NOTES

INTRODUCTION

1. See Cook 2006, 63.

2. Briefly discussed by Edwards 1995.

3. See Cook 2006, 63. Dawson 2002 is essential reading for the development of Christian figural reading.

4. See Mastrangelo 2008, 160–75.

5. Prudentius’s popularity continues to grow; there have been many significant studies published since the late twentieth century. As an entrée into his work, see, for books, Bardzell 2009, Gnilka 2000, Gosserez 2001, Lühken 2002, Malamud 1989, Mastrangelo 2008, Nugent 1985, Palmer 1989, Paxson 1994, Petruccione 1985, Roberts 1989 and 1993. Unfortunately, Dykes 2011 appeared too late for me to take into account in this book. For articles, see Ballengee 2002, Burrus 1995, Conybeare 2007, Goldhill 1999, James 1999, Kaesser 2002, Machosky 2003, Malamud 1990 and 2002, Miller 2000, Nugent 2000, Ross 1995. Important earlier works include Fontaine 1980 and 1981, Gnilka 1963, Haworth 1980, Herzog 1966, Lana 1962, Mahoney 1934, Smith 1976, Taddei 1981, Thraede 1965.

6. Palmer 1989, 22.

7. Cameron 1970, app. B; Charlet 1982; Fontaine 1980 and 1981; Costanza 1983.

8. Other examples of aristocratic conversi include Sulpicius Severus, Paulinus of Pella, and the author of the Carmen ad uxorem. Griffe 1965, 128–48 discusses the life of such converts. For the life and work of Paulinus, see especially Conybeare 2000 and Trout 1999.

9. Roberts 1993, 3; Fontaine 1981, 145–47.

10. Mastrangelo 2008, 10.

1. WRITING IN CHAINS

1. Nam et in hoc philosophi et oratores et poetae perniciosi sunt, quod incautos animos facile inretire possunt suauitate sermonis et carminum dulci modulatione currentium. [mella sunt haec venena tegentia] (Inst. 5.1.10–11). See Mastrangelo 2009, 317 for a discussion of this passage and the patristic attitude toward pagan poetry.

2. MacCormack 1998, 190; Mastrangelo 2009, 319.

3. Mastrangelo 2009, 327.

4. Argued by Mastrangelo 2008, especially chapter 2 and the epilogue; briefly recapitulated in Mastrangelo 2009, 324–25.

5. Miller 2001, 7. On figural reading, see Dawson 2002 and Fabiny 1992.

6. Chin 2008, 12.

7. Ibid., citing Sidonius Apollinaris 1.6.44.

8. Roberts 1989, chapter 3, especially 85–89.

9. Very similar scenes occur in Prudentius’s Pe. 11.111–22, the dismemberment of Saint Hippolytus, and Psy. 720–25, the death of Discordia. See Malamud 1989, 98–99. Henderson 1983 discusses violence in the Peristephanon.

10. Scholarship on violent spectacle in the Roman world includes Grig 2002, Shaw 1996, Futtrell 1997, Kyle 1998, Ballengee 2009, Frankfurter 2009. Enders 1999 focuses on the spectacle of violence in medieval theater, and includes a discussion of the classical background; for a collection of articles on violence in late antiquity, see Drake 2006b.

11. Miller 2001, 3.

12. Mitchell 1986, 8, quoted in Miller 2001, 5.

13. Theon, Progymnasmata, 118.1.7. I draw extensively on Webb 2009, chapter 3 in the following discussion of ekphrasis and enargeia. See also the excellent introduction to Miller 2009.

14. Cited and discussed by Webb 2009, 95–96.

15. Brilliant 1999, 224; I owe the reference to Miller 2005, 32.

16. Carruthers 1998 and 1990. Yates 1966 remains indispensable on the topic of memory and mnemonics.

17. Miller, 2005, 37–43, discusses Prudentius’s spiritual landscapes, though her emphasis is on landscapes in the Peristephanon.

18. For a discussion of Prudentius’s mnemonic technique, see Gosserez 2001, 189–92; the classical sources on the notion of memory place include Cicero, De orat. 2.353–54; Quintilian, Inst. 11.2.23; Ad Herennium 3.16–24; Martianus Capella, De nuptiis 410–30.

19. Discussed by Carruthers 1998, 147 (although she confuses the characters: “when Hope cuts off the head of Luxuria”).

20. Fontaine (1981, 188) remarks on “la puissance de l’imagination visuelle, qui est peut-être la faculté poétique majeure de Prudence.”

21. Grig 2002, 328.

22. Prudentius’s violence has upset many readers: as Levine (1991, 5) remarks, “They find his use of violence excessive, his mixing of genres anti-classical, and his shifts of tone generally troublesome.” On violence in the Psychomachia, see Paxson 1994, 66–71; Lewis 1958, 69; Mastrangelo 2008, 145–55; Nugent 2000, 20–25.

23. In addition to Enders 1999 and Carruthers 1990 and 1998, excellent scholarly studies of the art of memory include Yates 1966 and Coleman 1992.

24. There are many versions of this well-known anecdote, including Cicero, De oratore 2.351–55; Martianus Capella, De nuptiis 177g. See Enders 1999, 63–65. Here is Quintilian’s version: Simonides “had written an ode of the kind usually composed in honour of victorious athletes, to celebrate the achievement of one who had gained the crown for boxing. Part of the sum for which he had contracted was refused him on the grounds that . . . he had introduced a digression in praise of Castor and Pollux, and he was told that, in view of what he had done, he had best ask for the rest of the sum due from those whose deeds he had extolled. And according to the story they paid their debt. For when a great banquet was given in honour of the boxer’s success, Simonides was summoned forth from the feast, to which he had been invited, by a message to the effect that two youths who had ridden to the door urgently desired his presence. He found no trace of them, but what followed proved to him that the gods had shown their gratitude. For he had scarcely crossed the threshold on his way out, when the banqueting hall fell in upon the heads of the guests and wrought such havoc among them that the relatives of the dead who came to seek the bodies for burial were unable to distinguish not merely the faces but even the limbs of the dead. Then it is said, Simonides, who remembered the order in which the guests had been sitting, succeeded in restoring to each man his own dead” (Inst. orat. 11.2.11–13, trans. H.H. Butler).

25. Enders 1999, 68–69.

26. Carruthers 1998, 143; Katzenellenbogen 1964.

27. Miller 2001, 3.

28. Origen, Hom. in Ex. 1.1 (Heine) quoted in Miller 2001, 5.

29. Prudentius refers specifically to a number of his poems in the preface to his collected works: the Cathemerinon (line 37); the Hamartigenia and Apotheosis (line 39); the Romanus hymn, now printed as Pe. 10 but originally a separate poem (line 40); the Contra Symmachum (line 41); and the Peristephanon (line 42). As Shanzer 1989 argues, the Psychomachia is not referred to in the Praefatio’s list of Prudentius’s poems, and is probably to be dated after 405.

30. Of all his poems, the one that comes closest to focusing on contemporary events is the Contra Symmachum, a two-part poem in which Prudentius dramatizes the controversy that arose over the cessation of state payments for priesthoods and the removal of the statue and altar of Victory from the Senate house in Rome. Quintus Aurelius Symmachus had requested the restoration of the statue in 384, but was successfully defeated by the intervention of the influential bishop Ambrose of Milan. Cunningham 1976 discusses the contexts of Prudentius’s poems.

