1. This article has been with me for some time, and I have incurred a number of debts over the years; for their comments on an earlier version I am grateful to Elizabeth Harries, Charles Henderson, Jr., Peter White. For thorough criticism and encouragement both at that earlier stage and now I am indebted to my colleague George E. Dimock, Jr. and to my husband, William Oram.

For bibliography on Longus and on the ancient novel, see especially Sandy’s, Gerald N. bibliographical survey in CW 67 (1974), 321–359Google Scholar; McCulloh, William E., Longos (New York 1970), 133–137Google Scholar; Reardon, B. P., Courants littéraires grecs des IIe et IIIe siècles après J.-C. (Paris 1971Google Scholar); Anderson, Graham, Eros Sophistes: Ancient Novelists at Play (Chico, Calif., 1982), 174–180Google Scholar; Hägg, Thomas, The Novel in Antiquity (Oxford 1983) 235–250Google Scholar. See also the most recent and complete study of Longus’ work, Hunter, R. L., A Study of Daphnis and Chloe (Cambridge 1983CrossRefGoogle Scholar). The Greek text I have used unless otherwise noted is that of Schönberger, Otto, Longos: Hirtengeschichten von Daphnis und Chloe (Berlin 1960Google Scholar); translations are mine.

2. The next pastoral romance in prose after Longus’ probably mid-second-century A.D. work (for a recent discussion of the date, see Hunter [n.1 above), 1–15) appears to be Boccaccio’s Ameto (1341), as critics have frequently noted. To review the long-standing debate on Longus’ originality is beyond the scope of this paper. On the possibility of an earlier Greek tradition of bucolic prose fiction, see Hunter, 66; on oriental models, see now Anderson, Graham, Ancient Fiction: The Novel in the Greco-Roman World (London 1984Google Scholar). Anderson argues that the prototype for Daphnis and Chloe is to be found in ‘a number of interrelated Sumerian texts, of which the most significant is Dumuzi’s Dream, a highly formulaic poetic composition of some two-hundred-odd lines’, Daphnis being none other in origin that ‘the legendary Sumerian hero-king Dumuzi’ (6). Even if the discovery of the Sumerian texts alters the terms of the continuing debate on the origins of the novel, Anderson’s careful presentation of possible parallels (6–15) emphasizes for this reader the radical differences in Longus’ handling of the material and in the social and religious framework as well as the literary texture of Longus’ language and his indebtedness to a rich Greek and Latin tradition.

3. Pastoral has long been recognized as an ‘invasive’ and ‘expansionist’ genre: see the discussion and summary of earlier scholarship in Halperin, David M., Before Pastoral: Theocritus and the Ancient Tradition of Bucolic Poetry (New Haven and London 1983), 28ff.Google Scholar; for the daring catachresis of genres that characterizes Hellenistic poetry, see esp. 203ff. Daphnis and Chloe combines elements of pastoral poetry, prose adventure romance, mime, tragedy and — especially — comedy. (On comic elements, see the discussion below.) Longus’ style assumes a literate and sophisticated audience. Despite his light touch and the apparent simplicity of his language, Longus constructs a highly polished, tightly structured and stylistically complex work. It abounds in playful allusions to and imitations of Plato, the Greek historians, Sappho, fifth-and fourth-century drama, Hellenistic poets such as Theocritus, Bion, Moschus and — so far as we can tell — Philetas of Cos whose namesake is a character in the story. There are almost certainly echoes of Vergil and possibly of the Roman elegists as well. (See McCulloh [n.1 above], 72–78; Schönberger [n.1 above], passim; J. M. Edmonds in his introduction to the Loeb edition [London and Cambridge, Mass., 1916], vii-ix; Hunter [n.1 above], 59–83.) On Longus’ language and style, see Norden, Eduard, Die Antike Kunstprosa ² (Darmstadt 1958), i.437–439Google Scholar; Rohde, Erwin, Der griechische Roman und seine Vorlaufer ⁴ (Hildesheim 1960), 550–554Google Scholar; Valley, Gunnar, Über den Sprachgebrauch des Longus (Uppsala 1926); Hunter, 84–98Google Scholar, with a review of earlier scholarship. On the sophistication of author and audience, see Wolff, Samuel Lee, Greek Romances and Elizabethan Prose Fiction (New York 1912), 130Google Scholar; Paul Turner in his introduction to the Penguin translation (Harmondsworth 1956); Rosenmeyer, Thomas G., The Green Cabinet: Theocritus and the European Pastoral Lyric (Berkeley 1969), 121Google Scholar; Heiserman, Arthur, The Novel Before the Novel (Chicago and London 1977), 144Google Scholar; Halperin, 54f. Ben Edwin Perry’s argument for a ‘young and naive’ audience ‘of little education’ (in The Ancient Romances [Berkeley and Los Angeles 1967], 56; cf. 98) has won few if any adherents.

4. For instance, among other ancient Greek novelists, Achilles Tatius (Leucippe and Clitophon) puts the narrator in the story; Dio Chrysostom (The Hunters of Euboea) presents himself as an eye-witness — an approach Lucian parodies in his True History; Chariton (whose historical novel Chaereas and Callirhoe is set in fifth-century B.C. Syracuse, at least four and probably six centuries before the date of composition) introduces himself by name in a formula that recalls Herodotus and Thucydides; Heliodorus begins cinematically, presenting a tableau to the reader without identifying the apparently omniscient narrator or orienting the reader.

