¹ Bradley, D. R. (‘The Glosses on Bella Parisiacae Urbis 1 and 11’, Classica et Mediaevalia 28 (1967), 344–56)Google Scholar has questioned the correctness of the word urbs instead of polis (which would be more consonant with the poem's vocabulary) in the title currently employed. In fact Abbo uses both words. What is remarkable is the use of descidium in Æthelwold's booklist, as the word does not occur either in the poem or in its glosses.

² The inventory of Æthelwold's gifts (which includes twenty-one books) is recorded in a cartulary, now London, Society of Antiquaries 60. The charter is listed as no. 1448 in Sawyer, P. H., Anglo-Saxon Charters. An Annotated List and Bibliography (London, 1968)Google Scholar; it has been printed and analysed on several occasions. For a detailed discussion of its content, see Lapidge, M., ‘Surviving Booklists from Anglo-Saxon England’, Learning and Literature in Anglo-Saxon England, ed. Lapidge, M. and Gneuss, H. (Cambridge, 1985), pp. 33–89, at 52–5.Google Scholar The importance of this inventory has also been stressed by Gneuss, H., ‘Englands Bibliotheken im Mittelalter und ihr Untergang’, Festschrift für Walter Hübner, ed. Riesner, D. and Gneuss, H. (Berlin, 1964), pp. 91–121, at 98.Google Scholar Abbo's work is not mentioned in the Peterborough booklist of c. 1100 in Oxford, Bodleian Library, Bodley 163 (S.C.2016): see Lapidge, , ‘Surviving Booklists’, pp. 76–82Google Scholar. Three of the books listed in the abbey's fifteenth-century Matricularium (ed. James, M. R., ‘Lists of Manuscripts formerly in Peterborough Abbey Library’, Supplement to the Bibliographical Society's Transactions 5 (1926), 1–104, at 30–81 (nos. 125, 133, 136))Google Scholar are said to contain ‘Versus Abonis’ but, as with other catalogue entries, there is no way to decide if they refer to a work by Abbo of Saint-Germain-des-Prés or by Abbo of Fleury.

³ All known manuscripts but one contain only the third book of the poem and the title in Æthelwold's list could not be guessed either from the content of the third book or from any surviving incipit or explicit. The sole manuscript containing the whole poem is Paris, Bibliothéque Nationale, lat. 13833 (= P);and this is unlikely to have left the abbey of Saint-Germain-des-Prés before the French Revolution (except for a short period when it was in the hands of Pierre Pithou).

⁴ Abbo apparently completed his poem within ten years of the siege, in 896 or 897, since it contains no allusion to Odo's death (1 January 898). The beginning of the composition must be dated back to 888, since Odo (who was crowned on 29 February 888) is called ‘rex … futurus’ at 1.45. For the historical events described in the poem, see the classic (though dated) study by Vogel, W., Die Normannen und das fränkische Reich bis zur Gründung der Normandie (799–911) (Heidelberg, 1906; repr. Aalen, 1973), pp. 320–76Google Scholar for the years 885–96. Among recent works on the Viking raids, see the novel and stimulating approach by D'Haenens, A., ‘Les invasions normandes dans l'Empire franc au IX^e siècle’, Settimane di studio del Centro italiano di studi sull' Alto Medioevo 16 (1969), 233–98.Google Scholar

⁵ We do not know much about Abbo's life. We get some glimpses from his poem (he says that he was born in Neustria in a gloss to 1.624) and from the prefatory epistle. From the Necrologium S. Germani (ed. Molinier, A., Recueil des historiens de la France. Les obituaires de la Province de Sens 1. 1 (Paris, 1902), 253)Google Scholar we know that he died on 9 March but not in which year. He must have died after 922 since, at the request of Fulrad, bishop of Paris (921–7) and Frotarius, bishop of Poitiers, he wrote a collection of sermons which are contained (inter alia) in Paris, BN lat. 13023 and are still unedited, apart from five ptd Migne, Patrologia Latina 132, cols. 761–78 (from an earlier edition by D'Achery). Abbo's sermons had a wide circulation and were known in England: Wulfstan translated his ‘Sermo in cena domini ad poenitentes’ into English (Bethurum, D., The Homilies of Wulfstan (Oxford, 1957), pp. 366–72)Google Scholar and drew material from others.

⁶ All quotations and references are to Abbonis Bella Parisiacae Urbis ed. von Winterfeld, P., Monumenta Germaniae Historica, Poetae Latini Aevi Carolini iv. 1 (Berlin, 1899), 72–122Google Scholar; citations of the Scedula are to page and line of this edition.