31. Pelikan 1993, 40–74 and 200–31, discussed by Mastrangelo 2008, 82–83.

32. Mastrangelo 2008, 83.

33. Ibid., 87.

34. As Bardzell puts it, Prudentius’s artificial use of language in the Psychomachia “might not be artificial in the privative sense (i.e., an idiosyncratic deviation from normal language . . .), but rather artificial in an experimental sense, an attempt at rearticulating reality in a radically faithful way” (2009, 44).

35. Mastrangelo 2008, 169.

36. Benson 1966, 75, cited by Fyler 2007, 53.

37. Peristephanon 10 was probably originally published as a separate poem, and later included with the other poems of the Peristephanon. Its length sets it apart from the other poems in the collection. I have discussed Cyprian’s tongue and the ambivalent aspects of his rhetoric in Pe. 13 in Malamud 1989, 117–21. Miller (2005, 40) has a good discussion of the surreal aspects of Romanus’s tongue. See Frankfurter 2009, 229–30 on the erotic and voy eur is tic aspects of Cyprian’s martyrdom.

38. See James 1999 for an analysis of the Psychomachia as a Christianized form of the spectacle of the arena.

39. Loraux 1987; see also Sissa 1990, 53–70, and King 1983.

40. Nugent 2000, 22. See Loraux 1987, 50–55, and Cixous 1981, 41–55 for further discussion of this motif.

41. Roberts 1989, 28–29.

42. James 1999, 79.

43. Levine (1991, 12–30) discusses the relation between Prudentius’s excessive violence in the Peristephanon and his stylistic excesses.

44. On Prudentius’s sense of humor, which is an important aspect of his poetry, see Conybeare 2002, Goldhill 1999, Malamud 1989 (esp. chapter 5), and Levine 2005.

45. De Man 1986, 76, cited by Paxson, 1994, 69.

46. As Paxson says (emphasis mine), “Prudentius’ focus upon the imagery of the destruction of the face, therefore, is a literalized reverse of prosopopeia. It is the symbolic dismantling of the trope by which the text invents the figural characters who inhabit its actantial narrative. Prosopopeia, or more precisely, lexopeia, is self-reflexively thematized as a central topical ground in the Psychomachia. . . . At times, [the Virtues’] main semiotic function (as opposed to their obvious thematic role) is to reflect not the ethically abstract offices of virtuehood, but rather the self-reflexive figural armature of the entire Psychomachia. This ‘armature’ is the self-conscious omnipresence of the text’s foundational figural operator, personification” (Paxson 1994, 69–70).

47. Mastrangelo 2008, 154.

48. Ibid., 154–55.

49. Bardzell 2009, 51.

2. FIGURING IT OUT

1. Claudian, Prudentius’s younger contemporary, also uses the device of an iambic preface; see Gruzelier 1993 ad loc. It is not clear when the Greek titles were attached to the poems. Palla 1981, 23n49 argues for taking the Apotheosis and Hamartigenia as separate poems, not as two parts of the same work, based on the differences between the prefaces of the works.

2. Averil Cameron makes a point about Byzantine antiheretical texts that is worth extending to late antiquity in general: “Whether we like it or not as historians, writing heresy, in all its various forms, did occupy a major place in Byzantium—so much so indeed that a full treatment would in its way constitute a new history of Byzantium. This is far from having been written as yet. But meanwhile, at the very least, I suggest that one ought to read these compositions, so strange to our minds, as part of Byzantine pedagogy and the Byzantine sociology of knowledge, self-perpetuating constructions that helped to formulate thought and underpin social norms” (Cameron 2003, 484).

3. See below, 78–81, for a discussion of why Prudentius associates Cain with dualism.

4. The importance of rhetorical performance in Augustine’s career is briefly discussed by Webb 2009, 16. For the development of a Christian rhetoric in the fourth and fifth centuries, see Cameron 1991, Brown 1992.

5. See the essays of Clark, Miles, and Lim on the dialogue form in late antique Christian writing in Goldhill 2009. Goldhill (2009, 6) casts the Psychomachia as a type of dialogue in which there is no exchange of views: “Each virtue gets to deliver a speech, like a Homeric warrior on the battlefield, upholding Christian values, before vanquishing the enemy. It is more like an extended model of the martyr’s put-down to his enemies than an exchange. There is no uncertainty, the only questions are rhetorical, and the dominant model is the battlefield boast—where good can only defeat evil, not exchange views.” I would complicate this argument somewhat: the destabilizing use of language in the Psychomachia works against the simplistic framework, as the apparently stark differences between Vices and Virtues are undermined.

6. Lavarenne 1961, xi. See Stam 1940, 8–13; Evans’s (1972) introduction to Tertullian’s Adversus Marcionem, ix–xxvi; Palla 1981, 16–17. Marcion died in 160 and the heresy named after him was dead long before the end of the fourth century. As various scholars have pointed out, the arguments against Marcionism apply equally well to heresies much more prevalent in Prudentius’s day: Priscillianism, Arianism, and Manichaeism, which famously claimed the young Augustine as a convert and teacher. One of Marcionism’s salient features was its use of the contradictions between the Old and New Testaments as a theoretical tool to undermine orthodox doctrine: pitting the scriptures against one another was a particularly powerful rhetorical ploy.

7. Though Marcion is accused of believing that matter is inherently evil, this is probably an exaggeration of his views. McGowan 2001 provides a balanced assessment of the evidence. What ever Marcion’s own views were, it is likely that Prudentius’s understanding of them was influenced by apologists such as Tertullian (Adv. Marc. 1.16.1), Clement (Strom. 3.3.12), and others, who claimed that Marcion believed matter was evil.

8. Wyrick (2004, 299–301) discusses Marcion’s impact on canon formation in the early church and Tertullian’s response.

9. Miller 2001, 261. She adds: “Irenaeus understood Scripture and Tradition as an Author; what came to be called canon functioned for him to guarantee, to specify, and to close the contents of meaning. The Gospel of Truth, on the other hand, understood what came to be called canon hermeneutically; its concept of authority was rooted not in the past but in the present, and grew out of its understanding of the dynamics of language.”

10. Conybeare 2007, 229. My emphasis. While I agree with Conybeare on most points, I would point out that while Prudentius does not mention any nonscriptural texts, he makes obvious and frequent allusions to nonscriptural texts in the poem, even if he does not refer to them by name.

11. The preface to the Psychomachia offers the biblical story of Abraham as an interpretive model for reading the Psychomachia, a misleading move that has caused considerable confusion, since the typological allegory from scripture used in the preface is quite different from the personification allegory in the rest of the poem. “Prudentius’ allegory is calling attention not so much to metaphysics, but rather to meta-narrative. By translating the story of Abraham to the inner battle for the soul, the narrator is modeling an interpretive methodology, in which the reader learns to recognize narrative truths. These truths are understood as facts, or sequences of cause and effect, rather than objects or things” (Bardzell 2009, 42–43). Paxson (1994, 78–95) has an illuminating discussion of Job, the one scriptural figure who takes part on the primary narrative level of the Psychomachia. The other biblical figures mentioned exist outside the narrative frame, in what Carruthers 1998 would call the ornaments: similes or embedded narrations, such as the narration by the Virtue Hope (Spes) of the tale of David and Goliath in Psy. 291–99.

12. Carruthers 1998, 145.

13. Taddei 1981, 2–3; Tertullian, Adv. Jud. 5 (Tränkle, Wiesbaden, 1964); Ambrose, De Cain et Abel 2.6 (CSEL 32 pars. 1, Schenkle, 1893). Augustine, De civitate dei 15.7 offers several different interpretations of Cain’s error, concluding in quo autem horum Deo displicuerit Cain, facile non potest inveniri (CCSL 48, Dombart, 1955). See Herzog 1966, 122ff. for a discussion of patristic interpretations of the Cain and Abel story.

