e. ‘new place (Newark means ‘new building’) for the old burh’, suggesting to me the priory may be successor to the lost seventh century minster of Woking nearby. Blair reached the same conclusion, although for different reasons [8].

eg - 'island'

Increasingly, a Celtic earlier phase of several English monasteries is not in dispute. The rights of existing Celtic establishments at Glastonbury and Abingdon were restated by their Anglo-Saxon conquerors [9]. And here the OE element eg (‘island’) seems relevant. It occurs, like burh, in many monastic names including Chertsey, Ramsey, Romsey, Selsey, Bardney, Athelney, Thorney (Cambs, but also the site of Westminster Abbey). Eg is found also in archaic forms of some monastic names: Glastingei (Glastonbury), Laestingaeu (Lastingham), Heruteu (Hartlepool). Parallels to English eg would seem to be Welsh ynys (‘island’), as in Ynys Pyr (Caldey Island, ‘the ynys of St Piro') [10] and Scandinavian holm (‘small, island’), as in the monastic names Durham (dun holm), Holme Cultram (Cumb) and Hulme St Benet’s (Norfolk). Could eg, ynys and holm, under the influence of the early Irish missionaries, each have undergone during the Dark Ages a temporary ‘semantic shift’ and come to mean ‘(monastic) island (retreat)’? Cox notes that eg seems to refer to major estates, but that this sense dies out by the eighth century. The name sequence Glastingei (704), Glestingaburg (732-55) is perhaps telling.

That Celtic monks sought remote islands, headlands and marshland retreats is history. Mayr Harting, quoting Bowen, has described the remote locations of early Welsh monasteries as ‘ynys’ [11]. That the monks were the conscious inheritors of a pagan tradition is less well advertised. Ellis Davidson says ‘some names (those ending in -ey for instance) indicate cult centres on islands, like that of the old Nerthus in Tacitus’ account’ [12]. In Ireland, St Colm when he settled an island in Lough Derg is said to have found there Maccriche, a pagan man [13]. While from Gwent, to quote a church leaflet guide (with my emphases) ‘Llantilio Crossenny means the Church of St Teilo at Iddon’s Cross. Iddon was the local ruler in the sixth century who had been leading the struggle against Saxon invaders. Hearing that Teilo was at LLanarth nearby, he asked the holy man to aid him with prayers for victory. Teilo raised his cross here, on an ancient pre-Christian mound and after Iddon had defeated his enemies, this land was granted to St Teilo for the building of a church.’

Looking at eg then, churches where we might seek Stenton’s lost monasteries perhaps also include Rye (Sussex), Olney (Bucks), Gedney (Cambs), Witney (Oxon), Eyam (Derbs), Kersey and Eye (Suffolk). In Surrey we find Chertsey and Bermondsey (recorded seventh century monasteries), Molesey and Battersea (recorded in seventh century charters as monastic possessions), Titsey (site of a Romano-British temple), and in the Wealden forest the insignificant Puckney (‘Puck’s island’). (In the more minor Surrey names, however, eg seem to preserve the original meaning simply of ‘island’, as in the dialect Thames ‘eyot’).
After the Council of Hertford in 672, new Roman ecclesiastical organisation took over from Irish monastic influence. Minsters and their defined territories replaced the more independent Celtic monasteries [14]; is the replacement of eg by burh a reflection of this? Celtic-style monasteries either dissolved, or mutated into the new minsters, sometimes but not always retaining their old name. A Latin term, mynster, came in. Thus Westminster perhaps changed its name from Thorney, as befitted a newly important state church. While in the far south-west, areas relatively lately taken into Anglo-Saxon lordships, the English names of local centres (Axminster, Exminster, Charminster, Ilminster) seem likely to have been mynster from the start, their prior names being Celtic.

ingas - ‘people’

There must be a suspicion that -ingas fits into this sequence. Ingas (‘people’) is much beloved by place-name students. It is interpreted as recording the folk groupings of Germanic (‘Anglo-Saxon’) invaders. A different (or perhaps supplementary) interpretation is possible. Could -ingas place-names recall the ‘followers’ of early missionaries, or the ‘religious communities’ at monastic sites? These senses are admitted by Dodgson [15] in the case of Guthlacingas, the followers of Guthlac founder of Crowland Abbey in Lincolnshire at the beginning of the eighth century, and by Ekwall with Berclingas, ‘the monks of Berkeley’ in Gloucestershire. There are equivalents in Maildubiensis aecclesia, ‘the church of the people of Maildulf’ for Malmesbury, and Cuthbertfolk, ‘the people of St Cuthbert’ for the soke of the Bishops of Durham [16], (Cuthbert’s shrine being in the cathedral).

