Skip to main content Accessibility help
Hostname: page-component-8bbf57454-z2jr4 Total loading time: 0.61 Render date: 2022-01-25T20:08:32.609Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "metricsAbstractViews": false, "figures": true, "newCiteModal": false, "newCitedByModal": true, "newEcommerce": true, "newUsageEvents": true }

Suicide by Gladiator? The Acts of Perpetua and Felicitas in its North African Context

Published online by Cambridge University Press:  16 October 2019


The third-century martyr text that tells the story of Perpetua and her coreligionists as they witnessed to their faith in Carthage—the Passion of Perpetua and Felicitas—is much beloved by modern readers. It purports to be Perpetua's prison diary, and, as such, it is the earliest writing by a Christian woman that has been preserved. In the fifth century, two shorter texts—the Acts—were written that recount in quite different ways the experiences of these North African Christians. In this essay, I explore the ways these shorter accounts reimagine the original text as well as the possible motives for doing so. The fifth-century author did not merely shorten the Passion; he made marked changes to it. It presents, for instance, a dramatic change in the death scene: whereas the Passion narrates Perpetua drawing the gladiator's sword to her own throat, the Acts report that she was killed by a lion. An analysis of these revisions to the Passion suggest the author was working from an Augustinian perspective. The author revised the Passion to deny the Donatists’ claims to the Carthaginian martyrs. The author's revision of Perpetua's death, therefore, is best understood as a rejection of Donatist desires for martyrdom.

Copyright © American Society of Church History 2019 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


1 All translations are my own except where otherwise indicated.

2 I follow van Beek's sigla for both the Passio and the Acta. See van Beek, C. J. M. J., ed., Passio Sanctarum Perpetuae et Felicitatis, vol. 1 (Nijmegen: Dekker and Van de Vegt, 1936)Google Scholar. On the popularity of the Acta, see Heffernan, Thomas J., The Passion of Perpetua and Felicity (New York: Oxford University Press, 2012), 442443CrossRefGoogle Scholar; and Cotter-Lynch, Margaret, Saint Perpetua Across the Middle Ages: Mother, Gladiator, Saint (New York: Palgrave MacMillan, 2016), 56CrossRefGoogle Scholar.

3 Passio MS. 3b (MS Einsiedeln 250 [382]) also preserves the tradition that Perpetua and Saturus were killed by lions. This 12th cen. manuscript, according to Heffernan, descends from a Latin text that was influenced by the Acta. See Heffernan, Thomas, “The Legacy of Misidentification: Why the Martyrs in the Passio Sanctarum Perpetuae et Felicitatis were not from Thuburbo Minus,” Journal of Early Christian History 6, no. 3 (2016): 134CrossRefGoogle Scholar.

4 Robinson, J. Armitage, Texts and Studies: Contributions to Biblical and Patristic Literature, vol. 1, no. 2, The Passion of S. Perpetua (New York: Cambridge University Press, 1891), 16Google Scholar.

5 The Carthaginian martyrs are not the only exception. St. Lawrence, for instance, was—in the earliest traditions, it seems—beheaded. But the account of his roasting over the fire—popularized by Ambrose (De officiis 2.28; Ep. 37.36–37) and Prudentius (Peristephanon 2)—won the day and so, iconographically, he is shown with the gridiron. See Delehaye, H., “Recherches sur le légendier romain: la passion de S. Polychronius,” Analecta Bollandiana 51 (1933): 4950CrossRefGoogle Scholar.

6 For a comparable approach to the Acta, see Cobb, L. Stephanie, “Memories of the Martyrs,” in Melania: Early Christianity Through the Life of One Family, ed. Chin, Catherine M. and Schroeder, Caroline T. (Oakland: University of California Press, 2016), 111129Google Scholar.

7 While Acta A and Acta B are related—B most likely derived from A—I will not conflate the texts. Indeed, in some of the particulars I am interested in here, the two texts appear have somewhat different interests. And so, I focus here solely on Acta A. Petr Kitzler suggests that Acta B is more influenced by Augustinian interpretations of the Passio than is Acta A (From “Passio Perpetuae” to “Acta Perpetuae”: Recontextualizing a Martyr Story in the Literature of the Early Church [Berlin: de Gruyter, 2015], 105, 114Google Scholar). See also Cotter-Lynch, Saint Perpetua Across the Middle Ages, 45.

8 Other scholars have connected the Acta to the Donatist controversy, specifically as standing in the line of Augustinian interpretations of the Passio. See, for example, Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 105; Guazzelli, Giuseppe Antonio, “Gli ‘Acta brevia sanctarum Perpetuae et Felicitatis': Una proposta di rilettura,” Cristianesimo nella storia 30 (2009): 3435Google Scholar; and Shaw, Brent D., “Passion of Perpetua,” in Ancient Greek and Roman Society, ed. Osborne, Robin (New York: Cambridge University Press, 2004), 323Google Scholar. Gail Corrington Streete and Petr Kitzler argue that Augustine's sermons on Perpetua and Felicitas are, themselves, reworkings of the Passio to mitigate Donatist interests. See Streete, Gail Corrington, Redeemed Bodies: Women Martyrs in Early Christianity (Louisville: Westminster John Knox, 2009), 60Google Scholar; and Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 90–92. Catholic authors not only retooled martyr accounts to ensure they were not “misread” by Donatists, they also “Catholicized” martyr accounts that were originally Donatist texts, as Matteo Davit has recently shown: “Ecclesia martyrum. Analisi del corpus martirologico donatista” (PhD diss., Università degli Studi di Padova, 2013).

