^{page 1 note 1} Oxford, 1970. I employ the following abbreviations in this article: BCS = Cartularium Saxonicum, ed. Birch, W. de G., 4 vols. (London, 1885–1899)Google Scholar; Councils = Councils and Ecclesiastical Documents relating to Great Britain and Ireland, ed. Haddan, A. W. and Stubbs, W., 3 vols. (Oxford, 1869–1878)Google Scholar; Dümmler = Alcuini Epistolae, ed. Dümmler, Ernst, Monumenta Germaniae Historica, Epist.4 (Berlin, 1895)Google Scholar; EHD 1 = Whitelock, Dorothy, English Historical Documents 1: c. 500–1042 (London, 1955)Google Scholar; EMK = Early Medieval Kingship, ed. Sawyer, P. H. and Wood, I. N. (Leeds, 1977)Google Scholar; Jaffé = Jaffé, Philipp, Bibliotheca Rerum Germanicarum VI (Monumenta Alcuiniana), ed. Wattenbach, W. and Dümmler, E. (Berlin, 1873)Google Scholar; KCD = Codex Diplomaticus Aevi Saxonici, ed. Kemble, J. M., 6 vols. (London, 1839–1848)Google Scholar; Sawyer = Sawyer, P. H., Anglo-Saxon Charters: an Annotated List and Bibliography (London, 1968)Google Scholar. I am indebted to Thomas Charles-Edwards for some very stimulating criticisms of an early draft of this paper; I am very grateful also to John Bannerman, Peter Hunter Blair, Daniel Huws, Michael Lapidge and Peter Sawyer, who have all read drafts and given help and encouragement. At a late stage Professor Dorothy Whitelock commented comprehensively and gave much kind help: her learning saved me from a number of errors; we have not, however, been able to agree on all problems of interpretation of the evidence surveyed here.

^{page 1 note 2} Kingship, p. 1.

^{page 1 ote 3} ibid. pp. 24 and 25.

^{page 1 note 4} ibid. p. 28.

^{page 1 note 5} ‘Æðeling: Edling. Succession to the Throne in Early Mediaeval Wales’, forthcoming.

^{page 1 note 6} Asser's Life of King Alfred, ed. Stevenson, W. H. (Oxford, 1904; rev., 1959)Google Scholar, §§29.3, 38.8 and 42.2.

^{page 2 note 1} ‘Some Celtic Legal Terms’, Celtica 3 (1956), 224.Google Scholar

^{page 2 note 2} Kingship, p. 29.

^{page 2 note 3} ibid. pp. 29–30.

^{page 2 note 4} The eldest son, Athelstan, had apparently predeceased his father; he was dead by 855 at the very latest.

^{page 2 note 5} In the sense intended by Binchy, , Kingsbip, pp. 25–26Google Scholar, rather than by Niocaill, G. Mac, ‘The “Heir Designate” in Early Medieval Ireland’, Irish Jurist, n.s. 3 (1968), 326–9Google Scholar. Any attempt to invoke the following ‘English’ evidence to decide the meaning of rigdomnae must fail: in the Irish ‘Annals of Tigernach’ and Chronicum Scottorum for 629, the death (in a battle in Ireland between the Dál Riata and the Ulster Cruthin) is recorded of one ‘Oisiricc mac Albruit’, described as rígdomna Saxan: see Stokes, Whitley, ‘The Annals of Tigernach, Third Fragment, ad 489–766’, Revue Celtique 17 (1896), 180–1Google Scholar, and Chronicum Scotorum, ed. and trans. Hennessy, William M. (London, 1866), pp. 80–3Google Scholar (which reads ‘Osiricc mac Albirit‘). Byrne, F. J., Irish Kings and High-Kings (London, 1973), p. 259Google Scholar, says he was ‘probably one of the Bernician princes in exile at the Dál Riatan court’; cf. pp. 35 and 312 on rígdomnae. It is plain that rígdomna in this annal renders OE æþeling; but, since the entry contains an error of identification (a not uncommon feature of items unique to the Clonmacnois group of annals), it is not clear who this ætheling was and therefore in what relationship he stood to King Edwin (616–32/3). Osric son of Ælfric was a first cousin of Edwin and ruled as king of Deira for some months in 635/4 after Edwin's death (Bede, , HE 111.1Google Scholar); he cannot therefore be the victim of 629, as Bannerman, John, Studies in the History of Dalriada (Edinburgh, 1974), pp. 98–9Google Scholar, points out. The slain prince was probably one of Æthelfrith's sons: Bede (HE 111.1) tells us that they were in exile among the Irish (and Picts) during Edwin's reign. But in view of the difficulties in this annal the equation æbeling =rígdomnae is not helpful in defining either the English word or the Irish.

^{page 3 note 1} The rôle of Queen Sexburg, who ruled Wessex in 672/3 or 673/4, is too shadowy to be used as evidence; as Wallace-Hadrill, J. M., Early Germanic Kingship in England and on the Continent (Oxford, 1971), pp. 92–3Google Scholar, points out, early Anglo-Saxon England lacked the powerful queen-regents seen in continental Germanic kingdoms. Similarly, later Anglo-Saxon England lacked the female regent of the type seen in the Ottoman and Salian empire. On the rôle of the widowed queen in relation to the succession in early Germanic society, see Grierson, P., ‘Election and Inheritance in Early Germanic Kingship’, Cambridge Hist. Jnl 7 (1941–1943), 9Google Scholar; cf. Chadwick, H. Munro, Studies on Anglo-Saxon Institutions (Cambridge, 1905), p. 365.Google Scholar

^{page 3 note 2} Toller, T. Northcote, An Anglo-Saxon Dictionary. Supplement (Oxford, 1921), p. 22.Google Scholar

^{page 3 note 3} For example, Nithard, , Historiarum Libri Quattuor (iv.2), ed. Ernst, Müller, 3rd ed. (Hanover, 1907), p. 41Google Scholar, speaking of the continental Saxons, who writes ‘edhilingui … latina uero lingua … nobiles’. The same conclusion may be drawn from the Lex Thuringorum (Lex Angliorum et Uuerinorum) concerning the adaling (us): Leges Saxonum et Lex Thuringorum, ed. von Schwerin, C. F. (Hanover, 1918).Google Scholar

^{page 3 note 4} The Old English Version of Bede's Ecclesiastical History of the English People, ed. and trans. Miller, Thomas (London, 1890–1898), 1, 306Google Scholar, lines 5–6; cf. 1,130 (HE 11.12), where Edwin before he becomes king is ‘se geonga æþeling’ (regius iuuenis), and 1, 140 (HE 11.14), ‘monige æðelingas þæs cynecynnes’, translating regii uiri non pauci. At 1, 220 (HE 111.21), Peada, son of King Penda, is ‘ging æðeling good‘ (iuuenis optimus). In the plural, œþlingas generally renders nobiles (cf. continental usage, noted above, p. 3); e.g., duo iuuenes magnae indolis de nobilibus Anglorum (HE 111.27) is rendered as ‘twegen geonge æðelingas’ (1, 242).

^{page 4 note 1} King Alfred's Orosius, ed. Henry Sweet 1 (London, 1883), 44, line 24. Chadwick, Institutions, pp. 301, n. 2, and pp. 416–17, devoted a little attention to this terminology.

^{page 4 note 2} Compare the account of the Irish situation by Corráin, D. Ó, ‘Irish Regnal Succession: a Reappraisal’, Studia Hibernica 11 (1971), 7–39.Google Scholar

^{page 4 note 3} ASC 904 A, 905 CD: The Anglo-Saxon Chronicle: a Revised Translation, ed. Whitelock, D. et al. (London, 1961; rev., 1965), p. 60Google Scholar, n. 4; cf. below, p. 12, n. 1. But at the time of Edward the Elder's accession (899) members of Mercian royal lines were no doubt still accorded their proper titles.

^{page 4 note 4} But cf. below, p. 30: it is conceivable that the annalist had a polemical intention in using the term here. This is hardly a necessary deduction, however, especially in the light of the next item.

^{page 5 note 1} In London, British Library, Cotton Tiberius B. v, vol. 1, 23r. I have briefly discussed this item, “The Anglian Collection of Royal Genealogies and Regnal Lists’, ASE 5 (1976), 43.Google Scholar

^{page 5 note 2} In the Anglo-Saxon Chronicle, s.a. 617 E, which is an item belonging to the northern version DE (perhaps compiled at York between c. 970 and c. 1030), but now occurring only in the midtwelfth-century E (D having a lacuna here), we read of the seven eðelingas, the sons of King Æthelfrith of Northumbria. Also in the northern version (represented here by E and F), s.a. 443 (entered also by the Canterbury scribe of F as an addition in A), we read of Angel cynnes œðelingas who are invited by the Britons to help them against the Picts; the implication is that the English did not then have kings.

^{page 5 note 3} And in the charter (Sawyer, no. 1422) mentioned below, n. 6.

^{page 5 note 4} One can hardly imagine that he was Cnut's heir-apparent, and it seems unlikely that the chronicler is here taking a strong partisan line in favour of Eadwig's right to the English throne.

^{page 5 note 5} Sawyer, no. 1454; ptd as KCD, no. 693, and, with translation, in Adams, Henryet al., Essays in Anglo-Saxon Law (Boston, Mass., 1876), pp. 355–6Google Scholar. The Chronicle, s.a. 1002, notes the death of Æfic þæs cynges heahgerefan, who might be the same person, perhaps having advanced since 992 to a more exalted position.

^{page 5 note 6} Sawyer, no. 1422; KCD, no. 1302. Note that Leofwine is named here as discþegn of just this one ætheling: æþelinges discpen. Blair, Peter Hunter, An Introduction to Anglo-Saxon England (Cambridge, 1956), p. 214Google Scholar, misdates it to ‘c. 1030’; I have no evidence as to the situation in Cnut's reign.

^{page 5 note 7} The charter is re-edited and translated by Robertson, A. J., Anglo-Saxon Charters (Cambridge, 1939), pp. 146–9 and 392–4Google Scholar (no. LXXIV). The conjecture that the priests Eadwine and Wulfric among the witnesses were ‘perhaps chaplains to Edmund’ is hers (p. 394).

^{page 5 note 8} Sawyer, no. 937; KCD, no. 1312; EHD 1, 537–9 (no.123).

^{page 6 note 1} Noted by Whitelock, D. in Asser, ed. Stevenson, p. cxxxvGoogle Scholar. Hart, C. R., The Early Charters of Northern England and the North Midlands (Leicester, 1975), p. 34Google Scholar, conjectures that the land constituting the royal fisc was ‘shared between the king, his consort and the æthelings’.

^{page 6 note 2} In the legendary account of St Mildred of Thanet in the mid-eleventh-century London, British Library, Cotton Caligula A. xiv we find children as æthelings: ‘Wæron þa æþelingas befæste Egcbrihte cynge … wæs se cyng heora fæderan sunu Eorcenbrihtes’ (Leechdoms, Wortcunning and Starcraft of Early England, ed. and trans. Cockayne, Oswald, 3 vols. (London, 1864–1866) 111, 424Google Scholar, line 11); for the documents see Blair, P. Hunter, ‘Some Observations on the Historia Regum Attributed to Symeon of Durham’, Celt and Saxon, ed. Chadwick, N. K. (Cambridge, 1963; rev. 1964), PP. 78–82.Google Scholar

^{page 6 note 3} Their joint claim is, in effect, stated by the genealogy in the Anglo-Saxon Chronicle, s.a. 1067 D, which proudly recites the ancestry of Queen Margaret of Scotland. She was sister of Edgar and daughter of Edward æþeling, the title accorded him s.a. 1057 DE and 1067 D.

^{page 6 note 4} S.a. 1066 E, 1068 DE and 1069 DE he is œþeling; 1066 D has cild; 1067 DE and 1074 DE have both; after D ends in 1079, E calls him æþeling several times till 1106.

^{page 6 note 5} The word mæg seems to have no more specific connotation than that of ‘kinsman’.

^{page 6 note 6} I have discussed the Middle English evidence, which is important in connection with the borrowing of æþeling into Middle Welsh, in my paper cited above, p. 1, n. 5.

^{page 7 note 1} Very occasionally, in the well-known group of alliterative charters (on which see EHD 1, 340), we find the vernacular word æþeling used: Sawyer, nos. 566 (an Old English translation of an alliterative charter of Eadred, 955), 569 (Eadred, 955) and 1497 (a will which quotes from an alliterative charter). The word is used in these of Eadwig and Edgar. Another alliterative charter (Sawyer, no. 633: Eadwig, 956) has a unique use of regulus to describe Edgar: the language of charters of this group is often flamboyant and therefore it would be unwise to lay any stress on this unique description of Edgar's position in 956.

^{page 7 note 2} See Lapidge, M., ‘The Hermeneutic Style in Tenth-Century Anglo-Latin Literature’, ASE 4 (1975). 67–111.Google Scholar

^{page 7 note 3} The Chronicle of Æthelweard, ed. Campbell, A. (London, 1962).Google Scholar

^{page 7 note 4} There are eight occurrences, giving the inflections clito, -nem and -ne.

^{page 7 note 5} For a convenient collection of Anglo-Latin glossaries, see Anglo-Saxon and Old English Vocabularies, ed. Wright, Thomas and Wülcker, R. P., 2nd ed., 2 vols. (London, 1884Google Scholar). Ælfric's equation is found at 1, col. 155, line 21, col. 309, line 28, and col. 538, line 22 (from three different manuscripts); see also Ælfrics Grammatik and Glossar, 1: Texte und Varianten, ed. Zupitza, Julius (Berlin, 1880), 300Google Scholar (‘clito æþeling’). The form clyton (see below) appears in the mid-tenth-century English glossarial manuscript BL Cotton Cleopatra A. iii (Wright and Wülcker 1, col. 379, line 25), of which there is a later copy (now badly damaged) in BL Cotton Otho E. i (s. x/xi).

^{page 7 note 6} Sawyer, no. 912 (KCD, no. 672), of 1005.

^{page 7 note 7} The subscription of large numbers of clitones is commonplace in Æthelred's charters; see below, p. 8, n. 7. For another plural, see the Anglo-Latin Life of St Neot (in Oxford, Bodleian Library, Bodley 535, of s. xi²), which makes the equation æthelinga-ig =clitonum insula.

^{page 7 note 8} With the possible exception of the poem discussed by Stevenson, W. H., ‘A Latin Poem Addressed to King Athelstan’, EHR 26 (1911), 482–7CrossRefGoogle Scholar, which he assigns to 926 × 934; see below, p. 9.

^{page 7 note 9} Sawyer, no. 1417; there is a facsimile in Facsimiles of Anglo-Saxon Manuscripts, ed. Sanders, W. Basevi, 3 vols. (Southampton, 1878–1884) 11Google Scholar, Winchester College no. 2.

^{page 8 note 1} Sawyer, nos. 511 (Edmund, 941), 565 (Eadred, 955) and 571 (Eadred, 956, recte 953 × 955). These are BCS nos. 765, 905 and 931 respectively. They are nos. 85, 57 and 87 respectively in the catalogue by Hart, C., ‘The Codex Wintoniensis and the King's Haligdom’, Agricultural Hist. Rev. 18 (1970), Supplement, pp. 7–38Google Scholar, who declares them all authentic.

