Runes in the Lake District. A wander through runic inscriptions in the West of England.

A list of inscriptions, with translations.

Discussion on the corpus.

The runic inscriptions in this corpus were all located in the North-West of England within the district of Cumbria, specifically from the Solway Coast to Morecomb Bay. This area was known as a trading area during the Viking Age and part of the Viking diaspora in England, with Scandinavians raiding, trading and settling from the ninth to eleventh centuries; these Scandinavians were likely in continued contact with other Scandinavians from Ireland, the Isle of Man, and the Western Isles of Scotland (Fellows-Jensen 1985: 80-81). However, though there was a Scandinavian population in the West of the British Isles, they left very little written contemporary evidence for themselves outside of runic inscriptions. Therefore, this corpus is the primary evidence for not only runic writing and its status in the NW but also the language of those using them.

Before linguistic concerns are explored, the material aspect of the corpus will be considered to see what it can tell us about the rune-using community. The corpus indicates that the rune-using community (likely descended from the pagan Scandinavian immigrants) were Christian and used runes as a form of memorializing patronage to the church. All of the inscriptions are found within a Christian setting. There are differences, however: several of these inscriptions are deliberately built into the structure of the building or are clearly designed to be used within a Christian locale, for example E1 is a baptismal font used in Christian religious ceremonies, and E9 is a semicircular tympanum above the church entrance. However, E3 appears more lightly incised and could be graffiti made inside the church, and though E6 and E11 are deeply carved inscriptions on stone re-used within the structure of a religious building, the original location and functions are unknown. Though E6 is inscribed on a stone depicting Adam and Eve, it is unclear whether it was intended as part of the inscription, though there are Roman graphs on the stone, so it is not unlikely. So, the corpus demonstrate that runic epigraphy was deemed suitable by those attached to the church to be inscribed on religious property. Furthermore, E1 and E9 inscriptions show that runes are used for inscriptions that indicate patronage and wealth to the church. E9 states that Gamall léta þenna kirk ‘Gamall ?raised this church’, and E1 that Rikarþ he me iwrocte ‘Rikarþ, he made me…’. This is unsurprising: many runic inscriptions from the Viking Age and later are directly related to Christianity, with approximately 13% of runic material from the Viking age and early medieval period containing Christian prayers and invocations (Zilmer 2013: 121). Though the corpus lacks direct reference to Christianity in the inscriptions, the use of runes in making inscriptions concerning contributions to the church further confirm that runes were culturally acceptable within a religious setting, and even the primary mode of written communication at that specific time for those local communities, since there appears to be a lack of entire Roman letter inscriptions in the area being used in the same way. Even E6 is inscribed on a stone with the Christian iconography of Adam and Eve. Though it cannot be certain that all of the corpus was deliberately Christian in scope (E3 and E6 could be graffiti, and E11 could be reused stone), the two instances of specifically Christian usage (E1, E9) indicate that the rune-using communities in the NW were likely also Christian communities and that runes are used as a mode of memorial within religious settings.

The issue of patronage of the runic inscriptions highlights the personal names in the corpus. All but one inscription bares a personal name (pers. n.). These pers. n. are Rikarþ (E1), Dolfinn (E3), Gamall (E9) and Dotbert (E11), but these names are not without their interpretative difficulties. The pers. n. Gamall (E9) has been taken from rr.1-3 kml. It is infrequent, however, that a pers. n. would be abbreviated to that extent because it was likely incised to refer to the benefactor of the church, making the pers. n. one of the most important parts of the inscription. Some runic inscriptions from Scandinavia feature pers. n. abbreviations, such as DR 124 and 228 (Barnes 2012: 310); however, the other younger fuþark inscriptions from the British Isles do not feature abbreviations of personal names to that extent (see Barnes and Page 2012 Appendix II). Holman (1996:75) claims that abbreviating to that extent is so unusual that kml is not a pers. n. but instead the noun kuml ‘monument’. However, Holman’s interpretation relies on readings of rr.4-7 lïta as having a word divider between r.6 and r.7 and reading r.7 as i i instead of a a , making lït:i, so that the i is an ON preposition followed by an accusative þïna. This interpretation would solve the issue of the pers. n. abbreviated to E9’s extent. However, it is unlikely that a monument in stone such as E9 would be erected without a pers. n. as a sign of commemoration to the patron. An abbreviated form of the pers. n., though unusual, could have been inspired by inscriptions within the trading contact zone of the NW. An abbreviated form of ON pers. n. Gamall is convincing, then: the name is known from Cumberland place-names (EPNSC: 20-22) so would not be uncommon for the area. Another pers. n. that raises questions of interpretation is E11. Several interpretations by scholars have included Brown and Dickens’s (1929: 222) ‘Rotbert/Rodbert’, where Barnes and Page (2006) claim reading r.1 as r is “an act of desperation” (318). Holman (1996: 84) draws information from MS Add.10374; she suggests the initial ‘ rune represents /he/ based on her perception of rune-like graphs in the manuscript - she claims that these runes read as Heogebrt. Barnes and Page (2006: 319) takes this as unlikely: ‘it is hard to believe that these curious inventions can have influenced epigraphic practice in north western England’ (319). Indeed, since dotted runes were formed to expand the phonological range of the fuþark from the twelfth century, ‘ could represent the voiced [d] where t represents [t] r.3. If taken as [d] this would make this rendering the sole attestation of the pers. n. Dotbert in the British Isles (PASE 2019). So, some of the names in the inscriptions refer to those that inscribed the runes (E3; possibly E11), two inscriptions (E1, E9) are inscriptions based around patronage. The corpus, then, is pers. n. heavy with a preference towards named public commemoration (E1, E9).

The personal names also highlight the irregular morphology present in the inscriptions. By the tenth century, Scandinavian was in regular use in NW England (Higham 1985: 48-59), the area settled by Norwegian Vikings from colonies in Ireland, Scotland, the Western Isles, and the Danelaw (Fellows-Jensen 1985: 80-81). The geographical area where the inscriptions were found was an area dominated initially by immigration and trade, and later by new settlement of Scandinavians who likely spoke dialects of Norse. However, how long Scandinavian survived in the British Isles, as well as the nature of its survival, is an ongoing topic (Ekwall 1930: 17-30; Page 1971: 156-181) and this corpus can contribute.

E1, E3, E9 and E11 lack the nominative masculine inflected ending -r that would be expected for inscriptions made in ON. The diagnostically ON inscriptions, where the language appears to be ON through the vocabulary forms, E3 (þesa, stain) and E9 (þena, kirk), have the Scandinavian names Dolfinn and Gamall respectively. These names would expect to have -r ending on the pers. n. since the names are in the nominative. Other pers. n. such as Rikarþ (E11) and Dotbert (E11) are not necessarily Scandinavian in origin. Though we would not necessarily expect all of these names to be grammatically Scandinavianised then, the fact that these names have not been so suggests that at the time of the inscription, the language of these inscriptions featured a lack of nominative inflexional endings for pers. n.

There are further features of non-standard grammatical forms outside of the pers.n. especially concerning the demonstrative pronouns. E3 has two examples rr.12-19 þesarunr and rr.21-29 þ*sastain. Rr.12-19 could be taken as þessar runr, with the r used for both the ending of the first word and the beginning of the next, and with r.19 being an irregular ending of the plural rún. Rr.21-29 þ*sastain is also irregular: we would expect the acc. m. sg. þenna stein- instead, we have þessa. Furthermore, E9 has acc. m. þïna agreeing with f. kirk, which lacks its inflexional ending as kirkju. Therefore, all inscriptions in the corpus except E6 have non-standard grammatical forms, whether concerning pers. n., demonstrative pronouns or inflexional endings of nouns. In situations of intimate and prolonged language contact, one or more of the languages in contact may show the loss of grammatical inflexional endings and low status languages are more likely to exhibit these symptoms of language contact (Baugh and Cable 1978: 103; Haugen 1976: 285). This could suggest, then, that the corpus language(s) is a language of lower social status within a wider speech community in the NW, where OE was also present, though the corpus is not large enough to come to that conclusion firmly.

The vocabulary of the inscriptions also varies between ON and OE. E3 has the verb form urait ‘scratched, carved’. The initial vr- is unlikely in Scandinavian in the twelfth century (Barnes and Page 2006: 83). There is an alternative to rïta: OE wrītan retains the initial bilabial into the Middle English period and therefore E3 could be an attempt at rendering an OE verb, showing lexical influence from Old English. Concerning E1, rr.26-30 are open to much interpretation too. Howard (1803) read rr.29-30 as rn, so the word ȝernr is taken as an adverb or adjective in apposition to the pers. n. Rikarð. An interpretation of this word as OE georne is possible, with ME ȝerne ‘careful’, making the reading Rikarð he made me and to this splendour carefully brought [me]. Barnes and Page (2006: 283) disagree, preferring ON gjarn ‘eager’, and though the borrowing of this word into English brings in issues of the date of the borrowing of ON gjarn (Haugen 1976: 155). The use of ON gjarn makes the plausible interpretation Rikarð he made me and to this splendour eagerly brought [me], rendering ȝer:** as ȝerna^r.

There are further vocabulary items in the corpus that are unclear. E9 rr.4-7 are problematic: r.4 has been read as an S s (Gaythorpe 1903: 378) which would make rr.4-5 a variant on ON sete/seti ‘established’. However, most readings, including for this essay, make r.4 an l l, making the word lïta. Additionally, Collingwood (1902-4) and Holman (1996) read a divider between r.6, reading it as lït:i. Holman (1996) claims r.r.1-3 takes rr.16-26 as the subject of the verb, making it Hubert mason had a monument made in this church. The area around the supposed divider is eroded and speculation cannot be taken further on this account. Nonetheless, reading l l and unable to see the word divider, lét with an irregular -a ending is not impossible to explain. When taking into account the rest of the grammatical forms in this inscription, such as kirk with a missing -ju ending it is highly likely that lïta is an attempt at rendering the verb lét, since -a could be another irregular ending in the inscription.

The final vocabulary item that is uncertain is E6. A word cannot be established with certainty from E6 hnirm. It is unlikely to be a word of Scandinavian vocabulary since initial hn- had been simplified to n- in all Scandinavian dialects except in Icelandic by the time of E6’s inscription; therefore, unless we date E6 to the thirteenth century, where hn- appears in the Icelandic dialect, the runes are not able to be formed into a lexical item. Therefore, the majority of the corpus shows vocabulary borrowed from languages other than the inscription’s language. This, along with the irregular grammatical forms, is to the extent that the corpus brings into question the extent of persistent language contact in the NW and the status of the language used for the runic inscriptions, as well as the possibility of Scandinavian influence outside of the NW.

Finally, due to the suggested influence of book-hand on graph forms, E1 will be explored. There is a problem with identifying some of the graphs as runic or Roman: r.6, r.24, and r.32 have rounded bows that mirror the forms found in book-hand (though r.19 is more pointed) and show the carver is influenced by knowledge of book-hand. Indeed, it is questionable whether r.25, for example, is Roman or runic. Nonetheless, the inscription is clearly mixed: graphs such as r.3, r.8 and r.10 are diagnostically runic, r.11 and r.26 are instead book-hand. This demonstrates a mixing of orthography of Roman and runic on E1; this is not a rare phenomenon, since other finds such as Lindisfarne 24, Lindisfarne 25, Bailey and Cramp 1988: 203) also have the two scripts together. E6 also has possible Roman graphs: the right-hand side of the slab has been damaged but there were incised graphs to the right-hand side of the runic inscription. There appear to be two verticals that slope towards a join, possibly ‘A’; it should be noted that this mixture of script is not necessarily the same as the deliberate mixing found on E1. Nonetheless, the choice to mix Roman with runic on E1, when there is clearly a knowledge of the two scripts within the community, is difficult to explain. The position of these runes as Christian commemoration of wealth also challenges the idea that runes were a lower-class, demotic form of writing. Indeed, E1 and E9 especially challenge this: it could even suggest an esteem associated with runic writing.

Conclusion.

The twelfth century runic corpus subsequently provides information about the rune-using community of the twelfth century NW: suggesting linguistic contact with English through grammatical evidence, whilst maintaining Norse linguistic and orthographic inclinations, and a possible cultural prestige attached to runes within the Christian community of the area. The corpus provides evidence of a rune-using community in the NW long after the initial bringing of the younger fuþark to Britain in the ninth and tenth centuries, and provides evidence for the longevity of the script in this area, up until the twelfth and possibly thirteenth centuries. Though the data set of runic inscriptions in the NW is too small to make sweeping conclusions about the language spoken in this area, these inscriptions are some of the only evidence for local speech at this time, and they show language that could be heavily English-influenced Scandinavian, retaining diagnostically Scandinavian items but showing vocabulary influence from English and variations on standard grammatical forms that suggest language contact. Furthermore, the corpus is evidence for the function of the runic inscriptions at this time: parts of the corpus indicates a cultural prestige for runic writing, with runes used to present inscriptions commemorating donations to the church in full public display. The apparent knowledge of book-hand and its mixture with runic graphs proposes an equal status of runic and book-hand in the case of E1, whilst the presence of non-linguistically comprehensible inscriptions, and inscriptions of pers. n., proposes a knowledge of graph forms outside of church commemoration. Therefore, this corpus provides a wealth of information about the NW in the twelfth century, suggesting the continuation of the Norse language and the use of runes in a variety of functions.

Bailey, R., and Cramp, R. (1988). Corpus of Anglo-Saxon stone sculpture: general introduction and volume II, Cumberland, Westmorland and Lancashire North of the Sands. Oxford: Oxford University Press.

Barnes, M., and Page, R. (2006). The Scandinavian Runic Inscriptions of Britain. Uppsala: Uppsala universitetet.

Baugh, A., and Cable, T. (1978). A History of the English Language. London: Taylor & Francis.

Collingwood, W. (1902-4). ‘Runic tympanum at Pennington, Furness’, Saga-Book of the Viking Club 3 139-141.

Ekwall, E. (1930). ‘How long did the Scandinavian language survive in England?’, Bøgholm, N., Brusendorff, A., and Bodelsen, C. [eds.], A grammatical miscellany offered to Otto Jespersen on his seventieth birthday. Copenhagen: Levin & Munksgaard. 17-30.

Fellows-Jensen, G. (1985). ‘Scandinavian settlement in Cumbria and Dumfriesshire: the place-name evidence’. In Baldwin, J and White, I [eds] The Scandinavians in Cumbria. Edinbrugh: Edinburgh University Press. 65-82.

Gaythorpe, H. (1903). ‘The runic tympanum lately found at Pennington’, TCWAAS 3 373-379.

Haugen, E. (1976). The Scandinavian languages: an introduction to their history. Harvard: Harvard University Press.

Higham. N. (1985) ‘’The Scandinavians in north Cumbria: raids and settlement in the later ninth to mid tenth centuries’. In Baldwin, J and White, I [eds] The Scandinavians in Cumbria. Edinbrugh: Edinburgh University Press. 37-51.

Holman, K. (1996). Scandinavian Runic Inscriptions in the British Isles: Their Historical Context. Trondheim: Tapir.

Howard, H. (1803) ‘Observations on Bridekirk font and on the runic column at Bewcastle, in Cumberland’, Archaeologia 14: 113-118.

Mawer, A., Stenton, F., and Armstrong, B. (1950). The place-names of Cumberland. Cambridge: Cambridge University Press.

Page, R. I. (1971). ‘How long did the Scandinavian language survive in England? The epigraphical evidence’, Clemoes, P, and Hughes, K [eds.], England before the Conquest: studies in primary sources presented to Dorothy Whitelock. Cambridge: Cambridge University Press 165-81.

Zilmer, K. (2013) ‘Christian Prayers and Invocations in Scandinavian Runic Inscriptions from the Viking Age and Middle Ages’, Futhark 4.

Online resources.

Digimaps <https://digimap.edina.ac.uk/> Accessed 10 May 2019.

PASE <http://pase.ac.uk/> Accessed 12 May 2019.