The Venerable Bede, writing in the 730s, is a good source of -ingas place-names. These need to be distinguished from his use of -ingas for royal lineages like Oiscingas ‘the kings after Oisc’ (i.e. of Kent). Bede’s place-names, perhaps not surprisingly, occur in a religious context, like the monastery of Barking (Essex), but also as ‘districts’, as in Inngetlingum (perhaps Gilling near Richmond, Yorks). This set of lineage, religious and geographical meanings may be resolvable, in that monasteries often were the possessions of local aristocratic households [17]. That is, the household of a local thegn converted en bloc and in situ to the new Christianity. So, aristocratic households absorbed simultaneously the status of lineage (‘-ingas’), monastery and tribal territory (and were the instrument for the spread both of Christianity and of ‘Englishness’). The territory in essence is likely to have been the catchment area of the local pre-Germanic sacred centre (the original Hundred?). This explanation is not inconsistent with, but broadens into a wider historical context, the ‘-ingas’ ‘mini-states’ postulated by Bassett and Blair [18].

Check this against the eight known -ingas sites of Surrey. Significantly, all occur in early charters, and in each of seven Hundreds. They are Woking (seventh century monastery; Woking Hundred), Bintungom (Farnham charter and Hundred; Binton Farm is by Seale, ‘hall’; close to later Waverley Abbey), Eashing (short-lived royal Saxon fort whose remote location is a puzzle but lies next to Peper Harow, possibly ‘the pipers’ temple’, Godalming Hundred), Godalming (minster, perhaps supercedes Eashing; Godalming Hundred), Dorking (Roman station on Stane Street, and likely minster site; Wotton Hundred), Tooting (station on Stane Street; close to later Merton Abbey; Brixton Hundred), Tyting (beside the hilltop church of St Martha’s; Blackheath Hundred) and Getinges (Eaton Farm at Cobham, the A3 crosses the Mole here so conceivably the lost Elmbridge, ‘Mole bridge’, which names this Hundred). Poulton notes that in Surrey, minster territories and Hundreds appear to be co-terminous [19].

Were the Surrey-ingasplaces all seventh century monasteries? If so, might they be connected with Birinus, the fresh emissary from the Pope who became first bishop of Wessex (circa 635), of whom Bede recalls that he ‘built and dedicated churches and brought many to the Lord’, and who is said to have converted Surrey? His churches remain unidentified. One of his sponsors was king Oswald of Northumbria, which had overrun all England except Kent. Under these circumstances, and at this date (before the Council of Hertford), churches founded by the state probably would be hybrids between Roman minster territories and Celtic monastic organisation (complete with pagan undertones), and quite likely could have had a Northumbrian Anglian type of name. Thus it is interesting that -ingas place-names are found in Northumbria (Bede’s own land), East Anglia, Middle Anglia, Essex and eastern Wessex - sub-kingdoms all newly fixed up with Celtic-trained (bar Birinus) bishops - but are absent from east Kent. (Sussex remained pagan late, but has a rash of -ingas places that might be associated with the activities of the Northumbrian Wilfrid, first bishop at Selsey, in 681).

If, after the Council of Hertford, Celtic-style monasteries were amalgamated into administrative minster areas, the focus is now on proto-urban, rather than archaic holy sites. In the Farnham charter (signed at some place called Besingahearh, ‘pagan temple’) we may be witnessing this process, since the dependent ‘estates’ (perhaps actually dependent sacred centres) apportioned to the new minster are Cusan weoh (‘Cusa’s temple’), Bintungom (see above) and Churt (‘heath’, but which conceivably is ‘(shrine in the) heath’, as Chard in Somerset is ‘house in the heath’). The charter’s sponsor was Caedwalla of Wessex, a lively character who appears in both Bede and the Anglo-Saxon Chronicle. A ‘pagan’ and ‘usurper’ with a Celtic name, in the 680s he emerged from the forest of the Weald, devastated Wessex, the Isle of Wight, Sussex and part of Kent, before being converted by St Wilfrid and retiring to Rome. Was his base the wild Hants/Surrey border around Farnham (Liss in Hants is Celtic llys, ‘court / hall’)? By contrast, the Chertsey documents from the settled Thames valley focus on an established monastery, with dependent -ham (Germanic ‘home’) estates which today are villages and parish churches and are perhaps the successors of Romano-British agricultural estates [20]. The Chertsey charter cites Sonning (Berks) as the neighbouring ‘province’ [21], perhaps, that is, the next minster territory. Blything Hundred in Suffolk is perhaps a minster territory centred on Sigeberht’s monastery at Bythburgh (burh); Happing Hundred in Norfolk may similarly relate to Happisburgh.