^{page 8 note 2} Sawyer, no. 569.

^{page 8 note 3} Sawyer, no. 570.

^{page 8 note 4} Sawyer, no. 928. See Hart, , Charters of Northern England, pp. 244–5.Google Scholar

^{page 8 note 5} Sawyer, nos. 386–9 and 433. These charters use the phrase indolis clito, on which see below.

^{page 8 note 6} I count eleven charters allegedly of 956 which use this word: Sawyer, nos. 589, 591, 593, 594, 597, 605, 608, 616, 630, 637 and 666, although all but 597, 605 and 666 combine it with indoles; see below. Of these 616 is said to be spurious; none survives in an assuredly contemporary copy (although 594 may perhaps have that status). The form clito is found in an Ely charter (Sawyer, no. 572) attributed to Eadred and dated 956; Professor Whitelock tells me that this alliterative charter has been tampered with. Nor is the conclusion as to the date of this usage upset by Sawyer, no. 146 (BCS, nos. 272 and 273), a grant of Offa of Mercia to Worcester (793 × 796), witnessed by his son Ecgfrith. The charter survives in two Worcester cartularies: in BL Cotton Nero E. i, vol. 11, perhaps c. 1050 (BCS, no. 272), we read ‘Ego Ecgfriðus clito dono patris mei consensi’, but in BL Cotton Tiberius A. xiii, c. 1000 × 1050 (BCS, no. 273), ‘Ego Ecgferð e<i>usdem regis filius …’ The conclusion to be drawn is that the compiler of the Nero cartulary remodelled the charter in the idiom of his own day when the use of clito was normal: clito is an anachronism in an eighth-century context.

^{page 8 note 7} I note it in four charters attributed to Edgar: Sawyer, nos. 739, 745 and 746 (all of 966) and 783 (of 971); the last two are certainly spurious. It occurs also in twelve charters of Æthelred, extending in date from 984 to 1015: Sawyer, nos. 854, 893, 910, 911, 912, 921, 923, 925, 927, 931, 933 and 934; forms in other cases (and therefore with -n-) occur in Sawyer, nos. 916,917 and 929. Of all these 916 alone survives in a contemporary copy.

^{page 8 note 8} An apparent exception is Sawyer, no. 936, dated 1028, attested by Hardecnut clyto, but this pseudo-Æthelredian charter is spurious, whatever its genuine base may be, and the form of attestation no doubt derives from a diploma of Æthelred's reign.

^{page 9 note 1} Sawyer, nos. 589, 591, 593, 594, 608, 616, 630 and 637 (indolis clito). For these charters, see above, p. 8, n. 6.

^{page 9 note 2} Sawyer, nos. 614 (indolens, presumably by scribal error), 623, 629 and 661. The last has been denounced as a forgery.

^{page 9 note 3} This judgement is not, I think, disturbed by the following items where the word indoles/indolis is used. Sawyer, no. 576, a spurious charter ascribed to Eadred (as king of Mercia!) and dated 958, is confirmed by Eadwig and Edgar, indoles. A group of five spurious Exeter charters (Sawyer, nos. 386–9 and 453) ascribed to Athelstan are witnessed by Edmund, indolis clito. Sawyer, no. 934, an Æthelredian charter of 1015, is attested by Edmund ‘regiae indolis suboles’ and ‘Eaduuærd clito’. From the pre-Viking period the hotly disputed Sawyer, no. 90 (Æthelbald, 742, preserved in an early-ninth-century single sheet) is attested by ‘Æðelmod indolis Merciæ’, while Sawyer, no. 55 (a disputed Hwiccian charter of 757, surviving only in eleventh-century Worcester cartulary-copies) bears the extraordinary subscription ‘Ego Offa, nondum regno Mercionum a Domino accepto, puer indolis in prouincia Huicciorum constitutus, huic donationi eorum consensi …’

^{page 9 note 4} It was printed and discussed by W. H. Stevenson, ‘A Latin Poem Addressed to King Athelstan’, pp. 482–7. It has now been re-edited by Michael Lapidge; see his ‘Some Latin Poems as Evidence for the Reign of Athelstan’, ASE 9 (1980)Google Scholar, forthcoming.

^{page 9 note 5} The corrupt manuscript-reading is clitanam: clitonem should probably be restored. We do not know, of course, what the author's nominative singular usage would have been.

^{page 9 note 6} Frithegodi Monachi Breuiloquium Vitæ Beati Wilfredi et Wulfstani Cantoris Narratio Metrica de Sancto Swithuno, ed. Campbell, Alistair (Zürich, 1950), p. 30Google Scholar. I am indebted to Michael Lapidge, who is re-editing Frithegod, for the following translation of these lines: ‘Afterwards, the people bewailed the prophesied murder of Ælfuini, / giving an appropriate funeral to the clito(n) (who was) slaughtered as (he) deserved, / after whose burial the victory of the brother [ = Wilfrid] flowed forth.’

^{page 9 note 7} The Life of Bishop Wilfrid by Eddius Stephantis, ed. and trans. Colgrave, Bertram (Cambridge, 1927), pp. 36 and 50Google Scholar. Cf. Bede, HE IV.22 (20).

^{page 10 note 1} Note, however, Bede, HE iv.21 (19), ‘Aelfuini frater regis Ecgfridi, iuuenis’, rendered by the Old English translator (ed. Miller 1, 324) as ‘Ælfwine Ecgfriðes broðor þæs cyninges, geong æðeling eahtatynewintre’. Frithegod might have known the vernacular version, but more probably deduced clito(n) from his reading of Bede's own words, quoted here.

^{page 10 note 2} Stevenson, , ed., Asser, p. 299Google Scholar, draws attention to an entry (s.a. 922) in which John refers to Æthelweard, the younger son of King Alfred, as clito. He is called that also in the New Minster Liber Vitae of the first half of the eleventh century: see Liber Vitae: Register and Martyrologf of New Minster and Hyde Abbey, Winchester, ed. Birch, W. de G. (London and Winchester, 1892), p. 6Google Scholar; he appears here erroneously (as it would seem) as Edward's son (Stevenson, , ed., Asser, p. 299Google Scholar, n. 4).

^{page 10 note 3} For a summary of William's life, see Poole, A. L., From Domesday Book to Magna Carta, 2nd ed. (Oxford, 1955), pp. 121–8Google Scholar. For some important remarks about him, see Davis, R. H. C., King Stephen 1135–1154 (London, 1967), pp. 6–7.Google Scholar

^{page 11 note 1} Sawyer, nos. 165 and 177, of 811 and 814 respectively. Both survive in contemporary copies.

^{page 11 note 2} Pp. 14–18.

^{page 11 note 3} Cf. the usage among the sixth-century Franks, as reported by Gregory of Tours, Historia Francorum v.20; see Wood, Ian, EMK, p. 14Google Scholar. Can the West Saxon usage be tied in any way to Asser's report (§13) that the wives of West Saxon kings were not called ‘queen’ in the ninth century?

^{page 11 note 4} Sawyer, nos. 327, 329 and 331–3. Sawyer, no. 335 may be another example, although the charter, dated 862, is ascribed to Æthelred; its authenticity is uncertain; and in one copy Alfred appears as ‘filius regis’, in another as ‘frater regis’.

^{page 11 note 5} Sawyer, nos. 340 and 1201.

^{page 11 note 6} Sawyer, no. 1203, surviving in a contemporary copy.

^{page 11 note 7} Sawyer, no. 356.