5. Schönberger convincingly argues for this reading (over eikona graptēn) on stylistic grounds. Hunter (n.1 above), 43f., points out that the phrase can mean both a painted picture and a description of a painting; he suggests that historian erōtos may mean a painting as well as a story of love. For a learned and excellent discussion of mimesis, the link between poetry and painting, and the prologue more generally, see id., 38–52.

6. The xenoi, non-locals who have come to combine sight-seeing and a pilgrimage of sorts call to mind not only modern Greek tourists in; their own country but the women in Herodas’ fourth mime who have come to make votive offerings and sacrifice to Asclepius. One may note also that the rustic ‘hosts’ of the Methymnean youths are called xenodochoi by Longus (2.13.2).

7. The term exēgētēs (cf. periēgētēs) encompassed a range of meanings by Longus’ time, both religious and secular. (Pausanias uses the term to mean ‘local guide’ or tourist-and-museum-guide frequently: 1.13.8, 34.4, 35.8, 41.2, 42.4; 2.9.7, 23.6, 31.4; 4.33.6; 5.6.6, 10.7, 18.6, 20.4, 21.8f.; 7.6.5; 9.3.3. He uses it to mean an interpreter of dreams and portents at 5.23.6 and to designate one of the men charged with the care of sacrifices at 5.15.10.) For a recent view of a comparable role in transition, one may think of the monks selling postcards to busloads of tourists in the relatively recently inaccessible monasteries on the cliffs of Meteora in Greece. Chalk, H.H.O. (‘Eros and the Lesbian Pastorals of Longos’, JHS 80 [1960], 32–51CrossRefGoogle Scholar) sees in Longus’ preface a seriously-meant reference to an interpreter of sacred rites (36 n.7) and takes the mention of a votive offering as further reason to seek religious meaning in the text. Chalk’s interpretation of the work as a Mysterientext for the cult of Eros-Dionysus (in the tradition of Rohde [n.3 above); cf. Merkelbach, R., ‘Daphnis und Chloe: Roman und Mysterium’, Antaios 1, [1960], 47–60Google Scholar, and Roman und Mysterium in der Antike [Munich and Berlin 1962]) compels him to gloss over even those instances of secular humor he notes — see for instance 49 n.2 where he castigates Lykainion’s ‘cynicism about dreams from the Nymphs’ when she concocts a divinely-sent dream (amusingly like the dreams Chalk takes more seriously) in order to seduce Daphnis. Many have argued against the school of criticism which would see the work as a mystery-cuit roman-à-clef; see for instance Geyer, A., ‘Roman und Mysterienritual’, WJA n.f. 3 (1977), 179–196Google Scholar; Anderson (n.1 above), 111 n.1, 136 n.78, and more generally 1–11 and 41–49; cf. also Hunter (n.1 above), 114 n.93, 31–38, 46, and relevant notes, esp. Ill n.69. Northrop Frye’s succinct caveat about taking at face value the mythic-cultic structure of what may well be a secular text bears recalling: ‘The same structural principle may still be used where there is no longer any question of cult. Greene’s story Pandosto, the main source of Shakespeare’s Winter’s Tale, treats Apollo’s oracle at Delphi in a way which would tempt us to assume, if it had been written two thousand years earlier, that it was trying to promote the prestige of that oracle.’ (The Secular Scripture: A Study of the Structure of Romance [Cambridge, Mass., and London 1976], 13.Google Scholar)

8. A recollection of Theocritus 7.51 (to meludrion exeponasa); cf. Gow’s note ad loc. for the connotation of highly polished work. (Cf. Philetas’ words at 2.3.3 below).

9. The echo of Thucydides has been noted by Boissonade, who is cited by Valley (n.3 above), cited in turn by McCulloh (n.1 above), 121 n.13, who also notes that Turner, Paul (‘Daphnis and Chloe: An Interpretation’, G. & R. n.s. 7 [1960], 117Google Scholar) made the connection. McCulloh himself also sees a similarity to Plutarch’s On Isis and Osiris 27. Anderson (n.1 above) 135f. n.74, points out the playful paraphrase of Agathon’s discussion of Eros in Plato’s Symposium 197e, where the expressions paidia (‘fun’), spoudē metria (‘moderate [and ‘metrical’] earnestness’) and ho logos tōi theōi anakeisthō (‘let my speech be offered to the god as an anathēma’) seem to provide Longus with a model. Hunter (n.1 above), 47–50, discusses the association of to terpnon (of Horace’s discussion of voluptas at A.P. 333–346) with to muthōdes (ficta) and cites numerous parallels. In this context he discusses the Thucydidean distinction between the fictional and sweet or pleasant on the one hand (to muthōdes kai to hēdu) and what is non-fictional and useful (to mē muthōdes kai ōphelimon) as it occurs in Isocrates, Polybius and Plutarch. His conclusion (49), which differs in its thrust from mine, is that ‘by stressing the element of to terpnon with an allusion to Thucydides, Longus aligns his work with that of Herodotus’ while claiming for it the qualities of to terpnon, to muthōdes and to ōphelimon as well.