9 This is not to say, of course, that the Perpetua tradition itself had been forgotten. It is included in Jacopo da Varagine's Golden Legend. It is the Acta not the Passio that James draws upon.

10 Holstenius, Lucas, Passio sanctarum martyrum Perpetuae et Felicitatis. Prodit nunc primum e MS. Codice Sacri Casinensis Monasterij. Opera et studio Lucae Holstenij Vaticanae Basil. Canon. Et Bibliothecae Praefecti. Notis eius Posthumis adiunctis, ed. Poussin, P. (Rome, 1663)Google Scholar.

11 Holstenius, Lucas, Passio SS. Perpetuae et Felicitatis, cum notis Lucae Holstenii, Vaticanae Bibliothecae Praefecti. Item Passio Bonifacii Romani Martyris. Eiusdem Lucae Holstenii Animadversa ad Martyrologium Romanum Baronii. His accedunt Acta Sanctorum Martyrum Tarachi, Probi & Andronici. Ex codice MS. S. Victoris Parisiensis (Parisiis: Carolus Savreux, 1664)Google Scholar.

12 This is clearly the case with the Greek version of the Passio. Although Rendel Harris claimed priority for the Greek text, J. Armitage Robinson showed—more or less to everyone's satisfaction—that the Latin was, indeed, the language of original composition. Harris, Rendel, The Acts of the Martyrdom of Perpetua and Felicitas: The Original Greek Text (London: C. J. Clay and Sons, 1890)Google Scholar; and Robinson, Passion of S. Perpetua. As Brent D. Shaw notes, however, Harris himself later concluded that the Latin text was original (Shaw, “The Passion of Perpetua,” 322–323). The scholarship on the priority of the Latin text is succinctly surveyed by Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 213–229. Nonetheless, Bowersock and Dodds have maintained a preference for the originality of the Greek text, and Gonzales argues for a bilingual text. See Bowersock, Glen W., Martyrdom and Rome (New York: Cambridge University Press, 1995), 34CrossRefGoogle Scholar; Dodds, E. R., Pagan and Christian in an Age of Anxiety: Some Aspects of Religious Experience from Marcus Aurelius to Constantine (New York: Cambridge University Press, 1965), 50n1CrossRefGoogle Scholar; and Gonzalez, Eliezer, Fate of the Dead in Early Third Century North African Christianity (Tübingen: Mohr Siebeck, 2014), 7CrossRefGoogle Scholar. Generally speaking, the Greek text has not been a scholarly focus except insofar as it has been rejected as the original language of the Passio and in the ways testifies to the spread of traditions about the Carthaginian martyrs into the Greek East. The Greek text is utilized, however, by scholars such as Bremmer, who see in its later traditions some survivals of more authentic historical traditions than those preserved in the Latin text; thus, according to Bremmer, it may be used to clarify and even emend the Latin text. See Bremmer, Jan, “The Vision of Saturus in the Passio Perpetuae,” in Jerusalem, Alexandria, Rome: Studies in Ancient Cultural Interaction in Honour of A. Hilhorst, ed. Martínez, Florentino García and Luttikhuizen, Gerard P. (Leiden: Brill, 2003), 5758Google Scholar.

13 Recently, Petr Kitzler and Éric Rebillard have included the Acta in their respective works. Kitzler traces the textual history of the Passion in From “Passio Perpetuae” to “Acta Perpetuae”; Rebillard includes the Acta in his collection of martyr stories in Greek and Latin Narratives About the Ancient Martyrs (New York: Oxford University Press, 2017), 332349Google Scholar. Both authors consider the textual history of the Passio apart from judgments about “originality.”

14 Robinson, Passion, 15.

15 Shaw, “The Passion of Perpetua,” 314.

16 Shewring, Walter H., The Passion of SS. Perpetua and Felicitas: A New Edition and Translation of the Latin Text (London, 1931), xxivGoogle Scholar.

17 This is a longer version of a previously known sermon (Serm.282). See Schiller, Isabella, Weber, Dorothea, and Weidmann, Clemens, “Sechs neue Augustinuspredigten: Teil 1 mit Edition dreier Sermones.Weiner Studien 121 (2008): 227284CrossRefGoogle Scholar.