Modern names in ‘-ing’ or ‘-ing-’ are not all -ingas, but if the hypothesis presented here has merit then some or all place-names containing the dative form -inga- ought also have religious meaning. This is conceivable for an -inga-ham name like Lastingham. It is more than probable for names in -inga-hearg, -inga-eg, -inga-burh, -inga-hoh, like Gumeningaherg (Harrow on the Hill, Middlesex), Glastingei, Glestingaburg and Ivinghoe (see hoh, below). And could -inga-ham names on the Continent originate with the missionaries who went to Europe from East Anglia? [22].

ge - 'territory’

The picture for east Kent, missing under -ingas, might be completed by ge (‘territory’, as in modern German gau). Nearly all place-names in ge lie here, at or close to minster sites, and fit well a concept of minster territories based on geographical zones: Eastry (‘east’), Denge (‘marsh’), Sturry (‘on the river Stour’), Lyminge (‘of the Roman town Lemanis’). Kentish minster territories on the Roman model could date from Kent’s mid-seventh century isolation, or even from St Augustine's archbishopric at Canterbury from 597. The only known exceptions seem to support this. Ely’s monastery although in the fens tradionally was founded by St Augustine. Vange lies on the Essex bank of the Thames estuary opposite Kent. ‘Surrey’ (‘southern ge’) perhaps in reality is Southwark, a minster territory founded at the south gate of London by the kings of Kent?

stow - ‘religious meeting place’

A late major religious place-name element, accepted as such, is stow, ‘(religious) meeting place’ [23]. Examples are Stow on the Wold (Gloucs), Stow (Lincs), Stowmarket (ecclesia de Stou, formerly Thorney, Suffolk), Peterstow (earlier Lann petyr, Herefordshire), sancte Albanes stow (St Albans), but perhaps also Bristol (‘stow at the bridge’). Surrey has two cases of burh stow, at Bristow Farm in Frimley by Bagshot Heath, and Burstow church a mile from Thunderfield Castle (‘Thunor’s field’, see later) in the remote clayey Weald and inexplicably the site of a court of king Alfred. Do both hint backwards to lost pagan centres?


A specific suggestion regarding monasteries. When a place-name has a woman’s name in it, could it just be that the person refered is a member of the aristocracy who became an abbess? The most famous was Hilda of Whitby. Others might be the Bebbe who seems to appear in the name of Bamburgh (Northumberland), Tetta (abbess of Wimbourne) at Tetbury (Wilts), Wulfran at Wolverhampton, and Beage at Bibury (Glos) and Byland Abbey (Yorks) (or is this beag, ‘circle’, suggested by Barker for Beaminster in Dorset)? The essentially pagan tradition of female seers gradually dies out in English Christianity.

Pagan sites

Christian minsters were well established in England by the mid-eighth century, often at future town sites whose large parish churches survive to this day. But was this the end of pagan sites? Writing of Surrey, the medieval historian John Blair notes of Bisley church that it is ‘conceivable that some cult survived . . . around the nearby holy well of St John the Baptist where parishioners were still being baptised within recorded memory’ [24]. It is reasonable to surmise such sites to be relict pagan ones. As with the sanctification of Bisley’s well through the good offices of St John, do place-names give clues to the surviving cultural significance of pagan holy springs, trees, barrows and sacred hill-tops and enclosures?.


The cathedrals of Wells, Southwell and Bath relate to the springs (OE wielle, ‘that which wells up’) still found there. A Roman mausoleum has been found by the pools at Wells [25], while the miraculous hot springs at Bath were dedicated to the Celtic god Sulis. The pools at which Lichfield cathedral was sited include Stowe Pool (presumably stow, ‘religious meeting place’).

Parallels are the OE place-name elements aewiell, ‘source of a river’, as in Ewelme (Oxon), and wael, ‘a deep pool, whirlpool’, as in the name Medeswael associated with Peterborough Abbey and refering to a pool in the Nene. Lincoln contains Celtic llyn, the ‘lake’ in the Witham here.