10. George Dimock first made me aware of the multiple nuances of sōphronousi here, including the sly glance at titillation. Cf. also the use of sōphrosunē in Chariton (2.4.5, a lack of sexual involvement that Eros finds hubristic). By offering his book as a cure for love, Longus challenges — among others — Theocritus (11.1, and see the discussion and references in Schönberger [n.1 above], 164–5) and, more explicitly, Philetas, who in the guise of a shepherd utters the words [at Daphnis and Chloe 2.7], ‘There is no pharmakon against love, nothing you can drink or eat or chant in incantations, nothing except kissing and hugging and lying together naked.’) Within the story there is no remedy but physical gratification, but for the audience, with the author as guide, there is a cure. Hunter [n.1 above], 50f., gives numerous references to poetry perceived as cure but I think errs in concentrating too narrowly on the comic tone of Daphnis’ and Chloe’s love story and so dismisses Longus’ aesthetic claim. For Hunter the force of Eros in the story is not even potentially destructive (42): ‘Neither Longus nor his readers are ever likely to be “afflicted” with the experiences of Daphnis and Chloe. Longus here puts an ironic distance between himself and his narrative, a distance which we come to recognize as typical of him: it may be true that oudeis erōta ephugen ē pheuxetai (no one has escaped or will escape love), but on the evidence of this novel no one would wish to.’

11. By using the term propaideusei, Longus presents himself — or his book — as an instructor. For propaideia as preliminary training to prepare one for more specialized study, cf. Plato Rep. 536d; Aristotle, Pol. 1337al9; Lucian, Rh. Pr. 14. By the word propaideusei the narrator is linked verbally to Philetas who is described as paideusas in 2.7.

12. Forms of the verb (terpō) and the related adjective and noun (terpnos, terpsis) abound throughout the work, too numerous to list. One may perhaps recall the statement of another very self-conscious author, Vladimir Nabokov: ‘For me a work of fiction exists only insofar as it affords me what I shall bluntly call aesthetic bliss, that is a sense of being, somehow, somewhere, connected with other states of being where art (curiosity, tenderness, kindness, ecstasy) is the norm.’ (‘On A Book Entitled Lolita’, in Lolita [New York 1958) 286.Google Scholar) Interestingly enough, Nabokov makes his high claim for his art in the afterword to a work that was greeted as pornography. (This is, by the way, an aspect of critical reception Nabokov, like Longus, is not averse to playing with: cf. the first sentence of the fictional ‘editor’s’ foreword, with the alternate title The Confessions of a White Widowed Male. Cf. Rohde’s [n.3 above] view of Longus, 549n.2: ‘Nach meinem Gefühl sind solche, in Wahrheit mückerhafte erotische Experimente sehr viel anstössiger …’)

13. Longus is playing with a Platonic topos, beauty ensnaring the soul through the eyes (e.g. Phaedrus 249d-252c; Symp. 209d-e; 211b-c) exploited also by Heliodorus (3.5) and Achilles Tatius (1.4.and 9; 5.13), used again by Longus with a typical — but not unplatonic — twist at 1.24 where the ophthalmōn halōsis, ‘ensnarement through the eyes’, takes place at the sight of naked bodies. (For the temptation, and the philosopher’s ability to control the fiery passion aroused by physical beauty and turn his eyes to even higher beauty, see Plato’s Charmides 155d, where Socrates says of his encounter with the young Charmides, ‘he looked into my eyes with an indescribably irresistible look, and I saw what was down inside his garment and I caught fire …’) For Heliodorus, the sight (opsis) of beauty can vanquish sexual restraint (sōphrosunē): so the noble brigand Thyamis thinks it impossible to gaze on (blepein) Charicleia and hold himself in check (sōphronein, 1.24), and even a life dedicated to self-control and chastity does not enable the eyes of the soul to battle those of the body enough to keep the wise Calasiris immune from the destructive sexual passion (pathos erōtikon) aroused by a beautiful courtesan — only flight can save him (2.25).

14. Longus likes puns (see the references coveniently collected in Hunter [n.1 above), 88–90, with notes on 127–128). Although critics have signalled more arcane instances of word-play, however, no one seems to have noted this very obtrusive one. For ktēma as landed property, see also e.g. Demosthenes On The Crown 41; Acts of the Apostles 5.1; and esp. Alciphron 4.9.2 and 4.13.1 where the garden described as erōtikou … ktēmation ontōs, ou geōrgou, the ‘dear little estate of a connoisseur of love, not a farmer’, is very reminiscent of what we have in Longus: the surviving part of the letter begins with the words erastou chōrion Numphais thusian legousa opheilein (‘… her lover’s place, saying she owed a sacrifice to the Nymphs’) and describes the spot as twenty stadia (about two miles) distant from the city, calling it a leimōn or kēpos (a ‘meadow’ — the same word that appears in amatory scenes from Homer on — or a ‘garden’), with a small amount of land for sowing near the villa and dominated by cypress and myrtle trees. Longus’ efforts as an author are echoed not only by Philetas as gardener (see p.119 below) but in Lamon, who prepares his master’s estate — the very one we are discussing now — eis pasan theas hēdonēn, so that it will be a wholly and variously pleasuring sight (4.1.2). Philetas’ garden and words are full of platonic reminiscences; attempts to impose a Platonic interpretation on the gardens in Longus’ text are, however, unconvincing (see Forehand, Walter E., ‘Symbolic Gardens in Longus’ Daphnis and Chloe’, Eranos 74 [1976], 103–112Google Scholar who notes the reminiscences of Phaedrus and Symposium and comments on Lamon’s garden as an ideal pastoral landscape, but uses this material as a springboard into an unconvincing philosophical interpretation).

15. On Philetas of Cos, see Pfeiffer, Rudolf, History of Classical Scholarship: From the Beginnings to the End of the Hellenistic Age (Oxford 1968), 88–93Google Scholar. Schönberger (n.1 above), 163, collects and summarizes evidence for the identification and for Longus’ probable imitation of Philetas. Cf. also Hunter (n.1 above), 76–83.

16. Chalk (n.7 above), 36, gives a serious reading of Eros’ epiphany; cf. Schönberger (n.1 above), 164: ‘Longos fasst Eros als die Urpotenz des Werdens auf, nicht als den hellenistischen Putto.’ I agree with Anderson (n.1 above), 132 n.9, where he speaks of the ‘amusing effect of loading Platonic reminiscences on to a Hellenistic putto’, and I am also inclined to see as a playful manipulation of a literary commonplace the reminiscences of Plato’s Symposium (cf. id., 46; Hunter [n.1 above], 109 n.47). To a list of humorous allusions to the Symposium I would add the mythologizing encomium of Eros delivered by a Gnathon ‘thoroughly schooled in the mythology of Eros at dissolute drinking parties’ (pasan erōtikēn muthologian en tois tōn asōtōn sumposiois pepaideumenos, 4.17.3). More recently Anderson (n.2 above), 222f., sees in a Sumerian text (BM 120011) a precursor to Philetas in the garden. Even granting similarities, however, the difference in tone between the earlier text and Longus’ seems to me to call into question rather than sustain a seriously religious reading of Longus’ imitation, if that is what it is.

17. The locus classicus for a discussion of ‘soft’ and ‘hard’ pastoral is Lovejoy, Arthur O. and Boas, George, A Documentary History of Primitivism and Related Ideas in Antiquity (Baltimore 1935Google Scholar), esp. i.10ff. Longus breaks the illusion of ‘soft’ pastoral (and, I will argue, thereby heightens the reader’s pleasure in the illusion) by contrasting it not with an equally idealized ‘hard’ pastoral vision but with a cruder, debunking realism. Critics recognize that ‘the first condition of pastoral is that it is an urban product’ in the words of Kermode, Frank (English Pastoral Poetry from the Beginnings to Marvell [London 1952], 14Google Scholar; cf. Wolff (n.3 above], 122, and Poggioli, Renato, ‘The Oaten Flute’, in Harvard Library Bulletin 11 (1957], 147–184Google Scholar, esp. 147), and that the genre is built on contradictions. The tension between the narrative surface of pastoral and the artificiality of the genre, between the rustic setting and the highly sophisticated language (Segal, Charles, ‘Theocritus’ Seventh Idyll’, in Poetry and Myth in Ancient Pastoral: Essays on Theocritus and Vergil [Princeton, N.J., 1981], 114Google Scholar) is characteristic of pastoral and essential to its very meaning (cf. ibid., 4–6: ‘Pastoral’s obvious artificiality and artifice make explicit the frame that every work of art draws around experience… [while] the realistic side of pastoral… only sets off the power of art that transforms and transfigures the daily round’). Artifice embellishes nature (in Renaissance poetics as well as in ancient pastoral — see Kermode 11f) and the pastoral setting that results is no realistic landscape. For a very useful discussion of the dominant contrast in Theocritean pastoral between ‘external reality’ and ‘inner fantasy’, see Lawall, Gilbert, Theocritus’ Coan Pastorals (Cambridge, Mass, 1967Google Scholar), esp. 8–12, 85 and 101–107. An illuminating treatment of pastoralism can be found in Martin Price’s essay on Pope’s pastoral and mock-heroic poetry in To the Palace of Wisdom: Studies in Order and Energy from Dryden to Blake (New York 1964), esp. 143–154Google Scholar.

