1 The most extensive studies are Étienne Gilson, ‘Sur le Iesu dulcis memoria’, Speculum, 3 (1928), 322–34, repr. as ‘La mystique cistercienne et le Iesu dulcis memoria’ in his Les idées et les lettres (Paris, 1932), 39–57; André Wilmart, Le ‘jubilus’ sur le nom de Jésus dit de Saint Bernard, Ephemerides liturgicae: Analecta historico-ascetica, 57 (Rome, 1943), repr. as Storia e letteratura, 2 (Rome, 1944); and Heinrich Lausberg, Der Hymnus ‘Jesu dulcis memoria’, Hymnologische Studien, 1 (Munich, 1967). The poem is also mentioned in numerous studies and anthologies of medieval Latin religious verse, among them Frederic James Edward Raby, A History of Christian Latin Poetry from the Beginnings to the Close of the Middle Ages (2nd edn, Oxford, 1966), 329–30.

2 Wilmart mentions the music in two sentences of his 250-page study (p. 32); Lausberg's 500-page book devotes 12 of those pages (pp. 55–66) to the musical setting, of which six are taken up with a transcription: this forms the only published edition of the music so far available in print, but is unsatisfactory as a critical text since Lausberg normalized many of the manuscript variations, making the melody appear to follow a much more regular repetitive structure than it in fact does. Appendix 1 below presents a new transcription showcasing these variants, whose significance is analysed on pp. 19–24 below.

3 An edition of the song set in the context of the other songs preserved in British manuscripts of the twelfth and thirteenth centuries can now be found in Songs in British Sources, c.1150–1300, ed. Helen Deeming, Musica Britannica, 95 (London, 2013), 15–20 (edition) and 172–3 (commentary).

4 It is, of course, impossible to know if earlier manuscripts once existed that have not survived, or whether such lost sources would have contained music or not. But the text's references to the work of twelfth-century authors suggests that it is unlikely to have originated much before the earliest witness that is currently known; see below for further analysis of the textual and transmission histories.

5 The Feast of the Holy Name of Jesus was not officially incorporated into the Roman calendar until 1791, but had been in use in the Franciscan, Dominican and Sarum commemorations, as well as a number of other ‘unofficial’ devotions, for some three centuries by then. On these liturgical and devotional adoptions of the song, see Wilmart, Le ‘jubilus’, chapter 3.

6 Wilmart, Le ‘jubilus’, 111, notes that ‘the original poem had certainly not been written to serve the fulfilment of solemn liturgy […]. Furthermore, [those who selected only certain stanzas for liturgical use] caused it serious harm by keeping only scraps, chosen with little discernment’ (‘l'ancien rhythme n'avait certainement pas été écrit pour servir à l'accomplissement de la liturgie solennelle […]. De plus, on lui fait un grave tort en n'en retenant que des bribes, choisies avec peu de discernement’).

7 Wilmart's inventory of manuscripts is in Le ‘jubilus’, 10–47; see also chapter 3 of the same volume for his study of the later transmission of the poem in printed books, liturgical appropriations and modern editions.

8 Samuel Harrison-Thomson, ‘The Dulcis Jesu memoria in Anglo-Norman and Middle French’, Medium aevum, 11 (1942), 68–76; Wilmart, Le ‘jubilus’, appendices; Denis Renevey, ‘Anglo-Norman and Middle English Translations and Adaptations of the Hymn Dulcis Iesu memoria’, The Medieval Translator, 5 (1996), 264–83.

9 Walter Howard Frere, Bibliotheca musico-liturgica, 2 vols. (London, 1894–1932; repr. Hildesheim, 1967), i, 407; Edward Williams Byron Nicholson, Introduction to the Study of Some of the Oldest Latin Musical Manuscripts in the Bodleian Library, Oxford, Early Bodleian Music, 3 (London, 1909), lxxxvi and plate 68; Anselm Hoste, Bibliotheca Aelrediana: A Survey of the Manuscripts, Old Catalogues, Editions and Studies Concerning St Aelred of Rievaulx (Steenbrugge, 1962), 35, 112, 115, 125 and 127; Otto Pächt and Jonathan James Graham Alexander, Illuminated Manuscripts in the Bodleian Library, Oxford, 3 vols. (Oxford, 1966–73), iii, 25; Henry Octavius Coxe, Laudian Manuscripts, rev. Richard William Hunt, Bodleian Library Quarto Catalogues, 2 (Oxford, 1973), cols. 482–3.

10 For example, the manuscript includes a Passion of St Thomas Becket (fols. 96^r–100^v), which places its copying after his martyrdom in 1170, and a list of English kings originally ending with Henry II (fol. 129^v), which was hence presumably written before the accession of Richard I in 1189. This same list of kings, and a list of the kings of Scotland on the same folio, has been extended by the addition of further kings’ names in later hands.

11 A sense of the rarity of musical notation outside liturgical books may be gained by consulting Karl Drew Hartzell's Catalogue of Manuscripts Written or Owned in England up to 1200 Containing Music (Woodbridge, 2006), in which only a small handful of sources constitute neither liturgical books nor casually written jottings added to margins and blank leaves; GB-Ob Laud 668 is listed on p. 481. The challenges posed to scribes who incorporated music in books designed primarily for text are explored in Helen Deeming, ‘Observations on the Habits of Twelfth- and Thirteenth-Century Music Scribes’, Scriptorium, 60 (2006), 38–59.

12 See especially the discussions of Gilson, ‘Sur le Iesu dulcis memoria’, passim, and Wilmart, Le ‘jubilus’, 222–4.

13 A recent contribution arguing strongly for Aelred's authorship on the basis of the poem's content is Wolfgang Buchmüller, ‘Dulcis Iesu memoria: Poetische Christusmystik bei Aelred von Rievaulx’, Geist und Leben, 80 (2007), 436–52.

14 For example, Lausberg tentatively proposed Stephen Langton (c.1150–1228) as a possible author, arguing that this archbishop of Canterbury (the likely author of the sequence Veni sancte spiritus) was schooled in and drawn to Cistercian theology without being a member of the order (Der Hymnus ‘Jesu dulcis memoria’, 395–6). Lausberg's argument, however, partly rests on a different chronological view from that proposed here, in that he dates GB-Ob Laud 668 to the early thirteenth century instead of the late twelfth. The same chronological grounds cause Lausberg to rule out Aelred's authorship altogether.

15 The text presented in Appendix 1 preserves the readings of GB-Ob Laud 668; for a critical text, based on the collation of many textual sources, see the editions of Wilmart (Le ‘jubilus’, chapter 4) and Lausberg (Der Hymnus ‘Jesu dulcis memoria’, 491–502).

16 James Wimsatt, ‘The Canticle of Canticles, Two Latin Poems, and “In a valey of þis restles mynde”’, Modern Philology, 75 (1978), 327–45 (p. 335).

17 Mary Carruthers has pointed out that this verse was widely known in two alternative Latin versions during the Middle Ages: ‘gustate et videte quoniam suavis est Dominus’ and ‘gustate et videte quoniam bonus est Dominus’. Contemporary ambivalence over the propriety of describing the Lord as ‘sweet’ (a term that could have negative connotations in some contexts) may lie behind the alternative reading, ‘good’ (‘Sweetness’, Speculum, 81 (2006), 999–1013 (p. 1006)). Leofranc Holford-Strevens has proposed that the Latin word ‘suavis’ ought not to be translated as ‘sweet’, as if it were a direct synonym for ‘dulcis’: he prefers to render it as ‘pleasing’ (‘Suavis et morosus: The Ways of a Word’, Quomodo cantabimus canticum? Studies in Honor of Edward H. Roesner, ed. David Butler Cannata, Gabriela Ilnitchi Currie, Rena Charnin Mueller and John Louis Nádas (Middleton, WI, 2008), 3–14 (p. 4)). On the other hand, Franz Posset regards ‘suavitas’ and ‘dulcedo’ as synonymous when used in relation to God (‘The Sweetness of God’, American Benedictine Review, 44 (1993), 143–78 (pp. 148–9)).

18 Carruthers, ‘Sweetness’, 1007.

19 See also 12. 1–2: ‘Experti, recognoscite, / amorem pium pascite’ (‘You who know by experience, recognize him, / nourish pious love’).

20 Bernard of Clairvaux, Sermones super Cantica Canticorum, 22. 2, ed. Jean Leclercq, Charles Hugh Talbot and Henricus M. Rochais, Sancti Bernardi opera (hereafter SBO), 8 vols. (Rome, 1957–77), i, 130.

21 Augustine of Hippo, Sermones, 123. 4, quoted in Bruce Holsinger, Music, Body and Desire in Medieval Culture: Hildegard of Bingen to Chaucer (Stanford, CA, 2001), 80.

22 Bernard of Clairvaux, Sermones super Cantica Canticorum, 15. 6, ed. SBO, i, 86.

23 Carruthers, ‘Sweetness’, 1000.

24 Psalm 118. 103.

25 Carruthers, ‘Sweetness’, 1005.

26 Bernard of Clairvaux, Sermones super Cantica Canticorum, 85. 13, ed. SBO, ii (1958), 316. This passage, among many others, is discussed in Edith Scholl, ‘The Sweetness of the Lord: Dulcis and Suavis’, Cistercian Studies, 27 (1992), 359–66 (p. 362).

27 The textual history of this idea is traced in Franz Posset, ‘Sensing God with the “Palate of the Heart” According to Augustine and Other Spiritual Authors’, American Benedictine Review, 49 (1998), 356–86. See also Rosemary Drage Hale, ‘“Taste and See, for God is Sweet”: Sensory Perception and Memory in Medieval Christian Experience’, Vox mystica: Essays for Valerie M. Lagorio, ed. Anne Clark Bartlett, Thomas Bestul, Janet Goebel and William F. Pollard (Cambridge, 1995), 3–14.

28 Pope Gregory I, Homily 36, ed. Jacques-Paul Migne, Sancti Gregorii papae I cognomento magni opera omnia, ii, Patrologiae cursus completus, series latina, 76 (Paris, 1857), cols. 1265–74 (col. 1266).

29 Aelred of Rievaulx, De speculo caritatis, ed. Anselm Hoste and Charles Hugh Talbot, Aelredi Rievallensis opera omnia, 1 (Turnhout, 1971), Book I, chapter 1, no. 2.

30 Ibid., I, 1, no. 16.

31 Bernard of Clairvaux, Sermones super Cantica Canticorum, 28. 6, ed. SBO, i, 196.

32 Bernard of Clairvaux, Epistolae, 228. 3, ed. SBO, viii (1977), 99.

33 Benedict of Nursia, Regula, ed. Timothy Fry, RB 1980: The Rule of St Benedict in Latin and English with Notes (Collegeville, MN, 1981), Prologue, verse 19.

34 For a brief summary, see Hale, ‘“Taste and See”’, 13.

35 ‘Ineffabilis suavitas atque dulcedo’, Augustine of Hippo, Epistulae, epist. 11, par. 4; Sancti Aurelii Augustini Epistulae I–LV, ed. Klaus-Detlef Daur (Turnhout, 2004), 28.

36 ‘Inerrabili dilectionis dulcedine’, Benedict of Nursia, Regula, ed. Fry, Prologue, verse 49.

37 ‘Dulcedo ineffabilis et bonitas’, Bernard of Clairvaux, Sententiarum series tertia, 97, ed. SBO, vi/2 (1972), 155.

38 Philippians iv. 7: ‘et pax Dei quae exsuperat omnem sensum custodiat corda vestra et intellegentias vestras in Christo Jesu’ (‘and may the peace of God which passes all understanding keep your hearts and your minds in Christ Jesus’).

39 Lowry Nelson, Jr, ‘The Rhetoric of Ineffability: Toward a Definition of Mystical Poetry’, Comparative Literature, 8 (1956), 323–36 (pp. 327–8).

40 Wilmart, Le ‘jubilus’, 234–6.

41 The repeated rhyme-sounds are: ‘-ibus’ (stanzas 3, 8 and 42), ‘-ium’ (stanzas 4 and 29), ‘-ere’ (stanzas 5, 22 and 35), ‘-ulo’ (stanzas 6 and 7), ‘-ite’ (stanzas 12 and 37) and ‘-io’ (stanzas 15 and 34).

42 Though some of the repeated rhyme-sounds could be considered serendipitous, given the inflectional endings of Latin words and the further constraints of the metrical structure, this density of vowel-sound repetition is even more striking and seems likely to be evidence of poetic intent.

43 Augustine of Hippo, Enarrationes in Psalmos, Enarratio II on Psalm 32, ed. Jacques-Paul Migne, Sancti Aurelii Augustini opera omnia, iv/1, Patrologiae cursus completus, series latina, 36 (Paris, 1841), cols. 277–300 (col. 283); trans. James McKinnon, Music in Early Christian Literature (Cambridge, 1987), 155.

44 Ibid., Enarratio on Psalm 99, ed. Migne, Sancti Aurelii Augustini opera omnia, iv/2, Patrologiae cursus completus, series latina, 37 (Paris, 1846), cols. 1271–81 (col. 1272); trans. McKinnon, Music in Early Christian Literature, 158. Fuller exploration of Augustine's use of this term, and its position within his changing attitudes towards the value of music, may be found in Robert Boenig, ‘St Augustine's Jubilus and Richard Rolle's Canor’, Vox mystica, ed. Bartlett, Bestul, Goebel and Pollard, 75–86. On Augustine's wrangling with the morally ambiguous nature of song, see most recently Philip Weller, ‘Vox – Littera – Cantus: Aspects of Voice and Vocality in Medieval Song’, Music in Medieval Europe: Studies in Honour of Bryan Gillingham, ed. Terence Bailey and Alma Santosuosso (Aldershot, 2007), 239–62 (pp. 255–61).

45 Bernard of Clairvaux, Sermones super Cantica Canticorum, 15. 6, ed. SBO, i, 86. An appropriate English translation of ‘iubilus’ is very hard to find, and most writers leave it untranslated. In translating this passage, Carruthers opts for ‘jubilee’ and Raby for ‘joy’, but neither alternative (nor my own, ‘jubilation’) seems quite to capture the musical-vocal quality of the term as Bernard must surely have understood it; see Carruthers, ‘Sweetness’, 1000, and Raby, A History of Christian Latin Poetry, 329.