But are most instances of wielle, etc. religious (i.e. not just the obvious ones like Holywell), or do they refer to mundane water sources, as assumed? [27]. Surrey names suggest the former. Wielle, etc. is surprisingly uncommon: among parish names found only at Ewell (aewiel, a pool at the source of the Hogs Mill river in which Roman coin offerings have been found) [27], Carshalton (aewiell, source of the Wandle), and Camberwell - all in north-east Surrey, an area later known for its medicinal spas.

Consider also the rare Latin survival funta (‘fountain, spring’), Welsh ffynnon, and OE burna (‘spring, stream’). One set of names, Bidwell (Beds), and the adjacent Bedfont and Stanwell (Middlesex), containing byden (‘trough’) [28] and stan (‘stone’), seem likely to relate to the stone basins commonly found at holy wells. Burna in Surrey and elsewhere came to mean ‘bourne, stream’, but when used as a place- rather than a river-name the sense ‘spring’ seems evident. The minster name Sherborne (Dorset, ‘bright spring’) has its Surrey counterpart in Shirburn Spring, the former name of the Silent Pool near Guildford. This deep, clear legend-girt pool at the foot of the Downs closely resembles that at Wanborough (also near Guildford) where a Romano-British temple has been found; the name Shirburn appears to have transferred to the sub-minster at the village of Shere a mile from the Silent Pool.

Trees, stones, crosses

Sacred trees do not seem so well evidenced in place-names. Perhaps their significance faded earlier. Notable, however, are the overt religious connotations of a small minority of names with the OE element leah, a term whose origins are obscure but which seems normally to mean something like ‘wood’. Just a few catch the eye. Willey / Weoley, (Surrey, Worcs) means ‘leah with a heathen temple’. A sense ‘holy grove’ for such cases seems likely, especially when one adds the other Surrey names Thursley and Tuesley apparently refering to the worship of Thunor and Tiw. Another example is Thunder(s)ley in Essex [29]. Of similar type may be names in graf (‘grove’), as in Gravesend (Kent, Northants), Grafton (various counties), Bromsgrove (a minster, Worcs). Near Wing (Bucks), an -ingas place with a surviving early minster, Wingrave perhaps marks a prior holy site. Treo (‘tree’) is found in some important names, including Oswestry, Daventry, Coventry, and interestingly as Ekwall notes, in ‘several Hundred-names’. That ‘tree’ may be sometimes a term for a (wooden) religious upright or cross, is indicated by Croesoswald, the Welsh name for Oswestry. That Hundreds often did centre on natural trees, is shown by a name like Copthorne Hundred in Surrey, ‘at the pollarded hawthorn’. Some meeting places may have been marked by a (standing) stone or stone cross rather than a tree, as perhaps at Boston (‘Botolph’s stone’, Lincs), Holystone Abbey (Yorks), and the Surrey Hundred-name Brixton (‘Beorhtsige’s stone’).


The term beorg, ‘barrow’, is normally considered descriptive of a mere landscape feature. Continuing cultural significance is obvious where the name retains that of the individual whose burial mound it is. This happens with the term hlaw, ‘barrow’, as in Wilmslow (‘Wilhelm’s hlaw’) and Taplow (Taeppa’s, Bucks - see Eric Fitch in At the Edge No.1). Hlaw seems used of new feudal Germanic burial mounds like Taeppa’s of circa 620 (where the rediscovered adjacent church of 700 could be the burh of Berry Hill). That prehistoric mounds retained a role in the community is suggested by the survival of the Celtic-derived element cruc (Welsh crug), as in Crich (Derbs), Crick (Northants), Cricklade (Wilts), Crewkerne (Somerset; interestingly, ‘cruc house’, i.e. monastery?) and the Somerset and Surrey names Creechbarrow / Crooksbury (cruc beorg). (That crug in Wales meant barrow seems proven by the coincidence of crug names with tumuli evident for example on O.S. map 145 of the Preseli mountains of north Pembrokeshire). In Surrey, cruc crops up at the aforementioned Crooksbury Hill (beside the -ingas place Bintungom), Creek Copse (by Hascombe hillfort), and Cherchefelle Hundred (likely centred at one time on Thunderfield Castle, see below).