⁷ See, inter alia, Löwe, H., ‘Geschichtschreibung der ausgehenden Karolingerzeit’, DAEM 23 (1967), 1–30 at 13–14Google Scholar, repr. in his Von Cassiodor zu Dante (Berlin and New York, 1973), pp. 180–205, at 190–1; Ganshof, F. L., ‘L'historiographie dans la monarchic franque sous les Mérovingiens et les Carolingiens. Monarchic franque unitaire en France Occidentale’, Settimane di studio del Centro italiano di studi sull' Alto Medioevo 17 (1970), 631–85, at 655–6Google Scholar; Prinz, F., Klerus und Krieg im früheren Mittelalter, Monographien zur Geschichte des Mittelalters 2 (Stuttgart, 1971), 129–32Google Scholar; Zettel, H., Das Bild der Normannen und der Normanneneinfälle in westfränkischen, ostfränkischen, und angelsächsischen Quellen des 8. bis 11. Jahrhunderts (Munich, 1977), p. 44Google Scholar and passim. For a positive evaluation of the literary value of the first two books, see Ploss, E., ‘Das 9. Jahrhundert und die Heldensage. Eine kritische Betrachtung der Zeugnisse’, Zeitschrift für deutsche Philologie 89 (1970), 3–34, at 20Google Scholar, and Soubiran, J., ‘Prosodie et métrique des Bella Parisiacae Urbis d'Abbon’, Journal des Savants, 1965, 204–331.CrossRefGoogle Scholar

⁸ For example, Abbo gives precise descriptions of the Frankish siege engines; see Gillmor, C. M., ‘The Introduction of the Traction Trebuchet into the Latin West’, Viator 12 (1981), 1–8.CrossRefGoogle Scholar

⁹ Winterfeld silently emends the reading of the manuscript, scidula (which is that ptd by all other editors). All the Latin words from Abbo's poem and its glosses will be given in the form in which they appear in Winterfeld's edition.

¹⁰ There are at least 120 Greek loanwords (in some cases, e.g. congius, the etymology is still dubious, but we should also take into consideration the medieval etymologies: cf. aliqua (= alica: line 80) with Isidore, Etym. xvii.iii.9: ‘alica Graecum nomen est’), amounting to about half of the glossed words of the third book. On contemporary knowledge of Greek see, inter alia, Laistner, M. L. W., ‘The Revival of Greek in Western Europe in the Carolingian Age’, History 9 (1924), 177–87CrossRefGoogle Scholar; Bischoff, B., ‘Das griechische Element in der abendländischen Bildung des Mittelalters’, Byzantinische Zeitschrift 44 (1951), 27–55CrossRefGoogle Scholar, repr. in his Mittelaltcritche Studien 11 (Stuttgart, 1967), 246–75; Prinz, O., ‘Zum Einfluss des Griechischen auf den Wortschatz des Mittellateins’, Festschrift Bernhard Bischoff, ed. Autenrieth, J. and Brunhölzl, F. (Stuttgart, 1971), pp. 1–15Google Scholar; Le Bourdellès, R., ‘Connaissance du grec et méthodes de traduction dans le monde carolingien jusqu' à Scot Erigène’, Jean Scot Erigène et l'histoire de la philosophic, Colloques internationaux du Centre National de la Recherche Scientifique 561 (Paris, 1977), 117–23Google Scholar; Jeauneau, E., ‘Jean Scot Erigène et le grec’, Archivum Latinitatis Medii Aevi 41 (1979), 5–50Google Scholar; and the comprehensive treatment of the subject by Berschin, W., Griechisch-lateinisches Mittelalter. Von Hieronymus zu Nikolaus von Kues (Bern and Munich, 1980)Google Scholar.

¹¹ On the cleric as ‘Christ's heir’, see Isidore, Etymologiae vii.xii.1 (De clericis). Similar associations lie behind Abbo's gloss in the Scedula (p. 78.6): cleronomos: clericos. Cleronomiagrece, latine ereditas; inde cleronomus.i. heres dei. Cleronomus has no gloss in the third book, because it belongs to the ‘frame’ not to the ‘key’ words. In my view it was Abbo himself who glossed bk 111 as well as bks 1 and 11 (for a contrary view, see Bradley, ‘The Glosses’).

¹² ‘Let the belly be your adversary.’ The rest of the sentence echoes St Paul's Letter to the Philippians 111. 19. The Pauline epistles, with their long lists of moral advice, are frequently echoed by Abbo: cf. 111.29 and I Cor. v.8; 111.40 and I Cor. ix.24; 111.77 and II Cor. 111.14–15; 111.84 and I Cor. vii.8. Line 79 (‘leva ancile geras, dextra agoniamque fidei’) combines Ephes. vi. 16–17 with Vergil, Aen. vii. 188. In bk m there are numerous echoes from the Bible, Vergil and other authors.

¹³ No manuscript preserves the correct spelling agonotheta. Winterfeld records – unsystematically – only the major variant readings from the manuscripts he knows, which are, besides P, A (but he does not mention the prose version), C, H, Q, S, R, V and B (prose version): for these manuscripts see below. A thorough revision of Winterfeld's edition of the third book is an unavoidable first step towards a new edition of the prose version with Old English gloss.

¹⁴ See Weise, F. O., Die griechischen Wörter im Latein, Preisschriften gekrönt und herausgegeben von der fürstlich Jablonowski'schen Gesellschaft zu Leipzig 23 (Leipzig, 1882; repr. 1964), 332.Google Scholar

¹⁵ The meaning ‘fighter’ is attested in the Scholica graecarum glossarum (‘agonitheta: praeliator vel praemii auctor sive qui agoni praesidet’) and elsewhere. The Scholica item conflates a series of interpretamenta from the Liber glossarum: agonotheta is glossed as praemii auctor (AG 173), proeliator(AG 174) and qui agoni praesidet (AG 175), Glossaria Latina iussu Academiae Britannicae edita, ed. Lindsay, W. M., 5 vols. (Paris, 1926–1931; repr. Hildesheim, 1965).Google Scholar

¹⁶ Lemma and gloss occur both in the Scholica and in the Liber glossarum. For other occurrences, see Goetz, G., Corpus Glossariorum Latinorum a Gustavo Loewe incohatum, 7 vols. (Leipzig, 1888–1923; repr. Amsterdam, 1965)Google Scholar (hereafter CGL )v, 590.23 (from a glossary which draws from the Scholica) and v, 615.31.

¹⁷ There are other works of this period bearing glosses by their authors, such as the Gesta Berengarii Imperatoris (ed. P. von Winterfeld MGH, PLAC iv, 354–403), but none bears comparison with Abbo's third book. For a study of the glossing of bks 1 and 11, see Löfstedt, B., ‘Zu den Glossen von Abbos Bella Parisiacae Urbis’, SM 22 (1981), 261–6.Google Scholar

¹⁸ The glossed words are 296 out of a total of 698 (reckoning conjunctions with the exception of the enclitic - que). Some words display two (48 × ) or three (6 × ) glosses of the same or of a different type: e.g., lexical + lexical or lexical + grammatical, amounting to a total of 356 glosses.

¹⁹ 3 lines have no glossed words, 10 have one, 43 have 2, 41 have 3, 15 have 4 and 3 have 5.

²⁰ 326 out of 356. One gloss (111.78: noxis:.i. pro culpis) is both lexical and grammatical.

²¹ Consider for example: toparcha:.i. diabolus erebi (111.3); hostis: demonis (111.76); sinislri: diaboli (111.91); cloaca:.i. fossa Tartari (111. 4); baratrum: infernum (111.36) and antro: inferno (111.103). Analogous explications are given for sacrata per ora:.i. per ewangelistas (111. 108) and arcisterium:.i.singularitatem dei servitii (111.81).

²² Silemsis (= syllepsis, from Greek σύλλΨις): 111.46.

²³ For example, the name of the case is once given over an ambiguous form: anime: dativus (111. 100). All these glosses fall within categories commonly employed by medieval glossators. What is interesting is that the grammatical glosses are dropped more often than the lexical ones in the course of the text's transmission.

²⁴ Winterfeld has omitted two letters (μ over abbachus (111.33) and ø over anodiam (111.9)). In his edition he does not explain and treats inconsistently the relative position of multiple glosses and of the letters placed over the same word.

²⁵ Paris, BN lat. 13833 is not the autograph, as was suggested by Winterfeld but, as Bernhard Bischoff (pers. comm.) kindly informs me, ‘nur … eine (fast gleichzeitige) Abschrift saec. ix^ex oder ix/x’.

²⁶ MGH, PLAC iv, 73 and n. 7.

²⁷ 93 ‘key’ words (8 of them occurring twice) begin with A; 27 (2 occurring twice) with B; and 43 (4 occurring twice) with C; for the other letters the figures are smaller (D: 10, E: 12, F: 10, G: 7, H: 5, 1: 5, L: 6, M: 7, N: 6, O: 7, P: 14, Q: 5, S: 14, T: 11, U/V: 9, Y: 1, Z: 1, Ψ: 1 and Θ: 1), but see my remarks above, n. 9.

²⁸ For this aspect of the history of glossaries, see Lindsay, W. M., ‘The Affatim Glossary and Others’, Classical Quarterly 11 (1917), 185–200CrossRefGoogle Scholar. Scribal practices and the way they affected the diffusion of the Liber glossarum, are examined by Bishop, T. A. M., ‘The Prototype of the Liber glossarum’ Medieval Scribes, Manuscripts and Libraries. Essays presented to N. R. Ker, ed. Parkes, M. B. and Watson, A. G. (London, 1978), pp. 69–86.Google Scholar

²⁹ See Goetz, G. ‘Der Liber Glossarum’, Abhandlungen der philologisch-historischen Classe der königlich sächsischen Gesellschaft der Wissenschaften 13 (Leipzig, 1893), 213–88.Google Scholar

³⁰ Winterfeld's and Laistner's estimates were distorted by their limited knowledge of the Liber glossarum. Only excerpts from this glossary had been printed in CGL v, and Lindsay's edition in the first volume of the Glossaria Latina (1926) was posterior to Laistner's first articles.

³¹ Beside these, there are about twenty lemmata with similar glosses, while fifty of the words chosen by Abbo have interpretamenta different from those in the Liber glossarum. Both groups testify to the high rate of glossary-vocabulary employed in the third book.

³² Similarly, one third of the Scholica begins with the letters A–C. The numerous overlappings between the items of the Scbolica and of the Liber glossarum complicate research on Abbo's sources. There are 63 common items (41 of which have an identical gloss).

³³ This copy of the Liber glossarum comprises only the sections A–E.

³⁴ The gloss furiosus to cerritus (111. 32) is found only in the so-called ‘Second Amplonian’ (CGL v, 276.15) and may be drawn from this source. In the same glossary (CGL v, 287.31) we find a parallel to diamant: valde amant (111. 17) which is drawn ultimately from Servius's Commentary on Aen. viii.428.

³⁵ For example, strabo: welcus (111.54). The third book of the poem has been made object of a careful analysis by M. Arullani, ‘Un glossario in versi del IX secolo’, Attt dell' Accademia degli Arcadi 16 (1932), 23–85. Many of her conclusions are, however, undermined by ignorance of Laistner's articles and of Lindsay's edition of the Liber glossarum.

³⁶ Laistner, M. L. W., ‘Abbo of St-Germain-des-Prés’, Archivum Latinitatis Medii Aevi 1 (1924), 27–31Google Scholar, and ‘The Revival of Greek’, pp. 185–6.

³⁷ The first to put forward this attribution was Thomson, H. J. (‘Anaphus’, The Classical Rev. 34 (1920), 32–3)Google Scholar, who believed, like Laistner, that Vatican City, Reg. lat 215 came from Laon. See, besides the articles quoted in the preceding note, Laistner, M. L. W., ‘Candelabrum Theodosianum’, The Classical Quarterly 16 (1922), 107Google Scholar; idem, ‘Notes on Greek from the Lectures of a Ninth-Century Monastery Teacher’, Bull. of the John Rylands Lib. 7 (1923), 421–56; idem, ‘Martianus Capella and his Ninth-Century Commentators’, Bull. of the John Rylands Lib. 9 (1925), 130–8, at 131; idem, ‘Rivipullensis 74 and the Scholica of Martin of Laon’, Mélanges Mandonnet (Paris, 1930) 11, 31–7; and idem, Thought and Letters in Western Europe, A.D. 500 to 900 (Ithaca, 1957), pp. 215 and 244.

³⁸ Contreni, J. J., The Cathedral School of Laon from 850 to 930: its Manuscripts and Masters, (Munich, 1978), p. 114Google Scholar; idem, ‘The Formation of Laon's Cathedral Library in the Ninth Century’, SM 13 (1972), 919–39, at 938, n. 76; idem, ‘A propos de quelques manuscrits de l'école de Laonau IX^e siècle: découvertes et problèmes’, Le Moyen Age 78 (1972), 5–39, at 6 and n. 4; idem, ‘Le formulaire de Laon, source pour l'histoire de l'école de Laon au début du IX^e siècle’, Scriptorium 27 (1973), 21–9, at 23, n. 9; idem, ‘Martin Scottus (819–875) and the “Scholica Graecarum Glossarum”: a New Look at the Manuscripts’, Manuscripta 19 (1975), 70–1; idem, ‘The Biblical Glosses of Haimo of Auxerre and John Scottus Eriugena’, Speculum 51 (1976), 411–34, at 413 n. 18 and 426, n. 57; idem, ‘Three Carolingian Texts Attributed to Laon: Reconsiderations’, SM 17 (1976), 797–813, at 802–8. Contreni has changed his mind about the author of the Scholica: according to the last cited work we must look for the ultimate home of the compilation to Spain, the monastery of Ripoll providing the probable link with Saint-Germain-des-Pres.