^{page 11 note 8} Sawyer, nos. 359, 360, 362, 365, 366, 368, 370, 373 and 374. In 364 Æthelweard appears as ‘fr. R.’. Note also the specification mater regis and coniunx regis in Sawyer, no. 363. Many of these charters are dubious or spurious, however.

^{page 11 note 9} A list of Nomina filiorum regum, confined (with two exceptions) to kings' sons who did not themselves become kings, and starting c. 900, may be found in the New Minster Liber Vitae (ed. Birch, p. 14).

^{page 11 note 10} It is very tempting to associate this diplomatic development with the arguments flowing from the interpretation of Æthelwulf's will. See below, pp. 21–4.

^{page 12 note 1} He is very likely the Beornnoth dux of Burgred's charters. We may note that the northern annalist thought the death of his son Brihtsige so important that he made it the sole entry for 902; see the Historia Regum, Symtonis Monachi Opera Omnia, ed. Arnold, Thomas, 2 vols. (London, 1885) 11, 92.Google Scholar

^{page 12 note 2} Sawyer, nos. 854, 922 and 929. 854 is dated 984 but has a witness-list of 1005.

^{page 12 note 3} This consideration seems to me to be especially appropriate in the Chronicle entry s.a. 982 C.

^{page 12 note 4} In Ireland in the eleventh century attempts to define the rígdamnai (those eligible for succession) within the royal kindred were ‘accompanied by a noticeable narrowing of the circle of eligible candidates to the kingly succession’ (Byrne, F. J., ‘Tribes and Tribalism in Early Ireland’, Ériu 22 (1971), 156).Google Scholar

^{page 13 note 1} Sawyer, no. 811 (BCS, no. 1519), of 959 × 963.

^{page 13 note 2} Leechdoms, ed. Cockayne 111, 438, line 5.

^{page 13 note 3} The curious use of æþeling in the northern version of the Anglo–Saxon Chronicle, s.a. 972 DEF, requires a more extended exegesis. We read that ‘Eadgar eþeling was consecrated king … in the thirteenth year after he succeeded to the kingdom.’ Unless we take this to be a general, poetic usage (see above, p. 3), we must view it as a highly technical (or pedantic) expression – Edgar does not cease to be æbeling and become cyning until he has been consecrated, and this he chose to put off until his thirtieth year. But Janet Nelson (EMK, pp. 63–70) has given reasons for rejecting this theory of a delayed consecration, seeing the events of 97 3 as part of an imperial design; she is therefore unwilling (ibid. p. 65, n. 90) to assign any great importance to the word eþeling, viewing this usage as ‘a pedantic reflex’ on the part of the annalist. One could argue (as Professor Whitelock suggests to me) that the northern chronicler may simply have been influenced by a common formula, namely the description of the successor to the throne as an ætheling, even though in this case it was inappropriate at a consecration many years after the accession of the king. However, another and more technical context may offer promising possibilities. For what follows, cf. Nelson, J., EMK, p. 55Google Scholar; ‘Kingship, Law and Liturgy in the Political Thought of Hincmar of Rheims’, EHR 92 (1977), 258Google Scholar, n. 1; and ‘Ritual and Reality in the Early Medieval Ordines’, Stud, in Church Hist. 11 (1975), 51Google Scholar. Wolfram, Herwig, Intitulatio (Graz, 1967–1973) 1, 145 ff.Google Scholar, has drawn attention to the use of princeps in eighth-century titles to designate one who exercised royal potestas without the royal nomen. With this may be associated two occurrences of a liturgical usage found in royal inauguration rituals. In the ‘frühdeutsch Ordo’ the ruler is termed princeps (or electus) until the anointing and only thereafter is called rex; see Erdmann, C., Forschungen zur politischen Ideenwelt des Frübmittelalters (Berlin, 1951), pp. 83–7Google Scholar, and Ullmann, W., Festschrift Percy Ernst Schramm zu seinem siebzigsten Geburtstag, ed. Peter, Classen and Scheibert, P., 2 vols. (Wiesbaden, 1964) 1, 77–8Google Scholar, n. 24. In the coronation prayer used at Edgar's crowning in 973 the wording had been changed from the source's presul (for ‘king’) to ‘hunc praeelectum principem’. It is particularly useful in the present context to have such a usage attested for 973, for we can read the annal's eþeling precisely in the sense of princeps.

^{page 14 note 1} Kingship, p. 29.

^{page 14 note 2} Ibid; cf. Bosworth, Joseph and Toller, T. N., An Anglo-Saxon Dictionary (Oxford, 1882–1898), p. 22.Google Scholar

^{page 14 note 3} Toller, , Supplement, p. 22.Google Scholar

^{page 14 note 4} Stenton, F. M., Preparatory to ‘Anglo-Saxon England’, ed. Stenton, D. M. (Oxford, 1970), p. 119Google Scholar (cf. pp. 106–8); see also Chadwick, H. Munro, The Origin of the English Nation (Cambridge, 1907), pp. 25–7.Google Scholar

^{page 14 note 5} Who is described on five other occasions in the annal as ‘the ætheling’.

^{page 15 note 1} Although the possibility is not to be overlooked that this is dynastic propaganda of the ninth century. Loyn, H. R., Anglo-Saxon England and the Norman Conquest (London, 1962), p. 204Google Scholar, plainly suspects this. Cf. Chadwick, , Institutions, p. 287Google Scholar, n. 1.

^{page 15 note 2} This rests on the testimony only of DE (the northern version of the Chronicle), where the chronicler is otherwise fond of adding ‘his kinsman’, referring back to the preceding king, when he knows nothing of the ancestry of a new ruler.

^{page 15 note 3} Although, if Æthelweard's version has any evidential value, the reverse was the case: ‘Æthelheard … primoque anno regni sui instituit bellum aduersus clitonem Osuueo’ (11.13). The Latin text of ASC (F), s.a. 730 and 731, calls Oswald regulus; see Magoun, Francis P. Jr, ‘Annales Domitiani Latini: an Edition’, MS 9 (1947), 252.Google Scholar

^{page 16 note 1} Anglo-Saxon Chronicle, pp. 205 ff.

^{page 16 note 2} Although we may have a record of the pedigree of one claimant to the throne c. 800; see my “The Anglian Collection of Royal Genealogies’, pp. 48–9.

^{page 16 note 3} His paternity is uncertain; it rests on two twelfth-century sources.

^{page 16 note 4} There seems no reason to assume that this is at least one generation too short: over seven generations the average in both Alhred's and Cenred's lines is quite credible. However, at least one Northumbrian writer seems to have had his doubts about Alhred's claim: in the Historia Regum (ed. Arnold 11, 43), s.a. 765, we read, ‘Cui Alcred, prosapia Ide regis exortus ut quidam dicunt, successit in regnum.’

^{page 16 note 5} I say ‘overtly’, for there is no guarantee that those claiming descent from Ida were biologically his direct male heirs; what matters for our purpose is that their claim to such descent seems to have been accepted as a legal statement. Cf. Charles-Edwards, T. M., ‘Kinship, Status and the Origins of the Hide’, Past and Present 56 (08 1972), 32CrossRefGoogle Scholar, n. 54. For a case where an Anglo-Saxon annalist notes the non-royal origin of a Northumbrian king, using the phrase ungecynde cyning, see ASC s.a. 867, Chadwick, , Institutions, pp. 305 f.Google Scholar, and Hunter, Blair, Introduction, p. 197.Google Scholar

^{page 16 note 6} Stenton, F. M., Anglo-Saxon England, 3rd ed. (Oxford, 1971), p. 92Google Scholar. Hart, , Charters of Northern England, pp. 137–8Google Scholar (no. 153), wrongly takes Æthelwald Moll to be the brother of King Eadberht Eating and therefore a member of the line of Ida; but the Latin of Pope Paul's letter to Archbishop Egbert of York and King Eadberht (ed. BCS 1, 262–4, and Councils 111, 394–5, and trans. EHD 1, 764–5 (no. 184)), which is the supposed evidence for this relationship, makes it plain that Æthelwald is the brother of Abbot Forthred, whose appeal to Rome was the cause of the letter. I owe this reaffirmation of Stenton's position to Professor Whitelock.