Sacred hill-tops

Beowulf’s burial mound was sited on a hill-top. In such cases the OE element hoh seems to present itself. It is said to mean a ‘projecting ridge of land’, deriving from the word ‘heel’. Yet clearly some hoh sites are of special significance. Synods were held in 645 at Icanho and in 747 at Clofesho. These locations are not known, but it has been suggested that Clofesho may be the eighth century minster at Brixworth in Northants [30]; this church stands beside a mound. Icanho might be Iken, a flattish hill beside the Alde estuary in Suffolk, where recently the remains of a Saxon cross have been found at the church [31]. No less suggestive are hill-top names like Ivinghoe Beacon on the Chilterns, Trentishoe (‘circle hoh') on the Devon cliffs, the seventh century monastery of Hoo on a hill above the Medway estuary, and the famous ship burial tumuli at Sutton Hoo above the Deben estuary in Suffolk. Is hoh a religious term, or are hoh-named sites simply good places to investigate? What, for example, is at Plymouth Hoe? Houghton and Hutton (hoh tun) appear in several counties, perhaps not always at religious sites. In Surrey, hoh (as Hoe, etc.) seems partially to correlate with parishes containing ingas names. Were our postulated seventh century minsters cited adjacent to pre-existing religious centres, the focus Surrey’s original Hundreds?

Sacred enclosures

Old English names for stone circles and the like seem rarely to have survived. Exceptions may be Stonehenge (‘hanging stones’), Ringstead (‘place of the ring’, Norfolk, Northants), Trentishoe (trendels hoh, see above) and Bewholme (beagum, ‘at the rings’, Yorks). But look at OE ora, said to mean ‘border, margin, bank’. Again this clearly is a culturally significant term. The Anglo-Saxon Chronicle claims Cymenesora and Cerdicsora respectively to be the landing places of the Germanic conquerors of Sussex and Wessex. Since the latter name seems to contain a Celtic personal name [32], are we dealing rather with British coastal religious centres at which the invaders first seized ratification?

Does ora in fact describe the embankments at significant sites - for example those surrounding Bronze Age sacred hill-top sites? At Oare (Wilts) is a huge such enclosure, and the same name in Berks and Kent (as in Wilts), figure in charters. As to the continuing importance of Bronze Age hill-top enclosures, Cunliffe notes of Sussex that all three known rural Romano-British temples occur inside such [33]. In Surrey, ora is recorded in eight places, usually hill-tops and including Nore below Creek Copse (see above), a lost ‘wolf ora’ beside Thursley (see above), and Nore Hill in Chelsham where a prehistoric embankment has recently been discovered [35]. Surrey’s most obvious sacred hills are not named with ora; instead, the Hills of St Martha, St Catherine (formerly Drakehull, ‘dragon hill’) and St Ann are topped by chapels which overshadow any former embankments.


Just occasionally, a place-name may give a clue to the activities at religious sites. What, for example, were ‘pipers’ up to at Peper Harow temple in Surrey?

We note from Ekwall Mottisfont (‘pleader’s funta’, Hants), Botwell (‘healing spring’, Middlesex), Fritwell (‘well for auguries’, Oxon), Elwell (‘wishing well’, Dorset). Such essentially religious rituals have their counterpart in the civil activities Ekwall suggests at for example Thingoe Hundred (‘assembly mound’, Suffolk), and Playstow names (marking places for sporting games, or maybe for religious plays). Wanborough in Wilts, Ekwall notes ’. . . looks like . . . “wagon”, but it is not easy to see what a compound waegenbeorg could mean. Cf., however, Wagenberg . . . in Germany’. Do we here have a reference to pagan ceremony involving the progression to the barrow of a holy wagon (see Alby Stone’s article in At the Edge No.2)? Beside Nore Hill in Surrey is Worms Heath, ‘snake’s head’ - one of a type with Heronshead (‘eagle’s head’) and Evershead (‘boar’s head’) also in the county. In discussing such names, and others including Gateshead (‘goat’s-’, Durham) and Manshead Hundred (‘man’s-’, Bedfordshire), Dickson [34] cites the pagan Germanic habit of religious decapitation, but Green [35] notes equivalent rituals among the Celts.

