The Ten Towns of Coquetdale
Despite their nominal status as tenants in chief the Vescis played only a very minor role in Alwinton and the other townships of the parish. Alwinton was one of ten vills in the Alnwick barony, later known as the ‘Ten Towns of Coquetdale’, which the Vescis granted, or ‘subinfeudated’, to the neighbouring Umfravilles.
Eight of these comprised the vills in the eastern half of the parish of Alwinton listed above, including Alwinton township itself. These formed a compact block on the north side of the upper Coquet, opposite the Umfravilles’ stronghold of Harbottle Castle. The other two, Ingram and Fawdon, lay much further north in the Breamish valley, and were detached from the main block of the ‘Ten Towns’.
The Umfravilles did not hold all of these vills directly, as demesne manors, with the exception Shirmundesden and Alwinton, the two situated closest to Harbottle, and later, from the beginning of the 14th century, Fawdon. The remainder were granted to a variety of subordinate manorial/feudal lords.
A useful and detailed, though not necessarily comprehensive, summary is provided by the Inquisition Post Mortem of Robert de Umfraville in 1325 (Selected Sources and Surveys no. 2, iii; Cal IPM VI no. 607). Several of these lords held only a portion (moiety) of a particular township so that manorial and tenurial structure in many of the townships was complex and fragmented. Even in Alwinton itself, alongside the main Umfraville manorial holding, some land seems to have been granted to a local gentry family, the Horsleys of Farnham (Thirnam).
Hence, in the Inquisition Post Mortem of Gilbert de Umfraville II, earl of Angus in 1308 it was stated that ‘the lordship of the ten towns . . . . rendered nothing to the earl but knight's service and suit at his court of Hirbotell’ (Selected Sources and Surveys no. 2, ii; Cal IPM V no. 47), although the corresponding IPM for his successor, Robert, in 1325, also mentions a series of cornage payments and sums in lieu of castle guard at Alnwick. As ‘lords of the ten towns’ the Umfravilles thus formed an intermediate seigneurial tier between the Vescis, who were nominally the feudal tenants in chief, and the manorial lords. Effectively, they functioned as de facto tenants in chief, with the Vescis maintaining only residual feudal rights of knight service and castle guard at Alnwick.
Because of their apparently anomalous status, the Ten Towns of Coquetdale have given rise to considerable interest and speculation regarding their origins, and therefore merit further discussion. As noted above, the ten comprised Ingram (with its ‘members’ Reaveley and Hartside), Fawdon, ‘Chirmundesden’ (Peels), Biddlestone, Clennell, Netherton, Burradon, Sharperton, Farnham and Alwinton, and did not form a single geographical territory. Ingram and Fawdon in the Breamish Valley were separated from the main block of eight townships in Coquetdale by the four townships of Alnham Parish, namely Prendwick, Scrainwood, Unthank and Alnham, which were retained by the Vescis and held in demesne by them or subinfeudated directly to local manorial lords and free socage tenants (see the Historic Village Atlas: Alnham Report).
The apparent allocation of Breamish Valley townships to Coquetdale may appear a little odd today, but in this context Coquetdale seems to refer to the ancient administrative subdivision, or ‘ward’, in the county of Northumberland which also bore that title, rather than the vale itself. Medieval administrative geography could throw up some curious phenomena - Harbottle Castle, the capital of the ‘liberty of Redesdale’, was actually in Coquetdale, for instance - but similar examples could doubtless be found amongst the names of modern administrative districts.
The Ten Towns have aroused much interest because it is rather uncommon for such a large and compact block of manors to be subinfeudated to a neighbouring baronial lineage, the Umfravilles, whose status was equivalent to that of the Vescis, the baronial ‘tenants in chief’. In effect there would appear to be two levels of tenants in chief or superior tenant in these vills above the direct manorial lord. Furthermore, by the end of the medieval period, the inhabitants of the ten townships were tied by obligations of military service to Harbottle castle, the capital of the liberty of Redesdale (later the royal manor of Harbottle). This obligation is most clearly expressed in the 1604 Border Survey:
(The ten towns in Coquetdale) by their ancient custome owe their service to Harbotle in Rydsdale to be comaunded by the Capten there to serve in feild on horse or on foote in the Princes affaires for the defence of the Border lands (1604 Survey, 116).
It has been suggested that the customary service of the ten Coquetdale townships represented a relic of some Anglo-Saxon – perhaps even Anglian – military estate or district centred on Harbottle and embracing the Ten Towns, plus presumably the rest of upper Coquetdale at the very least (Anon. 1864; Dixon 1903, 177-8; NCH XV (1940), 472). Harbottle is one of a number placenames in northern England and southern Scotland which incorporate the Old English suffix -botl, generally translated as 'lord's hall'.