Nonetheless, critics of Daphnis and Chloe often tend to forget that ‘countryside’ and ‘city’ alike in Longus’ work are artificial constructs. So Turner, Paul (‘Novels, Ancient and Modern*, in Novel 2 [1968], 15–24CrossRefGoogle Scholar) assumes more realism than the text warrants (e.g. ‘the plot … is drawn from life, not literature’; by using historia and echoing Thucydides, Longus ‘implies that he has written an authentic history, recording a universal fact of human experience’; Daphnis and Chloe are ‘not stock pastoral types. The foundlings are so named purely to make it appear that they are the genuine children of their shepherd foster-parents’); similarly Kestner, Joseph (‘Ekphrasis as Frame in Longus’ Daphnis and Chloe’, CW 67 [1973/74], 166–171Google Scholar) speaks of the work’s ‘realism rather than its fantasy’ (168). Critics from Goethe (Johann Wolfgang von Eckermann, Gespräche mil Goethe, Sonntag den 20. März 1831) to Croiset (Histoire de la littérature grecque v.800) to Wolff (n.3 above), 124 — cf. Haight, Elizabeth Hazelton, Essays on the Creek Romances (New York 1943), 132Google Scholar — see incursions from outside and corruption from the city as threatening the pastoral landscape. Wolff, for instance, states that ‘… a plot that has become entangled with the complex evils of court or town [is allowed] to straighten itself out under the simpler conditions of country life, and reach a happy end’ (123). Such a view credits Longus’ portrayal of the ‘court or town’ with too much realism and also accepts a ‘soft’ pastoral view of ‘country life’ — one which Longus systematically undercuts — as realistic. Even critics who note the cruder realism with which Longus undercuts the illusion of utopian optimism (e.g. Schönberger [n.1 above), 159, on 1.20.1; McCulloh [n.1 above], 111; Anderson [n.1 above], 44f.) do not analyze why Longus adopts this strategy, and even those aware that there is no ‘black and white opposition between Town and Country’ (Hunter [n.1 above], 55) blur the distinction between fantasy and reality: ‘The ending of the novel suggests rather a balance between the two [sc. town and country) — although the children are partially drawn into the urban world, their pastoral life continues. Beyond… superficial differences… there is no fundamental gulf between the two worlds: both are full of people trying to steer a path through life’s hazards’. Such a statement assumes too much realism for both the pastoral landscape and the urban world Longus depicts: there is indeed no fundamental gulf, but that is because both are fundamentally artificial.

18. On boukolikos vs. pastoralis as a generic technical term, see Halperin (n.3 above), 8ff. and 75–84 with references. For boukolikos, poimenikos, aipolikos as generic terms in Theocritus, see id., 182–186. (One might perhaps note Daphnis and Chloe 4.15.4 where Daphnis is rewarded as a kalos aipolos kai mousikos.) Longus here seems to me to be using poimenikos as a technical term; cf. the double-entendre at 1.21.5 (poimenikēn paidian), the closing words of the fourth book (poimenōn paignia, with the play on the author’s paignion reminiscent of the technopaignia of Alexandrian poets); on Poimenika in the title, see Hunter (n.1 above), If.

19. Apples and pomegranates are traditional lovers’ gifts, but Nape does not care about literary references or sentimental conventions; again, the author shares a joke with his readers. What characterizes Longus is that the joke is as much at the expense of the sentimental reader as of the characters: by desentimentalizing them, he focuses with amiable self-irony on the urban aesthete’s idealizing expectations. (Cf. 3.33.4, where Longus alludes to Sappho 105a LP when he has Daphnis pluck an apple for Chloe; see Schönberger 160, who also calls attention to Plato epigr. 2d, Theocritus 5.88 and 6.6, Vergil Ecl. 3.64; McCulloh [n.1 above], 72ff,; Hunter [n.1 above], 73ff. and relevant notes; Foster, B.O., ‘Notes on the Symbolism of the Apple in Classical Antiquity’, HSCP 10 [1899], 39–55Google Scholar supplemented by Littlewood, A. R., ‘The Symbolism of the Apple in Greek and Roman Literature’, in HSCP 72 [1967], 147–181Google Scholar, with bibliography at 177ff.)

20. Goethe (n.17 above) focuses on this detail although he does not say why he singles it out: ‘Es ist so durchdacht, dass darin kein Motiv fehlt und alle von der gründlichsten besten Art sind, wie zum Beispiel das von dem Schatz bei dem stinkenden Delphin am Meeresufer.’ Dolphins are associated with Dionysus: one saved Arion, the inventor of the dithyramb, from drowning when pirates forced him to jump overboard (Herodotus 1.23f; the dithyramb is Dionysus’ hymn, and Arion was dressed in full regalia. Pirates who attempted to kidnap the god were turned into dolphins and are depicted in one of the paintings gracing Lamon’s garden (see the Homeric Hymn to Dionysus; Ovid Met. 3.597ff.; Apollodorus 3.37f.; Hyginus 2.17; Nonnus 45.105ff.). Longus’ use of the mythological element here is characteristically desentimentalizing; as William Oram commented to me, Longus also provides another image of his work here: out of corrupting, organic nature comes gold, and the author is the one who’ creates the treasure.

21. When the cows sink the boat by jumping overboard to answer the pipe and save not only themselves but Daphnis too from the pirates, the narrator comments on their gifts as swimmers (1.30.6). Some have thought the passage an interpolation, e.g. Castiglioni, , RFIC 34 (1906), 312Google Scholar, followed by Dalmeyda in the Budé edition (2nd ed., Paris 1960), who thinks the passage an ancient interpolation, contemporaneous with the text and satisfying the imperial taste for paradoxa; Schönberger (n.1 above), 161, sees the passage as a genuine paradoxographical digression, as does Philippides, Marios, ‘The Digressive Aitia in Longus’, CW 74 (1980/81), 193–199Google Scholar, esp. 194. The humor is noted by Turner (n.3 above), 11f., and Anderson (n.1 above), 132 n.20.

22. For critics’ views of external attacks on the countryside, see n.17 above. But in fact outsiders in a sense preserve the integrity of the artificial landscape: as we have seen, even the pirates, whose attack brings about Dorcon’s death, serve a positive function. Dorcon’s violence has threatened to tear the fabric of Longus’ fragile cosmos; by returning the plot to the mode of romance, the pirates mend the damage and restore harmony.

23. Hägg (n.1 above), 39 seems to me to miss the mark when he says Longus is contemptuous of Astylos ‘who goes to the country only to enjoy its exoticism’. Both the landscape within the story and the literary work itself owe their existence to precisely this taste for xenikē terpsis (the enjoyment of the exotic) in urban gentlemen like Astylos and the hunter-narrator. One may note that they, like the Methymnean youths, have come to hunt; but Longus is making no easy distinction between hunters and shepherds, of the kind one might expect in pastoral. (See Poggioli [n.17 above], 152: ‘venaticai attitudes consistently oppose the pastoral…’) The hunters delight in soft pastoral fantasy; the shepherds, like Daphnis, kill birds (3.6.2).

24. The Methymneans may not just be lying to save face. The way they see the disappointing rustics reflects the way idealized views of a ‘golden age’/‘noble savage’ tend to be wedded to and easily slip into the opposite perception of the savage as innately brutal and inhuman. So Prospero’s view of Caliban in Shakespeare’s Tempest reflects the bitterness of disillusionment and his earlier, equally one-sided perception seems to be reflected in Gonzalo’s soft pastoral vision. (It may not be frivolous to reflect on how modem-day tourists who first believe that all natives are the smiling folk on travel posters may come to fear, after one nasty experience, that these same folk are all machete-wielding desperadoes.) I am indebted to Caroia Greengard for an illuminating discussion on the liminality of the shepherd in ancient literature (‘Philoctetes’: Theatre in Crisis, forthcoming, Hakkert): from the Babylonian Enkidu to Homer’s Cyclops, fifth-century satyrs, even the boorish rustics of Theocritus, the gentle shepherd and the aggressive wild man are not systematically differentiated. But Longus may be doing something other, or at least more, than introducing a literary commonplace here: he may be letting the ‘real’ world intrude as he does elsewhere, and the readiness with which the youths are believed when they return home may suggest such realism. Outside the idealized pastoral setting, the status of herdsmen is ambiguous. In the novels of both Achilles Tatius and Heliodorus the brigands of the Nile Delta are called boukoloi. See Achilles Tatius 3.9.2 and 4.12.4. In Heliodorus brigands camp in a spot called Boukolia, the Pasturage (1.4); cf. 2.20 where boukoloi is a synonym for lēstai, ‘brigands’. In Xenophon’s Ephesiaka we find the term poimenes, ‘shepherds’, used similarly (3.12).

25. The paintings (graphai) on the altar are alluded to tersely: Semele giving birth, Ariadne sleeping, Lycurgus in fetters, Pentheus being torn limb from limb, Indians being vanquished, Tyrrhenians being transformed; also represented are satyrs treading grapes, maenads dancing, and Pan himself, sitting on a rock and piping for the satyrs and maenads. The scenes, and the stories they call forth for the reader, contain more violence than the language suggests: Dionysus was torn from Semele’s body by Zeus when she was destroyed by his epiphany as the god of thunder and lightning: the sleeping Ariadne, abandoned by Theseus on Naxos, rests after her bacchic frenzy, claimed by Dionysus; Lycurgus, king of the Thracian Edoni, opposed the god and was bound and blinded; Pentheus was torn apart by maenads who included his mother, punished by sparagmos for denying the god’s divinity; the Indians who resisted Dionysus were defeated as were the Tyrrhenian pirates who tried to sell him into slavery and were turned into dolphins by his power. Even in Longus’ very brief, allusive catalogue, the same elements are at work as elsewhere in the text: violence is contained and set apart from the story proper; the brutality of the myths is brought to mind but glossed over; and the artist ends the tableau not on a note of violence or srife but with the harmonious dance-music of the piper Pan, with the maenads not ripping apart a human victim but dancing like idealized pastoral figures, rather like the protagonists of the work in the second book.

26. Don Quixote is trampled by bulls in a counterfeit Arcadia (2.58), then by swine (2.68). For first making me think about Cervantes’ use of pastoral in Don Quixote I am indebted to Elizabeth Harries, ‘Don Quixote: Chivalric and Pastoral Games’, a paper delivered at Smith College on March 26, 1980.

27. Reading kainon rather than kenon with Schönberger. Whatever the reading, the point remains the same.

28. Hubris, which can connote assault and battery and not infrequently sexual violence, coupled with the reminiscence of Sappho 105c LP (a reference noted by virtually every critic of Longus’ work — see e.g. McCulloh [n.1 above], 77) makes the garden a victim of rape. (See also recently Gagarin, Michael, ‘The Athenian Law Against Hybris’ in G. W. Bowersock (ed.), Arktouros: Hellenic Studies Presented to B.M. W. Knox ed. Glen W. Bowersock [Berliń and New York 1979], 230–236Google Scholar, with bibliography 230 n.1).

29. The more realistic insects that occasionally intrude are less sympathetic, for instance the biting flies that plague Chloe and refuse to be driven away from the milk-pails (1.22.3).