A place-name theory

We need to put religious place-names in context. Early estates, according to Ford [36], centred on a particular focus and a particular name; most other names were generated as geographical and economic dependents. Perhaps we can now say that the core ‘estate’ name often was religious. (The ‘estate’ anyway being a series of Chinese boxes: sometimes a hall, sometimes the Hundred, sometimes the pagan centre, monastery or minster, later the manor or parish). Thus the fact that Chertsey Abbey’s charter lists a set of -ham estates says nothing about their date or origins, but more about the power of the Church to codify a system. As Sawyer says [37], Old English place-names probably were subject to periodic complete substitution, with stability achieved only through the influence of legal or tax documents, or by the construction of a church.

We have noted a progression in monastic names, from -eg through -burh to -mynster, and suggested this reflects religious power politics. A general view of the way place-names change is needed. Sometimes core names are replaced, but perhaps as often they simply are modified or translated, as in the series Glastingei, Glestingaburg perhaps from Celtic glasto-, ‘woad’, where the earlier Celtic name Ineswytrin (Latin/Welsh Ynys Vitrium) may likewise have meant ‘island of woad’. Sometimes an earlier name survives attached to a marginally different geographical site, as in Surrey probably at Peper Harow / Eashing, and at Cherchefelle / Thunderfield / Burstow. The ‘different sites’ may well be a church, and the pagan spring, barrow or enclosure it replaced.

Religion provided cultural continuity. It was doubtless their religious content that allowed the Romano-British terms cruc and funta to survive. John [38] has suggested Celtic place-names survive where a local Romano-British aristocracy survived late. Perhaps this ‘aristocracy’ sometimes was monastic or priestly. The cathedral name Lichfield retained the bones of the name of the Roman town of Leto cetum (Wall, Staffs) [40] when the local administrative focus shifted (reverted?) to priests at a sacred pool two miles away.

So, place-names relate to religious sites in different ways. An archaic name may not be inherently religious but, as in the case of Lichfield, may survive through ecclesiastical agency. Or, a place-name element may have gathered a (temporary?) religious meaning over time: examples being Welsh llan (originally ‘enclosure’, later ‘religious enclosure’), and perhaps the OE eg, ge, ingas, burh (interpretations easy to miss if you have no adequate religious paradigm to call on).

Few place-name elements define objects that are specifically religious: weoh, hearg (both ‘pagan temple’), mynster, cirice (both ‘church’), but probably also cruc, beorg, hlaw. Some objects may appear secular or topographical but perhaps most commonly can more accurately be seen as religious: those indicated by funta, wielle, hoh, ora. So it seems our place-names are permeated by religious references of one sort or another. For the reason, one could look to Higham. He sees post-Roman England as a largely constant population with a smattering of intrusive energetic ‘Anglo-Saxons’ and a gradually acculturating British nobility becoming ‘English’. Newly enthused and peripatetic Christian missionaries doubtless were a civilising influence by personal example, but in the end, ‘local group loyalties may have been more to cult centres than to specific dynasties.’ [39]