The suffix is perhaps the equivalent of the Latin term villa, which is used frequently in the works of Bede and his contemporaries to denote royal and ecclesiastical estates (cf. Higham 1986, 293). This class of placename has been considered to represent an early element in Anglian place-name formation, i.e. belonging to the 5th-6th centuries, but it has recently been the subject reconsideration by Barrow (1998, 67-9), who points out that its distribution across southern Scotland suggests some of these names could have originated later on, in the 7th-8th centuries. Such a defensive arrangement is not of itself implausible.
However the evidence from earlier documents presents a rather different picture from those of the sixteenth and seventeenth centuries. In the Inquisition Post Mortem for Robert de Umfraville, taken in 1325, the tenants holding all or part of the manors of Clennell, Biddlestone and Burradon were each listed as having to to pay sums for the guard of the castle of Alnwick, not Harbottle, as part of their dues (Cal IPM vi, no.607; see the Harbottle Village Report). This was repeated as a simple total - '30s for the ward of the castle of Alnewike' - when the inquest into Robert's holdings at death was retaken in 1331 (Cal IPM vii, no.390).
This suggests that the manorial tenants of the ten townships retained some military obligations to the Alnwick barony up until at least the early 14th century. Furthermore none of the 13th and 14th century inquisitions specify that the tenants of the ten townships had to perform castle guard at Harbottle.
Perhaps even more significantly, the obligatory military service performed by tenants of the Coquetdale was recorded in the 16th and 17th centuries, after centuries of association between the former Umfraville liberty of Redesdale and the Ten Towns, and following the profound transformation of the character of the Border as a result of the prolonged warfare and chronic insecurity prevalent during the late medieval period. In other words, the obligatory military service of the Coquetdale tenantry at Harbottle may result from the circumstances of the late medieval period rather than representing a fossilised relic of very much earlier arrangements.
More difficult to interpret are the 15d. cornage payments recorded in respect of several of the townships in the Inquisition Post Mortem of Robert de Umfraville in 1325 (Cal IPM VI, no. 607; Selected Sources and Surveys no. 2, iii). Cornage was a non-feudal tax, originally a cattle render, levied upon each vill and paid to the king by the barons in a block sum. These appear to be due to the Umfraville lord, rather than the king.
Moreover the cornage payments made by the Vescis for the Alnwick barony to Henry III in the 49th year of his reign (1265/6) apparently exclude any payments from the Ten Towns (Hinde 1857; Dixon 1985, I, 73-5). Perhaps the Umfravilles had managed to bring these payments within the vice-regal envelope of their liberty of Redesdale. At any rate the fact that the Umfravilles rather than the Vescis had control over these payments may indicate no more than the respective degree of authority to the two baronial lineages sustained in these vills. It is clear that in almost every significant respect the lordship of the Ten Towns rested with the Umfravilles, as emphasised in the Inquisition Post Mortem for Gilbert de Umfraville II in 1308 (Cal IPM VII, no. 47)
More recently O’Brien (2002, 66-7) has put forward a very different interpretation, arguing that the Ten Towns represent the territory of an early medieval ‘multiple estate’ or shire, which he labels ‘Bromic’. There is convincing evidence that the Breamish Valley once formed part of monastic estate held by St Cuthbert’s house of Lindisfarne, perhaps between the 7th-9th centuries. The Historia de Sancto Cuthberto, or ‘History of St Cuthbert’, a work probably compiled in the mid tenth century, included in a summary of the territory once held by the monastery of Lindisfarne (Lindisfarnensis terrae) ‘all the land lying to either side of the River Breamish (Bromic), right up to its source’ (HSC, 4; Craster 1954, 178; Hart 1975, doc. 152; Morris 1977, 89, map; Higham 1986, 288-9).
It is widely accepted that ‘it is difficult to regard this archaic description of the bounds of Lindisfarne territory as being other than authentic’ (Hart 1975, 137). However the relevant passage of the Historia de Sancto Cuthberto makes no mention of the monastic landholding embracing other areas further to the south, in upper Coquetdale and Alnham parish (which O’Brien also considers part of the shire of ‘Bromic’ although it never formed part of the Ten Towns). It simply refers to Lindisfarne’s lands embracing the full extent of the Breamish valley, right up to its source. In the light of this, it would therefore seem preferable to regard both theories with a degree of caution, unless new information emerges.