18 Bremmer, Jan N., “Felicitas: The Martyrdom of a Young African Woman,” in Perpetua's Passions: Multidisciplinary Approaches to the Passio Perpetuae et Felicitatis, ed. Bremmer, Jan. N. and Formisano, Marco (New York: Oxford University Press, 2012), 38CrossRefGoogle Scholar. See also Paul Monceaux, who assumes the Passio was a Montanist document and, thus, could not be the text Augustine refers to in his sermons. Rather, Monceaux posits that Augustine was “directly inspired by the Acts” (Histoire littéraire de l'Afrique chrétienne depuis les origines jusqu’à l'invasion Arabe, vol. 1, Tertullian et les origins [Paris: Ernest Leroux, 1901], 78n2; 81n5Google Scholar). In this case, the Acta, must be fourth-century texts. Guazzeli, though, argues that the Acta reflect Augustine's work with the Passio in De anima et eius origin: Augustine goes to great lengths in this treatise to explain Perpetua's visions about her brother, Dinocrates. Guazzeli suggests that the Acta take a cue from De anima and omit these visions altogether to avoid their theological problems. The influence of De anima on the Acta, then, set the terminus post quem for the Acta in the middle of the fifth century. (“Gli ‘Acta brevia sanctarum Perpetuae et Felicitatis,’” 36).

19 Shaw also dates the Acta to the fourth century (“Passion of Perpetua,” 312). Cotter-Lynch dates Acta A to the fourth century but posits that Acta B is a post-Augustinian text (Saint Perpetua Across the Middle Ages, 45).

20 Rebillard, Greek and Latin Narratives About the Ancient Martyrs, 301. The Liber Genealogus has a complicated textual history. According to Rouse and McNelis, the Liber originated among North African Donatists as an interpretation of biblical and early church history, which traced the genealogy of Donatists to Abel and that of Caecilianists to Cain—thereby explaining the history of persecution experienced by Donatists. It was composed in the late fourth or early fifth century and underwent various Donatist—and later Catholic—editions. Four of these recensions are preserved in the Liber’s seven manuscripts (Rouse, Richard and McNelis, Charles, “North African Literary Activity: A Cyprian Fragment, the Stichometric Lists and a Donatist Compendium,” Revue d'histoire des Textes 30 [2000]: 211233Google Scholar). Inglebert has argued that the earliest Donatist version was composed in 427 CE. and preserved in a singular manuscript family, represented by MS G (St. Gall Stiftsbibliothek 133) and MS C (Vitt. Em. 1325). It was revised in 438 CE, but the two manuscripts that preserved this recension are now lost. Yet another recension dates to 455 CE, which Inglebert identifies with Caecilianist polemics, and is preserved in MS L. But Inglebert identifies a Donatist gloss dating from 470 CE in MS Inglebert, L. Hervé, Les Romains chrétiens face àl'histoire de Rome. Histoire, christianisme et romanités en Occident dans l'Antiquitétardive (IIIe–Ve siècles) (Paris: Institut d’Études Augustiniennes, 1996)Google Scholar. See also Dearn, Alan, “Persecution and Donatist Identity in the Liber Genealogus,” in From Rome to Constantinople: Studies in Honour of Averil Cameron, ed. Amirav, H. and ter Haar Romeny, R. B. (Leuven: Peeters, 2007), 127136Google Scholar; and Torreira, Federico-Mario Beltrán, “Historia y Profecía en el Donatismo Tardío: El “Liber Genealogus,” Antigüedad y cristianismo: Monografías históricas sobre la Antigüedad tardía 7 (1990): 343352Google Scholar.

21 Rebillard, Greek and Latin Narratives About the Ancient Martyrs, 301.

22 Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 101–116.

23 If the Acta should be dated to the fourth century, my hypothesis that the text is anti-Donatist remains unaffected.

24 I have followed Jacqueline Amat's texts of the Acta in Passion de Perpétue et de Félicité suivi des Actes, Sources Chrétiennes, no. 417 (Paris: Les Éditions du Cerf, 1996). On the production of the Acta for liturgical purposes, see Shaw, “Passion of Perpetua,” 314; Heffernan, The Passion of Perpetua and Felicity, 81, 442; Robinson, The Passion of S. Perpetua, 15; and Kraemer, Ross and Lander, Shira, “Perpetua and Felicitas” in The Early Christian World, ed. Esler, Philip Francis, 2nd ed. (London: Routledge, 2017), 979Google Scholar. Most analyses assume a literary relationship between the Passio and the Acta. An important exception to the assumption of literary dependency is James Halporn, who argues: “the literary histories and texts are misleading in calling the Acta the Acta brevia or the abridged Acta (so, for example, Amat). Although the Acta deals with the same events as the Passio, and has many similarities, it is not derived directly from it, as even differences in the handling of the materials indicate” (Literary History and Generic Expectations in the Passio and Acta Perpetuae,” Vigiliae Christianae 45, no. 3 [1991]: 238n11Google Scholar).

25 The author may have expected (or hoped) that his text would be read liturgically. Indeed, Acta A instructs the “holy brothers” to share the memory of the martyrs faithfully by reading their acts in church (9.5).