^{page 16 note 7} Kirby, D. P., The Making of Early England (London, 1967), p. 165Google Scholar. He has stated this more strongly, ‘British Dynastic History in the Pre-Viking Period’, Bull. of the Board of Celtic Stud. 27 (1976–1978), 81Google Scholar, n. 4: ‘The royal kindred-group in early England appears … to have included all the descendants of a great-great-great-great-grandfather.’

^{page 17 note 1} The inference fails also for Mercia. Cenwulf was descended in the seventh generation from Pybba, but we do not know that the latter was ever king, as our knowledge of Mercian royal history begins only with his son Penda (apart from the genealogically unplaced Cearl, whose daughter was first wife of King Edwin of Northumbria (Bede, HE 11.14)). Cenwulf is descended in the eleventh generation from the royal founder, Icil.

^{page 17 note 2} ‘Northumbria in the Time of Wilfrid’, Saint Wilfrid at Hexham, ed. Kirby, (Newcastle upon Tyne, 1974), pp. 17, 20 and 21.Google Scholar

^{page 17 note 3} This seven-generation group receives no support, it seems to me, from the statement of Lancaster, L., ‘Kinship in Anglo-Saxon Society – I’, Brit. Jnl of Sociology 9 (1958), 256Google Scholar, that ‘male lineal ascendants could be traced, linguistically, to the distance of “great-great-great-great-grandfather”, who was known as sixta fæder’, quoted by Kirby, ‘British Dynastic History’, p. 81, n. 4. More relevant is Lancaster's general observation (p. 367): ‘Succession to particular positions through kinship is hard to trace, except for the kingship which by its nature is exceptional and moreover demands a discussion that would involve much of Anglo-Saxon political history. The only general conclusion that can be arrived at is that kinship with the holder of the throne was a first requirement for regular succession, and that although the king's appointment by election was in the hands of the witena gemot, a tendency towards the acceptance of primogeniture grew up over the period. Succession along a line of brothers, however, is clearly apparent in the genealogy of the house of Wessex.’ Kirby's further assertions (‘British Dynastic History’, pp. 81–2) that ‘Pedigrees, whether Anglo-Saxon or Celtic, have generally been accepted as evidence for historical periods and, when it is considered that a royal family would need to know its own pedigree in order to make a valid claim to the kingship within a … seven-generation group, the likelihood that a genealogy would be correctly preserved seems to be strengthened’ seem to me to contain at least four misstatements. On some general principles to be observed in the examination of genealogies, see Charles-Edwards, ‘Kinship’, p. 32, n. 54, and Dumville, D. N., ‘Kingship, Genealogies and Regnal Lists’, EMK, pp. 72–104.Google Scholar

^{page 17 note 4} In this the English seem to have differed from the continental Germans of the fifth and sixth centuries. As Wallace-Hadrill, J. M., The Barbarian West 400–1000, 3rd ed. (London, 1967), p.35Google Scholar, points out in the context of a discussion of the forged pedigree of Theodoric the Ostrogoth, ‘Recent research tends to cast doubt upon the traditional view that the Germans cared, above all else, that their chieftains should belong to the right dynasties; chieftains owed their power to their own right arms, and handed it on to their children if they could’ (cf. ibid. p. 55 on the Lombards), but for a slightly different emphasis, see his Germanic Kingship, pp. 9–11. For a general survey, see P. Grierson, ‘Election and Inheritance in Early Germanic Kingship’, pp. 1–22.

^{page 17 note 5} As Loyn, H. R., ‘Kinship in Anglo-Saxon England’, ASE 3 (1974), 200Google Scholar, rightly points out.

^{page 18 note 1} Cf. Charles-Edwards, ‘Kinship’, pp. 28–9: ‘Royal kindreds were agnatic descent groups, that is, a man was a member of such a kindred if he could trace his descent back in the male line to a particular ancestor like Oisc or Wuffa.’ For the same situation among the German aristocracy of the ninth to eleventh centuries, see K. Leyser, ‘The German Aristocracy from the Ninth to the Early Twelfth Century’, Past and Present 41 (December 1968), 32–3, who stresses the very large size of these family groups; Leyser refers (pp. 3 2 and 52) to the possible importance of comparative German evidence for the study of the Anglo-Saxon nobility.

^{page 18 note 2} It has been suggested that this was the case also in Mercia after the deposition of Ceolwulf I in 823; see Wainwright, F. T., Scandinavian England: Collected Papers, ed. Finberg, H. P. R. (Chichester, 1975). p. 64.Google Scholar

^{page 18 note 3} Kingship, p. 29.

^{page 18 note 4} Introduction, p. 198. I am greatly obliged to Dr Hunter Blair for specifying to me the passages which he had in mind. Cf. also Chadwick, , Institutions, pp. 360–1.Google Scholar

^{page 18 note 5} Cf. Wallace-Hadrill, , Germanic Kingship, pp. 89–90.Google Scholar

^{page 18 note 6} Cenred (704–9) was the son of Æthelred's predecessor Wulfhere (king 657–74). Another nephew of Æthelred, Berhtwald, is found as a sub-king of Æthelred, c. 681; see Whitelock, , Anglo-Saxon Chronicle, p. 212Google Scholar, and Chadwick, , Institutions, pp. 285Google Scholar, n. 1, and 299 (and n. 1). Plummer suggested as an outside possibility (Two of the Saxon Chronicles Parallel, ed. Plummer, C. (Oxford, 1892–1899Google Scholar; repr., 1952, with contr. by D. Whitelock), 11, 35) on the evidence of ASC 702 E that Cenred had been associated with Æthelred in the kingship since that date (cf. Whitelock, , Anglo-Saxon Chronicle, p. 25Google Scholar, n. 5), but this seems hard to reconcile with Bede's wording.

^{page 19 note 1} For remarks on the designation of a successor, see Wood, Ian, EMK, p. 18.Google Scholar

^{page 19 note 2} Historia Regum (ed. Arnold 11, 41), s.a. 758.

^{page 19 note 3} Thus Kirby, ‘Northumbria in the Time of Wilfrid’, p. 17, where he refers to his paper ‘Problems of Early West Saxon History’, EHR 80 (1965), 10–29Google Scholar, but this contains no additional evidence on the point, only speculation. Cf. Bede, , HE IV. 12Google Scholar (‘deuictis atque amotis subregulis, Caedwalla suscepit imperium’); the passage is discussed in Venerabilis Baedae Opera Historica, ed. Plummer, C., 2 vols. (Oxford, 1896) 11, 220–1Google Scholar. This is the only evidence.

^{page 19 note 4} HE IV. 16. Cf. the activity of the Frankish Clovis towards the end of his career, discussed by Wood, Ian, EMK, pp. 9–10, 15, 16 and 25Google Scholar; see Gregory, of Tours, , Historia Francorum 11, 42.Google Scholar

^{page 19 note 5} ‘Northumbria in the Time of Wilfrid’, p. 17. But if the evidence is solely that discussed above, pp. 14–15, it is insufficient to warrant this conclusion.

^{page 19 note 6} ibid. p. 21, but no evidence is offered. He is perhaps referring to the entries s.a. 740 and 750 in the Historia Regum (ed. Arnold 11, 38 and 39–40): in 740 Arwine son of Eadwulf was killed, but no agent is named; in 750 King Eadberht took Bishop Cyniwulf prisoner and dragged Offa son of Aldfrith from his refuge in the church of Lindisfarne.