30. The flute-playing Satyr Marsyas dared to challenge Apollo, the god of the lyre, to a music contest; he was flayed for his efforts, his skin hung up. On Pitys, see n.40 below.

31. The unsentimental way in which peasants view their surroundings seems like brutality to an urban sentimental audience. In the film Lacombe, Lucien, Louis Malle shocks viewers and forces them to question their assumptions at the outset: the protagonist, a peasant youth, is performing menial tasks in a bare, dark and subdued hospital ward; suddenly he sees a brightly colored bird, on a branch outside the window. As he approaches the window and eyes the bird, most viewers think he is responding to the bird’s beauty, turning to nature for spiritual sustenance, away from the grayness of the ward. When he takes out his slingshot and kills the bird, as most country boys would (cf. Daphnis at 3.6), he pretty much kills the audience’s fusions as well. (In a New York theater, there was an audible, collective gasp from the audience.)

32. On tragic devices, such as the repeated appearance of the messenger, see Wolff (n.3 above), 122; on the characters and conventions of bourgeois New Comedy, see Edmonds (n.3 above), xi; Schönberger 6f., and 175; Reardon, B. P., ‘The Greek Novel’, Phoenix 23 (1969), 291–309CrossRefGoogle Scholar, esp. 292f.; McCulloh (n.l above), 57ff; Anderson (n.l above), 43f. Heiserman (n.3 above), 140ff., notes comic elements but (as Hunter [n.l above], 119 n.31 points out) conflates in his discussion elements from new comedy and very broadly defined ‘comic’ features. See also Hunter’s discussion (67–70); he points out (68) that Longus exploits the reader’s familiarity with the plots of new comedy to give the reader a suspenseful frisson: will Daphnis and Chloe discover they are brother and sister?

33. habroteron: habros, ‘graceful’, ‘pretty’ is an epithet of Eros in Anacreon (17b) and the Graces in Sappho (128 LP); from the mid-fifth century B.C. it comes to be associated frequently with Persians and other orientals, and more generally with pretty, delicate objects. But by the time of the novelists the adjective — especially in the comparative degree — connotes stylish, voluptuous sexual attractiveness. Heliodorus (1.17) has Thisbe tell Demainete to turn herself out to advantage (kosmei sautēn) so as to appear sexually alluring (habroteron echousan hēkein prosēkei). In the Budé edition (Héliodore, Les Ethiopiques, ed. R. M. Rattenbury and T. W. Lumb [2nd ed., Paris I960)), the editors note that the adjective ‘implique à la fois une idée d’élégance et de mollesse séduisante et voluptueuse’. So in 2.21.1 of Heliodorus’ romance Knemon cuts his hair just enough to change his coiffure from that of a brigand to that of a Don Juan (apotemōn oun ho Knēmōn hoson eikos ēn elation koman lou lēstrikou ton habroteron); in the preceding sentence, brigands (lēstrikoi) are paired and contrasted with erōtikoi, or ‘seducers’, and in this sentence habroteros is used as a synonym for erōtikos. Cf. 4.7.11 of the same work, where habroteros = ‘groomed for amatory conquest’, and see editors’ note ad loc.

34. truphōn aei means something like ‘always enjoying the good things in life’, like the Methymnean youths who are described as discomfited truphomes (2.19.1; cf. 4.33.2 where truphē pollē is used of the luxurious trappings of Dionysophanes’ retinue; and 4.3.1 where the spectacular views from Dionysophanes’ garden contribute to the truphē, the voluptuous and luxurious pleasure experienced in the garden.) In the context of ouk apeiros erōtikēs lupēs, the connotation of truphē as voluptuous sensuality may also come to mind (cf. Aristophanes’ Lysistrata 387, 405, for example). So too megalophrēn means noble or generous but can also connote arrogance (cf. references s.v. in LSJ).

35. This says nothing about Astylos’ proclivities, but it does show the nonjudgmental attitudes of his society. In Achilles Tatius’ Leucippe and Clitophon the heterosexual hero accepts his cousin Clinias’ pederasty sympathetically (1.7ff.) and engages in a friendly debate on the relative merits of women and boys as sexual partners with another pederast, Menelaos (2.35f.).

36. The costly civic responsibilities of chorēgia (bearing all the costs of collecting, maintaining, instructing and equipping a dithyrambic, tragic or comic chorus for a festival) and triērarchia (bearing the costs and full responsibility for launching a warship, making it ready for service, keeping it efficient and in good repair while at sea and returning it to the state in seaworthy condition) were demanded of the wealthiest citizens in classical Athens and seem to have been paniculariy burdensome in the fourth century (see esp. Lysias 19, ‘On the Property of Aristodemus’; cf. 21. 1–5, 16, 22. I am indebted for these references to Matthew J. Dillon, Aristophanes’ ‘Ploutos’: Comedy in Transition [Diss. Yale 1984], 23ff.). What dramatic date Longus may have had in mind, however, is unclear, and any historical context or direct relevance to the economic realities of Longus’ audience are impossible to recover from the casual remark of a stock character from the stage of New Comedy.

37. On infanticide and the exposing of infants in the Hellenistic period, see Tarn, W. W., Hellenistic Civilization ³ (London 1952), 100–102Google Scholar.