26 Prinzivalli, E., “Perpetua the Martyr,” in Roman Women, ed. Fraschetti, A. (Chicago: University of Chicago Press, 2001), 139Google Scholar.

27 See Augustine, Serm. 280–282; and Quodvultdeus, Sermo de tempore barbarico.

28 The Liber Genealogus is preserved in four editions, denominated by Mommsen as G (Sangallensis), F (Florentini), L (Lucensis), and T (Taurinensis). Mommsen attributed the original text (G) to a Donatist author writing in 427 CE; a second edition (F) was produced—also of Donatist origin—in 438 CE; the final edition (L) was a Caecilianist (or Arian) revision, whose two recensions date to 455 CE and 463 CE. Whether “Catholic” or Arian, it is clearly not Donatist: this recension replaces the word “Christians” for “Donatists” in describing the persecution that occurred during the consulate of Stilicho. Since the Donatists did not refer to themselves this way, scholars conclude that the recension reflects the work of one of their opponents. The Taurinensis edition is abbreviated—ending at Jesus's birth—and remains undated. Following Monceaux, Dearn classifies G, F, and L as Donatist texts because all three report Mensurius's traditio (131). In addition, all three manuscripts end with Honorius's persecution of the Donatists in 405 CE. Inglebert, however, has argued that the 455 CE and 463 CE editions were Caecilianist appropriations of the Donatist Liber (Les Romains chrétiens face àl'histoire de Rome, 599), but he identifies a Donatist gloss, dating to around 470 CE, in the Caecilianist Lucensis recension. For discussion see Dearn, “Persecution and Donatist Identity”; Rouse and McNelis, North African Literary Activity; Inglebert, Les Romains chrétiens face àl'histoire de Rome; and Conant, Jonathan, “Donatism in the Fifth and Sixth Centuries,” in The Donatist Schism: Controversy and Contexts, ed. Miles, Richard (Liverpool: Liverpool University Press, 2016), 352353Google Scholar.

29 For a discussion of this “Unity Edict,” see Hogrefe, Arne, Umstrittene Vergangenheit: Historische Argumente in der Auseinandersetzung Augustins mit den Donatisten (Berlin: De Gruyter, 2009), 3753Google Scholar.

30 The story is told in the Donatist Acts of the Abitinian Martyrs. The accusations against Mensurius and other Caecilianists is also preserved in Augustine's writings. See, for instance, Against Petilian II.93.202; and Correction of the Donatists, I.1.4–5. The outline of the beginnings of the Donatist controversy are well-known. For a full discussion of the conflict—focused more on history than theology—see Hogrefe, Umstrittene Vergangenhei, esp. chap. 2 (esp. 350–354). Hogrefe argues that the specifics of the accusations made against Mensurius and Caecilianus cannot be historically verified.

31 Matteo argues that Donatist pneumatology—inherited from Tertullian—was based on the notion of a corporeal spirit, which was physically passed from one Christian to another (“Ecclesia martyrum,” 110). When Christians handed over scripture during persecution, so Donatist arguments went, they forfeited the spirit and, thus, their baptisms and ordinations were nullified. As the Donatist Parmenian said, “The one who has nothing to give, in what way can he give?” (Optatus, Against the Donatists 5.6, in Patrologia cursus completus, series latina [Paris: Migne, 1857–1912], 11:1056B [hereafter cited as PL]). While Donatists called their bishops “the givers” (Optatus, Against the Donatists 5.7), they referred to Caecilianist priests as “dead men” (Against Petilian 2.7.14).

32 PL 11:1403. In the Acts of the Council of Carthage, Petilian utilizes genealogical language to associate Caecilianists with traditores of the previous generation. He asks Augustine, for instance, if he is the “son of the [traditor] Caecilian.”

33 See discussion in Dearn, “Persecution and Donatist Identity,” 128–129.

34 Gesta conlationis Carthaginensis a 411, 3.22.

35 For example, Augustine, Against Petilian, II.19.42; II.23.51 II.65.145.

36 Matthew Alan Gaumer argues that Augustine aimed to undermine Donatist claims to Cyprian through his interpretations of the third-century bishop's primary values (Augustine's Cyprian: Authority in Roman Africa [Leiden: Brill, 2016], esp. 103146Google Scholar). I suggest the same type of work is being done here with the Passio. There is some evidence that Perpetua and Felicitas were even more popular than Cyprian. At least this appears to be of concern to Pontius in his Life of Cyprian. Pontius argues that Cyprian is more worthy of honor than laypeople and catechumens (like Perpetua and Felicitas) because he was not only a martyr but also a bishop (1.2).

37 All modern critical editions are based on MS 1 (Monte Cassino 204), which is a late 11th cen. text. For a complete description of the Passio manuscripts, see Heffernan, The Passion of Perpetua and Felicity, 369–430.

38 H (MS Jerusalem 1) dates to the tenth or eleventh cen.

39 Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 106.