^{page 19 note 7} Æthelwulf De Abbatibus, ed. and trans. Campbell, A. (Oxford, 1967), pp. 6–7Google Scholar; cf. Kirby, ‘Northumbria in the Time of Wilfrid’, p. 20.

^{page 19 note 8} Stenton, , Anglo-Saxon England, pp. 218–19Google Scholar, considers this briefly; Wallace-Hadrill, , Germanic Kingship, pp. 114–17 and 119–20Google Scholar, gives it greater attention; John, Eric, Orbis Britanniae and other Studies (Leicester, 1966), pp. 27 and 32–5Google Scholar, also discusses Ecgfrith's consecration. On the ordination of Irish kings, an issue raised by John (pp. 27 ff.), see Binchy, D. A., ‘The Fair of Tailtiu and the Feast of Tara’, Ériu 18 (1958), 118–19Google Scholar. Nelson, Janet L., “The Problem of King Alfred's Royal Anointing’, JEH 18 (1967), 160Google Scholar, denies that anointing was a factor either in 787 or at Eardwulf of Northumbria's consecration in 796, although her comments, EMK, p. 52, about the events of 787 are more restrained.

^{page 20 note 1} For comparable ploys in Visigothic Spain, see Wood, Ian, EMK, pp. 17–18Google Scholar. One might also compare the creation of Carolingian sub-kings; see Nelson, J., EMK, pp. 59–60.Google Scholar

^{page 20 note 2} Cf. Chadwick, , Institutions, p. 301Google Scholar, which usefully places this episode in the context of earlier examples of sons bearing the title of king in their fathers’ lifetimes. In charters up to 787 the order is usually (1) Offa, (2) his queen, Cynethryth, (3) Ecgfrith (when all three arc found subscribing); from 788 onwards we find Ecgfrith immediately following Offa – Sawyer, nos. 129, 130, 131, 132 and 1412.

^{page 20 note 3} Dümmler, no. 122 (pp. 178–80); Jaffé, no. 79 (pp. 348–52); and EHD 1, 786–8 (no. 202). When Wallace-Hadrill, J. M., Early, Medieval History (Oxford, 1975), p. 168Google Scholar, wrote that Alcuin attributed the collapse of Offa's personal empire after Ecgfrith's death ‘to moral causes; he knew of the bloody beginnings of Offa's reign’, he perhaps had this letter in mind; it seems to me to fit better the circumstances discussed here, although the other interpretation is not thereby excluded.

^{page 20 note 4} See above, p. 19, and below, p. 30.

^{page 20 note 5} Councils 111, 521–3, and EHD 1, 791–5 (no. 204): Cenwulf to Leo, 798.

^{page 20 note 6} Councils 111, 542–4, and EHD 1, 799–800 (no. 210).

^{page 20 note 7} Offa is by no means the only Mercian king we see in this light. Felix's Life of St Guthlac, §49 (ed. and trans. B. Colgrave (Cambridge, 1956), pp. 148 and 149), shows the future King Æthelbald (716–56/7) hard pressed by King Ceolred's persecution. They belonged to different dynastic segments and Ceolred no doubt felt Æthelbald to be a threat to his own family's rule. Æthelbald on the other hand does not appear to have been hostile during his reign to Offa's grandfather Eanwulf, to whom he granted land (Sawyer, no. 146; cf. also no. 117 for Eanwulf's activities).

^{page 20 note 8} With whom we know Eardwulf to have been closely connected; cf. Stenton, , Anglo-Saxon England, pp. 94–5Google Scholar. On Frankish consecrations from 751 to 848, see J. Nelson, ‘Kingship, Law and Liturgy’, p. 245, n. 3.

^{page 20 note 9} Historia Regim (ed. Arnold 11, 57–8), s.a. 796.

^{page 21 note 1} Nelson, J., EMK, p. 57Google Scholar, reminds us that in the eighth century ‘consecration’ referred to any statuschanging rite and did not necessarily imply anointing.

^{page 21 note 2} Dr Hunter Blair kindly drew my attention to the importance of this event.

^{page 21 note 3} Historia Regum (ed. Arnold 11, 52), s.a. 790; cf. Kirby, ‘Northumbria in the Time of Wilfrid’, pp. 25–6, for comment.

^{page 21 note 4} Historia Regum (ed. Arnold 11, 59),s.a. 798.

^{page 21 note 5} His reign is interrupted, at the very least for much of 808, by the usurpation of one Ælfwald (ii) (Plummer, Chronicles Parallel 11, 84; cf. Stenton, , Anglo-Saxon England, pp. 94–5Google Scholar). One is bound to note the identity of the latter's name with that of his predecessor who reigned 779–88 and whose children Oelf and Oelfwine were murdered by King Æthelred in 791 (Plummer, , Chronicles Parallel 11, 60).Google Scholar

^{page 21 note 6} Historia Kegum (ed. Arnold 11, 52), s.a. 790, where he is called dux.

^{page 21 note 7} For Alfred's will, where these are reported, see Select English Historical Documents of the Ninth and Tenth Centuries, ed. and trans. Harmer, F. E. (Cambridge, 1914), pp. 15–19, 49–55 and 91–103Google Scholar; for a new translation see EHD 1, 492–5 (no. 96). §16 of Asser's Life of King Alfred also provides relevant information (ed. Stevenson, pp. 14–16 and 209–12).

^{page 21 note 8} John, , Orbis Britanniae, pp. 36–44Google Scholar (and pp. 37–40 on Alfred's coronation). John's views are declared convincing by Wallace-Hadrill, , Germanic Kingship, p. 140Google Scholar. Independently of John, the problem of this royal anointing was discussed by J. L. Nelson, ‘King Alfred's Anointing’, pp. 145–63, who raises a number of other controversial issues which are fortunately not relevant to the problem to be discussed here.

^{page 22 note 1} ‘Some Celtic Legal Terms’, p. 224.

^{page 22 note 2} Orbis Britanniae, pp. 40–1.

^{page 22 note 3} ibid. p. 41, but this is an inference from the subsequent course of events.

^{page 22 note 4} Land Tenure in Early England: a Discussion of some Problems (Leicester, 1960; rev., 1964), p. 41Google Scholar, n. 1. Compare (and contrast) the succession-arrangements envisaged by the will of the Vandal king Gaiseric; see Wood, Ian, EMK, p. 10.Google Scholar

^{page 22 note 5} Some useful continental parallels are adduced by Gillingham, J. B., The Kingdom of Germany in the High Middle Ages (900–1200) (London, 1971), p. 18Google Scholar. In effect Gillingham's view is identical with John's on the importance of this point.

^{page 22 note 6} John has incautiously asserted, without discussion, that the yrfe must include the kingship (Orbis Britanniae, p. 40).

^{page 23 note 1} ibid. pp. 40–1.

^{page 24 note 1} Asser (ed. Stevenson), §§29, 38.

^{page 24 note 2} ibid. §42.

^{page 24 note 3} ‘Some Celtic Legal Terms’, p. 224.

^{page 24 note 4} Kingship, p. 29.

^{page 24 note 5} Cf. John's remarks, Orbis Britanniae, p. 20.

^{page 24 note 6} Sayles, G. O., The Medieval Foundations of England, 2nd ed. (London, 1950), p. 64CrossRefGoogle Scholar, rightly notes ‘the lack of any settled principle of succession to the throne’.

^{page 25 note 1} Harrison, K., The Framework of Anglo-Saxon History to A.D. 900 (Cambridge, 1976), p. 92Google Scholar (cf. Nelson, J., EMK, p. 51)Google Scholar, calls attention to our lack of knowledge of the political processes of succession.