³⁹ There is no reliable edition of the Scholica. Excerpts from Vatican City, Reg. lat. 215 have been printed in CGL v, 583–6. The edition of M. L. W. Laistner (‘Notes on Greek’, pp. 426–51) is based on a collation of this manuscript with London, BL, Royal 15. A. xvi.

⁴⁰ ‘Abbo of St-Germain-des-Prés’, p. 27.

⁴¹ 80 lemmata of the Scholica (+ 6 contained in the interpretamenta of other items + 1 in the miscellaneous notes following the compilation in the Vatican manuscript) correspond to Abbo's ‘key’ words, but only 14 of these do not recur in other glossaries: anabola, baen (Abbo has baben), bule, bioticus (biotticus), chrisis (crisis), catascopus (catasscopus), entole, ierarchia, ieron, machia, palinodian, sinteca, teche and ton meson (temeson). Another 14 recur elsewhere but only in the Scholica have the same interpretamentum as in Abbo's poem: antigraphus, apocrisiarius (apocrisarus), apozima, cliothedrum (cliotedrum), choraula (corcula), diamoron, diametrum, dipticae, ephyppia (effipia), eniheca (enteca), ergastulum, horoscopus (oroscopus), propoma and zelotipium. In many instances the identity is limited to the lemma and not to the gloss. All cases where a word (and/or its gloss) does not occur exclusively in the Scholica, especially those shared with the Liber glossarum, must be ruled out of consideration as a possible source for Abbo.

⁴² Ed. Wattenbach, W., Levison, W. and Löwe, H., 5 vols. (Weimar, 1952–1973) v, 581.Google Scholar

⁴³ For example, H. Waquet excluded the third book from his edition of the poem ‘à l'abord rebutant et au caractère trop spécial, monument moins de l'histoire de la France que de celle du pédantisme’ (Abbon, Le siège de Paris par les Normands, ed. H. Waquet (Paris, 1942; repr. 1964), p. xvi). See also Petitjean, J., ‘Abbon l'humble: son poème sur le siège de Paris par les Normands’, Annales de la faculté des lettres de Caen 4 (1888), 61–74.Google Scholar

⁴⁴ On the background of this style and its fortune in England, see Lapidge, M., ‘The Hermeneutic Style in Tenth-Century Anglo-Latin Literature’, ASE 4 (1975), 67–111Google Scholar; Abbo's work is discussed at 71–2 and 75–6.

⁴⁵ See the two poems mentioned by Lapidge, ‘The Hermeneutic Style’, at pp. 84–5 and ptd ibid. 103–5.

⁴⁶ Aimoin was the author of De Miraculis S. Germani libri duo, a reworking of two earlier works on the miracles performed by the saint: see Patrologia Latina 126, cols. 1027–50.

⁴⁷ Leclercq, J., ‘Le florilège d'Abbon de Saint-Germain’, Revue du mqyen âge latin 3 (1947), 113–40.Google Scholar

⁴⁸ Consider, for example, the severe condemnation of greed and lust, or the stress on the difference between clerics and laymen.

⁴⁹ Opusculum LV Capitulorum, in Patrologia Latina 126, cols. 279–648, at 448. The existence of this amusing expostulation against the use of abstruse vocabulary uttered by a contemporary was pointed out for the first time by Goetz, G., ‘Uber Dunkel- und Geheimsprachen im späten und mittelalterlichen Latein’, Berichte über die Verhandlungen der königlich sächsischen Gesellschaft der Wissenschaften 48 (Leipzig, 1896), 62–92, at 73–4Google Scholar; see also the remarks by Lapidge, M., ‘L'influence stylistique de la poésie de Jean Scot’, jean Scot Erigène et l'histoire de la philosophie, pp. 441–51, at 444Google Scholar. Another self-ironic complaint by Ratherius of Verona in his Qualitatis coniectura cuiusdam (Patrologia Latina 136, col. 524) has been pointed out by P. Dronke, ibid. p. 452.