1: Unless otherwise stated, all OE forms and interpretations are taken from The Concise Oxford Dictionary of English Place-Names, E. Ekwall, Oxford UP 4th edn 1960 or The Place-names of Surrey, J.E.B. Mawer et al, English Place-Name Society Vol. xi, Cambridge UP, 1934.
2: Anglo-Saxon England, F. Stenton, Oxford UP 1971 (3rd edn).
3: Bede’s Ecclesiastical History of the English People, B. Colgrave, R.A.B. Mynors (eds), Oxford UP 1969.
4: The Church in British Archaeology, R. Morris, Council for British Archaeology research report 47 1983.
5: ‘Place-Names of the Earliest English Records’, B. Cox, Journal of the English Place-Name Society Vol.8 1976.
6: Signposts to the Past, M. Gelling, Dent 1978.
7: K. Barker in The Early Church in Western England and Ireland, S.Pearce (ed), BAR 102, Oxford UP 1982.
8: Early Medieval Surrey: Landholding, Church and Settlement, J. Blair, Alan Sutton 1991.
9: The Anglo-Saxon Age, D.J. Fisher, Longman 1973; Britons and Saxons: the Chiltern Region 400 - 700, K. Rutherford Davis, Phillimore 1982.
10: The Age of Arthur: a History of the British Isles from 350 - 650, J. Morris, Weidenfeld & Nicholson 1973.
11: The Coming of Christianity to Anglo-Saxon England, Mayr Harting, Batsford 1972.
12: Gods and Myths of the Northern Europeans, H.R. Ellis Davidson, Pelican 1964.
13: East Clare Heritage Journal, Vol.2.
14: Fisher op. cit..
15: The Significance of the Distribution of the English Place-Names in -ingas, -inga- in South-East England, Medieval Archaeology x, 1966.
16: Ekwall op. cit. pxiv.
17: P. Wormald in Bede and Anglo-Saxon England, R.T.Farrell (ed), BAR British series 46, Oxford 1978.
18: The Origins of the Anglo-Saxon Kingdoms, S. Bassett (ed), Leicester UP 1989; Blair op. cit. 1991. Contrast this non-religious approach with Bede’s Hrypsaetna cirican, ‘the church of the people of the Hrype tribe’, for Ripon.
Analysis of the first elements of -ingas, -inga- names is relevant. Some personal names repeat themselves - are they peripatetic ecclesiastics? Like Teilo who became Bishop of Llandaff, it may even be possible to track missing persons through the ecclesiastical record. Could Bass or Basa at Basing (Hants, and in Bucks by a monastery near Runnymede) and/or Baschurch (Shrops), be the thegn Bass who retreated with the papal missionary Paulinus south from York in the 630’s, and/or the ‘priest Bass’ given in 669 the Roman fort at Reculver in Kent for a monastery? (Bede; The Anglo-Saxon Chronicle: a Revised Translation, D. Whitelock et al (eds), Eyre and Spottiswode 1961).
Some -ingas and -inga- names seem to contain totemic words (my thanks to Alby Stone for this suggestion): the helm (‘helmet’) and scyld ('shield') of the Helmingas and Scyldingas lineages of Scandinavia, and in English place-names the bill (‘sword, point of land’) of Billing-, the gara (‘spear, strip of land’) of Goring and the horn (‘horn, point of land’) of Horning-. In antique lineage names, totems or warrior nicknames might be expected. In place-names, the explanation could equally reflect some religious practice, or be topographical (as also perhaps in the ‘up’ of Upping- / Epping), eg. ‘monastery or people on the point / strip of land’ - reminiscent of Twynham, former name of Christchurch Priory in Hants, ‘(monastery) between the rivers’. Place is important in ancient religious sites, and we are perhaps dealing in geomancy (see discussion of hoh, later).
A British tribal area in the ‘Avon’ valley might be reflected in Avening (Gloucs), paralleling Axminster (Devon) and Sturry (Kent) in the Axe and Stour valleys (see discussion of ge). Cf. Bede’s Meanwara, ‘people of the Meon valley’, and the minster at Meon (Hants).
19: R. Poulton in The Archaeology of Surrey to 1540, J. & D.G. Bird, Surrey Archaeological Society 1987.
20: Pearce in Pearce (op. cit.), 1982.
21: Blair op. cit.
22: See D. Whitelock in Anglo-Saxon England Vol.1, P. Clemoes (ed), Cambridge UP 1972.
23: M. Gelling in Pearce op. cit. 1982.
24: Blair op. cit.
25: W. Rodwell in The Early Church in Western Britain and Ireland, S. Pearce, BAR 102, Oxford UP 1982.
26: Place-names in the Landscape, M. Gelling, Dent 1984. Cf. The Living Stream, J. Rattue, Boydell Press 1995.
27: Surrey Archaeological Society Bulletin No.265, 1992.
28: Gelling op. cit. 1984.
29: Higham suggests these Germanic gods would have been attractive to local British groups, as being close to Celtic paganism and less oppressive than Roman hierarchies. Rome, Britain and the Anglo-Saxons, N. Higham, Seaby 1992. (The next incumbant of the holy places was Roman Christianity - but Christianity liberated by the Irish paganism of its messengers).
30: D. Parsons in Baukunst des Mittelalters in Europa, F.J. Much (ed), Stuttgarter Gesellschaft fur Kunst und Denkmalpflege 1988 (reprinted for Brixworth parish church).
31: Iken parish church guide, 1996.
32: J. Morris op. cit.
33: The Regni, B. Cunliffe, Duckworth 1973.
34: Appendix 1 in Mawer op. cit.
35: The Gods of the Celts, M. Green, Alan Sutton 1986.
36: W.J. Ford in Medieval Settlement, P.H. Sawyer (ed), Arnold 1976.
37: P.H. Sawyer in Sawyer op. cit.
38: Orbis Britanniae and Other Studies, E. John, Leicester UP 1966.
39: Higham op. cit.

Originally published in At the Edge No.3 1996.

All rights reserved.

England, Pagan Sites, Trees, Stones, Crosses,
Filed under: