¹ The Homilies of the Anglo-Saxon Church: the First Part containing the Sermones Catholici or Homilies of Ælfric, ed. Thorpe, B., 2 vols. (London, 1844–1846) I, 1.Google Scholar References to the First Series of Catholic Homilies ( = CH I) will be to vol. I of Thorpe's edition by page number and, where appropriate, line numbers. References to the Second Series ( = CH II) will be to Ælfric's Catholic Homilies: the Second Series, Text, ed. Godden, M., EETS ss 5 (London, 1979), except that references to vol. II of Thorpe are provided additionally in the headings for each of the three Second Series homilies analysed in order to facilitate cross-referral to published source studies which predate Godden's edition.Google Scholar

² ‘and sometimes Haymo’.

³ Smetana, C.L., ‘Ælfric and the Homiliary of Haymo of Halberstadt’, Traditio 17 (1961), 457–69CrossRefGoogle Scholar, and The Homilies of Ælfric: a Supplementary Collection, ed. Pope, J.C., 2 vols., EETS os 259–60 (London, 1967–1968), 169, with discussion on pp. 155–61. Pope observes (p. 160)Google Scholar that although Ælfric used Haymo's homiliary frequently, it was seldom his primary authority. In this respect his use of Haymo differed from his use of patristic writers (mainly via Paul the Deacon's homiliary) and — as I argue here — from his use of Smaragdus. It should be noted that the Haymo in question is now recognised as being Haymo of Auxerre (died 865/6) and not Haymo of Halberstadt (778–853) as given in Smetana's title: see Barré, H., Les homéliaires carolingiens de l' école d' Auxerre, Studi e Testi 225 (Vatican City, 1962), 33–42.Google Scholar

⁴ This is not, of course, to say that Ælfric always named his sources, or that his identifications always agree with modern scholarship. There are also occasions when Ælfric was selective in representing what a source text says, as Malcolm Godden has shown: ‘Ælfric's Saints' Lives and the Problem of Miracles’, Leeds Stud. in Eng. n.s. 16 (1985), 83–100. But it remains true that Ælfric did not make spurious claims.Google Scholar

⁵ Smetana, C.L., ‘Ælfric and the Early Medieval Homiliary’, Traditio 15 (1959), 163–204.CrossRefGoogle Scholar

⁶ For Ælfric's use of Haymo, see above, n. 3. On the subject of Frankish homiliaries generally, see Grégoire, R., Les homéliaires du moyen âge, Rerum Ecclesiasticarum Documenta, Series Maior, Fontes 6 (Rome, 1956)Google Scholar; Barré, , Les homéliaires carolingiensGoogle Scholar; and McKitterick, R., The Frankish Church and the Carolingian Reforms, 789–895 (London, 1977), pp. 80–114.Google ScholarSee also Clayton, M., ‘Homiliaries and Preaching in Anglo-Saxon England’, Peritia 4 (1985), 207—42, who provides a useful survey as part of her English-oriented discussion.CrossRefGoogle Scholar

⁷ Smetana characterizes Haymo's exegetical work as a homiliary, but he describes Smaragdus's Expositio as ‘a compilation of brief extracts from the Fathers’: Smetana, C.L., ‘Ælfric and the Early Medieval Homiliary’, p. 181Google Scholar, n. 6, and ‘Ælfric and the Homiliary of Haymo of Halberstadt’, p. 457.Google Scholar If we are to attempt a fair assessment of Ælfric's possible dependence on Smaragdus, it is important to realise that Smetana has given us a misleading description of his work: the ‘extracts’ from the Fathers are by no means always ‘brief’ and a given exegesis may well draw upon and interweave a wide range of patristic materials, selecting and adapting as appropriate. The Expositio should therefore not be regarded as being in a different category from Haymo's homiliary or other exegetical materials in the patristic tradition. The only editions of these two works are: Smaragdus, , Collectiones in epistolas et evangelia, PL 102, cols. 14—551Google Scholar [more usually referred to as Expositio libri comitis]; Haymo, , Homiliae de Tempore, PL 118, cols. 11–746.Google Scholar It is possible that both were originally intended for monastic use: for Smaragdus, see below, pp. 206–8Google Scholar; for Haymo, see Clayton, ‘Homiliaries and Preaching’, pp. 211–12.Google Scholar

⁸ Haymo is cited by name in CH I.viii (Thorpe, p. 120) and xxxiv (Thorpe, p. 510).Google Scholar