^{page 25 note 2} ‘Northumbria in the Time of Wilfrid’, pp. 1–34, esp. 17 ff. But he seems to me often to risk overstepping the limits of the evidence, for example in his association of the deposition of King Ceolwulf with the expulsion of Bishop Acca from Hexham in 751/2 (p. 24). One case which Kirby does not discuss is that of Archbishop Eanbald ii of York (796–808 × 837), who is shown clearly by two of Alcuin's letters to be heavily involved in Northumbrian politics: he harbours the king's enemies, protects their possessions and has a large retinue of thegns; he is also, of course, a large landowner (Dümmler, nos. 232 and 233 (pp. 376–9) = Jaffé, nos. 173 and 174 (pp. 620–4) = EHD 1, 795–7 (nos. 207 and 208); Dümmler 232 is in Councils (111, 534–6)).

^{page 25 note 3} ‘Northumbria in the Time of Wilfrid’, p. 17.

^{page 25 note 4} Die Briefe des Heiligen Bonifatius und Lullus, ed. Tangl, Michael, 2nd ed. (Berlin, 1955), pp. 146–55Google Scholar (no. 73); Councils 111, 350–6; and EHD 1, 751–6 (no. 177), esp. 755.

^{page 25 note 5} Plummer, , Baedae Opera Historica 11, 306, 314 and 386.Google Scholar

^{page 25 note 6} Alcuin to King Æthelred of Northumbria in 793 (Dümmler, no. 16 (pp. 42–4); Jaffé, no. 22 (pp. 180–4); Councils 111, 492–5; and EHD 1, 775–7 (no. 193)).

^{page 25 note 7} Dümmler, no. 21 (pp. 58–9); Jaffé, no. 25 (pp. 193–4); and EHD 1, 784–5 (no. 199).

^{page 26 note 1} These are the two kings to whom Charlemagne was best disposed.

^{page 26 note 2} Dümmler, no. 109 (p. 156); Jaffé, no. 66 (pp. 305–6); and EHD 1, 785–6 (no. 200).

^{page 26 note 3} Cf. Plummer's analysis (Chronicles Parallel 11, 63).

^{page 26 note 4} Dümmler, no. 16 (pp. 42–4); Jaffé, no. 22 (pp. 180–4); Councils 111, 492–5; and EHD 1, 775–7 (no. 193). Alcuin writes to King Æthelred and his optimates, after the sack of Lindisfarne by Vikings in 793; ‘Non dico, quod fornicationis peccata prius non essent in populo. Sed a diebus Aelfwaldi regis fornicationes adulteria et incestus inundauerunt super terram, ita ut absque omni uerecundia etiam et in ancillis Deo dicatis hec peccata perpetrabantur.’ But Stenton, , Anglo-Saxon England, p. 93Google Scholar, says that ‘Ælfwald was remembered as a “just and pious king” [a quotation from the Historia Regum, s.a. 779], and his death, which was the result of a private conspiracy, was followed by a grievous degeneration of morals in the north. [Here a footnote refers to Alcuin's letter.] He was the last Northumbrian king for whom any ancient writer expressed admiration.’ Stenton seems to me to have missed the force of Alcuin's remarks, being misled by the entries in the Historia Regum, s.a. 779, 781 and 788 (ed. Arnold 11, 46–7 and 52), which show every sign of being interpolations made at Hexham (where Ælfwald was buried and culted) in the twelfth century. I know of no certainly early source which expresses admiration for him.

^{page 26 note 5} Dümmler, no. 129 (pp. 191–2); Jaffé, no. 86 (pp. 369–72); and Councils 111, 509–11. This is something of an inconsistency in view of his attitude to the line of Ida.

^{page 26 note 6} We find this stressed in royal diplomas too. Wallace-Hadrill, , Germanic Kingship, pp. 111–12Google Scholar, and Early Medieval History, p. 156, draws attention to Sawyer, no. 105 (BCS, no. 195), issued in 764 to the bishop of Rochester by Offa, styled ‘rex Merciorum, regali prosapia Merciorum oriundus atque omnipotentis Dei dispensatione eiusdem constitutus in regem’. For Offa's sense of his own royal ancestry, see Stenton, , Preparatory, pp. 380–1Google Scholar. Much later, in 941, Edmund is styled ‘ex regali progenie Deo annuente regenteque … rex ordinatus’ by Sawyer, no. 478 (BCS, no. 769), a charter to John, which, Orbis Britanniae, pp. 54–5Google Scholar, draws attention.

^{page 27 note 1} Wallace-Hadrill, , Germanic Kingship, pp. 119–20.Google Scholar

^{page 27 note 2} ibid. p. 135 and n. 49. Wallace-Hadrill notes Archbishop Hincmar's emphasis in his Metz sermon on Charles the Bald's hereditary claim, as descendant both of the Merovingian Clovis and of St Arnulf, and refers to Schramm, P., Der König von Frankreich 1 (Weimar, 1960), 27Google Scholar. On royal anointing, cf. Ullmann, Walter, Principles of Government and Politics in the Middle Ages, 2nd ed. (London, 1966), pp. 58–9.Google Scholar

^{page 27 note 3} Ed. D. N. Dumville, ‘The Anglian Collection’, pp. 25–50.

^{page 27 note 4} I owe this point about the Northumbrian pedigrees to Professor Whitelock. The position in the Mercian pedigree is rather more complicated. Neither Cenred (704–9) nor his father Wulfhere (657–74) appears, but the line may be represented by Wulfhere's brother Æthelred (674–704). Likewise Ceolred (709–16) is omitted, although his father Æthelred's pedigree is given; one wonders if the notorious Ceolred was quietly but deliberately overlooked.

^{page 27 note 5} §12: Dümmler, no. 3 (pp. 19–29, at 23–4); Councils 111, 447–62, at 453–4; and EHD 1, 770–4 (no. 191), at 771.

^{page 27 note 6} A sample of §12 reads, in Professor Whitelock's translation: ‘We decreed that in the ordination of kings no one shall permit the assent of evil men to prevail, but kings are to be lawfully chosen by the priests and the elders of the people, and are not to be those begotten in adultery or incest; for just as in our times according to the canons a bastard cannot attain to the priesthood, so neither can he who was not born of a legitimate marriage be the Lord's anointed and king of the whole kingdom and inheritor of the land.’

^{page 27 note 7} §12, ibid.

^{page 28 note 1} Dümmler, no. 101 (pp. 146–8); Jaffé, no. 58 (pp. 290–3); Councils 111, 498–9 (incomplete text); and EHD 1, 782–4 (no. 198).

^{page 28 note 2} Wallace-Hadrill, , Germanic Kingship, p. 134Google Scholar, refers to ‘the ancient doctrine … that a bad ruler was God's judgment on a sinful people, and as such had better be endured than dispensed with’. He notes (p. 144) that King Alfred shared this view, as is shown by his version of Gregory's ‘Pastoral Care’. For juristic means of dealing with a bad ruler, see Ullmann, W., The Carolingian Renaissance and the Idea of Kingship (Cambridge, 1969), pp. 66–7.Google Scholar

^{page 28 note 3} Which may be influenced by Irish thinking on the subject; cf. Binchy, , Kingship, p. 10,Google Scholar and Audacht Morainn, ed. and trans. Kelly, Fergus (Dublin, 1976), pp. xiii–xixGoogle Scholar. But while defeat in battle and various natural disasters result from unjust rule, there is no explicit statement that a king might be deposed for such proofs of his iniquity.