⁵⁰ On Laon in the ninth century see, beside the works by Contreni cited above, n. 38, Merlette, B., ‘Ecoles et bibliothèques à Laon, du déclin de l'antiquité au développement de l'universite’, Enseignement et vie intellectuelle (IX^e–XVl^ee siècle). Actes du 95^e Congrès National des Sociétés Savantes (Paris, 1975) 1, 21–53Google Scholar; S. Martinet, ‘Les arts libéraux à Laon au IX^e siècle’, ibid. 1, 55–62; E. Jeauneau, ‘Les écoles de Laon et d'Auxerre au IX^e siècle’, Settimane di studio del Centra italiano di studi sull Alto Medioevo 19 (1971), 495–522; J. J. Contreni, ‘The Irish “Colony” at Laon during the Time of John Scottus’, jean Scot Erigène, pp. 59–67, and several other studies in the same Laon colloquium.

⁵¹ ‘Disdain all the mud, the banquets, so that the journey may purify you. Live half-guilty if you cannot (live) guiltless.’

⁵² See Weise, , Die griechischen Wörter, p. 527.Google Scholar

⁵³ Without entering into the much disputed question of the difference between occasional borrowings and institutionalized ones, I do not think that the words in question here can even be considered ‘casuals’: see Gusmani, R., Saggi sull' interferenza linguistica 1 (Florence, 1981), pp. 16 and n. 18 and 25–6 and nn. 40–1.Google Scholar

⁵⁴ See the remarks by Berschin, Griechisch-lateinisches Mittelalter, esp. pp. 44 and 161–2.

⁵⁵ Abbo might have been led into error by an association with πηλός.

⁵⁶ In the last part of the third book the ‘frame’ vocabulary shows many parallels with the vocabulary of hymns (almus, laus, perhennis, sacrattts, scandere, testis) and the poem ends with a sort of doxology (111. 113–14) reminding the reader of the equal rank of the three persons of the Trinity.

⁵⁷ We know ten manuscripts of bk 111 of the Bella Parisiacae Urbis: Amiens, Bibliothèque Municipale 110 (= S)(Selincourt, s. xiii); Cambridge, Corpus Christi College 326 (= K)(Christ Church, Canterbury, s. x²); Cambridge, University Library, Gg. 5. 35 (= C) (St Augustine's, Canterbury, s. xi^med); Edinburgh, National Library of Scotland, Adv. 8. 6. 12 (= E)(Thorney, s. xi^ex); London, British Library, Harley 3271 (= A)(s. xi¹); London, BL, Harley 3826 (= H)(? Abingdon, s. x/xi); London, BL, Royal 3. A. vi (= R)(Reading Abbey, 5. xiii); Paris, Bibliothèque Nationale, lat. 5570, fols. 64–6 (= Q)(? Italy, s. x); Valenciennes, Bibliothèque Municipale 298 (= V)(Saint-Amand, s. xi); and Erfurt, Wiss. Allgemeinbibl., Amplon. 8⁰8 (s. xii¹). The last of these has previously been unknown to students of Abbo.

⁵⁸ Six manuscripts were written in England: see Lapidge, ‘The Hermeneutic Style’, pp. 75–6; for K, C, E, A and H see Gneuss, H., ‘A Preliminary List of Manuscripts written or owned in England up to 1100’, ASE 9 (1981), 1–60 (nos. 12, 93, 252, 435 and 438).Google Scholar

⁵⁹ The popularity of the poem is attested by echoes of it in the anonymous Ecbasis cuiusdam captivi and in the Carmina by Radbod of Utrecht. An inverse relationship between the Waltharius (which was once thought to echo Abbo) and the poem was put forward by Schumann, O., ‘Waltharius-Probleme’, SM 17 (1951), 177–202, at 180Google Scholar. His hypothesis has received both approval (Onnerfors, A., Die Verfasserschaft des Waltharius-Epos aus sprachlicher Sicht (Opladen, 1979), p. 68)CrossRefGoogle Scholar and rebuttal (Schieffer, R., ‘Zu neuen Thesen überden Waltharius’, DAEM 36 (1980), 193–201, at 196)Google Scholar. On the possible knowledge of Abbo's poem in Breton milieux, see Waquet, H., ‘Abbo de Saint-Germain-des-Prés et la Ville d'ls’, Bulletin philologique et historique du Comité des travaux historiques et scientifiques (1951–1952), 75–81Google Scholar. The first line of the third book has also found a place in medieval collections of witty sententiae: see Walther, H., Lateinische Sprichwörter und Sentenzen des Mittelalters in alphabetischer Anordnung, 5 vols. (Göttingen, 1963–1967) 1, no. 2826.Google Scholar