⁹ See below, p. 213.Google Scholar

¹⁰ Smetana, , ‘Ælfric and the Early Medieval Homiliary’, pp. 165–80Google Scholar, prints a schedule of 244 entries in the original homiliary as reconstructed by Wiegand, F., Das Homiliarium Karls des Grossen auf seine ursprüngliche Gestalt hin untersucht, Studien zur Geschichte der Theologie und der Kirche 1.2 (Leipzig, 1897). However, as Smetana points out (pp. 202–4)Google Scholar, Ælfric's copy of this homiliary was probably a version modified by the addition of some items and the omission of others. For the importance of Paul the Deacon's homiliary in Anglo-Saxon England, see Clayton, , ‘Homiliaries and Preaching’, pp. 217–20.Google Scholar

¹¹ For example, in contrast with Paul the Deacon's homiliary, it does not figure by name in Gneuss, H., ‘A Preliminary List of Manuscripts written or owned in England up to 1100’, ASE 9 (1981), 1–60Google Scholar, although Clayton, , ‘Homiliaries and Preaching’, pp. 217–19Google Scholar, shows that two of the manuscripts listed (Gneuss, nos. 211 and 230) do draw on Haymo's homiliary: Canterbury, Cathedral Library, Add. 127/1 with Maidstone, Kent County Archives Office PRC 49/2, dated by Gneuss and Clayton as s. xi¹, and Durham, Cathedral Library, B. II. 11, dated by Gneuss and Clayton as s. xi². In the first of these it is the Maidstone leaves which preserve the Haymo material and, as Ker, N.R. acknowledged in Medieval Manuscripts in British Libraries: II, Abbotsford-Keele (Oxford, 1977), pp. 315–16, it is not certain that they belong to Canterbury Cathedral Library, Add. 127/1, although this does not detract from their value as evidence for the knowledge of Haymo in Anglo-Saxon England.Google Scholar

¹² See above, n. 3.

¹³ The selection of Smaragdus's work known in Anglo-Saxon England itself testifies to the influence of the Reform movement. On Smaragdus's life and work, see Rädle, F., Studien Zu Smaragd von Saint-Mihiel, Medium Ævum: PhilologischeStudien 29 (Munich, 1974), the main part of which is a detailed study of the Expositio libri comitis and especially of its sources.Google Scholar

¹⁴ The evidence of surviving manuscripts suggests that the English reform used only the textus receptus version of the Rule, which resulted from the reforms of Benedict of Aniane. For a summary of the textual traditions of the Rule and a discussion of the importance of Smaragdus's Expositio in Regulam Sancti Benedicti and the supplementation of the Rule within the reform tradition, with particular emphasis on the work of Æthelwold, see Wuljstan of Winchester: The Life of St Æthelwold, ed. Lapidge, M. and Winterbottom, M. (Oxford, 1991), pp. li–lx.Google Scholar For more detail, see Gretsch, M., Die Regula Sancti Benedicti in England und ibre altenglische Übersetzung, Texte und Untersuchungen zur englischen Philologie 2 (Munich, 1973)Google Scholar and ‘Æthelwold's Translation of the Regula Sancti Benedicti and its Latin Exemplar’, ASE 3 (1974), 125–51.Google ScholarÆthelwold's use of Smaragdus is discussed in these works by Gretsch on pp. 257–62 and 144–6 respectively.Google Scholar

¹⁵ Witters, W., ‘Smaragde au moyen âge: la diffusion de ses ecrits d'apres la tradition manuscrite’, Études ligériennes d' histoire et d' archeologie medievales, ed. Louis, R. (Auxerre, 1975), pp. 361—76, at 364 and 366—7.Google Scholar This survey of extant manuscripts of the Expositio libri comitis is deficient, however, in not including the two surviving copies from Anglo-Saxon England (see below, Appendices 1 and 2) and Witters is consequently wrong in stating that England is not represented in the dissemination. Clayton, , ‘Homiliaries and Preaching’, p. 217Google Scholar, is likewise in error in stating that ‘There is no trace of Smaragdus.’ A cumulative list of manuscripts of the Expositio is provided by Souter, A. in ‘Prolegomena to the Commentary of Pelagius on the Epistles of St. Paul’, JTS 7 (1906), 568–75, at 570–1Google Scholar, ‘Contributions to the Criticism of Zmaragdus's Expositio libri comitis’, JTS 9 (1908), 584–97, at 585Google Scholar, ‘Further Contributions to the Criticism of Zmaragdus's Expositio libri comitis’, JTS 23 (1922), 73–6, at 73—4Google Scholar, and ‘A Further Contribution to the Criticism of Zmaragdus's Expositio libri comitis’, JTS 34 (1933), 46–7Google Scholar; and there is a corrected list in Rädle, , Studien Zu Smaragd, pp. 120–7.Google Scholar Souter stated early in his investigations that the work ‘had passed almost entirely out of use after the tenth century’ (‘Contributions’, p. 586), but we can now see from the manuscript evidence that this was not so.Google Scholar