14. Vetus Latina: Nonne, si recte offeras, recte autem non dividas, peccasti? Quiesce! Cf. the Vulgate, which does not mention division: Dixitque Dominus ad eum quare maestus es et cur concidit facies tua? nonne si bene egeris recipies sin autem male statim in foribus peccatum aderit sed sub te erit appetitus eius et tu dominaberis illius. See Stam 1940, 41–42 and 120–21; Grasso 1972, 124–35; Palla 1981, 122–23.

15. Not until Prudentius has moved from the iambic trimeter preface to the dactylic hexameter body of the poem is there some attempt at explanation. Taddei argues that H. 6–7 (terrarum tibi forma duplex obludit, ut excors/dividuum regnare Deum super aethera credas) imply that Prudentius’s preoccupation is with dualism, and that he wishes to associate the notion of division with sin.

16. Josephus, Jewish Antiquities 1.2.2. See Fyler 2007, 29 for a discussion of the tradition of Cain as master of division.

17. According to Fyler 2007, 28, Philo, commenting on Genesis 4:17, where Cain sets up the first city, explains that this is not literally true, but rather means that Cain was setting up a creed of his own, and the buildings were plausible arguments and sophistic devices.

18. Taddei 1981, chapter 1, provides a summary.

19. Nichols 2010, 122, in a review of Fyler 2007.

20. Joshua Levinson, in his anaysis of late antique rabbinic exegetical narrative, frames the problem as one of competing genres, in which narrative undermines exegesis: “The exegetical narrative is composed of a story which simultaneously represents and interprets its biblical counterpart. As a hermeneutical reading of the biblical story presented in narrative form, its defining characteristic lies precisely in this synergy of narrative and exegesis. As exegesis, it creates new meanings from the biblical verses, and as narrative it represents those meanings by means of the biblical world. As exegesis, it is subservient to the biblical narrative, but as a story in its own right, it creates a narrated world which is different from its biblical shadow. It is obvious that the combination of these two elements creates a certain dissonance. Narrative and exegesis are two very different methods of persuasion, based upon divergent, if not opposing, presuppositions of ‘author-ity.’ It is specifically this tension between sameness and difference, subservience and creativity, which establishes the genre’s identity” (emphasis added).

21. Gnilka (2000, 674–90) has argued persuasively, though not conclusively, that H. praef. 35–47 is an interpolation. He believes that the identification of Marcion with Cain is misleading, for the poem is more an attack on dualistic theologies in general than on Marcion in particular. For my argument, it does not matter whether the text’s turn to allegorical exegesis comes immediately after the story of Cain, or whether it includes the identification of Marcion with Cain; in either case the narrator must guide the reader toward proper interpretation.

3. SEEKING HIDDEN TRUTH

1. Augustine systematically explores the complexities of reading and interpretation in De doctrina; see Stock 1996, 190ff. for a good summary of De doctrina’s argument.

2. Gosserez 2001, 34–52 discusses this passage and the related images of sun, mirror, and eyes. See also van Assendelft 1976, 18ff., and Gnilka 1980, 411–46, on Prudentius’s use of allegory.

3. See Taddei 1981, 74ff. for further discussion of this allegory.

4. Silvas Academi refers to Plato’s school, the Academy.

5. Mastrangelo 2008, 133–36 discusses the importance of Platonist doctrines of the soul, particularly of Plotinus in Ad Marcellam for the praefatio of the Psychomachia, and notes in particular the importance of the allegory of the cave for the notion of the body as a prison from which the soul strives to be released.

6. E.g., Seneca, NQ 1.11.1–3; Cicero, Rep. 1.11–15; Livy, 28.11.3 and 40.21.12; Pliny, NH 2.31.99.

7. There is no conclusive verbal parallel, but the association of double vision with madness is suggestive.

8. See Sextus Empiricus, Against the Professors 7.247–52; Webb 2009, 116.

9. Webb 2009, 118–19.

10. Gosserez also cites Philo, De decalog. 12, and Plutarch, Ad principem iner. 5.

11. Taddei (1981, 75) comments on Romans 1:20 in her discussion of the metaphor of sight. The context of the Pauline passage is relevant to the Hamartigenia, which takes as its subject the origin of sin: in it Paul lays out a catalog of the sins of the unrighteous, who fail to see the truth of God. Sin is linked to a failure to see the truth.

12. See Cook 2006, 35–38, for sources and a discussion of Augustine’s comments.

13. Cited ibid., 31, from the edition of Holtz (1981, 672).

14. Aristotle, Rhet. 3.11.6, Loeb edition, trans. J.H. Freese, cited by Cook 2006, 33.

15. This is indeed, as Augustine says, an obscure allegory. It seems to indicate insatiable need. Augustine is unusual in saying there are three rather than two daughters (De trin. 15.9). The Vulgate has sanguisugae duae sunt filiae dicentes adfer adfer tria sunt insaturabilia et quartum quod numquam dicit suffici. (“The horse leech has two daughters, saying ‘Give, Give!’ There are three things that are never satisfied and a fourth that never says ‘enough.’”) The horse leech (sanguisuga) is a form of leech that attaches to the flesh with two suckers; some rabbinical commentators, however, take the word ‘alukoh to be the name of a prophet rather than the word for leech (see Plaut 1961, 305).

16. Cook 2006, 40–41. Augustine stresses the unlikeness of man to God again a little later: “Wherefore, since we have found now in this enigma so great an unlikeness to God and the Word of God, wherein yet there was found before some likeness, this, too, must be admitted, that even when we shall be like Him, when ‘we shall see Him as He is’ (and certainly he who said this was aware beyond doubt of our present unlikeness), not even then shall we be equal to Him in nature” (De trin. 15.16.26, trans. Haddan).

4. FALLING INTO LANGUAGE

1. The phrase is from Juvenal 14.136, describing the insanity of the miser.

2. Goldhill makes clear in his introduction that his assertion that Christianity meant the end of dialogue in antiquity is intended to provoke contemporary scholarly dialogue. See the contributions of Clark, Miles, and Lim in the same volume for quite different analyses of dialogue forms and the social place of dialogue in the early church.

3. Herzog 1966, 94.

4. Invidia is etymologically linked with videre, “to see.” The verbal form invideo means literally “to look askance.” Skewed vision, as I discuss below, is of great thematic significance to Prudentius, as it is to Milton. Envy’s inability to tolerate the sight of the gaudia iustorum foreshadows the vision of the souls of the damned and the just at the end of the poem.

5. Palla 1981,168 reviews other uses of the rare adjective anguiferum.

6. Palla 1981, 166 provides a number of parallel passages from patristic literature on the snares of the devil. On the association of sophistry with tricky nets, see Detienne and Vernant 1978, 42, and Cook 2006, 244–45.

7. The version of Genesis quoted by Augustine (De civitate dei 22.3) uses the term gigans to refer to Nimrod (Chus autem genuit Nebroth, hic coepit esse gigans super terram; hic erat gigans venator ante dominum Deum), as does the Septuagint; the Vulgate does not: “ipse coepit esse potens in terra et erat robustus venator coram Domino.”

8. Quaestiones in Genesin 2.81–82, cited by van der Toorn and van der Horst 1990, 17.

9. Philo, De gigantibus 65–66, cited by van der Toorn and van der Horst 1990, 18.

10. Justin Martyr, Cohortatio ad Gentiles 28; Origen, De principiis 1.5. Leonard 1990, 88–90 discusses the tradition of the names Satan and Lucifer. See also Pagels 1995, 39–43. Satan was more of a divine colleague than an evil adversary in Job.

11. Fragment Targum ad Gen. 10:9, cited by van der Toorn and van der Horst 1990, 24.

12. “And so this giant is to be recognized as a ‘hunter against the Lord.’ And what is meant by the term ‘hunter’ but deceiver, oppressor, and destroyer of the animals of the earth? He and his people therefore, erected this tower against the Lord, and so gave expression to their impious pride; and justly was their wicked intention punished by God. . . . So that man, who would not understand God when He issued His commands, should be misunderstood when he himself gave orders. Thus was that conspiracy disbanded, for each man retired from those he could not understand, and associated with those whose speech was intelligible; and the nations were divided according to their languages, and scattered over the earth as seemed good to God” (Augustine, Civ. Dei. 16.4, trans. Dods).

13. Fyler 2007, 37, who cites Pseudo-Clement Recognitions 1.30 (PG 1:1224–24).

14. Leonard 1990, 87.

15. Ibid., 53–56.

16. The name Lucifer does appear once in the Cathemerinon (C. 12.32), but in a context that suggests it refers to the morning star rather than the devil.

17. His pride and jealousy lead him to insist that he is his own creator in H. 170–74.

18. Edwards 1995, 448.

19. For other examples, see Palla 1981, 217; Bartelink 1967; and Fontaine 1964.

20. Ambrose (Expositio evangelii secundum Lucam 7.73, ed. Schenkl, CSEL 32.4, 1902, p. 312, 16), and Augustine (Quaestiones evang. 2.19 [PL 35, 1340]).

21. Est vera secta? I have translated this phrase, which literally means “Is our doctrine true?” as “Are we on the true path?” to preserve the etymological play on secta, sectabitur, sequor. There is also a play on different meanings of recta: rectam fidem, “right faith,” line 2; and rectam tramitem, line 12, “the right (correct) path,” but also “the straight path,” as opposed to the twisting paths of sin.

22. There is another play on words here: hostilis manus first appears to mean “an enemy hand,” but the second manus, construed with latronum, has to mean “band.”

23. De doctrina 1.10–11; Camargo 1998, 399n7 has a good bibliography on Augustine’s transformation of the Neoplatonic topos of the journey.

24. Bartelink 1967 traces the topos of the devil as a robber or highwayman in patristic literature and provides an extensive list of references.

25. At non caede viri tanta perterrita Lausus./pars ingens belli, sinit agmina: primus Abantem/oppositum interimit, pugnae nodumque moramque (Aen. 10.426–28). “Lausus, a major force in the war, did not allow carnage/Pallas inflicted to panic his troops. He started by killing/Abas, a knot to frustrate any blade, who rose up to oppose him” (trans. Ahl). The phrase is difficult to translate. Literally it means “a knot and delay of battle,” and forms part of the sustained imagery of knots and snares throughout the poem. I have resorted to paraphrase (“a slow and difficult struggle”) in my translation.

26. Ahl’s translation.

27. Mastrangelo (2008, 100) notes that Superbia leaves out Eve and mentions only Adam putting on clothing—an omission magnified in the Hamartigenia, as I discuss below. Superbia also fails to note that originally Adam and Eve put on garments made of fig leaves. It is God who provides them with garments of skin.

28. I do not know whether Prudentius could have known this, but it interesting that in the Haggadic tradition sources say that Nimrod either received from his ancestors the garments of skin given to Adam by God or tried to kill Esau in order to attain them (van der Toorn and van der Horst 1990, 26).

29. Defined by Cicero (Orat. 79): “verecundus erit usus oratoriae quasi supellectilis. supellex est enim quodam modo nostra, quae est in ornamentis, alia rerum alia verborum.” Cf. Quint., Inst. 8.28. Seneca uses it, like Prudentius, in a way that combines both its original and transferred meanings: “an tu existimas reprendendum, qui . . . pretiosarum rerum pompam in domo explicat: non putas eum qui occupatus est in supervacua litterarum supellectile?” (Epist. 88.36).

30. Elegantly explored in Conybeare 2002, esp. 184–86.

31. “Eandem sententiam milliens alio atque alio amictu indutam referunt,” Fronto, Ant. (De eloquentia) (ed. Haines, 2:104) (157N).

32. E.g., Quintilian, Inst. 10.1.87 and 5.14.30; Fronto, Aur. (ed. Haines, 1:36) (46N); Cicero, Fin. 1.10 and 5.13; De orat. 1.80 and 3.185.

33. On the imagery of cosmic dissolution in H. 315–34 (Lat. 236–50), see Lapidge 1980, and Malamud 1989, 72–78. On anachronism as an epic technique, see Zissos and Gildenhard 1999.

5. UNDER ASSAULT

1. Tertullian, De cultu feminarum 1.1.

2. See especially Richlin 1995, Rimell 2005, Wyke 2002, and Sharrock 1991.

3. Rimell 2005, 194.

4. This double valence is reflected in another word used for makeup, medicamina, which carried associations with medicines, drugs, and poisons. Makeup comes in the form of unpleasant ointments to be smeared on the face, and Ovid describes how sickening these ointments can be in his Remedy for Love (Remedia amoris): “Tum quoque, compositis cum collinet ora venenis/ad dominae vultus (nec pudor obstet) eas./Pyxidas invenies et rerum mille colores,/et fluere in tepidos oesypa lapsa sinus./Illa tuas redolent, Phineu, medicamina mensas:/non semel hinc stomacho nausea facta meo est” (351–56). A number of ancient texts associate makeup with prostitutes and adulterers (e.g., Xenophon, Oeconomicus 10.2–13, where he chides his wife for appearing in makeup; Seneca, Controv. 2.7.3–4; Plutarch, Coniugalia praecepta 142a).

5. Tertullian, De cultu feminarum 1.2–3.

6. Enterline 2000, 62–65 has an excellent discussion of Ovid’s strategic play on the analogy between bodies and rhetorical forms, with particular attention to the rhetorical and corporeal meanings of figura and forma. Figura is frequently used as a rhetorical term (see OLD, s.v. 11).

7. Note Cicero’s use of the term supellex, which Prudentius used to describe the “rich resources” of the fallen world in line 281 (Lat. 207).

8. “Nam ut mulieres esse dicuntur nonnullae inornatae, quas id ipsum deceat, sic haec subtilis oratio etiam incompta delectat; fit enim quiddam in utroque, quo sit venustius, sed non ut appareat. Tum removebitur omnis insignis ornatus quasi margaritarum; ne calamistri quidem adhibebuntur; fucati vero medicamenta candoris et ruboris omnia repellentur; elegantia modo et munditia remanebit. . . . Verecundus erit usus oratoriae quasi supellectilis. Supellex est enim quodam modo nostra, quae est in ornamentis, alia rerum alia verborum.” Cicero, Orat. 79–80. Translation slightly modified from Hubbell’s Loeb translation.

9. See Sen., Epist. 114.16 for discussion of unchaste sententiae, 114.21 for a description of Maecenas. Richlin 1997 is an excellent analysis of the interconnections between rhetoric and gender in Roman oratory, with many more examples from primary sources. On rhetoric and masculinity, see Gunderson 2000, and the influential Gleason 1995/2008.