40 See Augustine, Serm 313E.4 and the Donatist Passion of Cyprian. Gaumer offers an extended analysis of the ways Augustine appropriated Cyprian for the Caecilianist cause in Augustine's Cyprian. Caecilianists and Donatists laid claim to other Christian martyrs, as well. For a discussion of the battles over who stands in the Pauline line, see Eastman, David L., Paul the Martyr: The Cult of the Apostle in the Latin West (Atlanta: Society of Biblical Literature, 2011), 178186Google Scholar.

41 On Baptism 1.1. For the importance of this treatise in Augustine's (and for the larger Caecilianist cause) anti-Donatist writings, see Gaumer, Augustine's Cyprian, esp. chap. 4.

42 For a discussion of the ways the Passio functioned as a template for later African martyr texts, see Shaw, “Passion of Perpetua” 323; Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 65–72.

43 Caecilianist Christians utilized the Passio as a model for their subsequent martyr texts, as well. See, for instance, the Passion of Marian and James and the Passion of Montanus and Lucius.

44 In Paul the Martyr, Eastman discusses how this tactic is reflected in a mosaic from Uppenna: by adding the Donatist martyr Saturninus to the mosaic alongside Peter and Paul, it claims the apostles for the Donatist church.

45 Trans. Tilley, Maureen A., Donatist Martyr Stories: The Church in Conflict in Roman North Africa (Liverpool: Liverpool University Press, 1996), 78CrossRefGoogle Scholar.

46 Kitzler, From “Passio Perpetuae” to “Acta Perpetuae,” 103–104.

47 See discussion in Shaw, Brent D., Sacred Violence: African Christians and Sectarian Hatred in the Age of Augustine (New York: Cambridge University Press, 2011), 598612CrossRefGoogle Scholar. Shaw notes that for Caecilianist Christians, the history of persecution can be divided into three eras. First was the period of physical torture and martyrdom, which ended with Constantine's conversion. Having been defeated, Satan turned to more subtle techniques in the second age, which had the goal of leading Christians astray from the true Church. The third age of persecution is the future battle with the antichrist. Donatist Christians, however, believed the first age had not yet ended. For this reason, Christians who died for their faith were martyrs precisely as were their predecessors (Sacred Violence, 598).

48 Bremmer, “The Vision of Saturus,” 71–72.

49 Heffernan, The Passion of Perpetua and Felicity, 273.

50 The clerics’ act is also reminiscent of Perpetua's fathers’ subordination, especially in Passio 5.5, 6.6, and 9.2. See discussion in Cobb, L. Stephanie, Dying to Be Men: Gender and Language in Early Christian Martyr Texts (New York: Columbia University Press, 2008), 97102CrossRefGoogle Scholar.

51 Bremmer, “The Vision of Saturus,” 66. That the bishop and priest are alive and actively serving their North African community is also suggested by the angels’ exhortation to the bishop to “rebuke” his people (13.6).

52 See, for instance, Augustine, De Unitate Ecclesiae. Honorious's legislation against the Donatists in 405 CE shows a growing desire within the Caecilianst church to force unity where they had previously failed at negotiating it. Under these edicts, Dontists were forbidden to baptize, unable to take donations, and—the coup de grâce for the Caecilianist cause—declared heretics rather than merely schismatics. Kitzler suggests the Acta omit Saturus's vision because it refers to events contemporary with the Passio, which are not relevant to the later church (From “Passio Perpetuae” to “Acta Perpetuae,” 109). See also Salisbury, Joyce E., Perpetua's Passion: The Death and Memory of a Young Roman Woman (New York: Routledge, 1997), 162Google Scholar.

53 Augustine, Ep. 66.1. That the Caecilianist church remained in solidarity with the larger church was an important piece of Augustine's arguments against the Donatist schism. See, for example, Against Cresconius, where Augustine leverages the bishop-martyr Cyprian against the Donatists (II.43, 45, 48; III.35, 73; IV 6). It is the unity of the church, moreover, that Augustine highlights in his appropriation of Cyprian. Whereas the Donatists follow Cyprian in his practice of rebaptism (Augustine admits), they fail to follow him in his more important teaching: that of preserving the unity of the church (for example, On Baptism 2.15, 2.17, 2.20). The sentiment is ubiquitous, as, for instance, in Ep. 76.2, where Augustine implores the Donatists to “return to unity.”

54 Augustine, Answer to Petilian II.68.153. See also II.78.172; II.93.205, 212; II.96.219; II.98.224.

55 Heffernan, Passion of Perpetua and Felicity, 291–292.

56 Tilley, Donatist Martyr Stories, x. In their understanding of intercessory powers, the Donatists may have simply reflected the popular views of the time in North Africa, but it is one that Augustine rejects. Thus, the Passio stands at odds with Augustine's views, whereas Acta A better reflects Augustinian thought. See also Bremmer, “Saturus's Vision,” 67. Bremmer goes on to suggest that the rose arbor featured in Saturus's dream also refers to the superiority of the martyrs over the clerics: “In early Christianity, red roses were symbolic of the blood of the martyrs, just as their thorns could symbolize the martyrs’ torture . . . . By associating themselves with the rose, then, Saturus and Perpetua stressed their preeminent position” (“Saturus's Vision,” 68); see also, Heffernan, The Passion of Perpetua, 291–292, 295.

57 Tilley rightly notes that we should not assume that Caecilianist rhetoric around Donatist desires for martyrdom is historically accurate (Sustaining Donatist Self-Identity: From the Church of the Martyrs to the Collecta of the Desert,” Journal of Early Christian Studies 5, no.1 [1997]: 21–35Google Scholar). For an indepth analysis of the history versus rhetoric around Donatism, see Hogrefe, Umstrittene Vergangenhei. Here—and in what follows—I am not making a historical case for Donatist practice but, rather, drawing on the reputation of Donatists and the resultant expectations of a fifth-century audience. Some late ancient authors attribute the more fanatical quests for martyrdom to a fringe group of Donatists called the Circumcellions. Nevertheless, the reputation of the Donatists—from the Caecilianist perspective—was that of a group that sought out martyrdom.

58 After her first vision, Perpetua tells her brother that they will endure martyrdom and from that time on she has “no hope in this age” (4.10). In 11.4 Saturus reminds Perpetua that God has kept his promise of martyrdom.

59 Petilian also appealed to Matthew 12:35.

60 Augustine, Answer to Petilian 2.25.58.

61 Optatus, Against the Donatists 5.1, in PL 11:1044C.

62 Optatus, Against the Donatists 5.1, in PL 11:1046A.

63 Optatus, Against the Donatists 5.1, in PL 11:1047B.

64 See Everett Ferguson's thorough treatment of Augustine's anti-Donatist arguments about baptism in Baptism in the Early Church: History, Theology, and Liturgy in the First Five Centuries (Grand Rapids, Mich: Eerdmans, 2009), 795803Google Scholar.

65 Augustine, Answer to Petilian I.2, 5. That Christ or the Holy Spirit is the power behind the sacraments, which makes them efficacious, is a constant refrain in Augustine's anti-Donatist writings. See also, for instance, Against Cresconius, II. 16–19, 34. For a detailed discussion of this tract, see Hogrefe, Umstrittene Vergangenheit, 61–106.

66 As the first evangelist describes them in Matt. 23:13–26.

67 Augustine, Answer to Petilian II.73.162.

68 Augustine, On Baptism I.1.2.

69 Augustine, Answer to Petilian II.86.191.

70 Augustine, On Baptism II.14.19.

71 These condemnations may be found throughout Augustine's treatise On Baptism. For instance: 1.1.2; 1.12.18; 2.14.19.

72 David Biale discusses the relationship of baptism to martyrdom in both Judaism and Christianity in Blood and Belief: The Circulation of a Symbol Between Jews and Christians (Berkeley: University of California Press, 2007), 7580Google Scholar. See also Jeanes, Gordon, “Baptism Portrayed as Martyrdom in the Early Church,” Studia Liturgica 23 no. 2 (1993): 158176CrossRefGoogle Scholar.

73 The author of Acta A implies here that Felicitas died by sword. This is likely a survival from the Passio where she does, indeed, die by the sword.

74 It is impossible to tie this to Augustine's insistence on the “cause” of martyrdom being more important than the fact of death, but the statement is certainly consistent with this Augustinian stance. This distinction was one of the ways Augustine attempted to differentiate the church's early martyrs from the Donatist's recent claims to martyrdom.

75 Augustine's statements in On the Soul III.12 are relevant here. In arguing against Vincent Victor, he takes up the case of the thief on the cross next to Jesus. Augustine accounts for Jesus’ promise that the thief will be with him in paradise by positing a number of possibilities: perhaps the thief was purified by martyrdom; or maybe the water from Jesus’ side sprinkled on him as a baptism; or might the thief have been baptized in prison? In any case, baptism is critical for salvation to Augustine, but martyrdom can substitute for it. Augustine also takes up Perpetua's second and third visions—absent in the Acta—here to reject the notion that prayers can posthumously save an unbaptized person. Augustine suggests that Dinocrates could have been baptized before his death. His pagan father might then have pulled him back into the non-Christian life, and, thus, Perpetua's intercession brought him relief.

76 For example, Augustine preached On Faith and the Creed to a gathering of bishops in Hippo Regius in 393 CE. In this sermon he explicated the creedal statement regarding one baptism: “I confess one baptism for the remission of sin” (confiteor unum baptisma in remissionem peccatorum). Similarly, in On Baptism, Augustine insists that there is “one baptism” (baptisma unum) just as there is only one God, one Christ, and one faith (5.26.37). Optatus taught in Against the Donatists that the heavenly wedding would be celebrated only by those who preserved the “singular baptism” (baptisma singulare; V.10).

