1. There has been, of course, a good amount of work done on the Attis; but, in view of its importance and length, a comparison of the number of articles written about it with the long lists of those done on Poem 64 and even some of the short poems will support my point about ‘relatively little attention’. See Leon, H. J., ‘A Quarter Century of Catullan Scholarship (1934–1959)’, CW 53 (1960), 104–113Google Scholar, 141–148, 173–180, 281–282; and Thomson, D. F. S., ‘Recent Scholarship on Catullus (1960–1969)’, CW 65 (1971), 116–126Google Scholar. A look at some of the books dealing with Catullus is even more revealing: Wheeler, A. L., Catullus and the Traditions of Ancient Poetry (Berkeley, 1934Google Scholar), and Ross, D. O. Jr., Style and Tradition in Catullus (Cambridge, Mass., 1969CrossRefGoogle Scholar), do not discuss the Attis at all; Havelock, E. A., The Lyric Genius of Catullus (reprinted, with new foreword; New York, 1967Google Scholar), mentions it only briefly (78 and 190, note 72). Quinn, K., The Catullan Revolution (revised impression; Cambridge, 1969Google ScholarPubMed), limits its discussion to lines 44–50 and line 55 (64); Quinn’s more recent Catullus: An Interpretation (London, 1972Google Scholar) contains a fuller but far from satisfying discussion of the poem (249–251 and 261), along with several incidental references to it (12, 204, 219, 258, 259, 260, 277, and 282). One of the favorable aspects of Schäfer, E., Das Verhältnis von Erlebnis und Kunstgestalt bei Catull (Wiesbaden, 1966Google Scholar), is that it contains a full discussion of Poem 63 on 95–107.

10. The basic ancient authorities for the Cybele-Attis myth and cult are Pausanias 7.17.10–12; Arnobius, Aduers. Nat. 5.5ff.; Ovid, Fasti 4.223ff.; Sallustius, De Dis et Mundo 4; Julian, Orat. 5.165bff.; and, of course, Catullus 63. Modern treatments include Graillot, H., Le culte de Cyhèle mère des dieux à Rome et dans l’empire romain (Paris, 1912Google Scholar); Cumont, F., The Oriental Religions in Roman Paganism (‘authorized trans.’ New York, 1956), 46–72Google Scholar; Lambrechts, P., Attis: Van herdersknaap tot God (with resumé in French; Brussels, 1965Google Scholar); and Vermaseren, M. J., The Legend of Attis in Greek and Roman Art (Leiden, 1966Google Scholar). Grimal, P., Dictionnaire de la mythologie grecque (4th ed. rev., Paris, 1969Google Scholar), gives a full list of the ancient sources and cites useful modern works on the subject. Finally, there is the influential theory of Sir James Frazer, well-enough stated in chapters 34 through 36 of the abridged edition of The Golden Bough (New York, 1922Google ScholarPubMed); Frazer’s has become the accepted position, and Lambrechts presents the first full argument against his views on Attis.

11. See, for example, Fordyce 261, and Elder, ‘Attis’, 395. Both state that Catullus had to ‘alter’ the myth in order to present Attis as a man and not as a god.

12. Elder, ‘Attis’, 395. Weinreich and T. Oksala are good on the poem as ‘psychological experience’. For the poem as ‘dramatization’, see Oksala, P., ‘Das Geschlecht des Attis bei Catull’, Arctos 6 (1970), 91–96;Google Scholar and Guillemin, A., ‘Le Poème 63 de Catulle’, REL 27 (1949), 149–157.Google Scholar Oksala argues that the oscillation of genders in the poem accentuates the ‘tragic conflict’ within Attis; Guillemin interprets the poem as a three-act tragedy which bears definite likenesses to Eurpides’ Bacchae.

13. Jakobson, R., ‘Two Aspects of Language and Two Types of Aphasic Disturbances’, in Jakobson, R. and Halle, Ms., Fundamentals of Language (The Hague, 1956), 55–82.Google Scholar See also Jakobson’s, ‘Linguistics and Poetics’, Style and Language Ced. by T. E. Sebeok; Cambridge, Mass., 1960), 350–377.Google Scholar On the relationship between the linguistic theories of Jakobson and the ethnological method of Lévi-Strauss, see Leach, E., Claude Lévi-Strauss (New York, 1970), 33–52.Google Scholar

14. Barthes, R., Eléments de sémiologie (Paris, 1969), 135Google Scholar. See Leach, 45–47.

15. Barthes, , S/Z (Paris, 1970Google Scholar). See also his ‘Introduction à l’analyse structurale des récits’, Communications 8 (1966), 1–27.

16. Genette, G., Figures III (Paris, 1972), 41–273.Google Scholar

17. Lévi-Strauss, C., Anthropologie structurale (Paris, 1958), 227–256.Google Scholar

18. Todorov, T., Littérature et signification (Paris, 1967), 69–77Google Scholar; and Grammaire du Décameron (The Hague, 1969Google Scholar). See also his Poétique de la prose (Paris, 1971Google Scholar) and Théorie de la littérature (Textes des Formalistes russes réunis et presentés par T. Todorov; Paris, 1965).Google Scholar

19. Fordyce, 262.

2. There have been some important recent attempts to relate Poem 63 to the experience of the poet. Besides Schäfer, see Oksala, T., ‘Catullus Attis-Ballade: Über den Stil der Dichtung und ihr Verhältnis zur Persönlichkeit des Dichters’, Arctos 3 (1962), 199–213Google Scholar. Much more extreme and much less valuable is Harkins, P. W., ‘Autoallegory in Catullus 63 and 64’, TAP A 90 (1959), 102–116Google Scholar. There is also a ‘Freudian’ discussion in Rankin, A. V., ‘Odi et Amo: Gaius Valerius Catullus and Freud’s Essay on “A Special Type of Choice of Object Made by Men”,’ The American Imago 19 (1962), 437–447Google Scholar. My objections to Rankin are the same as my general objections to most Anglo-American ‘Freudian’ critics: in my view their perspective is too narrow and their methodology is inadequate. I find the work of the French Freudians much more fruitful: see, for example, Laplanche, J., Hölderlin et la question du père (Paris, 1961Google Scholar).

20. Whigham, P., The Poems of Catullus: A Bilingual Edition (Berkeley, 1969), 20.Google Scholar I have, of course, important reservations concerning the second part of Whigham’s statement, i.e., ‘treated psychologically, in terms of Catullus’ own experience’. See above, note 6.

21. Schäfer, 73. He is referring to Lieberg, G., ‘L’ordinamento ed i reciproci rapporti dei carmi maggiori di Catullo’, RivFC 36 (1958), 23–47Google Scholar. See also McGushin, P., ‘Catullus’ Sanctae Foedus Amicitiae’, CP 62 (1967), 85–93Google Scholar; G. Williams, Tradition and Originality in Roman Poetry, 404–412; and especially, Sandy, G. N., ‘Catullus 63 and the Theme of Marriage’, AJP 92 (1971), 185–195Google Scholar, and ‘The Imagery of Catullus 63’, TAP A 99 (1968), 389–399.Google Scholar

22. Lévi-Strauss, C., Le totémisme aujourd’hui (Paris, 1965), 22–23.Google Scholar

23. I have taken the liberty of filling out the contrasts mentioned by Fordyce by giving both sides of the oppositions where he sometimes gives only one side, leaving the other there by implication.

24. See Barthes, , Critique et vérité (Paris, 1966), 65ff.Google Scholar

25. Even a superficial glance will show that many of these oppositions might be arranged in a different manner. This, as a matter of fact, is part of the argument of my essay: they interlock and intersect to such an extent that it is impossible to classify them neatly.

26. See lines 2, 14, 20, 22. 30, 49–55, 58–60, 64, 70–71, and 91–93.

27. See lines 18, 33, 51–52, 68, and 90.

28. See lines 13, 53–54, 70–72, 78–86, and 89.

29. See lines 3, 5–6, 8, 9, 12, 13, 17, 18, 20, 23, 25, 27, 35, 68, 69, 76, 90, 91–93.

3. Elder, J. P., ‘Notes on Some Conscious and Subconscious Elements in Catullus’ Poety’, HSCP 60 (1951), 101–136Google Scholar; and Quinn, K., ‘Docte Catulle’, in Critical Essays on Roman Literature: Elegy and Lyric (ed. by Sullivan, J. P.: London, 1962, 31–63Google Scholar), and The Catullan Revolution.

30. See line 9.

31. See lines 4, 11, 31, 38, 44, 46, 54, 57, 78–86, 89, and 91–93.

32. See lines 34, 38, 44, 47, 78–86, and 91–93.

33. See lines 18 and 45–46.

34. See lines 3, 32, 39ff., and 54.

35. See lines 1, 16, 40, 47, and 48.

36. For a summary of the problem of masculine and feminine agreements, see Quinn, Commentary, p. 287. He concludes: ‘In all instances … a feminine agreement is metrically possible. Most editors since Lachmann restore the feminine in some at any rate of seven passages where the masculine occurs in the MSS. The arguments for leaving some masculine agreements are (1) it stops the trick from becoming too mechanical; (2) Attis is of uncertain sex (27 notha mulier), rather than female; (3) the masculine agreement is dramatically more effective in describing Attis before his emasculation (51 miser), or during his rebellion (78 hunc, 80 qui — and perhaps 88 tenerum and 89 ille, though Mynors prints teneram and illa).’ These and other reasons for retaining the masculines seem convincing enough to me, so they shall be retained in the citations from the poem given in this paper, which otherwise follow Mynors’ text (Oxford, 1958; reprinted from corrected sheets, 1960).

37. The fact that the feminine form for the followers of Cybele appears only in Catullus 63 and in the Hephaistion fragment which has been ascribed to Callimachus has been made much of by those who look for an Alexandrian source for the Attis. But there is really no hard evidence for ascribing the fragment to Callimachus; even if there were there would be no real justification for inferring that Catullus 63 is a translation or adaptation of such an original. See Quinn, Commentary, 288; Fordyce, 264–265; and Elder, ‘Attis’, 394, note 2.

38. Quinn, Commentary, 283–284, notes that the structure of the poem can be seen in two ways. We can take it as having a basic narrative structure of two eleven-line vignettes (1–11 describing Attis’ frenzied self-castration and 39–49 describing his repentance); each ‘vignette’ is followed by a tragic speech (12–26, where Attis addresses the Gallae; and 50–73, Attis’ soliloquy); the first speech is followed by a ‘pendent narrative passage’ (27–38, describing the rush up Ida), and the second by an epilogue relating Attis’ rebellion and recapture (74–77, rebellion; 78–83, Cybele’s speech; 84–90, recapture; 91–93, prayer to Cybele). The poem can also be viewed as having a three-part structure: 1–38, frenzy; 39–73, repentance; 74–90, back to frenzy. I think that this second arrangement is more adequate to the poem, and my analysis will take the poem as a sort of three-act tragedy, somewhat like that suggested by Mile Guillemin.

39. For a good development of the meaning of ‘masculine’ and ‘feminine’ as I use them here, see Neumann, E., Amor and Psyche (trans, by Mannheim, R.; New York, 1956Google Scholar). Though influenced by the theories of Jung, Neumann manages to avoid a good deal of the ‘male chauvinism’ indulged in by Jung and Freud. See Jung, C. J., Two Essays on Analytical Psychology (trans, by Hull, R. F. C.; New York, 1956), 217–223Google Scholar; and Freud, S., New Introductory Lectures on Psychoanalysis (trans. by Strachey, J.; New York, 1965), 112–135Google Scholar. For a good critique of Freud's attitude to women, see Rieff, P., Freud: The Mind of the Moralist (Garden City, N. Y., 1961), 191–204.Google Scholar

4. See Putnam, M. C. J., ‘The Art of Catullus 64’, HSCP 65 (1961), 165–205Google Scholar; and Curran, L. C., ‘Catullus 64 and the Heroic Age’, YCS 21 (1969), 169–192Google Scholar.

40. Lines 39–43, reading excitum in 42 for Lachmann’s excitam. Fordyce, 267, explains the allusion made in line 43; Pasithea is, in Homer, Il. 14.267ff., one of the Graces, who is promised to Sleep by Hera. Her only other extant appearance is an epigram by Antipater (A. P. 9.517.5–6). He then draws the unwarranted inference that ‘this piece of far-fetched mythological erudition points to an Alexandrian source’. For one thing, the allusion might have been considerably less ‘far-fetched’ to Catullus and his readers than it is to us!

41. Compare Vergil, Aen. 1.124–127, on Neptune calming the storm, and 223–220 and 254–256, on the serenitas of Jupiter. See Pöschl, V., The Art of Vergil (trans, by Seligson, G.; Ann Arbor, 1962), 14–24.Google Scholar

42. Reading ipse of the MSS. instead of Guarinus’ ipsa.

43. Note the mention of olive and wild olive in Odyssey 5.474–493, where Odysseus is, as it were, midway between the ‘primaeval’ island of Calypso and the highly developed civilization of the Phaeacians.

44. Lines 68–72, italics added.

45. Lines 76–90, reading hunc in 78, qui in 80, tenerum in 88, and ille in 89. Mynors follows Lachmann’s teneram in 88 even though he has doubts about it: ‘an recte’, he says of the conjecture, ‘dubitari potest’.

46. Lines 90–93.

47. I am indebted to Professor Thomas Barry for suggesting the word ‘infects’ to describe what is happening here.

48. Barthes, Critique et vérité, 67.

49. See above, 3–4, on Lambrechts’ thesis about Attis.

5. Style and Tradition in Catullus. Ross’s method of dealing with Catullus is greatly influenced, as he himself acknowledges, by Axelson, B., Unpoetische Wörter: Ein Beitrag zur Kenntnis der Lateinischen Dichtersprache (Lund, 1945Google Scholar); for a telling critique of Axelson’s method, see Williams, G., Tradition and Originality in Roman Poetry (Oxford, 1967), 743–750Google Scholar. Tränkle, H., ‘Ausdrucksfülle bei Catull’, Philologus 111 (1967), 198–211CrossRefGoogle Scholar, takes an ‘Axelsonian’ view of Poem 63. Tränkle, however, while admitting that the poem is full of ‘traditional elements’, argues that these are considerably ‘refined’ by the poet: in his view, then, Poem 64 is both traditional and neoteric.

50. A point well-made by Elder, ‘Attis’, and by P. Oksala.

51. Barthes, , Sur Racine (Paris, 1963), 23Google Scholar. On the relationship of sexuality and domination, see Freud, , Three Essays on the Theory of Sexuality (4th. ed., trans, by Strachey, J.; London, 1953Google Scholar); Rieff, ‘Sexuality and Domination’, Freud: The Mind of the Moralist, 163–204; and Wollheim, R., ‘Sexuality’, Sigmund Freud (New York, 1971), 113–136.Google Scholar

52. See Lieberg, G., Puella Divina (Amsterdam, 1962), particularly 152–263Google Scholar, on Catullus 68.

53. Once again, accepting the thesis of Lambrechts that Attis was a man and not a god.

54. Od. 17.322–323. Translation by R. Fitzgerald (Garden City, N. Y., 1963), 320.

55. On the meaning of Calypso and Ogygie, see The Odyssey of Homer (ed. with general and grammatical intro., commentary, and indexes by Stanford, W. B.; London, 1950), I, 292 and 294.Google Scholar

56. See the vivid description of Calypso’s island in Od. 5.55–74.

57. Od. 9.28–33. Fitzgerald, 146.

58. Od. 6.182–185. Fitzgerald, 104.

59. See Dimock, G., ‘The name of Odysseus’, Hudson Review 9 (1956), 52–70CrossRefGoogle Scholar.

6. See, for example, Putnam’s review of Schäfer, , CW 61 (1967), 52Google Scholar, which notes that Schäfer is reluctant to take Poem 63 as a ‘personal document’ and goes on to criticize this reluctance: ‘… Freud and Frazer might comment that Poem 63 is but a strange version of what S. defines as the central theme of the longer poems, “die Hochzeit und ihr Strahlungskreis”.’ I would contend that it is not in the best Freudian tradition to make unmediated leaps from the language of poetry to the data of personal experience, and Schäfer’s reluctance to do so seems quite correct to me. Yet Putnam has touched upon the fundamental difficulty with approaches like Schäfer’s: an initial relation of the poetry to the experience of the poet tends to force the interpreter to ask the kind of questions Putnam accuses Schäfer of avoiding. If a discussion of Catullus’ poetry chooses to focus on its relationship to his experience, it is difficult to justify a refusal to take that approach as far as it will go. Yet there is, I believe, at least one valid way of discussing Catullus 63 without getting trapped in this dilemma, and I shall try to develop such a discussion in this essay.

60. Od. 5.1–2. Fitzgerald, 81.

61. For an interesting contemporary delineation of the relationship between female nurse and male patient, see Kesey’s, Ken novel, One Flew Over the Cuckoo’s Nest (New York, 1962Google Scholar). The story of Tithonus’ ‘cruel immortality’ first appears in the Homeric Hymn to Aphrodite 218ff., so there is no definite evidence that the writer of the Odyssey knew it. If he did not, then the ‘reminder’ I have mentioned was not intended as such by the poet. So be it; but it ‘reminds’ us nevertheless!

62. See Hyppolite, J., Genèse et structure de la phénoménologie de l’esprit de Hegel (Paris, 1946), 163–171Google Scholar, on ‘le combat pour la reconnaissance’.

63. Od. 10.291–301. Fitzgerald, p. 174. See Segal, C. P., ‘Circean Temptations: Homer, Vergil, Ovid’, TAPA 99 (1968), 419–442.Google Scholar

64. Od. 10.321–324. Fitzgerald, 175.

65. Od. 10.337–344. Fitzgerald, 175.

66. Lines 44–49, reading ipse of the MSS. in line 45.

67. I am speaking of the journey described in the poem, whether or not such a journey was actually ever made by the poet himself.

68. Barthes, Sur Racine, 67.

69. Observe, for example, the nuances of Odysseus’ interchanges with Athena, where he often assumes a passive role in the face of the goddess’s masculine strength. In his encounter with the Cyclops, he is the civilized man of culture, yet the Phaeacians are too civilized for him. One of the great symbols of the Odyssey, the hidden olive-tree trunk which is rooted in the soil of Ithaca, the sema which forces Penelope to recognize his true identity, identifies Odysseus as the man of culture who is still connected to nature. See Od. 23.181–204.

7. Wilamowitz, , ‘Die Galliamben des Kallimachos und Catullus’, Hermes 14 (1879), 144–199Google Scholar. See also his Hellenistische Dichtung in der Z.eit des Kallimachos II (Berlin, 1924), 291–295Google Scholar, which modifies his earlier position by admitting that the middle section of the poem is Catullan. See also Weinreich, O., ‘Catulls Attisgedicht’, Melanges Franz Cumont (Brussels, 1936), 463–500Google Scholar; Bardon, H., L’Art de composition chez Catulle (Paris, 1943), 29–32Google Scholar; Schuster, M., ‘C. Valerius Catullus’, RE (1948), 2376–2377Google Scholar; and Fordyce, C. J., Catullus: A Commentary (Oxford, 1966), 262Google Scholar.

70. See Girard, René, La violence et le sacré (Paris, 1972Google Scholar), and my review of this book in MLN 87 (1972), 986–998.Google Scholar

71. I wish to thank Professor John J. Peradotto for his comments and suggestions regarding previous drafts of this paper. Final responsibility for the views expressed herein does, of course, remain my own.

8. Catullus: The Poems (ed. with intro., revised text and commentary by Quinn, K.; London, 1970), 283Google Scholar. This book will hereafter be referred to as: Quinn, Commentary. See also Elder, J. P., ‘Catullus’ Attis’, AJP 68 (1947), 394–403Google Scholar: note 2, p. 394, reviews the evidence for Wilamowitz’ position and rejects it.

9. Sellar, W. Y., The Roman Poets of the Republic (reprint of the 1905 reissue of the 3rd ed. of 1869; Oxford, 1932), 461Google Scholar. The first edition of the book appeared in 1863.