10. A theme explored with relish by Juvenal (Sat. 2.44–51, 6.286–93, and 11.162–63).

11. As Richlin (1995, 186) notes, the use of makeup marks sexual difference, so its “improper” use within a culture must blur that difference. “A man wearing make-up and long hair, or a woman wearing no make-up and short hair, can in this culture provoke the fascinated question, ‘is that a boy or a girl?’ Once the signs on top are blurred, the real question comes out: what is the bottom line?” Tertullian shares the same anxiety; in his attack on makeup he includes a passage forbidding men “to cut the beard too sharply; to pluck it out here and there; to shave round about (the mouth); to arrange the hair, and disguise its hoariness by dyes; to remove all the incipient down all over the body; to fix (each particular hair) in its place with (some) womanly pigment; to smooth all the rest of the body by the aid of some rough powder or other: then, further, to take every opportunity for consulting the mirror; to gaze anxiously into it” (De cultu feminarum 2.8.2)—in other words, to elide the differences between male and female bodies and to act like females.

12. On vas as a term for the male genitalia, see Adams 1982, 42–43, who cites a number of examples including one contemporary with Prudentius (Augustine, Civ. 14.23). The phrase resectam particulam here may come from Horace, Carm. 1.16.13–16: “Fertur Prometheus addere principi/limo coactus particulam undique/desectam et insani leonis/uim stomacho apposuisse nostro.” Horace is describing Prometheus’s creation of man. Once again, Prudentius melds the Promethean creation story with the biblical account (Gen. 2:21–23), this time using it as a model for the creation of woman from Adam’s flesh. See Palla 1981, 196.

13. Chiappiniello 2009, 181 argues that in this passage Prudentius, unlike his predecessors, “does not castigate women for leading men astray but adopts a different approach by satirizing soft (cf. 282 mollescere) and effeminate men. . . . The onus of ensuring moral conduct in life rests entirely on men, who, by God’s commandment, are women’s guide.” Fontaine 1969, 64 also analyzes this passage and Prudentius’s insistence on the weakness of female character and the need for men to control women.

14. Tertullian devotes a chapter of De cultu feminarum to attacking luxurious textiles (1.8).

15. For a particularly good example, see Cath. 3.26–30, where Prudentius speaks of his muse weaving mystic crowns of dactyls, and Perist. 3.206–10, in which he offers garlands of (wilted) dactyls to the martyr Eulalia. As I have argued elsewhere, the tension between imagery of binding and imagery of dissolution is a predominant motif of both the Peristephanon and the Psychomachia (Malamud 1989, especially 75–78 and 172–80).

16. O fortunatos nimium, sua si bona norint,/agricolas quibus ipsa procul discordibus armis/fundit humo facilem victum iustissima tellus,/si non ingentem foribus domus alta superbis/mane salutantum totis vomit aedibus undam,/nec varios inhiant pulchra testudine postis/inlusasque auro vestis Ephyreiaque aera,/alba neque Assyrio fucatur lana veneno,/nec casia liquidi corrumpitur usus olivi.

17. See Gunderson 2009, 119, who discusses the language applied to rhetorical ornmament: “Dressing up one’s oratory is always liable to charges that it has been overdone: superfluity and effeminacy are the main vectors of criticism. Quintilian argues that the proper rhetorical adornment is always both functional and virile. Thus the well-made argument of the good man is akin to the virile beauty evinced by the muscular bodies of the athletes.”

18. Webb 2008, 59.

19. Cassiodorus, Variae 4.51, p. 178, ll.61–64 (= Bonaria Fasti no. 446), cited by Webb 2008, 64: “Idem corpus Herculem designat et Venerem, feminam praesentat in mare, regem facit et militem, senem reddit et iuvenem, ut in uno credas esse multos tam varia imitatione discretos.” Webb 2008, 62 describes the masks and costumes of pantomime dancers.

20. Translation Laistner 1951.

21. See Miller 2005, 30 for a discussion of this passage.

22. Translation Melville 1997.

23. On the tension between didactic and martial epic, see Von Albrecht 1997, 2:1362.

24. “Prudentius associates scripture with the notion of Moses as the chronicler of creation (Apoth. 219, 234–35, and 302). Similarly, Augustine at Conf. 12.26 argues that Moses wrote words (signa) for those who understand the hidden meaning of scripture” (Mastrangelo 2008, 157; see also 42).

25. “Sometimes he juxtaposes Sapientia and Deus, referring to them now as fellow creators and now as apophatic divine beings with no discernible origins (H. 345, 164). In another passage the poet invokes Sapientia as an offspring of the Father and therefore an ontological equal of Christ (Hymn on the Trinity 2). In all three passages, a Christian context is clear, but their debt to pagan metaphysics is clear from their concern with the ontology of the godhead” (Mastrangelo 2008, 223n32).

26. This is exactly the same condition in which Boethius appears at the opening of The Consolation of Philosophy. His soul is sick, and, weighed down by chains around his neck, he gazes at the earth (1m2) and his vision is obscured and needs to be cleansed by Philosophy (1p2.6). On the cleansing of Boethius’s eyes, see Goins 2001.

27. Rosenmeyer 1999, 19–47. As Rosenmeyer points out, science writers in antiquity argued that man’s semen is made up of marrow, but in erotic contexts the image of marrow as a locus eroticus is used almost exclusively of women (and where it is not, it occurs in passages where the male figuratively takes on the woman’s role).

28. Paxson 1994, 78–95 discusses the problematic relationship between allegorical levels in the Psychomachia and the extent of the interaction between the personifications and the scriptural figures within the narrative, and has a particularly interesting discussion of the figure of Job. See also Barney 1979, 63, and Van Dyke 1985, 34–35 for the semiotic tension generated by the relationship between the biblical or typological dimension of the poem and the personificational one.

29. Bloomfield 1943, 89 cites Origen, homily 12 in Lib. Jes. Nave.

30. These personifications do not appear in the Psychomachia, where the destabilizing elements of language that they represent are concentrated in the figure of Discordia.

31. Bardzell makes the point that even Sapientia in the Psychomachia does not possess intellection, the divine ability to have comprehensive knowledge, but relies instead on reason: “Sapientia, her name notwithstanding, appears to be engaging here in rational inquiry, rather than pure intellection. Thus, the construction of the temple does not indicate that the battles are over and the Vices defeated once and for all” (2009, 51–52).

32. For this point, that the assignment of Sapientia to different sides (Vice in H., Virtue in Psy.) reveals the instability characteristic of language, it makes no difference which text was written first. Obsession with linguistic instability is a characteristic of all of Prudentius’s work. Shanzer 1989, however, has persuasively argued that the Psychomachia should be dated after 405, and thus subsequent to the Hamartigenia. She also notes (360n106) that the mustering of the Vices by the devil in the Hamartigenia is similar to the Psychomachia, and suggests that it “can be seen as a preliminary version of the imagery to be used in the Psych.

6. GENERATION OF VIPERS

1. Athanasius, in his Orations against the Arians, makes this point at some length: “For brutes and men, after a Creator has begun them, are begotten by succession; and the son, having been begotten of a father who was a son, becomes accordingly in his turn a father to a son, in inheriting from his father that by which he himself has come to be. Hence in such instances there is not, properly speaking, either father or son, nor do the father and the son stay in their respective characters, for the son himself becomes a father, being son of his father, but father of his son. But it is not so in the Godhead; for not as man is God: for the father is not from a father; therefore doth He not beget one who shall become a father: nor is the Son from effluence of the Father, nor is He begotten from a father that was begotten; therefore neither is He begotten so as to beget. Thus it belongs to the Godhead alone, that the Father is properly father, and the Son properly son, and in Them, and Them only, does it hold that the Father is ever Father and the Son ever Son” (Athanasius, Ar. 1.21).

2. Cf. Prudentius’s argument against the possibility of two Gods: “Either God/is one and holds the highest power, or/the two that now exist are both diminished:/both cannot have supremacy. It’s clear/that nothing is supreme if it’s not one,/omnipotent, since separate things claim power/each for itself, rejecting the other’s rule,/and so are not supreme and not almighty./Dispersed authority is not complete:/one cannot have a thing another has” (H. 28–37).

3. As Salvatore 1958, 16–24 has noted, this passage is a tour de force of textual contaminatio: Prudentius draws from a number of models from classical poetry and prose, though he acknowledges no specific sources, only the ethici and physici. His sources include Herodotus 3.109; Pliny, HN 10.62.169–70; Petronius poem 32; Lucretius’s description of physical love in De rerum natura 4; and Horace, Odes 1.37.26–28 (on the death of Cleopatra). See Palla 1981, 254–61 for more parallels. There is also a passage in Horace’s Satire 2.8 (42–44) where diners are presented with a pregnant lamprey to eat; commentators have noted that in antiquity it was believed that lampreys mated with vipers. Miller 1998, 274–75 discusses the satirical implications of the lamprey pregnant with the viper’s brood, and compares it to Prudentius’s pregnant viper.

4. Cited by Ladner 1995, 127, who uses the Bern Physiologus, Cod. 318 of the Burgerbibliothek.

5. Aulus Gellius uses the figure of the mother bear to describe Vergil’s painstaking process of composition: “Ut illa bestia fetum ederet ineffigiatum informemque lambendoque id postea quod ita edidisset conformaret et fingeret, proinde ingenii quoque sui partus recentes rudi esse facie et inperfecta, sed deinceps tractando colendoque reddere iis se oris et vultus lineamenta” (Noctes Atticae 17.10). Cf. Aeneid 8.634 and Georgics 2.407.

6. Gender confusion is a hallmark of Prudentius’s poetry. See Malamud 1990 on Prudentius’s violent masculine virgin martyrs, and Nugent 2000, 16 on the transsexual nature of the Virtues and Vices in the Psychomachia.

7. Confusion or conflation of the mouth and the vagina is long-standing in ancient medicine. “The view that a female’s speech was influenced by and in turn indicative of her sexual experience is enshrined in the linguistic double meaning of the Greek word stoma, meaning both oral and genital mouth or lips. . . . Galen, for example, compared the clitoris to the uvula, for the clitoris protects the uterus from the cold in the same way that the uvula protects the trachea; the logic runs that a female who opens one of her mouths is thought to open the other” (Morales 1999, 50). On the conflation of throat and vagina, see Sissa 1990, 52–67; Hanson and Armstrong 1986. Nugent 2000 insightfully links this conflation to the striking fact that all but one of the Vices in the Psychomachia die from wounds to the mouth or throat.

8. Of course, Sin and her offspring in Milton have many forerunners in the literary tradition; see Fowler 1998 ad loc. Spenser’s Errour is one of the most exuberant: “Her scattred brood, soone as their Parent deare/They saw so rudely falling to the ground,/Groning full deadly, all with troublous feare,/Gathred themselves about her body round,/Weening their wonted entrance to have found/At her wide mouth: but being there withstood/They flocked all about her bleeding wound,/And sucked up their dying mothers blood,/Making her death their life, and eke her hurt their good” (FQ bk. 1, canto 1, st. 25).

9. Kilgour 2005, 308.

10. Ibid., 339, quoting Burrow 1993, 269.

11. Kilgour 2005, 309.

12. Conybeare 2007, 237.

13. Ibid., 234. She argues that John 8:31–32 is relevant here as well.

14. Arius associated with Eve and the serpent (Athanasius, Ar. 1.7). As Burrus puts it, “A writhing effeminacy of the word is the shifty mark of the illegitimacy by which he, Arius—motherless like all the sons of this work—is unmasked as the offspring not of God the Father but of Satan the antifather, disseminator of the ‘mania’ of heresy” (Burrus 2000, 54).

15. Burrus 1996, 467–68.

16. Ibid., 469. It is worth noting that the contamination of the mouth, or more precisely, confusion of the oral and genital areas, is also a figure in Ambrose’s rhetoric against Arian, in particular as it applies to language, as Burrus points out (ibid.): “It is a small step from the monstrous to the more graphically grotesque, a step that Ambrose seems to take easily as he recounts the death of the arch-heretic Arius: ‘For Arius’ bowels gushed out . . . and so he burst asunder in the midst, falling headlong and besmirching those foul lips wherewith he had denied Christ.’ Ambrose invites contrast between the grotesque figure of Arius and the sublimated eroticism of the following representation of the evangelist John: ‘Whom, then, are we to believe?—St. John, who lay on Christ’s bosom, or Arius, wallowing amid the outgush of his very bowels?’ he asks. In John, not heresy but the male body of orthodoxy is feminized in an asceticizing rejection of grotesque masculinity.” Easterling and Miles (1999, 101) cite an anecdote from Eunapius of Sardis (Historici graeci minores, fr. 54, Blockley 1981–83) that contains similar imagery associating language with gushing bowels, about an unnamed tragoidos in the time of Nero whose acting powers had an unfortunate effect on a population unfamiliar with tragic performances.

17. Harpham 1986, 249.

7. SIGNS OF WOE

1. Partner 2007, 142: “As the rain comes down like tears, and man falls, Milton gives us a typographical depiction of the fatal cadence.”

2. See Fyler 2007, 16–18 for further references.

3. Fyler 2007, 10–17 has a good summary of the tradition of commentary on the problem of Eve’s name.

4. Cf. Pucci’s (1977, 100) discussion of Pandora, whose creation represents the birth of rhetoric (my emphasis): “The text implies both the human dawn unmarked by imitation and rhetoric and a turning point that initiates the beautifying, imitative, rhetorical process. In this way, the text reproduces the split between a language identical to reality and a language imitative of reality.”

5. Butler 2003 discusses the relation drawn between Eve and Pandora by Milton and Tertullian.

6. Lot is the hero of this tale but represents another variant on the model of reproduction without a wife: after the loss of his wife, his two daughters live with him in a cave, get him drunk, and commit incest with him; each bears a son as a result (Genesis 19:30–38). This allows Lot to reproduce without blame (he is unaware of his actions) through his daughters, and thus produce offspring who are more like himself.

7. Patristic commentators on the passage interpret the transformation of Lot’s wife as indicating that the convert to Christianity should turn his back on his earlier life, or as symbolizing the weakness of the flesh compared to the strength of the spirit (Augustine, Civ. dei 10.8, 16.30; Origen, In gen. hom. 5.2). For further discussion, see Palla 1981, 275 and Herzog 1966, 143.

8. See Hollis 1997 for a discussion of ancient sources of the Niobe story.

9. Salvatore 1958, details the verbal similarities between the passages; he sees Prudentius outdoing Ovid at his own game: “Prudenzio si mostra ancor piu amator ingenii sui, ancor piu intemperante di Ovidio” (268).

10. Gnilka 2000, 68–90 argues that lines 745 (caute sigillati longum salis effigiata), 747–48 (et flexam in tergum faciem paulumque relata/menta retro, antiquae monumenta regentia noxae), and 765–68 are later interpolations. I do not find his argument conclusive, but my main point stands even if one accepts his excision.

11. Hardie 2002, 251 discusses how the repetition of the name Niobe in Ovid’s text reinforces the theme of self-alienation and absence.

12. Pausanias saw the figure of Niobe on Mount Sipylus: “Th is Niobe I myself saw when I had gone up to Mount Sipylus. When you are near it is a beetling crag, with not the slightest resemblance to a woman, mourning or otherwise; but if you go further away you will think you see a woman in tears, with head bowed down” (1.21.3, trans. Jones). Josephus similarly claims to have seen the statue of Lot’s wife in the desert: “But Lot’s wife continually turning back to view the city as she went from it, and being too nicely inquisitive what would become of it, although God had forbidden her so to do, was changed into a pillar of salt; for I have seen it, and it remains at this day” (Jewish Antiquities 1.204, trans. Thackeray).