77 MS 3b (Einsiedeln 250 [382]) is a twelfth-century text from Our Lady of the Hermits in Einsiedeln. See discussion in Heffernan, The Passion of Perpetua and Felicity, 370–375, 389–394. This is a defective manuscript in that it conflates the Passio with the Acta, a fact discussed by Heffernan in The Passion of Perpetua and Felicity, 370.

78 Heffernan argues that these two textual traditions—MS 3b and the Greek—descend from a common exemplar (Passion of Perpetua and Felicity, 375).

79 The author of Acta B changes the text quite drastically at this point. For our purposes, it is worth noting that he exchanges sacrificial or baptismal language for more explicit ritual/clerical language: Felicitas was happy to be “consecrated by her own blood” (9.2).

80 For a thorough discussion of the phrase salvum lotum in ancient bathing culture, see Franz Joseph Dölger, who discusses the important mosaic in Brescia (=Inscription 5725 in Herman Dessau, ed. Inscriptiones Latinae Selectae. Berlin: Weidmannsche Verlagsbuchhandlung, 1995). Dölger, Gladiatorenblut und Martyrerblut, Eine Szene der Passio Perpetuae in kultur -und religionsgeschichtlicher Bedeutung,” Bibliothek Warburg, Vorträge 1923–1924 (1926): 198201Google Scholar; See also Dölger, , “Tertullian über die Bluttaufe: Tertullian De baptismo 16,” in Antike und Christentum. Kultur- und Religionsechichtliche Studien, Band ii (Münster in Westfalen: Aschendorffsche Verlagsbuchhandlung, 1930), 117141Google Scholar.

81 See, for instance, Tertullian, On Baptism 16.

82 non facit poena sed causa. See discussion below.

83 Heffernan argues that the Greek author “completely misses the pun employed by the Latin” (Passion of Perpetua and Felicity, 359). That is, in the Latin text, the crowd shouts out what appears to have been a traditional greeting at the bathhouse: salvum lotum! The phrase was used to wish the bather good health as a result of the bath. The Greek author, then, missing the pun of salvum and salvus, translates the Latin literally. Alternatively, Heffernan suggests the Greek author could be using ὑγιὴς figuratively—reflective of the author's “interest in ‘theologizing’” (Passion of Perpetua and Felicity, 359). I am less inclined than Heffernan to discount the Greek translator's linguistic and theological dexterity. On the Latin greeting, see Dickey, Eleanor, The Colloquia of the Hermeneumata Pseudodositheana, vol 2, Colloquium Harleianum, Colloquium Montepessulanum, Colloquium Celtis, and Fragments (New York: Cambridge University Press, 2015), 132Google Scholar.

84 See, for instance, Cobb, Dying to Be Men.

85 There are good reasons to be skeptical of the descriptions of Donatists as eager for martyrdom—not least because these claims originate from their opponents (namely Optatus and Augustine). See Tilley, “Sustaining Donatist Self-Identity,” 21–35; Tilley, Donatist Martyr Stories; and Dearn, Alan, “Voluntary Martyrdom and the Donatist Schism,” Studia Patristica 39 (2003): 2732Google Scholar.

86 Carol Straw discusses the problems early Christian martyrs posed for later church theologians, like Augustine in Martyrdom and Christian Identity: Gregory the Great, Augustine, and Tradition,” in The Limits of Ancient Christianity: Essays on Late Antique Thought and Culture in Honor of R. A. Markus, ed. Klingshirn, William E. and Vessey, Mark (Ann Arbor: University of Michigan Press), 250266Google Scholar. She writes, for instance, “the zealous and sanguinary heroism of the Christian early martyrs could all too easily become the rogue independence of fanatics splintering the church” (251).

87 On the history of the category “voluntary martyrdom”—or what G. E. M. de Ste Croix labels “religious suicide”—and its occurrence in the ancient world, see: de Ste Croix, G. E. M., “Voluntary Martyrdom in the Early Church,” in Christian Persecution, Martyrdom, and Orthodoxy (New York: Oxford University Press, 2006), 153200CrossRefGoogle Scholar; Moss, Candida R., “The Discourse of Voluntary Martyrdom: Ancient and ModernChurch History 81 no. 3 (2012): 531551CrossRefGoogle Scholar; and Middleton, Paul, “Early Christian Voluntary Martyrdom: A Statement For the Defense,” Journal of Theological Studies n.s., 64, pt. (2013): 556573CrossRefGoogle Scholar. For an earlier—yet still classic—discussion, see de Ste Croix, G. E. M., “Aspects of the Great Persecution,” Harvard Theological Review 47 no. 2 (1954): 75109CrossRefGoogle Scholar and a rebuttal of Ste Croix's conclusions in Buck, P. Lorraine, “Voluntary Martyrdom Revisited,” Journal of Theological Studies n.s. 63 no. 1 (2012): 125135CrossRefGoogle Scholar. Middleton offers the important observation that modern scholars tend to discuss “voluntary” martyrdom in moralistic terms. While this attitude is reflective of some early Christian authors, it does not adequately relate the complexity of the phenomenon in early Christian discourse and history.