^{page 28 note 4} Note, for example, his words to the Mercian ealdorman Osberht – Dümmler, no. 122 (pp. 178–80); Jaffé, no. 79 (pp. 348–52); and EHD 1, 786–8 (no. 202) – not only did Ecgfrith die by God's vengeance for Offa's bloodletting, but ‘I am afraid that our king, Eardwulf, must quickly lose the kingdom on account of the insult he did to God, dismissing his own wife, and publicly taking a concubine, as it is reported. Let your dearest king [Cenwulf] beware of this.’

^{page 28 note 5} Much useful material is gathered and discussed by Chadwick, , Institutions, pp. 355–66Google Scholar; however, although his conclusions are eminently sensible, the framework of ‘the national council’ within which he discusses it is now seen as an unacceptable anachronism. For Frankish situations in which aristocratic factions clearly emerge, see Wood, Ian, EMK, pp. 10–11, 13, 14 and 17.Google Scholar

^{page 29 note 1} The events of 704/5 are well discussed by Kirby, ‘Northumbria in the Time of Wilfrid’, p. 10. Bede, in his verse Life of St Cuthbert, hails Osred on his accession as nouus Iosia: Bedas metrische Vita Sancti Cuthberti, ed. Jaager, Werner (Leipzig, 1935), p. 100Google Scholar, line 554 (cf. p. 4). He was no doubt soon disappointed; Osred's accession and slaying are reported without comment in the Historia Ecclesiastica (V. 18 and 22 respectively).

^{page 29 note 2} I have restricted myself to those discussed by Kirby, ‘Northumbria in the Time of Wilfrid’, pp. 18,19 and 25.

^{page 29 note 3} ibid. p. 17.

^{page 29 note 4} Compare the comments of D.Ó Corráin, ‘Irish Regnal Succession’, p. 30, on the Irish situation.

^{page 29 note 5} Help on this score may well be derived from an examination of the extent of legal liability within the kindred during the period. Beyond this immediate (but, by our standards, extended) family group, faction probably began and alliances probably therefore became necessary.

^{page 29 note 6} Kirby, ‘Northumbriain the Time of Wilfrid’, p. 22.

^{page 29 note 7} See above, p. 19.

^{page 30 note 1} Pp. 19–20.

^{page 30 note 2} And if John's reading of Æthelwulf's provisions be correct (Orbis Britanniae, pp. 36–44, esp. 44), only a member of the immediate royal kindred would have been likely to have the wealth and influence to be a successful candidate.

^{page 30 note 3} Cf. Kirby, D. P., ‘Hywel Dda: Anglophil?’, Welsh Hist. Rev. 8 (1976–1977), 6–7Google Scholar and n. 43.

^{page 30 note 4} The importance of churchmen in dynastic politics must be reckoned with at any period of Anglo-Saxon history. Not only may they have had an ideological viewpoint, but as nobles with consequent aristocratic concerns (such as property, whether their own or their church's, family interests, and their own personal standing or advancement) they will very naturally often have been party to factional activities.

^{page 30 note 5} As in the subscriptions to the New Minster foundation charter; see Sawyer, no. 745 (and the spurious no. 746). Cf. John, , Orbis Britanniae, pp. 274–5.Google Scholar

^{page 30 note 6} For the arguments of 975 against Edward, see the discussion by Nelson, J., EMK, p. 67.Google Scholar

^{page 31 note 1} A careful study of the events and the personalities of these years may be found in Fisher, D. J. V., ‘The Anti-Monastic Reaction in the Reign of Edward the Martyr’, Cambridge Hist. Jnl 10 (1950–1952), 254–70.CrossRefGoogle Scholar

^{page 31 note 2} Cf. Chadwick, , Institutions, p. 294Google Scholar. In the very different circumstances of the pre-Viking period the dynasties of the several kingdoms were often linked in the same way (as John reminds us, Orbis Britanniae, p. 20). In an important study (which will nonetheless require much modification), Cyril Hart has shown us something of the interlinking of personal and blood-ties with dynastic politics in tenth-century England: ‘Athelstan “Half King” and his Family’, ASE 2 (1973), 115–44Google Scholar. He asserts (p. 115) that ‘most of the ninth- and tenth-century ealdormen were scions of cadet branches of the royal house’; cf. Chadwick, , Institutions, p. 294Google Scholar (cf. p. 352), who makes a similar point rather more cautiously and with reference only to the tenth century.

^{page 31 note 3} For example, Æthelgeard, one of the most powerful Wessex thegns of the mid-tenth century, appears to have enjoyed a special relationship with the royal family: he knew the æthelings in their childhood and King Eadwig calls him his carus. See C. Hart, ‘The Codex Wintoniensis’, p. 16, and references cited there, nn. 2–4.

^{page 31 note 4} See Hart, C., ‘Athelstan “Half King” and his Family’, ASE 2 (1973), 116, 123–4 and 129–30Google Scholar. As Hart notes (p. 124), this might help to explain Edgar's attachment to the monastic ideals which the Half-King's family seems to have promoted. But for a less generous view, at least of Æthelwine, see Fisher, ‘The Anti-Monastic Reaction’, pp. 266–7.

^{page 32 note 1} §58.1: Die Gesetze der Angelsachsen, ed. and trans. Liebermann, Felix, 3 vols. (Halle, 1898–1916) 1, 350Google Scholar, and The Laws of the Kings of England from Edmund to Henry I, ed. and trans. Robertson, A. J. (Cambridge, 1925), pp. 204 and 205Google Scholar. For this part of the code Wulfstan was using Alfred's laws, but §3 of those makes no mention of the ætheling.

^{page 32 note 2} §2: Liebermann, , Gesetze 1, 460Google Scholar. It is one of five short texts concerning status; they are translated EHD 1, 431–4.

^{page 32 note 3} §§4 (cf. IV Athelstan 6.1–2), 11 (cf. viii Æthelred 5 and I Cnut 31.2) and 12 (cf. Alfred §15): Liebermann, , Gesetze 1, 470–1Google Scholar. In none of the source-passages is the ætheling mentioned.

^{page 32 note 4} For Wulfstan's authorship of these law codes, see Sermo Lupi ad Anglos, ed. Whitelock, Dorothy, 3rd ed. (London, 1963), pp. 23 ff.Google Scholar

^{page 32 note 5} But see the discussion of the wergild of the aetheling by Chadwick, (Institutions, pp. 17–18Google Scholar; 76; 163, n. 1; 167–8; and 307, n. 1), where a rather different view is expressed. However, Chadwick seems to have generalized to the whole period the evidence provided by Norðleoda laga.

^{page 32 note 6} Eric John has put forward the view that Æthelwulf's tenurial arrangements (cf. above, pp. 21–4) would thenceforth set the king apart, by his considerable wealth, even from many members of his family group (Orbis Britanniai, p. 44). This would accordingly be reflected in the status of an aetheling.

^{page 32 note 7} If the new provisions in these legal documents do indicate an advance in the status of an sctheling, then certain termini can be established for these developments. Breach of his peace (i.e. fighting in his presence) has become an offence since Alfred's time; his right to grant a special period of sanctuary is later than the promulgation of IV Athelstan; but special provisions for breach of his protection may be very new, for the corresponding parts of viii Æthelred and I Cnut do not mention them.

^{page 33 note 1} This assumes that in the circumstances of 1066 Edgar would have been unacceptably inexperienced, even if his youth did not by itself tell against him; and he would presumably have lacked the wealth and influence that a closer relative of the king – or, on the other hand, someone like Harold Godwinesson (and here continental parallels such as the Capetians are relevant) – would possess.

^{page 33 note 2} One must also ask whether Edward was entitled to designate a successor from outside the (agnatic) royal kindred.

^{page 33 note 3} And perhaps others, too, in the eleventh century; but in that dynastically varied century it is difficult to discern the difference between eligibility and determination.