38. On bouleuō, bouleusis, bouleusas as terms connoting legal responsibility, see Said, Suzanne, La faute tragique (Paris 1978), 169fGoogle Scholar. On hekōn used to refer to a premeditated act, see Vernant, Jean-Pierre, ‘Intimations of the Will in Greek Tragedy’, in J.-P. Vemant and P. Vidal-Naquet, Tragedy and Myth in Ancient Greece, tr. Lloyd, Janet (Sussex and New Jersey 1981), 28–62Google Scholar, esp. 38 n.24.

39. Critics have noted ways in which the aitia are connected thematically to the text. Frequently noted are Chloe’s crown of pine (1.24.1) parallelling Phatta’s (1.27.2); Daphnis’ words to Chloe as he begins the Phatta myth (‘There was once a maiden, maiden’: ēn parthenos, parihene); Chloe’s answering Daphnis ‘just like (an] echo’, kathaper ēchō (3.2.1). See for instance Schönberger (n.l above), 161–162; Chalk (n.7 above), 40–42; Deligiorgis, Stavros, ‘Longus’ Art in Brief Lives’, Philological Quarterly 53 (1974), 2–9Google Scholar; Philippides (n. 21 above) 193–199; McCulloh (n.l above), 65f.; Hunter (n.l above), 52–57. McCulloh (65) notes that all three aetiological transformation myths involve music, and ‘all contrast with the main narrative by ending unhappily, with an increasing element of active violence.’ Hunter (53) seems to follow McCulloh closely, although he does not mention him at this point; he stresses the transforming role of music in all three tales (where it appears either as an erotic or as an anti-erotic element) and the increasing violence and menace. Both critics finally read the tales differently from the way 1 do. McCulloh is very good on the sparagmos of Echo at 3.23 (‘The dismemberment… through its association with Orpheus and Dionysus, connotes the other side of Eros-Dionysus: destructive violence, a foreshadowing of Lampis’ ravage of the garden, and of the violent scenes depicted in the Dionysus shrine in that garden. Here, however, the violence pointedly contrasts with Daphnis’ unwillingness to hurt Chloe.’ — op. cit., 66) but draws no further conclusions. His treatment of the desecration of Lamon’s garden by Lampis is shaped by a serious religious interpretation of the work (cf. Chalk, 46f.): to him the reminiscence of Sappho at 4.8 suggests ‘a universal erotic destructiveness and violence, which will have its place in a fundamentally religious economy of life involving rebirth and new growth after loss’ (77). Philippides (197) also sees the contrast between the violence of the Echo myth and the ‘events of the plot of the third book, which is characterized by the total absence of physical violence’, but goes on, in a loosely anthropological-religious framework, to blame the victim: ‘In the descriptions of nature stress is placed on fertility and abundance, and the marriage between the hero and heroine is Finally arranged, but Echo dies because of her total denial of Eros.’ Hunter (53) notes that the sparagmos of Echo ‘is a version found to my knowledge nowhere else’, and is a ‘“Dionysiac element”, more specifically reminiscent of the fate of Orpheus at the hands of Thracian maenads'. (Like Lampis, I would note, these maenads are rejected hopefuls.) He notes that Echo’s story, like Orpheus’, ‘proclaims the power of music over the forces of destruction’, but does not, to my mind, explain convincingly why Longus includes such a violent variant of the Echo myth or how the increasing violence of the aitia functions when he says the ‘increasing savagery of these stories foreshadows the loss of Chloe’s virginity’ and notes the ‘Longan irony with which Daphnis grows less rather than more insistent as time goes on, whereas Chloe shows no obvious reluctance to yield…’ (540. Hunter also notes that the ‘violence and desecration recall the garden of Lamon’ but sees the desecration of the garden as symbolic of the ‘new life which Daphnis and Chloe are about to encounter and, more particularly, the dangerously perverse and urban threat which Gnathon poses’.

40. The story of Pitys and Pan is told by Nonnus and Lucian; a variant version that first appears in a tenth-century compilation but may be a good deal older gives Boreas and Pan as Pitys’ rival suitors and has her destroyed by Boreas when she chooses Pan. See RE 1881–2 (Hanslik) s.v. Pitys (2) and 1221–1225 (Oder) s.v. Geoponika.

41. As a feminist critic has pointed out, being taken over by her pursuer as his symbol is the ultimate defeat of the victim who undergoes metamorphosis to preserve her integrity: for an excellent discussion of how Ovid treats Daphne’s transformation I am endebted to Mary Kay Game), ‘Apollo and Daphne: The Making of a Sign’, paper delivered at the annual meeting of the American Philological Association, Toronto, December 1984.

42. ‘The Tale of Foolish Curiosity’ in Part One of Don Quixote shares another feature with Longus’ i. .t aetiological tales: its tragic outcome, contrapuntal to the happy endings of the other love stories that find their way into the main plot, serves to foreground the comedy and romance conventions Cervantes is manipulating. As Northrop Frye (n.7 above) points out, inset tales with ‘a narrative movement opposite to that of the main story… establish the main story as one of a category of stories, giving it broader significance than it would have had as an isolated story’ (12). Relevant also to Longus’ strategy as I see it is Frye’s discussion (135) of tragic minor characters in Ivanhoe: they remind the reader of the fate Scott’s genre has spared the protagonists.