88 Dearn, Alan, “Donatist Martyrs, Stories and Attitudes,” in The Donatist Schism: Controversy and Contexts, ed. Miles, Richard (Liverpool: Liverpool University Press, 2016), 97Google Scholar. This is not to say, as Dearn reminds us, that Donatists constructed their own group identity in this way.

89 Straw argues that Augustine represents a watershed moment in shifting the theology of martyrdom from individual acts to corporate ones: “when Augustine replaces heroic free will with grace, he lends support to an institutional definition of martyrdom in which grace is mediated through the agency of the church” (“Martyrdom and Christian Identity,” 265). The Donatist controversy forced Augustine to argue that autonomy—a quintessential virtue of the second and third centuries—must give way to “obedience to institution” (Straw, “Martyrdom and Christian Identity,” 265).

90 See, for example, Augustine, Serm. 359B.19.

91 See, for example, Augustine, Serm. 274, 283, 325, 327, 335, 335G, 359B.

92 Cotter-Lynch, Saint Perpetua, 48.

93 Council of Carthage, Canon 2. See discussion in Salisbury, Joyce E., The Blood of the Martyrs: Unintended Consequences of Ancient Violence (New York: Routledge, 2004), 120CrossRefGoogle Scholar; Minois, Georges, History of Suicide: Voluntary Death in Western Culture, trans. Cochrane, Lydia G. (Baltimore: Johns Hopkins University Press, 1999), 27Google Scholar.

94 Croix, De Ste, “Voluntary Martyrdom in the Early Church,” in Christian Persecution, Martyrdom, and Orthodoxy (New York: Oxford University Press, 2006), 192CrossRefGoogle Scholar.

95 Optatus, Against the Donatists 3.4.

96 Augustine, Epistle 185.3.12.

97 Shaw discusses the Donatists’ preference for the term agonistici over circumcellio in Sacred Violence, 633–638.

98 On the problem of the label “Circumcellion” and a discussion of the radical dissident group, see Shaw, Sacred Violence, 633–720. Shaw argues that to be a circumcellion was to participate in particular kinds of activities; it is a “behavioral category” not a particular belief system or an indicator of membership in a particular group (Sacred Violence, 656). Moreover, Circumcellions saw their violent acts as divine justice. They wielded clubs, which they called Israels and saw themselves as agents of God's retributive justice on earth.

99 Droge, Arthur J. and Tabor, James D. explore suicide in the ancient world in A Noble Death: Suicide and Martyrdom Among Christians and Jews in Antiquity (San Francisco: Harper Collins, 1992)Google Scholar.

100 This is not to say, as Shaw reminds us, that the Circumcellions were necessarily religious radicals who courted death. Rather, it is likely they saw themselves as victims of imperial and ecclesiastical persecution. “The collective self-killings,” Shaw writes, “could then be interpreted as symptoms of a despair in which the dissidents saw themselves increasingly hemmed in by their enemies—necessary sacrifices that were martyrdoms in their eyes” (Sacred Violence, 726–27). This seems to be the case for Gaundentius, for example.

101 For a discussion of the various positions on suicide in the early Church, see G. E. M. de Ste Croix, “Voluntary Martyrdom”; Minois, History of Suicide; and van Hooff, Anton, “From Voluntary Death to Self-Murder: The Dialogue on Self-Killing Between Antiquity and Christian Europe,” in Alma Parens Originalis?: The Receptions of Classical Literature and Thought in Africa, Europe, the United States, and Cuba, ed. Hilton, John and Gosling, Anne (Oxford: Peter Lang, 2007), 269290Google Scholar.

102 On Lactantius's arguments concerning suicide and homicide, see Shaw, Sacred Violence, 725.

103 See also Augustine, On Patience 13.10.

104 Augustine, Against Gaudentius 1.1.1. Gaudentius's own explanation of the situation, as preserved by Augustine, differs radically: he explains that they will stay in the church, and they will die there only if the imperial forces employ violence against them. That is, Gaudentius does not portray the Donatists as eager for death; rather, he and his congregants will submit to persecution and death if it is brought upon them. (Against Gaudentius 1.6.7). See discussion in Shaw, Brent D., “State Intervention and Holy Violence: Timgad/Paleostrovsk/Waco,” Journal of the American Academy of Religion 77 no. 4 (2009): 853894CrossRefGoogle Scholar.

105 Augustine, Against Gaudentius 1.27.30.

106 See Shaw who discusses the ways the martyr was a particularly problematic figure in the Caecilianist-Donatist struggle over self-death (Sacred Violence, 724–727).

Send article to Kindle

To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Suicide by Gladiator? The Acts of Perpetua and Felicitas in its North African Context
Available formats

Send article to Dropbox

To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

Suicide by Gladiator? The Acts of Perpetua and Felicitas in its North African Context
Available formats

Send article to Google Drive

To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

Suicide by Gladiator? The Acts of Perpetua and Felicitas in its North African Context
Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *