Monday, 28 March 2016

Al-Idrisi's twelfth-century map and description of eastern England

The aim of this post is simply to share an interesting mid-twelfth-century map and description of eastern England compiled by the great Muslim scholar Muhammad al-Idrisi for his geographical encyclopedia Nuzhat al-mushtaq, or Tabula Rogeriana, written for Roger II of Sicily and completed in 1154.

Britain according to al-Idrisi in the twelfth century, orientated with north at the bottom of the image; click for a larger view. Note, Scotland is the 'peninsula' at the bottom of the map, whilst the long thin peninsula on the right is Cornwall, which is said to 'resemble a bird's beak'. The original sectional maps have been combined in this image and the Arabic script transliterated/expanded by Konrad Miller; the original Arabic atlas map of eastern England, from a copy contained in the Oxford MS Pococke 375, is included at the bottom of this post for comparison. The towns named can be identified as follows, running clockwise from top-left: Dover (with London and Oxford inland along the Thames), Hastings, Shoreham, Southampton (with Winchester inland from this), Wareham, Dorchester, Salisbury, Durham (wrongly placed on the west side of the island), York, Grimsby, Boston (with Lincoln inland along the Witham), Norwich, and Yarmouth (image: Wikimedia Commons).

Al-Idrisi was a descendant of the eleventh-century Ḥammūdid dynasty of North Africa and Spain (Al-Andalus), which claimed in turn to derive from the Idrisids of Morocco, and his Nuzhat al-mushtaq is one of the great geographical works of the medieval period. In this work, written at Palermo, Sicily, he gathered together a vast array of information on the various regions of the world known to him, illustrated by a series of 70 maps, including a brief description of eastern England north of the Thames that runs as follows:
This [England] is an island resembling the head of an ostrich, and contains flourishing cities, lofty mountains, flowing rivers, and level ground. There is abundant fertility in it. Its inhabitants are hardy, resolute, and prudent. The winter there is of long duration... 
[Gernemutha=Yarmouth] is a handsome town beside the sea... From the town of Gernemutha [Yarmouth] to the town of Norwicca [Norwich] is ninety miles. The town of Norwicca [Norwich] is distant ten miles from the sea, and from there to Grimsby is a hundred and fifty miles by sea. From the said town of Gernemutha [Yarmouth] the sea[-coast] curves round in a circle, but still tending northwards. From the said town of Grimsby to the town of Evrvic [York] is eighty miles. The latter lies at a distance from the Ocean, and on the border of the peninsula of Scotia, which is contiguous with the island of l'Angleterre. It is a long peninsula stretching northwards of the larger island; it has no flourishing cities, towns or villages; its length is a hundred and fifty miles. From the town of Evrvic [York] to the estuary of the river of [Boston] is a hundred and forty miles, and [Boston] is a fortress situated on this river twelve miles upstream from the sea. From the aforementioned town of Grimsby to the the town of Nicolas [Lincoln] inland is a hundred miles; the river flows through the midst of it and flows out of it towards the town of Grimsby, but flows into the sea on the south of the latter, as we have mentioned before. From the inland Nicholas [Lincoln] to the town of Evrvic [York] is moreover ninety miles, and from thence to the town of Donelme [Durham] eighty miles northwards.(1

Map of places in eastern England between the Thames and the Tees that are mentioned in al-Idrisi's Nuzhat al-mushtaq, above (image: C. R. Green).

Detailed view of the east coast of England from Konrad Miller's redrawn and transliterated version of al-Idrīsī's map, with north at the bottom. The green river running left to right is the Witham, which is shown flowing past the fortress of Boston and then through the city of Nicholas, i.e. Lincoln, which is depicted as lying close to the centre of England. Further north, towards the bottom of the map, are shown Grimsby and York, after which comes the southern part of the 'peninsula' of Scotland; the northern bank of the Humber and Northumbria are missed out entirely, except for Durham, which is wrongly mapped on the west coast of Britain (image: Wikimedia Commons).

Although this description adds little to the historical record, and some of the distances appear rather dubious, it is nonetheless interesting both as an illustration of the spread of knowledge of this part of England by the mid-twelfth century and for the places that are chosen to be mentioned. With regard to the latter point, it seems clear from both the text and the map that the area from Yarmouth to York was the part of the east coast of Britain with which al-Idrisi's source of information was familiar. There is, for example, nothing depicted or mentioned to the south of Yarmouth until one reaches the mouth of the Thames and moreover little evidence for any knowledge of any sites north of the Humber aside from Durham (which is wrongly mapped on the western side of England, not the east), with the northern bank of the Humber being omitted entirely so that York is consequently placed on the coast and close to the border with Scotland. Similarly, it is noteworthy that the only river depicted between the Thames and Scotland is the Witham, which is rightly said to flow through the midst of Lincoln (it runs between the old Lower City and its medieval southern suburbs), and the statement that Boston lower down the Witham is a 'fortress' is also of interest, given that the riverside town was indeed moated/ditched in this period. Likewise, the description of the river at Lincoln as not only flowing into the sea near Boston, but also flowing 'out of [Lincoln] towards the town of Grimsby', equally seems to reflect an awareness of the local situation, as the two were indeed connected by inland waterways in the twelfth century, with one being able to travel by boat from the Witham at Lincoln north-westwards along the Foss Dyke and then down the Trent and the Humber through to Grimsby after 1121, when the Foss Dyke was renovated and made navigable again by Henry I. Needless to say, the fact that al-Idrisi's source thus appears to have had a greater familiarity with Norfolk and Lincolnshire than other parts of the east coast is perhaps a point of some interest, both generally and from the perspective of those concerned with the history of these regions.

A sixteenth-century copy of al-Idrīsī's original Arabic map of eastern England, Scotland and the west coast, from Oxford's MS Pococke 375, fol. 310b-311a (image: Bodleian Library).


1     A. F. L. Beeston, 'Idrisi's Account of the British Isles', Bulletin of the School of Oriental and African Studies, 13.2 (1950), 265–80 at pp. 278, 279–80.

The content of this post and page, including any original illustrations, is Copyright © Caitlin R. Green, 2016, All Rights Reserved, and should not be used without permission.

Saturday, 26 March 2016

The Hwicce of Rutland? Some intriguing names from the East Midlands

The following post offers a little idle speculation on those names from the East Midlands that make reference to the Hwicce, an Anglo-Saxon people who are better known as the inhabitants of a well-documented, seventh- to eighth-century kingdom based in the Worcestershire and Gloucestershire area. In particular, it asks whether these names could have anything to tell us about the early history of the Hwicce and, if so, what this might possibly be?

The seventh- and eighth-century kingdom of the Hwicce and place-names that definitely or possibly contain the group-name Hwicce, but which are located outside of the bounds of the kingdom; click image for a larger view and see the text below for a key to the names. Note, the names and the kingdom of the Hwicce are set against a reconstruction of a suggested likely coastline during the pre-Viking era (image: C. R. Green).

The above map depicts both the boundaries of the kingdom of the Hwicce and the eight place- and district-names that contain the group-name Hwicce or a similarly formed personal name derived from the group-name and also lie outside of documented territory of the Hwicce.(1) The names in question are as follows:
  1. Witchley Warren/Wicheley Heath—now the name of a farm in Edith Weston parish, Rutland, but originally the name of an extensive common area between Normanton, Edith Weston, Empingham and Ketton, labelled as Wicheley Heath on Speed's 1603–11 map of Rutland, Bowen's 1756 map, and other early maps. Early forms include Wicheslea, Whicchele, Wicheley etc, recorded from 1185 onwards.(2)
  2. Hwicceslea east hundred—first recorded in the Northamptonshire Geld Roll of 1072–8; see further number three, below.
  3. Hwiccleslea west hundred—first recorded in the Northamptonshire Geld Roll of 1072–8. In Domesday Book, both 'Witchley' hundreds were treated as a single unit (Wiceslea Hund', Wicelea Wapent') and included in Northamptonshire, although they are usually believed to have been originally part of the Anglo-Saxon-era Rutland. 'Witchley Hundred' is unrecorded after 1086 and its territory—roughly a third of the total area of Rutland—was subsequently divided into the medieval East Hundred and Wrangdike Hundred of Rutland. The hundred-name Wicelea/Witchley is, needless to say, identical in etymology to nearby Witchley Warren/Wicheley Heath (above)—although the farm currently bearing that name actually lies just outside of both East and Wrangdike hundreds—and Witchley Hundred is believed to be named from the original, extensive Wicheley Heath, with this being the original meeting-place/common for the eleventh-century hundred(s). With regard to the meaning of the name 'Witchley', the apparent genitival ‑s‑ in some of the early forms has led to the suggestion that this name was formed around an unrecorded personal name derived from the group-name Hwicce; however, these genitival forms disappear at an early date and both Cox and Insley suggest that the totality of evidence points rather to an original *Hwicca-lēah/*Hwiccenalēah, 'the woodland of the Hwicce' or similar.(3)
  4. Whissendine—located in the north-west of Rutland, early spellings include Wichingedene (1086), Wissendene (1206), and Wissinden' (1238). This name can be read either as an ‑inga‑ name formed around the above personal-name plus Old English denu, 'valley' or, more likely according to both Cox and Gelling, as an original *Hwiccena-denu, 'valley of the Hwicce'; certainly, the latter suggestion would seem a better fit for the majority of the surviving forms.(4)
  5. Wichley Leys—a minor name within Whissendine parish, first recorded in 1861; appears to be identical in meaning to Witchley Warren/Wicheley Heath, above, and is accepted as such by Cox.(5)
  6. Whiston—a parish in Northamptonshire, around 20 miles from the Rutland concentration of names; potentially early forms include Wychenton (974) and Hwiccingtune (974). On the basis of the second form, it is sometimes suggested that this is an ‑ingtūn name formed around an unrecorded personal name derived from the group-name Hwicce, as above; however, it has more recently been noted that the second form is from a late copy of a dubious charter and that this place-name and its other forms are more plausibly explained as reflecting an original *Hwiccenatūn, 'the village/settlement of the Hwicce'.(6)
  7. Wychwood Forest—a large area in western Oxfordshire on the boundary of the kingdom of the Hwicce; first mentioned as Huiccewudu in 841, 'the forest of the Hwicce'.(7)
  8. Wychnor—located in Staffordshire; the earliest forms are Hwiccenofre and Wicenore, probably representing either 'the flat-topped ridge of the Hwicce' or 'the river-bank of the Hwicce'?(8)
These place- and district-names are, of course, potentially of some significant interest. In particular, it is intriguing to note that, aside from Wychwood and Wychnor (both of which are relatively close to the documented territory of the Hwicce), all of the other names are located in the East Midlands. Two explanations have been offered for this situation. The first is that these names are evidence of an undocumented north-eastwards movement of people from the West Midlands kingdom of the Hwicce into the East Midlands, perhaps during the seventh–ninth centuries.(9) The second explanation is that these names instead reflect a situation wherein the Hwicce were originally settled in the East Midlands during the fifth to sixth centuries and then moved south-westwards into their seventh- to eighth-century kingdom at some point before the seventh century.(10) So, which of these contrasting theories offers the most plausible explanation of the evidence that we have?

Map of Rutland, showing the Rutland Hwicce names discussed above; the names in italics are those of the Witchley East Hundred and Witchley West Hundred, as recorded in the Northamptonshire Geld Roll of 1072–8, and the approximate position and extent of 'Wicheley Heath' is based on the maps of 1603–11 and 1756, along with the location of the surviving Witchley Warren Farm in Edith Weston parish. The grey lines reflect the late eleventh-century hundred boundaries, with the dotted line representing the division between the later East and Wrangdike hundreds, which is thought to perpetuate that between Witchley East Hundred and Witchley West Hundred (image: C. R. Green).

With regard to this question, it can be tentatively suggested that there is at least a potential case for considering the second of the above scenarios as the more likely of the two: that is to say, that the East Midlands names could somehow reflect an early, pre-seventh-century presence and territory of the Hwicce in this region. In the first place, if these names do indeed all contain the group-name Hwicce—as has been supported by both Insley and Cox (11)—then this is a rather different matter to them simply being names referencing individual members of the Hwicce who bore a personal name derived from their group-name. In particular, it would imply that we are not simply dealing with individual 'Hwiccians' present in the East Midlands, something that might accord well with the notion of a seventh-century or later north-eastern movement to this region from Worcestershire and Gloucestershire, but rather with a significant body of people living here, sufficient for them to be locally known as the Hwicce and give their name to multiple sites.

Second, the evidence we possess could be further taken to suggest that the Hwicce actually occupied a fairly sizeable territory in the East Midlands, not merely a handful of settlements. So, not only is their name apparently preserved in that of the eleventh-century Whitchley Hundred that met at Wicheley Heath, 'the woodland of the Hwicce', a district that encompassed around a third of modern shire of Rutland, but there are also other Hwicce names within Rutland, with yet another potential 'Witchley' name in the north-west of this small county—Wichley Leys, in Whissendine parish—and a probable *Hwiccena-denu, 'valley of the Hwicce', there too. In this context, it might be tentatively wondered whether this concentration of onomastic evidence might not in fact point towards the Hwicce being a group that somehow notionally inhabited the whole of Rutland at one point? Certainly, Phythian-Adams and Cox have both argued that Rutland as a whole represents a very ancient territory, with roots in the pre-Roman era, and one that retained its integrity until at least the later ninth century, when it was temporarily administratively divided, which is perhaps suggestive. Although such a hypothesis as the above is difficult to verify, it would arguably explain the concentrated distribution of most of the eastern Hwicce names, and it is furthermore intriguing to note that the point at which the three early Rutland hundreds all met actually lies only half a mile or less to the west of the original Wicheley Heath.(12)

Needless to say, if the Hwicce may thus have inhabited or controlled a sizeable Anglo-Saxon territory in the East Midlands—of whatever extent—as well as in the West Midlands, this would appear to accord well with Smith and Cox's notion that the Hwicce were initially based in the east during the fifth–sixth centuries prior to a move westwards and the foundation of their seventh- to eighth-century kingdom in the West Midlands.(13) Such a direction and chronology of travel would also find echoes in other evidence for 'secondary' movements westwards and northwards from the East Midlands and East Anglia during the fifth–sixth centuries too. For example, a combination of linguistic, archaeological and historical evidence suggests that the Anglo-Saxon kingdom of Northumbria may well have had its roots in groups originally based in the Lincoln region.(14) Similarly, attention has often been directed to place-name evidence that seems to indicate that the Mercian royal Iclingas were initially based in East Anglia and/or Middle Anglia before moving to their later middle Trent heartlands, and there is a case to be made for connecting the Hreope/Hrepingas—arguably one of the major population groups of the original Mercian kingdom, alongside the Tomsætan and the Pencersætan—with southern Lincolnshire too.(15) Finally, further support for the idea that groups from eastern Britain moved into the area of the documented kingdom of the Hwicce is offered by the Worcestershire names Conderton (Cantuaretun), Whitsun Brook (Wixenabroc) and perhaps Phepson (Fepsetnatune), which imply the presence of members of the Cantware of Kent and the Middle Anglian Wixan and Feppingas here in addition to the Hwicce.(16)

In conclusion, the Hwicce names of the East Midlands and especially Rutland clearly form an interesting group. Although it is impossible to prove definitively, A. H. Smith and Barrie Cox's suggestion that these names could reflect a situation whereby the Hwicce originally controlled a territory in this area prior to the establishment of their documented seventh- to eighth-century kingdom in the West Midlands would seem to be at least partially defensible. Not only would such a scenario explain—and perhaps fit better with—the distribution, concentration and nature of the majority of these names, but it would also appear to have a potential context in other suggested pre-seventh-century movements of groups out of the East Midlands and East Anglia. Of course, if the above is correct, then the notion that the Hwicce were named from the topography of their territory in the West Midlands and/or created de novo by the Mercian kings clearly cannot be sustained.(17) The other two major theories as to the origins of the Hwicce would, however, still work—namely that they were either a well-established Anglo-Saxon group bearing 'a very old folk-name, perhaps going back to the pre-migration age', or a group bearing an Anglicized post-Roman British name—although an examination and assessment of these lies beyond the scope of the present post.(18)


1     The boundaries of the kingdom of the Hwicce shown on the map are after P. Sims-Williams, Religion and Literature in Western England, 600–800 (Cambridge, 1990), fig. 1 and p. 5; see also J. Blair, Anglo-Saxon Oxfordshire (Stroud, 1998), p. 50. Sims-Williams offers a brief overview of the history of the kingdom of the Hwicce in Religion and Literature, pp. 29–39, concluding that 'by the end of the eighth century the kingdom of the Hwicce, as a kingdom, was extinct' (p. 39).
2     B. Cox, The Place-Names of Rutland (Nottingham, 1994), pp. 221–2. 
3     On the etymology and relationship to Witchley Warren/Wicheley Heath, see J. Insley & A. Scharer, ‘Hwicce’, Reallexikon der Geremanischen Altertumskunde, 11 (2000), 287–96 at pp. 288–9; Cox, Place-Names of Rutland, p. 222; C. Phythian-Adams, 'Rutland Reconsidered', in A. Dornier (ed.) Mercian Studies (Leicester, 1977), pp. 63–86 at p. 76–8; and A. Pantos, '"On the edge of things": the boundary location of Anglo-Saxon assembly sites', Anglo-Saxon Studies in Archaeology and History, 12 (2003), 38–49 at p. 40. For East Hundred and Wrangdike Hundred as the descendants of the eleventh-century Witchley Hundred(s), see Cox, Place-Names of Rutland, pp. xlviii, 130; O. S. Anderson, The English Hundred-Names (Lund, 1934), pp. 129–30; and I. B. Terrett, 'Rutland', in H. C. Darby & I. B. Terrett (eds.), The Domesday Geography of Midland England, second edition (Cambridge, 1971), pp. 359–83 at pp. 359 and fig. 124.
4     Cox, Place-Names of Rutland, pp. 55–6; M. Gelling & A. Cole, The Landscape of Place-Names (Stamford, 2000), p. 118; Insley & Scharer, 'Hwicce', p. 288; V. Watts, The Cambridge Dictionary of English Place-Names (Cambridge, 2004), p. 672.
5     Cox, Place-Names of Rutland, p. 61; Insley & Scharer, 'Hwicce', p. 288.
6     Insley & Scharer, 'Hwicce', p. 289; A. D. Mills, A Dictionary of English Place-Names (Oxford, 1991), p. 356; Watts, Cambridge Dictionary of English Place-Names, p. 672.
7     Watts, Cambridge Dictionary of English Place-Names, p. 706; Insley & Scharer, 'Hwicce', p. 289.
8     D. Horovitz, A Survey and Analysis of the Place-Names of Staffordshire, 2 vols. (PhD Thesis, University of Nottingham, 2003), vol. I, p. 46; vol. II, p. 649.
9     Sims-Williams, Religion and Literature in Western England, 600–800, p. 30; C. Hart, ‘The Tribal Hidage’, Transactions of the Royal Historical Society, fifth series, 21 (1971), 133–57 at p. 138; F. M. Stenton, ‘The historical bearing of place-name studies: the English occupation of southern England’, in D. M. Stenton (ed.), Preparatory to Anglo-Saxon England (Oxford, 1971), pp. 266–80 at pp. 269–70.
10     See A. H. Smith, The Place-Names of Gloucestershire IV (Cambridge, 1965), p. 42; A. H. Smith, ‘The Hwicce’, in J. B. Bessinger & R. P. Creeds (eds.), Franciplegus: Medieval and Linguistic Studies in Honour of Francis Peabody Magoun Jr (New York, 1965), pp. 60–2; and Cox, Place-Names of Rutland, pp. xxv–xxvi, xxx, 55–6, 61, 221–2. See also Green, Britons and Anglo-Saxons: Lincolnshire AD 400-600 (Lincoln, 2012), p. 264; Insley & Scharer, ‘Hwicce’, pp. 288–9; and D. Hooke, The Anglo-Saxon Landscape: The Kingdom of the Hwicce (Manchester, 1985), p. 14.
11     See especially Insley & Scharer, 'Hwicce', pp. 288–9, and Cox, Place-Names of Rutland, pp. xxv, xxx, 56, 61, 222.
12     Phythian-Adams, 'Rutland Reconsidered'; C. Phythian-Adams, 'The emergence of Rutland and the making of the realm', Rutland Record, 1 (1980), 5–12; Cox, Place-Names of Rutland, Introduction. The description of the physical relationship between Wicheley Heath and the meeting-place of the Rutland hundreds of Asloe, Martinsley and Witchley is based on the depiction of the hundred boundaries and Wicheley Heath on John Speed's 1603–11 proof map of Rutland, made for his The Theatre of the Empire of Great Britaine (Amsterdam, 1611/12), available online from the University of Cambridge at Interestingly, by far the largest early Anglo-Saxon cemetery in Rutland, Empingham II, also lies only around 250 metres from the point that the three Domesday hundreds met.
13     Smith, Place-Names of Gloucestershire IV, p. 42; Smith, ‘The Hwicce’, p. 64; Cox, Place-Names of Rutland, pp. xxv–xxvi, xxx, 56, 61, 222.
14     For a detailed analysis, see especially 'Lindisfarne, the Lindisfaran and the origins of Anglo-Saxon Northumbria', in Green, Britons and Anglo-Saxons: Lincolnshire AD 400-600 (Lincoln, 2012), at pp. 235–65 and also p. 153, and also Green, 'Lindisfarne and Lindsey', Anglo-Saxon, 2 (2009), 1–19 [awaiting publication].
15     On the Mercian ruling family and its possible links to Middle/East Anglia, see, for example, E. Martin, ‘The Iclingas’, East Anglian Archaeology, 3 (1976), 132–4; T. Williamson, The Origins of Norfolk (Manchester, 1993), pp. 71–2; and J. N. L. Myres, The English Settlements (Oxford, 1986), p. 185, though compare N. P. Brooks, ‘The formation of the Mercian kingdom’, in S. Bassett (ed.), The Origins of Anglo-Saxon Kingdoms (London, 1989), pp. 159–70 at p. 164. On the Hrepingas/Hreope, see the tentative discussion in Green, Britons and Anglo-Saxons, pp. 252–4.
16     Sims-Williams, Religion and Literature in Western England, p. 32; D. Hooke, Worcestershire Anglo-Saxon Charter Bounds (Woodbridge, 1990), pp. 125–9, 167–9, 190–3; E. Ekwall, The Concise Oxford Dictionary of English Place-Names, fourth edition (Oxford, 1960), pp. 120, 365, 514; Watts, Cambridge Dictionary of English Place-Names, p. 154.
17     Sims-Williams, Religion and Literature in Western England, p. 30; Insley & Scharer, 'Hwicce', p. 295; M. Gelling, ‘The place-name volumes for Worcestershire and Warwickshire: a new look’, in T. R. Slater & P. J. Jarvis (eds.), Field and Forest: an Historical Geography of Warwickshire and Worcestershire (Norwich, 1982), pp. 59–78 at p. 69. Note, I ignore the idiosyncratic theories as to the origins of the Hwicce expounded in S. J. Yeates, The Tribe of Witches: the Religion of the Dobunni and Hwicce (Oxford, 2008), which are widely rejected by academic commentators, aside from to note that this theory would also not be compatible with the idea of the Hwicce having been based in the Rutland area in the early Anglo-Saxon period.
18     The quotation regarding the possibility that the name Hwicce may go back to the 'pre-migration age' is from Sims-Williams, Religion and Literature in Western England, pp. 29–30 at p. 29; see also Smith, 'The Hwicce', p. 62; Smith, Place-Names of Gloucestershire IV, p. 33; and Insley & Scharer, 'Hwicce', p. 295. The alternative, British theory was recently proposed in R. Coates, 'The name of the Hwicce: a discussion', Anglo-Saxon England, 42 (2013), 51–61.

The content of this post and page, including any original illustrations, is Copyright © Caitlin R. Green, 2016, All Rights Reserved, and should not be used without permission.