¹⁶ The manuscripts are Oxford, Bodleian Library, Barlow 4 and Worcester, Cathedral Library, F. 91, for which see below, Appendix 2. I am grateful to the Bodleian Library and to the Dean and Chapter of Worcester Cathedral for allowing me to examine them. Witters, ‘Smargde au moyen âge’, p. 361Google Scholar, believes that the work was intended for the collatio. McKitterick, , The Frankish Church and the Carolingian Reforms, p. 96Google Scholar, considers that this collection, in common with others, was originally intended to aid the literate in their private meditation, but that such works later came to be used extensively for oral instruction. For objections to McKitterick's view, see Clayton, , ‘Homiliaries and Preaching’, p. 216.Google Scholar On the standardizing function of Smaragadus's Expositio, which would have made it particularly attractive to Ælfric, see Bischoff, B., Latin Palaeography: Antiquity and the Middle Ages, trans. Cróinín, D.Ó and Ganz, D. (Cambridge, 1990), p. 204CrossRefGoogle Scholar, and The Life of St Æthelwold, ed. Lapidge, and Winterbottom, , p. lxi.Google Scholar

¹⁷ The Homilies of Ælfric, ed. Pope, , p. 155.Google Scholar

¹⁸ Förster, M., ‘Über die Quellen von Ælfrics exegetischen Homiliae Catholicae’, Anglia 16 (1894), 1–61.CrossRefGoogle Scholar

¹⁹ Smetana, ‘Ælfric and the Early Medieval Homiliary’.

²⁰ The Homilies of Ælfric, ed. Pope, , pp. 150–77.Google Scholar I shall not be examining here any of the homilies edited by Pope but, in view of his strong patristic bias (see below, pp. 209–10), it will obviously be necessary to include them in subsequent extensions of the present argument.Google Scholar

²¹ Pope acknowledges that we still have to answer the question of ‘what authors and works Æfric used directly and to what extent he gained access to them by means of selective compilations’: The Homilies of Ælfric, p. 163. It is precisely this issue which the present paper seeks to address.Google Scholar

²² Förster, , ‘Über die Quellen’, p. 59.Google Scholar For an earlier suggestion that Ælfric used a homiliary, see Earle, J., Anglo-Saxon Literature (London, 1884), pp. 215–16.Google Scholar

²³ Förster, , ‘Über die Quellen’, pp. 39–43.Google Scholar

²⁴ The Homilies of Ælfric, ed. Pope, , p. 171.Google Scholar

²⁵ Smetana, , ‘Ælfric and the Early Medieval Homiliary’, pp. 196–7.Google Scholar

²⁶ See below, pp. 225–8.Google Scholar

²⁷ Förster had already revealed this possibility by noting instances where material in the Fathers was also in Smaragdus, although he failed to pursue this and allowed the source identification to stand as a patristic one: see, for example, ‘Über die Quellen’, §§54, 57, 60, 87 and 88. There were also instances where, as we shall see in the analyses below, Förster gave priority to the patristic source, even though the Smaragdus text was virtually identical except for limited variations which Ælfric followed. The implications of this are likewise that Smaragdus was an alternative possibility as a source for more than just the variation, although Förster resisted this conclusion.

²⁸ Förster, , ‘Über die Quellen’, p. 25 (§84).Google Scholar For Bede's commentary on Luke, references throughout are by page and the line number to Bedae Venerabilis opera. Pars II: Opera exegetica, ed. Hurst, D., CCSL 120 (Turnhout, 1960), 1–425.Google Scholar

²⁹ Smetana, , ‘Ælfric and the Early Medieval Homiliary’, pp. 193–4. Wiegand's reconstruction of Paul the Deacon's original homiliary as given by Smetana (see above, n. 5) uses an older system for designating the liturgical day than the one used by Smaragdus and Ælfric. It may well be that the modern system was used in Ælfric's own copy of Paul the Deacon but, if not, it would often have been easier for Ælfric to find an item in Smaragdus's homiliary than in Paul's. In the present instance Paul's homiliary gives the Bedan text for the Fourth Sunday after the Feast of St Laurence (feast day August 10), whereas in both Ælfric and Smaragdus it is for the Seventeenth Sunday after Pentecost. These Sundays would only coincide when Easter was as early as 22 or 23 March.Google Scholar

³⁰ This problem of intertextuality is particularly acute when comparing Paul the Deacon and Smaragdus, since Smaragdus's exegesis, unlike Haymo's, incorporates lengthy verbatim passages from the same texts which were anthologized by Paul.

³¹ For Smaragdus's homily, see PL 102, 464—6.Google ScholarRädle, , Studien Zu Smaragd, p. 216Google Scholar, gives Bede as the sole source, except for noting that the etymology of Nain is derived ultimately from Jerome. Rädle observes that this detail is also present in a homily on this lection which was added to Paul the Deacon's homiliary (Homily 179: PL 95, cols. 1415–22, at 1416).Google Scholar This homily is attributed tentatively to Ambrose in PL, but it is actually by Heiric of Auxerre: see Barré, , Les homéliaires carolingiens, p. 157.Google Scholar As one would expect, this agrees in general terms with Bede/Smaragdus and shares some phraseology, but it differs significantly from Bede/ Smaragdus in details, progression and scale. Ælfric's homily is also inevitably in general agreement with Heiric's text but it does not correspond with it as it does with Bede/ Smaragdus. In any case, we do not know that it was in Ælfric's copy of Paul the Deacon's homiliary, and, on the basis of a detailed comparative analysis, I see no grounds for supposing that Ælfric used it, particularly in view of the certain availability of Bede/ Smaragdus, which provides a far better match. For further evidence against Ælfric's use of Heiric here, see below, n. 40.

³² PL 102, col. 464D (Smaragdus) and pp. 490–2 (Ælfric).Google Scholar

³³ PL 102, col. 465B-C (Smaragdus) and p. 492Google Scholar lines 22–4 (Ælfric), compared with Bede (ed. Hurst, , p. 158, lines 2297–306). Smaragdus, followed by Ælfric, simplifies by omitting the whole of Bede's lengthy second sentence, ‘Desiste, inquit … astrueret.’Google Scholar

³⁴ Compare Ælfric, p. 492, lines 13–21 with Bede, p. 158, lines 2278–90. If the passage had been included in Smaragdus's homily, it would have occurred at PL 102, col. 465B, following ‘… formetur Christus’. Its omission is not an error in the PL edition since it is also not in Barlow 4.Google Scholar

³⁵ Source authors are named in Paul the Deacon's preface, but the authorship must also have been indicated by each homily, because throughout the Catholic Homilies Ælfric was able to name the patristic writer when authenticating a particular interpretation for which Paul's homiliary was the immediate source.

³⁶ See below, Appendix 2 and Rädle, , Studien Zu Smaragd, pp. 138–42.Google Scholar

³⁷ Rädle, , Studien Zu Smaragd, p. 138.Google Scholar See also ibid. p. 140 for comments on Smaragdus's failure to distinguish between ultimate and immediate sources (although it should be noted that this is not a ‘failure’ in medieval terms, since the decision to validate by reference to ultimate authorities was purposeful, as we have already seen to have been the case with Ælfric). In ‘Contributions’, p. 584Google Scholar, Souter suggested that Smaragdus borrowed the system of marginal attribution from Bede. For Bede's practice, see Laistner, M.L.W., ‘Source-marks in Bede Manuscripts’, JTS 34 (1933), 350—4. It is striking that Laistner's examples are from manuscripts of Bede's commentaries on Mark and Luke, which Smaragdus frequently used.CrossRefGoogle Scholar

³⁸ The collations were made by Souter in a series of articles listed above, n. 15.

³⁹ See below, Appendix 2.

⁴⁰ Souter, ‘Contributions’, p. 595. The attribution to Bede is a compelling reason for rejecting the homily by Heiric as a source (see above, n. 31) because there the etymology is attributed (correctly from a modern point of view) to Jerome.Google Scholar

⁴¹ Förster, , ‘Über die Quellen’, p. 25 (§84).Google Scholar

⁴² Smetana, , ‘Ælfric and the Early Medieval Homiliary’, p. 194.Google ScholarThe part of Bede's commentary under consideration is p. 188, line 865–p. 194, line 1101, in Hurst's CCSL edition.Google Scholar

⁴³ PL 102, cols. 162–4.Google Scholar

⁴⁴ Rädle, , Studien Zu Smaragd, p. 217, identifies Smaragdus's source as Augustine's Tractatus in euangelium Iohannis, although he notes that this is much abbreviated.Google Scholar

⁴⁵ In his edition of The Homilies of Ælfric, pp. 303–10Google Scholar, Pope discusses the possibility that Ælfric's various treatments of the three miraculous resurrections were directly influenced by Augustine (principally the Tractatus in euangelium Iohannis), by Bede (the commentaries on Mark and Luke) and by Heiric's homily in the PL edition of Paul the Deacon's augmented homiliary (see above, nn. 31 and 40). He rejects Heiric (p. 308), but accepts Bede and Augustine's Tractatus. For the Tractatus he notes that Ælfric abbreviates. So also does Smaragdus (see above, n. 44), although Pope does not raise the question of an intermediate source. The discussion highlights the problem of intertextuality and illustrates Pope's patristic bias, but it has the value of confirming that the association of the three resurrections was a standard one.Google Scholar

⁴⁶ Förster, , ‘Über die Quellen’, p. 9 (§57). All references to Gregory's homilies will be to XL Homiliae in Evanglia, PL 76, cols. 1075–316. The homily under consideration here is at cols. 1077–81.Google Scholar

⁴⁷ Smetana, , ‘Ælfric and the Early Medieval Homiliary’, p. 195.Google Scholar

⁴⁸ ‘Then there shall be a new heaven and a new earth.’

⁴⁹ PL 102, cols. 518–19. The quotation from Revelation is at col. 519C-D. Rädle, , Studien Zu Smaragd, p. 215, gives Bede as Smaragdus's sole source.Google Scholar

⁵⁰ It occurs on p. 371, line 333, of the CCSL edition.Google Scholar

⁵¹ Förster, , ‘Über die Quellen’, pp. 22–3Google Scholar (§79). Bede's homily will be cited by page and line number from Bedae Venerabilis Opera. Pars III: Opera Homiletica, ed. Hurst, D., CCSL 122 (Turnhout, 1955).Google Scholar

⁵² Smetana, , ‘Ælfric and the Early Medieval Homiliary’, p. 196.Google Scholar The tradition of the six ages is discussed by Tristram, H.L.C., Sex aetates mundi: Die Weltzeitalter bei den Angelsachsen und den Iren: Untersuchungen und Texte (Heidelberg, 1985).Google ScholarShe examines Ælfric's treatments of this theme on pp. 125–31Google Scholar although, in identifying the source for CH II. iv, she simply cites Förster and Smetana (p. 126, n. 22).Google ScholarCH II. iv is also discussed by Gatch, M.McC., Preaching and Theology in Anglo-Saxon England: Ælfric and Wulfstan (Toronto and Buffalo, 1977), but he likewise accepts the sources identified by Smetana (p. 77, n. 3).Google Scholar Tristram and Gatch both include in their references a possible supplementary source in Haymo identified by Smetana; on this see below, pp. 220–3.Google Scholar

⁵³ For Smaragdus's exegesis, see PL 102, cols. 84–90.Google Scholar

⁵⁴ Rädle, , Studien Zu Smaragd, pp. 216—17. The divisions which Rädle makes in relating Smaragdus's exegesis to its actual sources do not correspond to my comparison between Bede and Smaragdus because the degree of intertextuality within the exegetical tradition means that correspondences may continue beyond the point where, on other grounds, one can show that the actual source has changed.Google Scholar

⁵⁵ We cannot refer to Worcester F. 91 for this homily because it was in the part of the manuscript now lost (see below, Appendix 2).

⁵⁶ Rädle, , Studien Zu Smaragd, p. 216, agrees that the correspondences with Bede noted thus far are passages for which Bede is the actual source. In Migne's text the Bedan attributions are succeeded by an attribution to Augustine at col. 85D, but this does not necessarily indicate that Smaragdus's immediate source had changed.Google Scholar

⁵⁷ Smaragdus's immediate source here, as noted by Rädle, ibid, p. 217, is Alcuin's Commentaria in sancti Joannis evangelium, but there are textual similarities because Augustine's Tractatus in lohannis underlies Alcuin and Bede. For Bede's use of Augustine here, see p. 97Google Scholar in Hurst's edition, note to line 73. Souter, ‘Contributions’, p. 589Google Scholar, reports manuscript attribution to Augustine at ‘Nondum venit hora’ and he implies that the slightly earlier attribution to Augustine in Migne's text, at ‘Neque enim’, col. 85D, is an error. There is no doubt, however, that, despite the acknowledgement of Augustine (the ultimate source), Smaragdus's immediate source was Alcuin's modified version of Augustine.

⁵⁸ ‘The teacher Bede said that the Lord condescended to go to an earthly marriage as a guest and he hallowed it with the first sign of his miracles, because he wished to show that those marriages are praiseworthy which are made for the procreation of children rather than for lust.’

⁵⁹ ‘According to the spiritual meaning …’

⁶⁰ In my article ‘Ælfric's Use of Etymologies’, ASE 17 (1988), 35–44, at 38Google Scholar, I followed published source studies in taking Bede to be Ælfric's main source for this homily. I now think that he used Smaragdus, but this does not affect the general point I was then making, since Smaragdus, like Bede, provides etymologies only for Cana, Galilee and Babylon. Ælfric's inclusion of an etymology for Jerusalem (for which, see below, p. 223) thus remains as an example of his tendency to augment the etymological content of his main source.Google Scholar

⁶¹ As with the first half of the homily, this does not imply that Bede was Smaragdus's main source: see Rädle, , Studien Zu Smaragd, p. 217. Correspondences may result from the high level of intertextuality within the tradition as, for example, at PL 102, col. 89B, ‘Ergo Dominus … reservavit’, where the words are Bede's, but where Smaragdus's immediate source is Alcuin.Google Scholar

⁶² Smetana, ‘Ælfric and the Homiliary of Haymo of Halberstadt’, pp. 463—4.Google Scholar His position is, of course, based on the assumption that Bede is Ælfric's principal source for this homily. For Haymo's homily, see PL 118, cols. 126–37.Google Scholar

⁶³ ‘or the washing of hands’.

⁶⁴ Haymo (col. 132A-C) also uses the Cain and Abel story, but in a different way, which is not reflected in Ælfric's homily.

⁶⁵ Smetana, , ‘Ælfric and the Homiliary of Haymo of Halberstadt’, p. 464.Google Scholar

⁶⁶ Ælfric's Catholic Homilies, ed. Godden, , p. xliii.Google Scholar For the origin of this and other Latin notes in this manuscript, see Godden, , p. lxxxii.Google Scholar

⁶⁷ ‘… at which [scil. the end of the world] all sexual intercourse will cease and our mortal body will be turned to immortality …’

⁶⁸ ‘Bede the expositor said.’

⁶⁹ ‘The holy Gregory expounded for us the mystery of this gospel, beginning thus.’

⁷⁰ ‘The exposition of this gospel speaks in yet more various ways about these water-vessels and their meanings, but we fear that you will disregard those meanings already mentioned, if we tell you more about them.’

⁷¹ See p. 221.Google Scholar

⁷² ‘The expositor says.’

⁷³ For Souter's collations of the marginal attributions, see the articles listed above, n. 15.

⁷⁴ See p. 209.Google Scholar

⁷⁵ Souter, , ‘Contributions’, p. 589, notes that the attribution to Gregory in Migne should begin one sentence earlier, at ‘Mirum quomodo’. It is here that the attribution is given in Barlow 4.Google Scholar

⁷⁶ Rädle, , Studien Zu Smaragd, p. 214.Google Scholar

⁷⁷ For the attribution to Gregory by marginal letter abbreviation in Bede's commentary, see p. 176 in Hurst's CCSL edition. Both Bede and Smaragdus by this means provided visible evidence that they were transmitters of patristic orthodoxy; see further above, n. 37.Google Scholar

⁷⁸ For the sources, see Bede's commentary, pp. 176—7 (notes).Google Scholar

⁷⁹ Smetana, , ‘Ælfric and the Homiliary of Haymo of Halberstadt’, p. 467.Google Scholar

⁸⁰ Die Hirtenbriefe Ælfrics, ed. Fehr, B., Bibliothek der angelsächsischen Prosa 9 (Hamburg, 1914), repr.Google Scholar with supplementary introduction by Clemoes, P.A.M. (Darmstadt, 1964)Google Scholar, and Councils and Synods with other Documents relating to the English Church, I, A.D. 871—1066, ed. Whitelock, D., Brett, M. and Brooke, C.N.L. (Oxford, 1981), pp. 191–226Google Scholar, show, in their source identifications, the range of regulatory literature with which Ælfric was familiar.

⁸¹ On Thorpe's error in presenting this as a separate homily for the Wednesday in Easter Week, see Godden, M., ‘The Development of Ælfric's Second Series of Catholic Homilies’, ES 54 (1973), 209–16, at 210–11.Google Scholar

⁸² ‘… now you are gathered here. We imagine that you will not all be present here on that day when we shall read that gospel.’

⁸³ ‘Gregory expounded this gospel and considered why Peter or any of them took up again what he abandoned earlier.’

⁸⁴ Förster, , ‘Über die Quellen’, pp. 6–7 (§52). Gregory's homily is in PL 76, cols. 1183–8.Google Scholar

⁸⁵ Smetana, , ‘Ælfric and the Early Medieval Homiliary’, p. 198.Google Scholar

⁸⁶ For discussion, below, see p. 231.Google Scholar

⁸⁷ Rädle, , Studien Zu Smaragd, p. 218. The Alcuinian source, from the Commentaria in sancti Joannis evangelium, is in PL 100, cols. 995–8.Google Scholar

⁸⁸ Ibid. p. 218.

⁸⁹ Alcuin does not omit the sentence but he does have the phrasal variation which is the evidence that Smaragdus drew on him rather than Gregory directly.

⁹⁰ ‘The number of fish has more meaning than you can understand.’

⁹¹ Rädle, , Studien Zu Smaragd, p. 218, believes that col. 246B ‘Adiecit.. erant’ is from Augustine.Google Scholar

⁹² Cross, J.E., in ‘Ælfric – Mainly on Memory and Creative Method in Two Catholic Homilies’, SN 41 (1969), 135–55Google Scholar, and ‘The Literate Anglo-Saxon-On Sources and Disseminations’, PBA 57 (1972), 67–100, examines homilies in which Ælfric seems to be making connections between sources on the basis of a good ‘associative memory’. The arguments advanced in this article are not a basis for assuming that memory was of no significance to Ælfric, but they do suggest that a proper examination of his acknowledged Frankish intermediaries will reveal particular instances where the apparently Ælfrician assocation is already made, or where the association is easily explicable if one supposes a working method dependent on having three overlapping compilations to hand.Google Scholar

⁹³ I should like to record my thanks to Professors Peter Clemoes, Helmut Gneuss and Malcolm Godden for their comments and advice throughout the preparation of this article. I am, in addition, especially grateful to Professor Gneuss for his generosity in providing me with bibliographical information relating to the manuscripts discussed and for contacting the late Professor Bernhard Bischoff on my behalf with regard to the date and origin of Barlow 4.

⁹⁴ The information in this Appendix is as given by Gneuss, ‘A Preliminary List of Manuscripts’, unless otherwise stated.

⁹⁵ Bullough, D.A., ‘The Continental Background of the Reform’, Tenth-Century Studies: Essays in Commemoration of the Millennium of the Council of Winchester and ‘Regularis Concordia’, ed. Parsons, D. (London and Chichester, 1975), pp. 20—36, at 26.Google Scholar For a description, see Bishop, T.A.M., English Caroline Minuscule (Oxford, 1971)Google Scholar, no. 3. The manuscript contains anno tations in ‘Hand D’ which may be that of Dunstan himself: see Lapidge, M., ‘Schools, Learning and Literature in Tenth-Century England’, Sett Spol 38 (Spoleto, 1991), 951–1005, at 977.Google Scholar

⁹⁶ It is dated s. x by Bishop, English Caroline Minuscule, p. 2.Google Scholar

⁹⁷ Bishop, English Caroline Minuscule, no. 8.

⁹⁸ Lapidge, M., ‘Surviving Booklists from Anglo-Saxon England’, Learning and Literature in Anglo-Saxon England: Studies Presented to Peter Clemoes on the occasion of his Sixty-Fifth Birthday, ed. Lapidge, M. and Gneuss, H. (Cambridge, 1985), pp. 33–89, at 58–62 (no. 8 in list and commentary).Google Scholar

⁹⁹ There is also a fragment corresponding to PL 102, cols. 324A-6B as a flyleaf in Cambridge, Corpus Christi College 199. It is stated by Rädle, (Studien Zu Smaragd, p. 124) to be s. ix^med. I am grateful to Helmut Gneuss for informing me that, in the opinion of Bernhard Bischoff, the leaf is to be dated s. ix^med or s. ix² and that it comes from north-eastern France, probably from the area around Rheims. It is not included in the Gneuss list, however, because it may not have come to England before the second half of the sixteenth century. The manuscript with which it is bound is a late eleventh-century copy of Augustine's De trinitate.Google Scholar

¹⁰⁰ A Summary Catalogue of Western Manuscripts in the Bodleian Library at Oxford, ed. Madan, F., Craster, H.H.E. and Denholm-Young, N. (Oxford, 1937) II.1, 1044–5.Google Scholar

¹⁰¹ CatalogusLibrorum Manuscriptorum Bibliothecae Wigorniensis made in 1622–1623 by Patrick Young, ed. Atkins, I. and Ker, N.R. (Cambridge, 1944), p. 19.Google Scholar

¹⁰² Its date is given as s. xiⁱⁿ in the Gneuss ‘Preliminary List’ (no. 539), although I am told by Professor Gneuss that this is a typographical error for s. xi^ex. In The Life of St Æthelwold, ed. Lapidge, and Winterbottom, , p. lxi, n. 92Google Scholar, it is also (following Gneuss) said to be s. xiⁱⁿ and it is described as having been ‘written in England’, with which compare ‘Worcester’ in the Gneuss list, but it should be noted that Gneuss does not distinguish here between origin and provenance. It is s. xi in the Bodleian Summary Catalogue, p. 1044Google Scholar, and in Ker, N.R., Medieval Libraries of Great Britain, 2nd ed. (London, 1964), p. 208.Google ScholarIt is s. xiex in Atkins and Ker, Catalogus, p. 19Google Scholar and Schenkl, H., Bibliotheca patrum latinorum Britannica I, Sitzungsberichte der philosophisch-historischen Classe der kaiserlichen Akademie der Wissenschaften 121 (Vienna, 1890), IX. Abhandlung, no. 243.Google Scholar

¹⁰³ Souter, , ‘Prolegomena’, p. 570.Google Scholar

¹⁰⁴ Pächt, O. and Alexander, J.J.G., Illuminated Manuscripts in the Bodleian Library, 3 vols. (Oxford, 1966–1973) III, 6 (no. 40).Google Scholar

¹⁰⁵ Rädle, , Studien Zu Smaragd, p. 121.Google Scholar

¹⁰⁶ See below, p. 237.Google ScholarSouter, , ‘A Further Contribution’, p. 46, believes that the Boulogne manuscript must have ‘about as complete a list of these symbols as any MS of Zmaragdus now surviving’.Google Scholar

¹⁰⁷ See above, n. 37, for comment on the origin of these attributions and their occurrence in the manuscript tradition.

¹⁰⁸ Bishop, English Caroline Minuscule, no. 18. Bishop also notes the loss of a leaf between fols. 98 and 99.Google Scholar

¹⁰⁹ Hill, J., ‘Missing Leaves from Worcester Cathedral Library Manuscript F.91’, N&Q 236 (1991), 1–2.Google Scholar

¹¹⁰ Floyer, J.K., The Catalogue of Manuscripts preserved in the Chapter Library of Worcester Cathedral, rev. S. G. Hamilton, Worcester Historical Society (Oxford, 1906), p. 46. The work is listed as ‘Expositiones lectionum ex epistolis et evangeliis’, but it is not ascribed to any author.Google Scholar

¹¹¹ Schenkl, H., Bibliotheca patrum latinorum Britannica X, Sitzungsberichte der philosophisch-historischen Classe der kaiserlichen Akademie der Wissenschaften 139 (Vienna, 1898), IX. Abhandlung, no. 4319.Google Scholar

¹¹² Ker, , Medieval Libraries, p. 211.Google Scholar

¹¹³ Rädle, , Studien Zu Smaragd, p. 124, including n. 91 on the same page.Google Scholar

¹¹⁴ Bishop, , English Caroline Minuscule, p. 16 (no. 18) where he suggests further that it might come from the third quarter of the century. Bishop's is the fullest published description of the manuscript.Google Scholar

¹¹⁵ Gneuss, ‘Preliminary List’, no. 762.

¹¹⁶ In comparison with Barlow 4, Worcester F. 91 was lightly punctuated. The heavy punctuation in Barlow 4 would have facilitated oral delivery and, if this were the original intention (see the suggestion by Witters above, n. 16), it could be seen as a mark of the manuscript's relative closeness to the point of origin. The strengthening of the punctuation in Worcester F. 91 is probably also a reflection of the use made of the manuscript.

¹¹⁷ Worcester F. 91 also differs from Barlow 4 in its layout. Unlike Barlow 4, it is not written in double columns; the concordia which concludes many of the exegeses is not distinguished by spacing but is written run-on; and it uses marginal signs to signal biblical quotations, including the part of the lection being quoted at the start of its detailed exegesis.

¹¹⁸ For the information about Hedio's edition and Pitra's editorial practices in preparing the PL text, I am indebted to Souter, ‘Contributions’, pp. 584—5Google Scholar, and to the comments in Rädle, , Studien Zu Smaragd, especially pp. 127–8.Google Scholar

¹¹⁹ Souter, , ‘A Further Contribution’, p. 46.Google Scholar

¹²⁰ Witters, ,‘Smaragd au moyen age’, p. 367.Google Scholar

¹²¹ Rädle, , Studien Zu Smaragd, p. 120.Google Scholar

¹²² See above, n. 15, for a list of Souter's publications on this subject.

¹²³ Some of the variations I have noted are simply orthographical. Since orthographical variants are also present in the medieval manuscripts, these cannot be regarded as substantive errors, even though one is conscious that many of the variants in the printed text arose through Hedio's exercise of editorial practices which would not be acceptable today.