⁶⁰ For evidence from various fields see, inter alia, Wallace-Hadrill, J. M., ‘The Franks and the English in the Ninth Century: Some Common Historical Interests’, History 35 (1950), 202–18CrossRefGoogle Scholar; John, E., ‘The Sources of the English Monastic Reformation: a Comment’, RB 70 (1960), 197–203Google Scholar; Gneuss, H., Hymnar und Hymnen im englischen Mittelalter, Buchreihe der Anglia 12 (Tübingen, 1968), 55–74CrossRefGoogle Scholar; Bishop, T. A. M., English Caroline Minuscule (Oxford, 1971)Google Scholar; and Rella, F. A., ‘Continental Manuscripts acquired for English Centers in the Tenth and Early Eleventh Centuries: a Preliminary Checklist’, Anglia 98 (1980), 107–16.Google Scholar

⁶¹ Recent studies have stressed the differences between Aldhelm's vocabulary and style and that of hermeneutic writers like Abbo; see Winterbottom, M., ‘Aldhelm's Prose Style and its Origins’, ASE 6 (1977), 39–76Google Scholar and Marenbon, J., ‘Les sources du vocabulaire d'Aldhelm’, Archivum Latinitatis Medii Aevi 41 (1979), 75–90.Google Scholar

⁶² For other examples of the transmission of a Latin text with its apparatus of Latin glosses from the Continent to England (although none bear comparison with Abbo's third book for the density of the glossing and its relationship to the text), see Lapidge, M., ‘The Study of Latin Texts in Late Anglo-Saxon England: 1. The Evidence of Latin Glosses’, Latin and the Vernacular Languages in Early Medieval Britain, ed. Brooks, N. (Leicester, 1982), pp. 99–140.Google Scholar

⁶³ The statistics which might be adduced in support of this statement are too complex to be set out here; I hope to print them on another occasion.

⁶⁴ It contains 111. 1–32, 34–5, 37–44, 50–3, 55–9, 63–4, 66–7, 69–73, 75–6, 79, 81–95, 103–8 and 96.

⁶⁵ The lemmata correspond to 197 ‘key’ words (one lemma has no gloss). The rate of reduction of the glosses, which is higher than that of the other manuscripts, is due to the nature of the rearrangement.

⁶⁶ A manuscript of the tenth century from St Augustine's, Canterbury: see H. Gneuss, ‘A Preliminary List’, no. 326.

⁶⁷ Other examples of separate collections of glosses and lemmata drawn from the same work are: the glossary drawn from Bede's Historia Ecclesiastica (Book 1, chs. x–xxii) in London, BL, Cotton Tiberius C. ii, and that from Bede's metrical Vita S. Cuthberti in Cambridge, Corpus Christi College 183. There are glossaries drawn, entirely or in part, from Aldhelm's works: e.g. in Oxford, Bodleian Library, Auct. F. 2.14 (S.C. 2657), Bodley 163 (S.C. 2016) and Einsiedeln, Stiftsbibliothek 32 (the last item is erroneously entered in The Plan for the Dictionary of Old English, ed. Frank, R. and Cameron, A. (Toronto, 1973) under C. 32. 6).Google Scholar

Anecdota Oxoniensia, Mediaeval and Modern Ser. 15 (Oxford, 1929), 103–12 (both with variant readings from Oxford, St John's College 154 in the apparatus, whose text is also used to supply the words missing from the Harley gloss). All my quotations from the prose version and the Old English glosses are from Stevenson's edition.

to supply the words missing from the Harley gloss. All my quotations from the prose version and the Old English glosses are from Stevenson's edition.

⁶⁹ This version has been ptd, with a good many errors, in Leechdoms, Wortcunning and Starcraft of Early England, ed. Cockayne, T. O., 3 vols., Rolls Ser. (London, 1864–1866) 1, lxiii–lxviGoogle Scholar. Failure to recognize that the text printed in this volume was the other version of Abbo's poem brought about a series of misunderstandings, e.g. in the entries of BT, BTS and CH and consequently in the estimates of the hapax legomena by Waldorf. The erroneous distinct entry in The Plan (C.1. 1–2) of Zupitza's and Stevenson's editions, which differ only in a few editorial choices, is followed by A Microfiche Concordance to Old English ed. Healey, A. diPaolo and Venezky, R. L. (Toronto, 1980)Google Scholar.

⁷⁰ There are two OE glosses embedded in the text in a list of jewels and six more in a list of names of birds (all drawn from Isidore's Etymologiae) in Paris, BN lat. 1750, ptd Riehle, W., ‘Uber einige neuentdeckte altenglische Glossen’, Anglia 84 (1966), 150–5Google Scholar. The commentary of Hrabanus Maurus to the Pentateuch in St Gallen, Stiftsbibliothek 283 displays Old High German and Old English contextual glosses. Other contextual glosses are found in the commentary to Leviticus in Karlsruhe, Landesbibliothek Aug. 231 (119) (both erroneously entered under D 49 in The Plan). There are also a few glosses in the text of the Psalms in Vatican City, Pal. lat. 68 (entered in The Plan as B. 8. 3).

⁷¹ The inverse arrangement – Old English interpretations written as interlinear glosses – is found in two manuscripts of Ælfric's Grammar: Cambridge, Trinity College R. 9. 17 (819), fols. 1–48 and London, BL, Royal 15. B. xxii.

⁷² This arrangement might have been determined also by the recognition of the glossary-like value of Abbo's third book or by the propaedeutic function of this prose rendering which in Harley 3271 immediately precedes the text in its original metrical version.

⁷³ The two prose versions derive from a common ancestor: they exhibit the same word order, commonadditions (e.g. sit…sit…tibi: line 4), omissions (e.g. ΨΙΧΗ in line 38 and ΘΗΟ in line 111) and substitutions (e.g. obrissis, which is probably a former gloss, instead of crisis at line 39). But they could not be copied directly from one another: while the mutilated condition of the version in St John's College 154 rules out one of the possibilities, the other is ruled out by the omissions (which are sometimes rectified by additions above the line: e.g. liba and its gloss drinc at line 43) in Harley 3271 and seems confirmed by errors and divergencies (e.g. accipito for ac capito at line 40) which are not shared by St John's College 154.

⁷⁴ See Gneuss, Hymnar und Hymnen, pp. 91–101, 135–55 and 194–206, and Korhammer, M., Die monastischen Cantica im Mittelalter und ihre altenglischen Interlinearversionen. Studien und Textausgabe, Texte und Untersuchungen zur Englischen Philologie 6 (Munich, 1976), 128–38Google Scholar. For a valuable analysis of the rules underlying prose versions see Korhammer, M., ‘Mittelalterliche Konstruktionshilfen und altenglische Wortstellung’, Scriptorium 34 (1980), 18–58.CrossRefGoogle Scholar Syntactic patterns analogous to those of prose versions are provided by syntactical glosses; see Robinson, F. C., ‘Syntactical Glosses in Latin Manuscripts of Anglo-Saxon Provenance’, Speculum 48 (1973). 443–75.CrossRefGoogle Scholar

⁷⁵ The prose version of Abbo's third book and its relationship with the Old English glosses will be analysed in a forthcoming article in AIUON.

⁷⁶ ‘Let (God's) precept adorn you and let the necklace (adorn) the kings and the nobles.’ I assign to baben the meaning found in the Scholica, but the story of this word, which occurs for the first time in the Vulgate (I Mach. xiii.37and 51 and John xii. 13) would require a whole article by itself.

⁷⁷ Reges (acc.pl.), or a corrupt regis, has been taken for a gen. sing.

⁷⁸ ‘Decline the two-wheeled cart and take the victor's crown and the throne.’ For this meaning of cliotedrum, see exedra vel cliotedum: sedes episcopalis (CGL v, 618.51).

⁷⁹ There is only one (brief) article in print on the OE gloss to Abbo: L(iebermann), F., ‘Zur angelsächsischen Abbo-Glosse’, ASNSL 112 (1904), 391.Google Scholar

⁸⁰ Gretsch, M. (Die Regula Sancti Benedicti in England und ihre altenglische Ubersetzung, Texte und Untersuchungen zur Englischen Philologie 2 (Munich, 1973)Google Scholar) remarks that ‘Eine Gesamtdarstellung der verschiedenen Techniken der Ubersetzung in der ae. Literatur wäre noch zu schreiben’ (p. 238); beside heranalysis at pp. 235–306, which is extremely valuable see, for many enlightening remarks, Sauer, H., Theodulfi Capitula in England. Die altenglischen Ubersetzungen, zusammen mil dem lateinischen Text, Texte und Untersuchungen zur Englischen Philologie 8 (Munich, 1978), 118–64.Google Scholar

⁸¹ See my remarks in ‘II Colloquio di Ælfric e il colloquio di Ælfric Bata’, ‘feor ond neah’. Scritti di Filologia germanica in memoria di Augusto Scaffidi Abbate, ed. Lendinara, P. and Melazzo, L., Annali della Facoltà di Lettere e Filosofia dell'Università di Palermo, Studi e ricerche 3 (Palermo, 1983), 173–250.Google Scholar