13. As is the case with the metamorphosis of Actaeon, which is the occasion for contesting interpretations from the crowd: Rumor in ambiguo est; aliis violentior aequo/visa dea est. alii laudant dignamque severa/virginitate vocant: pars invenit utraque causas (Met. 3.253–55).

14. Conybeare 2007, 226.

15. Compare the contest with David and Absalom, where the battle between father and son is conceived as “signis contraria signa paternis” (H. 567); and H. 79–80, where Prudentius expresses his discomfort with comparing God to any sign: “non conferre deo uelut aequiparabile quidquam/ausim nec domino famulum componere signum” (Conybeare 2007, 232n16).

16. As Catherine Conybeare (2007, 227) has noted, the Hamartigenia, like other Roman epic poems, is a medium of “studied pseudo-orality.” Roman epic is a genre that purports to be sung, though in fact it is written. “The text of the Hamartigenia is typical,” she says, “in that it is dotted with feigned markers of orality, most notably the magnificent prayer with which it closes, but also its references to fabulae, its repeated first person interventions, and its direct address to interlocutors outside the text.”

17. Kaesser 2002, 158–71 analyzes the layers of exegesis in Prudentius’s account of the martyrdom of Saint Cassian. In the poem, the poet describes an encounter in a shrine with a verger who explains the painting of Cassian’s martyrdom. Kaesser argues that the point of the exegetical frame is to close the “hermeneutic gap” opened up by the medium of painting and “diminish, if not dissolve entirely, the potential misunderstanding of painting.” Like Goldhill (2006), I find it more likely that the point is to increase rather than diminish the complexity of the ekphrasis.

18. Mastrangelo 2008, 163–64.

19. See Salvatore 1958, 266–69 for a discussion of verbal parallels with Ovid’s Niobe. Niobe’s transformation into a statue is one of several episodes in the Metamorphoses involving humans and statues, including the famous story of Pygmalion and his metamorphic statue; the story of Narcissus, who is compared to a marble statue; and Perseus’s transformation of his enemies into statues using the severed head of Medusa. Hardie (2002, chapters 3 and 6) provides a comprehensive discussion of Niobe in the context of other metamorphoses into statues.

20. See Goldhill 2000 on Prudentius and his views on art, including (62–63) his defense of Theodosius’s decision to preserve classical sculptures (Contra Sym. 1.501–5).

21. Whitmarsh 2002, 111–12 quoting Theon, On Ekphrasis = p. 2.118 Spengel), and citing the important discussion of ancient ekphrasis in Webb 1997a and 1997b. For enargeia, see section 1 above, pp. 60–61.

22. Whitmarsh 2002, 112.

23. The identification of Orpah as the mother of Goliath also appears in Jewish tradition; she is identified in Sotah 42b of the Babylonian Talmud as the mother of four Philistine giants (including Goliath), but in this tradition her piety toward Naomi is rewarded. See Palla 1981 ad loc.

24. As Doniger (2000, 260) points out, “Boaz is of Naomi’s generation, not Ruth’s; he is Naomi’s brother-in-law, the brother of her husband Elimelech. If Naomi is the sister-in-law whom Boaz should marry, then when Naomi sends Ruth to Boaz, she is substituting the desirable younger generation for the appropriate older generation, in a reversal of Rachel/Leah and Shelah/Judah, where the appropriate (or, in the case of Judah, necessary) older generation was substituted for the desirable younger generation. I would suggest that Obed is ‘Naomi’s baby’ because he should have been born (physically) to Naomi or (officially) to Naomi’s dead son.”

25. E.g., Xenophon, Memor. 2.1.20ff; Cicero, Tusc. 1.30.72; Servius ad Aen. 6.136; Philo, Spec. leg. 4.108 and 1.12; Lactantius, Inst. 6.3–4; Jerome, Contra Rufinum 3.39 (PL 23, 508c); Ausonius, Professores 11.5. A scriptural parallel appears at Matthew 7:13–14. The metaphor of the road is widespread in Greek and Latin. For a thorough list of citations and scholarship on the topos, see Palla 1981, 285–86. Knox 1999 discusses the related image of the narrow road in Lucretius.

26. Bleary vision was something of a topos in Horatian satire. See Gowers 1993, 60, where she cites Horace, Sat. 1.5.30, 1.1.120, 1.7.3, and 1.3.25. Prudentius too uses the imagery of blurred vision at H. 85ff. as a metaphor for the misunderstanding of a dualist heretic; later, at 865, he offers an extended description of the supernaturally acute vision of the good soul after death.

27. De gen. ad lit. 12, esp. 12.6.15. See Miller 2005, 32–33 for a discussion of Augustine’s theory of three kinds of vision.

28. Lühken 2002, 76 notes this allusion.

29. In contrast to the devil, who is described as decolor, as is Envy at H. 186.

8. IN AENIGMATE

1. Bynum and Freedman 2000, 7.

2. Straw 2000, 23.

3. Taddei (1981, 212–20) notes that Prudentius’s diction throughout this opening section is juridical in nature, infused with words derived from Roman jurisprudence: vindicat (24), proprium (23), ius (25 and 110), potestas (20), imperio (24), dicio (19 and 108), testamur (27), probat (52), adscitus (53), signum, adsignare . . . iura (105), coheredem (110), ratio (180), and negat (180).

4. Codex Theodosianus 1.16.6, of 331 CE, cited by Harries 1999, 214.

5. Passio ss. Perpetuae et Felicitatis, 10.1–13 (SC 417:134–40). Vita Caecilii Cypriani 12 (CSEL 3.3cii–civ); Passio ss. Mariani et Iacobi = P. Franchi de’ Cavalieri, La Passio ss. Mariani et Iacobi = Studi e Testi 3 (Rome: Tipografia Vaticana, 1900).

6. Seneca, Ep. 14.2.

7. Shaw 2003, 541 citing Opus imperfectum in Matthaeum, Hom. 54.31 (PG 56:941). I have quoted Shaw’s translation.

8. Tert., Apol. 1.1.10–13 (CCL 1:86–87), cited by Shaw, 544n33.

9. Shaw provides extensive evidence that Christians did indeed internalize the judicial process; a particularly striking example is that of Jerome’s famous dream in which he is accused of being a Ciceronian, not a Christian: “When suddenly, taken up in the spirit, I was hauled before the judge’s tribunal [ad tribunal iudicis], where there was so much light and such a great shining from the radiance of those who were standing about that, throwing myself on the ground, I did not dare even to look up. When I was asked my status [interrogatus conditionem], I replied that I was a Christian. The one who was presiding as judge intoned: ‘You are lying,’ he said, ‘You are a Ciceronian, not a Christian. Your treasure is there, where your heart is.’ I remained rooted to the spot, tongue-tied. Amidst the lashes of the whip [verbera]—for the judge had ordered that I be beaten—I was tortured more by conscience than by any torturer’s firebrands, and I considered that little piece of verse in my mind: Who will confess you in the fires of hell itself? I began to cry and shouted out: ‘Pity me, my lord, have mercy on me.’ Amidst these pleas of mine the lash of the whip resounded. Then the spectators who were standing round about fell on their knees before the tribunal and began to implore the judge that he should have pity on my youth, that he should make allowance and forgive my mistake, and that he should only impose the penalty of crucifixion if I ever again read the books of the gentiles.” Jerome, Ep. 22.30 (CSEL 54:189–91), Shaw’s translation.