46 For example, in the title of Wilmart's study, and more generally in literary and theological criticism, particularly of the early twentieth century.

47 Frederick Brittain called it ‘the Office Hymn form par excellence’ in his introduction to the extract from Dulcis Jesu memoria included in his anthology The Medieval Latin and Romance Lyric to A.D. 1300 (Cambridge, 1937), 109–11.

48 The extract printed by Brittain is that selection of nine stanzas often referred to as ‘The Rosy Sequence’ which appears, in the liturgical position of a sequence in a Mass for the Holy Name, in the edition of the Sarum Gradual printed in 1532. Brittain's introduction points out that only this excerpt (and not the whole 42-stanza poem) should properly be referred to with this modern title, and also that the ‘more liturgically correct home’ in which excerpts from Dulcis Jesu memoria were usually found was among the hymns for the Office (The Medieval Latin and Romance Lyric, 109).

49 Fuller analysis of the arguments may be found in Richard L. Crocker, The Early Medieval Sequence (Berkeley and Los Angeles, CA, 1977).

50 Lori Kruckenberg, ‘Neumatizing the Sequence: Special Performances of Sequences in the Central Middle Ages’, Journal of the American Musicological Society, 59 (2006), 243–318.

51 Quoted (with a tabular comparison of the texts of Guillaume and Hugh) in Kruckenberg, ‘Neumatizing the Sequence’, 251.

52 Ibid., 265–6.

53 Ibid., 280.

54 Margot Fassler, Gothic Song: Victorine Sequences and Augustinian Reform in Twelfth-Century Paris (Cambridge, 1993).

55 The editorial policies employed in Appendix 1 are as follows: music and text follow the readings of GB-Ob Laud 668 entirely (unlike Lausberg's transcription, which underlays a critical text, collated from several manuscripts, to the music uniquely preserved in GB-Ob Laud 668); the distinction between virga and punctum in the original notation is not preserved in the transcription, both being presented as closed noteheads without stems, but occurrences of the ‘wave-note’, a note-form somewhat resembling the quilisma of plainchant notation though differing from it in usage in this and other non-liturgical manuscripts, are recorded using a wavy notehead similar to the form found in the manuscript; notes joined into a single neume- or ligature-form in the manuscript are slurred in the transcription. Stanza numbering, capitalization and punctuation are editorially supplied in line with modern convention, though original spelling is retained.

56 Similar procedures of musical variation between two adjacent stanzas may be observed in stanzas 15–16 and 41–2.

57 Lausberg, Der Hymnus ‘Jesu dulcis memoria’, 55–9.

58 Similar melodic and cadential procedures can be identified in many of the Victorine sequences, with those using the G final offering numerous points of comparison with phrases in Dulcis Jesu memoria; see the anthology in Fassler, Gothic Song, 416–41.

59 These 13 stanzas are nos. 1–3, 9–14, 21–2, 31 and 37; note also that different note-groups (as opposed to one note per syllable – by far the norm across the song) occur in these positions in stanzas 4–5, 15–16, 35 and 39–40.

60 The passage is found in Giraldi Cambrensis opera, ed. John Sherran Brewer and James Francis Dimock, 8 vols., Rolls Series (London, 1861–91), ii (1862), 119–20, and discussed in Weller, ‘Vox – Littera – Cantus’, 242–5.

61 Sam Barrett, ‘Music and Writing: On the Compilation of Paris Bibliothèque Nationale Lat. 1154’, Early Music History, 16 (1997), 55–96 (p. 93).

62 Paul Saenger, Space between Words: The Origins of Silent Reading (Stanford, CA, 1997).

63 Ibid., 141.

64 Although this proposition has not been extensively researched, several studies have drawn attention to the (surely related) increased importance of visual memory in music copying from the twelfth century onwards; see, for example, Michel Huglo, Les livres de chant liturgique, Typologie des sources du moyen âge occidental, 52 (Turnhout, 1988), 127–30, and James Grier, ‘Scribal Practices in the Aquitanian Versaria of the Twelfth Century: Towards a Typology of Error and Variant’, Journal of the American Musicological Society, 45 (1992), 373–427.

65 Augustine of Hippo, Sermones, 149; ed. Migne, Sancti Aurelii Augustini opera omnia, v, Patrologiae cursus completus, series latina, 38 (Paris, 1845), cols. 800–7 (col. 801).

66 Bernard of Clairvaux, Sermones super Cantica Canticorum, 7. 5, ed. SBO, i, 34.

67 Augustine of Hippo, De musica libri sex, book 5; quoted in Holsinger, Music, Body and Desire, 65.