¹ See the list of manuscripts in Ker, N. R., Catalogue of Manuscripts Containing Anglo-Saxon(Oxford, 1957), at pp. xv–xix.Google Scholar

² Cf. Wrenn, C. L., ‘“Standard” Old English’, TPS (1933), 65–88;Google ScholarLuick, K., Historische Grammatik der englischen Sprache 1.1 (Leipzig, 1921)Google Scholar, §§ 21 and 25; Brunner, K., Altenglische Grammatik. Nach der Angelsächsischen Grammatik von Eduard Sievers, 3rd ed. (Tübingen, 1965)CrossRefGoogle Scholar, § 2, n. 2; and Campbell, A., Old English Grammar (Oxford, 1959)Google Scholar, §§ 16–17.

³ Sisam, K., Studies in the History of Old English Literature (Oxford, 1953), pp. 119–39.Google Scholar

⁴ Ibid. p. 125.

⁵ Ibid. p. 138.

⁶ Grammar, § 18.

⁷ Luick, Historische Grammatik 1.1, § 25 (‘some of the Anglian peculiarities seem to have had a downright poetic colouring’).

⁸ See esp. Schabram, H., Superbia. Studien zum altenglischen Wortschatz. Teil I: Die dialektale und zeitliche Verbreitung des Wortguts (Munich, 1965)Google Scholar, and Wenisch, F., ‘Sächsische Dialektwörter in The Battle of Maldon’, IF 81 (1976), 181–203;Google ScholarSpezifisch anglisches Wortgut in den nordhumbrischen lnterlinearglossierungen des Lukasevangeliums (Heidelberg, 1979)Google Scholar; and ‘Judith – eine westsächsische Dichtung?’, Anglia 100 (1982), 273–300.Google Scholar For their attitude towards Sisam's concept, see esp. Schabram, , Superbia, p. 129,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, p. 328.Google Scholar For the concept of a dialectally mixed poetic language, see studies of Homeric Greek, such as Meister, K., Die homerische Kunstsprache (Leipzig, 1921;CrossRefGoogle Scholar repr. Darmstadt, 1966), esp. ch. 11; Ruijgh, C. R., L'Élément achéén dans la langue épique (Assen, 1957)Google Scholar; Wathelet, P., Les Traits éoliens dans la langue de l'épopée grecque (Rome, 1970)Google Scholar; and Hiersche, R., Die Sprache Homers im Lichte neuerer Forschungen (Innsbruck, 1972)Google Scholar. For these bibliographical references, I am indebted to Professor Oswald Panagl, University of Salzburg; a convenient summary of recent research is provided in his review of Wathelet's book in Kratylos 19 (1974), at 74–84.Google Scholar

⁹ Luick, Historische Grammatik 1.1, §§ 133–59, esp. 143 and 153.

¹⁰ Ibid. § 146; Sievers–Brunner, Altenglische Grammatik, § 85, n. 1; and Campbell, Grammar, § 143. For the place-name evidence, see esp. Ekwall, E., Contributions to the History of Old English Dialects (Lund and Leipzig, 1917)Google Scholar, and Hallqvist, H., Studies in Old English Fractured ‘ea’ (Lund, 1948)Google Scholar. A detailed survey of studies on breaking and retraction is given by Crowley, J., ‘The Study of Old English Dialects’ (unpubl. Ph.D. dissertation, Univ. of North Carolina, Chapel Hill, 1980), pp. 232–51.Google Scholar As Luick and others have pointed out, a-spellings in early West Saxon and early Kentish texts can be best explained by assuming the influence of Anglian orthographic tradition. For discussion of an alternative view, see below, n. 24.

¹¹ Stanley, E. G., ‘Spellings of the Waldend Group’, Studies in Language, Literature, and Culture of the Middle Ages and Later, ed. Atwood, E. B. and Hill, A. A. (Austin, Tex., 1969), pp. 38–69.Google Scholar

¹² Ibid. pp. 48–66, esp. 65–6.

¹³ See Ker, , Catalogue, nos. 39 (A = Cambridge, Corpus Christi College 173) and 180Google Scholar (G = London, British Library, Cotton Otho B. xi); Parkes, M. B., ‘The Palaeography of the Parker Manuscript of the Chronicle, Laws and Sedulius, and Historiography at Winchester in the Late Ninth and Tenth Centuries’, ASE 5 (1976), 149–71;Google ScholarWhitelock, D., English Historical Documents c. 500–1042, Eng. Hist. Documents 1, 2nd ed. (London, 1979), 109–11 and 124;Google Scholar and Lutz, A., Die Version G der Angelsächsischen Chronik. Rekonstruktion und Edition (Munich, 1981),Google Scholar esp. chs. 1, 2 and 6.

¹⁴ A Microfiche Concordance to Old English, compiled by A. diPaolo Healey and R. L. Venezky (Toronto, 1980).Google Scholar

¹⁵ For the language of versions A and G, see esp. Cosijn, P. J., Altwestsächsische Grammatik, 2 vols. (The Hague, 1883–1886);CrossRefGoogle ScholarSprockel, C., The Language of the Parker Chronicle, 2 vols. (The Hague, 1965–1973)CrossRefGoogle Scholar; and Lutz, , Die Version G, chs. 5 and 7.Google Scholar

¹⁶ 709 G Aldhelm … Aldelm … Aldhelme (Lutz, , Die Version G, p. 23).Google Scholar

¹⁷ 731 G Ealdbelming, Ealdbelm (ibid. p. 26).

¹⁸ 734 G Beda, A Bieda (ibid. and Two of the Saxon Chronicles Parallel, ed. Plummer, Charles (Oxford, 1892–1899) 1, 44).Google Scholar

¹⁹ For versions B ( = London, British Library, Cotton Tiberius A. vi, written at Abingdon in 977/8) and C ( = London, British Library, Cotton Tiberius B. i, Abingdon, mid-eleventh-century), see Ker, Catalogue, nos. 188 and 191; Whitelock, English Historical Documents, pp. 111–13;Google Scholar and The Anglo-Saxon Chronicle: a Collaborative Edition. Vol. 4: MS B, ed. Taylor, S. (Cambridge, 1983), pp. xi–lxii.Google Scholar For the language of B and C, see Campbell, Grammar, § 16, and MS B, ed. Taylor, pp. lxiii–cvi. The remaining three Chronicle versions (D = London, British Library, Cotton Tiberius B. iv, (?)Worcester, mid- and later-eleventh-century; E = Oxford, Bodleian Library, Laud Misc. 636, Peterborough, early- and mid-twelfth-century; and F=London, British Library, Cotton Domitian viii, Canterbury, c. 1200; cf. Ker, Catalogue, nos. 192, 346 and 148, and Whitelock, , English Historical Documents, pp. 113–16)Google Scholar do not contain the poems in question, and versions E and F are, moreover, too late to be considered Standard Late West Saxon; version D displays ‘a considerable non-West Saxon element’ in its orthography and inflexion (Campbell, Grammar, § 17).

²⁰ Incidentally, versions B and C agree with G also in having Aldhelm referring to the bishop (but ea-spelling with reference to the Northumbrian king of the same name) and the Anglian spelling Beda referring to Bede; see above, p. 56.

²¹ Cf. Lutz, , Die Version G, pp. clxxxi–clxxxiii.Google Scholar For Stanley's own doubts about his hypothesis, see ‘Spellings of the Waldend Group’, p. 66.

²² I have somewhat modified the figures from the Concordance for the occurrences of w(e)aldend and gew(e)ald in the manuscripts of the Anglo-Saxon Chronicle (see below, next n.). The figures for w(e)aldan do not include occurrences of gew(e)aldan; for this, see below, n. 24.

²³ The seven extant Chronicle manuscripts show various degrees of textual interdependence, which makes it difficult to set down clear principles for concording. The decision to concord versions A, C, D and E (which can be considered as more or less independent texts for much of their later sections; see Whitelock, , English Historical Documents, pp. 109–17)Google Scholar for the Microfiche Concordance is basically right, though this leaves us with some inconsistencies regarding the earlier entries, where A and C together represent one version and D and E another. In contrast to the prose entries of the Chronicle the poems have been concorded from only one version. The figures in my table 4 include the occurrences of gew(e)ald from the prose entries of all seven Chronicle manuscripts and the occurrences of w(e)aldend from all manuscripts that contain any poem (for details on the forms in the Chronicle poems, see below, table 6).

²⁴ The contrasting spelling evidence from poetic texts – waldend, but geweald and wealdan – shows clearly that a-spelling in waldend cannot be attributed to phonological factors, e.g. to progressive assimilation of the vowel to labio-velar /w/, as has been proposed by Kurban, N., ‘The Evolution of the Written Standard for Late West Saxon, from the Ninth to the Twelfth Centuries’ (unpubl. Ph.D. dissertation, Indiana Univ., 1978), pp. 63–79 and 228–57.Google Scholar Kurban assumes retraction to /a/ for late West Saxon in accented syllables with a labial onset – unless this syllable is preceded by a syllable with a palatal vowel. This hypothesis might account for frequent a-spelling in waldend in contrast to practically consistent ea-spelling in geweald in late Old English manuscripts, but it would not account for equally consistent ea-spelling in wealdan (the figures in my table 4 do not include occurrences of gew(e)aldan) and in weald, ‘forest’ (see above, table 1), nor for the spelling distinction between waldend in poetry and wealdend in prose.

²⁵ The figures in table 5 (and in the table below, in the Appendix) are based on the manuscripts selected for the Microfiche Concordance; I have made no attempt to gather additional material from other manuscripts (except for the Chronicle ones; see above, n. 23). For the texts selected for table 5, cf. Schabram, , Superbia, pp. 62–6, 88–93, 108–9 and 111–13,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 20–5, 28–9, 31–2, 35 and 59–60.Google Scholar Wenisch's reservations with respect to the West Saxon origin of some of the texts need not concern us here, since it was obviously the date (and place) of the extant manuscript that determined the spelling. Thus ea-spelling was also used regularly in late West Saxon copies of earlier prose texts from various dialect areas, as can be demonstrated by the manuscripts of the Old English Psalter (see section a of the table below, in the Appendix; cf. Schabram, , Superbia, pp. 21–34,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 65–8)Google Scholar, and by the Alfredian texts (see section b of the table below, in the Appendix): those preserved in late West Saxon copies have consistent ea-spelling (Gregory's Dialogues and the Soliloquies), whereas those preserved in earlier manuscripts show a mixture of a-/ea-spellings (the Pastoral Care, the Boethius, the Orosius and the Chronicle (A version)). A similar mixture of spellings can be observed in the Blickling and Vercelli Homilies (see section c of the table below, in the Appendix), both assumed to be originally Anglian texts that were inconsistently transposed into late West Saxon (cf. Schabram, , Superbia, pp. 73–87,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 30, 72–8 and 327;Google Scholar for similar spelling evidence in the poetic texts of the Vercelli Book and in the Metres of Boethius, see below, table 6). With reference to sections b and c of the table below, in the Appendix, it should be pointed out that a-spelling occurs more frequently in the agent noun than in the abstract noun and the verb. The evidence of the works of Wulfstan (see section d of the table below, in the Appendix), including the laws attributed to him (see Sermo Lupi ad Anglos, ed. Whitelock, D., 3rd ed., rev. (Exeter, 1976), pp. 23–6)Google Scholar, is difficult to explain: the three a-spellings in waldend might be attributed to the late date of the manuscripts in which they occur (cf. occasional a-spellings in late versions of the Psalter and in late Chronicle entries), but it seems remarkable that in Wulfstan's works a-spelling is recorded only for the agent noun, which is, moreover, used in an alliterative formula (3 × ‘waldend and wyrhta’ – but also 4 × ‘wealdend and wyrhta’).

²⁶ With the exception of Genesis B; see Schabram, , Superbia, pp. 124–9,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 85–6 and 328.Google Scholar

²⁷ Cf. Sievers–Brunner, Altenglische Grammatik, § 2, n. 6; Campbell, Grammar, § 17; Schabram, , Superbia, pp. 42–8;Google ScholarWenisch, , Spezifisch anglisches Wortgut, pp. 41, 46–7 and 327Google Scholar – all with reference to the prose texts. It is, however, very doubtful whether the two poems are contemporary with the prose texts; cf. The Anglo-Saxon Minor Poems, ed. Dobbie, E. V. K., The Anglo-Saxon Poetic Records 6 (New York, 1942), pp. cxv–cxviii.Google Scholar

²⁸ See The Battle of Maldon, ed. Scragg, D. G. (Manchester, 1981), pp. 23–8;Google ScholarGneuss, H., Die ‘Battle of Maldon’ als historisches und literarisches Zeugnis, Bayerische Akademie der Wissenschaften, phil.–hist. Klasse, Sitzungsberichte (Munich, 1976), pp. 55–8;Google Scholar and Wenisch, F., ‘Sächsische Dialektwörter in The Battle of Maldon’, IF 81 (1976), 181–203.Google Scholar

²⁹ See Ker, , Catalogue, no. 216;Google ScholarSchabram, , Superbia, pp. 124–5,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 82–3 and 328,Google Scholar and ‘Judith – eine westsächsische Dichtung?’. Cf. also The Dating of ‘Beowulf’, ed. Chase, Colin (Toronto, 1981).Google Scholar

³⁰ See Ker, , Catalogue, p. 408;Google Scholar it has already been pointed out by Stanley that ‘Book I’ and ‘Book II’ of the Junius manuscript belong to two different spelling traditions (‘Spellings of the Waldend Group’, p. 58).

³¹ See Ker, , Catalogue, no. 167.Google Scholar Both the Metres and the prose version – originally Alfredian West Saxon compositions – preserve many early West Saxon features; see Schabram, , Superbia, p. 127;Google ScholarWenisch, , Spezifisch anglisches Wortgut, pp. 31, 87 and 328;Google Scholar and above, n. 25.

³² For the manuscript, see Ker, , Catalogue, no. 394.Google Scholar The poems are believed to be originally Anglian compositions, and so are most of the prose texts contained in the manuscript; see Schabram, , Superbia, pp. 77–87,Google Scholar and Wenisch, , Spezifisch anglisches Wortgut, pp. 72–8 and 327–8.Google Scholar For the spelling of the Vercelli Homilies, see above, n. 25.

³³ Only those manuscripts and texts have been selected for table 6 that contain several occurrences of one of the words of the waldend group – or instances of at least two of the three words considered. Taken together, these manuscripts and texts contain nearly all poetic occurrences of the three words: waldend 205 (total 212), wealdend 71 (82), gewald 6 (6), geweald 72 (76), waldan 9 (10) and wealdan 56 (62); see above, table 4.

³⁴ At first sight (e)aldor, ‘ruler’, would seem to belong to this group of words too, but the spelling evidence from the poetic texts is contradictory. Some of the scribes who consistently wrote waldend used also the ‘poetic spelling’ aldor (e.g. the first scribe of the Junius manuscript 31 ×), but other scribes preferred regular late West Saxon ealdor (Exeter Book 10 × and Battle of Maldon 5 ×). The Chronicle evidence is divided (B, C1 × aldor, A, G1 × ealdor) and the Beowulf manuscript has both 6 × aldor and 8 × ealdor. Apparently the scribes made no difference in spelling between (e)aldor m., ‘ruler’, and (e)aldor n. (?f.), ‘life’ (see above, table 1). This divided spelling evidence for (e)aldor, ‘ruler’, in late West Saxon poetic texts and manuscripts could perhaps be explained by the wide range of meaning of the word. In contexts such as ‘lifes (e)aldor’, ‘sigores (e)aldor,’ ‘wuldres (e)aldor’ and ‘þioda ealdor’ it seems to have been closely related to w(e)aldend both in meaning and in poetic potential. (Cf. prose contexts such as ‘se deofol bið ealdor and wealdend’) (Oxford, Bodleian Library, Junius 121, line 2897). In contexts such as ‘se biscop and se ealdor’, ‘mynstres ealdor’ (both very common in monastic rules), ‘þære ciricean ealdor’ and ‘hundredes ealdor’, however, it had a number of rather technical meanings that must have been familiar to scribes living in a monastery. Cf. its use in glosses for ducatus, principalis, tribunus, centurio etc.

³⁵ I should like to thank Professor Hans Schabram (Göttingen) and Dr Roland Torkar (Würzburg) for their valuable comments on an earlier version of this paper.