10. Moreira 2006, 150, citing Clement, Stromata 4.24.

11. Brown 2000, 46.

12. Ibid., 49.

13. Translated by Marcus Dods. From Nicene and Post-Nicene Fathers, first series, vol. 2, ed. Philip Schaff (Buffalo: Christian Literature, 1887). Revised and edited for New Advent by Kevin Knight. http://www.newadvent.org/fathers/1201.htm.

14. Mastrangelo 2008, 108.

15. Plato, Phaedo 62b and Cratylus 400c. See Courcelle 1965 for an analysis of the topos of the body as prison in Platonic and Christian literature.

16. I departed slightly from the literal sense of the Latin in my translation, but tried to keep the nurse-nursling relationship clear. I call Fides a nurse, which Prudentius does not; instead, he identifies the soul as the alumna (nursling or foster child) of Fides.

17. The language here is erotic: cubile, line 856, refers to the marriage bed; querulo is an adjective typical in Latin love elegy.

18. It is typical of Prudentius’s highly ambivalent treatment of females that even this passage, which represents the culmination of his allegorical presentation of the soul, associates the feminine soul not only with the mystical brides of the Song of Songs and of Revelation 19, but also, disconcertingly, with an explicitly erotic pagan epigram attributed to Petronius (Anth. Lat. 705). See Salvatore 1958, 18n9; Palla 1981, 297. Even the presence of Fides, the faithful nurse, suggests unsettling parallels with classical literature—both Phaedra and Myrrha are assisted by their faithful old nurses in their pursuit of incestuous love. Prudentius, of course, is not alone in deploying highly erotic imagery to describe virgins, and there is much written on this topic. See Burrus 1995, Grig 2005, Averil Cameron 1989, Brown 1988, Clarke 2006, Uden 2009, and Malamud 1989 and 1990.

19. The association of blindness with lack of faith is graphically expressed in the Psychomachia, when Fides tramples on the eyeballs of her opponent, Paganism (Ps. 32–33), in the first battle scene of the poem. See Bardzell 2009, 44.

20. Brown 2000, 50.

21. See Goldhill 2001, 183–84; Webb 2008, 204–5 for different reactions to Tertullian’s views on spectacles.

22. Levine 1991, 21, in the context of his discussion of the long, violent, and grotesque Peristephanon 10.

23. Webb 2008, 207.

24. Moreira 2006, 152.

25. Straw 2000, 23.

26. Brown 1992, 208; Van Slyke 2005, 63. The bibliography on violence in the amphitheater is vast. In addition to Brown and Van Slyke, see Drake 2006, Wiedemann 1992, Kyle 1998, Walter 2004, Barton 1989 and 1993, and Coleman 1998 and 1990 for good introductions to this complicated subject.

27. Van Slyke 2005 provides numerous examples, e.g., Augustine, Sermo 14 (PL 46.864–65) invites his congregation to exchange public spectacles for the edifying spectacle of the martyrdom of Saint Cyprian. See also Chrysostom, In Ioannem Hom. 60 (PG 59.333), who proposes contemplation of the martyrs or of David (Webb 2008, 206).

28. Frankfurter 2009, 231.

29. Goldhill 1999, 82, which presents an excellent analysis of both Pe. 9 and Pe. 3, the martyrdom of Eulalia. Both martyrdoms involve writing the marks of torture on the body. For writing on the body in the Peristephanon, see Ross 1995, Ballengee 2002, and Malamud 1989.

30. Ballengee 2009, 93: “Prudentius creates a certain excess in his writing in his use of literary, rhetorical tropes, by which he hopes to gain for himself as poet the same benefits as the martyr.”

31. Augustine, Sermo Denis 14.3, as quoted in Goldhill 1999, 81.

32. Prudentius displays a strong dislike of schoolmasters and rhetoricians in the preface to his collected works: aetas prima crepantibus/flevit sub ferulis. mox docuit toga/infectum vitiis falsa loqui, non sine crimine (Praef. 7–9).

33. Discussed by Miles 1983, 142.

34. Kilgour 2005, 308, cited above.

35. Mastrangelo (2008, 83) sees the narrative allegory of the Psychomachia as a response to such a perceived crisis of meaning; it deploys typologies that point beyond the text to biblical texts and Christian teachings that make it “possible for the poet to communicate ideas that are incommunicable through normal object or referent language.”

36. In addition to poena levis, Prudentius echoes lenito in line 963 (lenta).

37. Malamud 1989, which prompted an interesting article by Alan Cameron (1995) arguing against the presence of the anagram. Nugent, too, expressed her doubts about the anagram with characteristic wit: “Well, maybe. But if my decoder unscrambles the same line as: LETTERS VALUED IN ARCANE POEM, what are we to make of it?” (1991, 326). As Morgan (1993, 143) put it in discussing a related form of wordplay, “Credulity about acrostics is so closely associated with crankdom that legitimate scholars risk contumely and scorn if they are tempted to believe.” My response to Nugent’s point is that ancient readers, so far as we know, had neither decoder rings nor sophisticated soft ware to unscramble anagrams, and so would have to be guided by textual clues if they were to have any hope of detecting them. See Somerville 2010 for a discussion of the clues to the signature acrostic that Vergil leaves for the reader at G. 424–35.

38. Bleisch 1996 reveals an interesting example from Callimachus, with an excellent analysis of its significance and its history. For other examples, see Fowler 1983, Cameron 1995, Courtney 1990, Levitan 1979 and 1985 (the latter on the unquestioned late antique master of figured verse, Optatian), and Morgan 1993.

39. Cameron (1995, 483) objects to the anagram for several reasons, including that the name Clemens is absent from it. However, since the name is clearly signaled by the appearance of clementer in the surface text, there is no reason for it to reappear in the anagram.

40. As Cameron notes (1995, 482). My original argument was faulty; I argued that the sphragis should read glorificent me: Aurelio prudente se clamante, ignoring the fact that grammar requires glorificent to be construed with the previous line.

41. Cameron 1995, 482–83: “There is a (to me) uncomfortable conflict between the surface conclusion of the poem (in which the poet sees himself as too much of a sinner to hope for paradise; that is for martyrs alone) and the anagrammatic conclusion, in which he boastfully proclaims his name.”

42. STARS (3.552–56), SATAN (9.5010–14), WHY (9. 703–6), and WOE (9.1002–5).

43. Partner 2007, 129.

44. Ibid., 137, citing Fowler’s notes ad loc.

45. On the SATAN acrostic, see Klemp 1977, 91; Leonard 1990, 136; Fyler 2007, 14–16; and Partner 2007 passim. See also Schnapp 1991, 280, on Dante’s use of the acrostic in the Commedia.

46. The close visual similarity between the English words “Eve” and “Eye” may well have struck Milton as significant.

47. Partner 2007, 140.

48. Ibid.

49. Compare H. 944–45, where the dead body is called conclamata . . . materies. Like clamo, conclamo can denote crying out both in approbation and in ritual mourning.

50. On Pe. 14 as Prudentius’s signature poem in the collection of the Peristephanon, see Malamud 1989, chapter 6.

51. I argued in Malamud 1989, 172–80 that Prudentius saw himself as a Promethean figure, and that he interprets his own name, Prudentius, as a translation of the Greek Prometheus. There is also a significant wordplay here on liber, “book,” and liber, “free.”

52. Ross 1995, 331.

53. For the image of building the Holy City with words, see Roberts 1993, 186–87.

54. Levine 1991, 24–25; he has a good discussion of the matrix of serpents, sexuality, and procreation in the poem, which he links with similar themes in Pe. 10, a poem in which the tongue is excessively important.