¹ Lunn-Rockliffe, S., ‘Commemorating the usurper Magnus Maximus: ekphrasis, poetry, and history in Pacatus’ Panegyric of Theodosius’, Journal of Late Antiquity 3 (2010), 316–36Google Scholar, at 321–3; Omissi, A., ‘Damnatio memoriae or creatio memoriae? Memory sanctions as creative processes in the fourth century a.d.’, CCJ 62 (2016), 170–99Google Scholar, at 176. For the laws abolishing Maximus’ enactments, see Cod. Theod. 14.15.6–8.

² H. Leppin, ‘Coping with the tyrant's faction: civil-war amnesties and Christian discourses in the fourth century a.d.’, in J. Wienand (ed.), Contested Monarchy: Integrating the Roman Empire in the Fourth Century a.d. (Oxford, 2015), 198–214 highlights the fairly consistent nature of Roman practice in this regard, with a generalized amnesty accompanied by retributive punishment against an ill-defined inner circle. It was vital to find oneself outside that circle.

³ See page 17 below.

⁴ H. Chadwick, Priscillian of Avila: The Occult and the Charismatic in the Early Church (Oxford, 1975), 8, 117, 121, 148; J. Matthews, Western Aristocracies and Imperial Court, a.d. 364–425 (rev. edn; Oxford, 1990), 181; S.N.C. Lieu, Manichaeism in the Later Roman Empire and China (Wissenschaftliche Untersuchungen zum Neuen Testament 63) (Tübingen, 1992²), 148; R.M. Errington, Roman Imperial Policy from Julian to Theodosius (Chapel Hill, 2006), 209.

⁵ P.R. Coleman-Norton, Roman State & Christian Church: A Collection of Legal Documents to a.d. 535, 3 vols. (London, 1966), 2.399–403.

⁶ O. Günther, Corpus Scriptorum Ecclesiasticorum Latinorum, vol. XXXV. Epistulae imperatorum pontificum aliorum inde ab a. CCCLXVII usque ad a. DLIII datae Avellana quae dicitur Collectio (Leipzig, 1895–8).

⁷ Much the best and fullest account remains N. McLynn, Ambrose of Milan: Church and Court in a Christian Capital (Berkeley, 1994), 170–219, although it has now been complemented by the insightful M.S. Williams, The Politics of Heresy in Ambrose of Milan: Community and Consensus in Late Antique Christianity (Cambridge, 2017), 214–86, which largely rejects Ambrose's characterization of the dispute as one over doctrine, and frames it rather as a contest over authority and the control of public space. See also Barnes, T.D., ‘Ambrose and the basilicas of Milan in 385 and 386: the primary documents and their implications’, ZAC 4 (2000), 282–99CrossRefGoogle Scholar; J.H.W.G. Liebeschuetz, Ambrose of Milan: Political Letters and Speeches (Liverpool, 2005), 124–36 (with 136–73 a translation of the three ‘letters’ of Ambrose—one is in fact a sermon—that constitute our key primary witnesses for this episode).

⁸ Terminology is difficult but important. Though Ambrose and others among the Nicene party might denounce their opponents as ‘Arians’, this label is in fact a poor fit for the non-Nicene party in Milan, who did not subscribe to Arius’ teachings concerning the created nature of the Son. Accordingly, when referring to the non-Nicene party in Milan, I use the term ‘Homoian’, which more accurately reflects the tenets of their theological position, though on occasion I use the term ‘Arian’ when referring to the assertions of those who, like Ambrose and Maximus, used it to slander their opponents. On the term ‘Homoian’, see U. Heil, ‘The Homoians’, in G.M. Berndt and R. Steinacher (edd.), Arianism: Roman Heresy and Barbarian Creed (Farnham, 2014), 85–115.

⁹ This law, astoundingly, has survived and was included in the Theodosian Code: Cod. Theod. 16.4.1.

¹⁰ This was Auxentius of Durostorum who ought not to be confused with Auxentius of Milan, Ambrose's predecessor in the bishopric who was also—rather unhelpfully—a Homoian: cf. Williams (n. 7), 252–81.

¹¹ See note 86 below.

¹² Easter in 386 was almost certainly celebrated in Milan on 5 April, the date prescribed both by the Roman eighty-four-year cycle and the Alexandrian nineteen-year cycle: cf. A. Mosshammer, The Easter Computus and the Origins of the Christian Era (Oxford, 2008), 214–15. M.V. Escribano Paño, ‘Maximus’ letters in the Collectio Avellana: a comparative study’, in R. Lizzi Testa and G. Marconi (edd.), The Collectio Avellana and its Revivals (Newcastle upon Tyne, 2019), 65 dates the letter to ‘some time after April 386’, without—to my mind—addressing the arguments I give here that suggest contemporaneity with the events they describe.

¹³ Escribano Paño's recently published book-chapter (n. 12) follows broadly the same interpretative line that I take here, which I would consider the mainstream interpretation. Neil McLynn's forthcoming chapter ‘Tyrants, Arians, and Manichees: Magnus Maximus in the Collectio Avellana’, in A. Evers (ed.), Emperors, Bishops, Senators: The Evidence of the Collectio Avellana (Leuven, forthcoming) takes Maximus’ expressed concern for Valentinian within the two letters more clearly at face value—unusual for McLynn—and sees a certain fatherly concern in their tone. Both authors see the two letters (Coll. Av. 39 and 40) as probably having been sent at the same time, which is not an argument I would seek to pursue (perfectly plausible though it is). See also the following note.

¹⁴ Matthews (n. 4), 181 sees the letter as ‘remarkably patronising’ but none the less a potential indicator of ‘a certain easing of relations between Milan and Trier’; Barnes (n. 7), 296–7 sees the letter, as I do, as a more or less open threat; most recently McLynn (n. 13) has argued that Maximus’ words ought to be taken at face value and that the letter is one of honestly intended advice by Maximus seeking not to overawe Valentinian but to win him over.

¹⁵ Amm. Marc. 31.10.18–19. Other sources make reference to Gratian having showed undue favouritism to his Alan bodyguards (Aur. Vict. Epit. 47.7; Zos. 4.35.2–3), which hardly seems grounds for military revolt, though cf. A. Omissi, Emperors and Usurpers in the Later Roman Empire (Oxford, 2018), 21–34. For modern accounts of the reign of Maximus, see J.R. Palanque, ‘L'empereur Maxime’, in Les Empereurs Romains de l'Espagne (Paris, 1965), 255–67; Matthews (n. 4), 173–82, 223–7; A.R. Birley, ‘Magnus Maximus and the persecution of heresy’, BRL 66 (1983–4), 13–43; Omissi (this note), 263–90.

¹⁶ Ambr. Ep. 30[24].4. Maximus’ Gallic issues featured legends such as RESTITVTOR REIPVBLICAE and REPARATIO REIPVBLICAE (RIC IX 1–70), common legends for emperors who saw their reigns as correctives to previous misgovernment.

¹⁷ On reasons for caution over accepting a more traditional and simplistic formulation that Valentinian and his mother were themselves Arians, see Williams (n. 7), 221–3.

¹⁸ Cf. Ambr. Ep. 76[20].12. On Maximus’ relationship to Theodosius and his father Theodosius the Elder, see Amm. Marc. 29.5.6, 29.5.21; Pan. Lat. 2(12).24.1, 2(12).43.4, with the commentary of C.E.V. Nixon and B.S. Rodgers (edd.), In Praise of Later Roman Emperors: The Panegyrici Latini: Introduction, Translation, and Historical Commentary, with Latin Text of R.A.B. Mynors (Berkeley, 1994), 479 n. 83; Zos. 4.35.3.

¹⁹ Aur. Vict. Epit. 47.7; Fast. Vind. prior 502 (= MGH AA.9 297); Oros. 7.34.10; Prosper, Chron. s.a. 384 (= MGH AA.9 461); Socrates, Hist. eccl. 5.10; Sozom. Hist. eccl. 7.11; Sulp. Sev. Chron. 2.49.5; Zos. 4.35.4–6.

²⁰ For the quotation, see Ambr. Ep. 30[24].7. The main evidence for this embassy comes from this letter (though it is also alluded to at Ambr. Ep. 75[21].20). For modern accounts, see McLynn (n. 7), 160–3; N. Dörner, ‘Ambrosius in Trier’, Historia 50 (2001), 217–44; Y.-M. Duval, ‘Les ambassades de saint Ambrose auprès de l'usurpateur Maxime en 383 et 384’, in J.-M. Carrié and R. Lizzi Testa (edd.), “Humana sapit”: Études d'Antiquité tardive offertes à Lellia Cracco Ruggini (Turnhout, 2002), 239–51.

²¹ Vera, D., ‘I rapporti fra Magno Massimo, Teodosio e Valentiniano nel 383–4’, Athenaeum 53 (1975), 267–301Google Scholar; H.R. Baldus, ‘Theodosius der Grosse und die Revolte des Magnus Maximus: das Zeugnis der Münzen’, Chiron 14 (1984), 175–92; Matthews (n. 4), 176–82; Lunn-Rockliffe (n. 1), 316–36; Omissi (n. 15), 263–9.

²² Cf. Escribano Paño (n. 12), 56–7, who interprets these passages similarly.

²³ Sulp. Sev. Chron. 2.49–51; Dial. 2.6, 3.11; V. Mart. 20.

²⁴ See page 14 below.

²⁵ The sources of this are complex, and the religious motivations for this war are likely back-writing on the part of Constantine. For consideration of these sources, see Elliott, T.G., ‘Constantine's explanation of his career’, Byzantion 62 (1992), 212–34Google Scholar.

²⁶ Socrates, Hist. eccl. 2.22.5; cf. Philostorgius, Hist. eccl. 3.12; Sozom. Hist. eccl. 3.20.1; Theodoret, Hist. eccl. 2.9; Theoph. 5849.

²⁷ See page 6 above. Ambrose clearly felt (and wished to display) his loyalty to Gratian, which found textual expression in his commentary on Psalm 62(61), rich with evocation of the treachery and calumny that sent the Christ-like Gratian to his death at the hands of the wicked Maximus: Raschle, C.R., ‘Ambrosius’ Predigt gegen Magnus Maximus. Eine historische Interpretation der explanatio in psalmum, 61 (62)’, Historia 54 (2005), 49–67Google Scholar.

²⁸ McLynn (n. 7), 160.

²⁹ Dating this second embassy has proved very difficult, with any date within the range 384–387 a possibility. On this, see page 11 with n. 38 below. Williams (n. 7), 215–19 usefully explores the reasons for Ambrose's continued allegiance to Valentinian.

³⁰ The best account of Priscillian's life in English is still Chadwick (n. 4). See also C. Stancliffe, St. Martin and his Hagiographer: History and Miracle in Sulpicius Severus (Oxford, 1983), 278–83; Birley (n. 15), 18–36; M. Conti (ed. and transl.), Priscillian of Avila: The Complete Works (Oxford, 2010), 1–12.

³¹ Lieu (n. 4), 148–50.

³² Sulp. Sev. Chron. 2.47.1–4; Chadwick (n. 4), 12–15, 26–7; Birley (n. 15), 19–20.

³³ Sulp. Sev. Chron. 2.47.5–7; Chadwick (n. 4), 35–6.

³⁴ Like many commonplaces of our discipline, one can find this assertion in Edward Gibbon's pages: The History of the Decline and Fall of the Roman Empire, 6 vols. (ed. J.B. Bury) (London, 1896–1900), 3.153.

³⁵ Maximus refers to the events of the trial as ‘recently’ (40.4 proxime) made known.

³⁶ Sulp. Sev. Dial. 3.13; Birley (n. 15), 29 notes that emendations of the manuscript reading have been suggested here.

³⁷ Prosper, Chron. s.a. 385 (= MGH AA.9 462); Hydatius, Chron. 291.16 (see R.W. Burgess [ed.], The Chronicle of Hydatius and the Consularia Constantinopolitana: Two Contemporary Accounts of the Final Years of the Roman Empire [Oxford, 1993], 77).

³⁸ Ambrose states that he refused communion with the Gallic bishops responsible for the trial and actually saw the bishop Hyginus of Cordoba being led away into exile (Ep. 30[24].12), so he clearly arrived when the trial was either ongoing or recently concluded. Liebeschuetz (n. 7), 349 n. 7 argues that, since the subject of Maximus and Ambrose's conversation is largely the events of 383, the embassy ought probably to be dated early. Barnes (n. 7), 293–4 finds in Ambrose's letter to his sister Marcellina (Ep. 76[20]) a reference to the second mission, and so dates it to 384. Chadwick (n. 4), 132–8 concedes that any date in the period 384–387 is possible, but deems 386 the most likely, whilst Birley (n. 15), 30–3 argues for the year 387.

³⁹ Both Escribano Paño (n. 12), 65–6 and McLynn (n. 13) assert that Coll. Av. 39 and 40 were sent at the same time, though without providing any particular evidence. This is certainly possible, but there is no reason to see it as certain.

⁴⁰ F. Millar, The Emperor in the Roman World (31 b.c. – a.d. 337) (London, 1977), 507–16; C. Rapp, Holy Bishops in Late Antiquity: The Nature of Christian Leadership in an Age of Transition (Berkeley, 2005).

⁴¹ E.g. N. Lenski, ‘Constantine and the Donatists: exploring the limits of religious toleration’, in M. Wallraff (ed.), Religiöse Toleranz: Moderne Idealien im Spiegel antiker Realien. Colloquium Rauricum XIV (Berlin, 2015), especially 104–9.

⁴² T.D. Barnes, Athanasius and Constantius: Theology and Politics in the Constantinian Empire (Cambridge, Mass., 1993), 101–5.

⁴³ For which, see K. Zechiel-Eckes, Die erste Dekretale: Der Brief Papst Siricius’ an Bischof Himerius von Tarragona vom Jahr 385 (JK 255): Aus dem Nachlass mit Ergänzungen herausgegeben von Detlev Jasper (MGH Studien und Texte Band 55) (Hannover, 2013).

⁴⁴ See page 5 above.

⁴⁵ Coleman-Norton (n. 5), 2.399 describes the letter as the emperor's ‘defence against the pope's protest about the former's execution of some Priscillians’. Escribano Paño (n. 12), 66–72 is more guarded, but none the less asserts that Siricius must have made enquiry about the execution.

⁴⁶ He introduces the trial of Priscillian with the phrase ‘as for the crimes which it has recently been made known (quid adhuc proxime proditum sit) that the Manichaeans are committing’, which suggests less that he is responding to an issue that Siricius himself raised than that he is addressing a topic so widely known that he can expect Siricius to have heard about it, but that the bishop had not himself brought up; a more idiomatic rendering (though more distant from the Latin text) would be ‘you will have heard, no doubt, that the Manichaeans confess to a crime’.

⁴⁷ Ambrose was hardly as conciliatory as I propose that Siricius was being here. During his second mission to the court of Maximus, he refused all communion with the bishops involved in the Priscillianist trial, for which Maximus lost all patience with him and ordered him to leave (Ambr. Ep. 30[24].12).

⁴⁸ As Richard Flower points out in the introduction to his Emperors and Bishops in Late Roman Invective (Cambridge, 2013), 17–18, what makes Athanasius, Hilary of Poitiers and Lucifer of Cagliari such commanding objects of study is that they had the nerve to engage in open and direct criticism of a living emperor (and all three notably underwent exile for their resistance to the imperial will).

⁴⁹ E.g. Birley (n. 15), 14, 24, 36; McLynn (n. 7), 161 n. 10.

⁵⁰ Wiemer, H.-U., ‘Akklamationen im spâtrömischen Reich: zur Typologie und Funktions eines Kommunikationsrituals’, AKG 86 (2004), 27–73Google Scholar.

⁵¹ J. Nelson, Politics and Ritual in Early Medieval Europe (London, 1986), 259–64; G. Dagron, Emperor and Priest: The Imperial Office in Byzantium (Cambridge, 2003), 59–83.

⁵² Constantine, according to Bishop Eusebius, refused to don his imperial robes again after his baptism (Vit. Const. 4.62). Theodosius had been baptized in the autumn of 380 during a severe illness, from which it was feared that the emperor would die (Socrates, Hist. eccl. 5.6; Sozom. Hist. eccl. 7.4). In 392, as Valentinian II's options slowly narrowed, he reached out to Ambrose of Milan to beg for baptism, perhaps aware that he was soon likely to die, whether by his own hand or by another's (Ambr. Ep. 25[53]). A movement to earlier baptism across the fourth century is detectable, with Valentinian having been baptized while still a private individual and Valens receiving this sacrament whilst emperor but in no danger of imminent death; cf. N. McLynn, ‘The transformation of imperial churchgoing in the fourth century’, in S. Swain and M. Edwards, Approaching Late Antiquity: The Transformation from Early to Late Empire (Oxford, 2004), 250–8. None the less, the uniting of an accession with this Christian ritual is without precedent.

⁵³ Ambrose of Milan made much, in a letter to Theodosius of 388/9, of the fact that Maximus had been denounced as a Jew (Ambr. Ep. 74[40].23), and Theodosius’ orator, Pacatus, devoted much attention to Maximus’ impiety: Omissi (n. 15), 280–2.

⁵⁴ McLynn (n. 13). Likewise, Escribano Paño (n. 12), 51: ‘The reason why two letters from a tyrannus would be incorporated into the CA remains an insufficiently addressed matter.’

⁵⁵ Maximus’ laws were explicitly declared void (Cod. Theod. 14.15.6–8, with my own comments, below).

⁵⁶ See the comments on Athanasius, page 12 above.

⁵⁷ In general, works on memory sanctions tend not to draw our attention to the sheer amount of surviving material for the obvious reason that this is not their object of focus. For some quantification of epigraphic erasures, see C. Crespo Pérez, La Condenación al Olvido (damnatio memoriae): La deshonra pública tras la muerte en la politica Romana (Siglos I–IV d.C.) (Madrid, 2014), especially 54–9.

⁵⁸ The ghosts of a number of Licinius’ laws can be detected in the code (see S. Corcoran, ‘Hidden from history: the legislation of Licinius’, in J. Harries and I. Wood [edd.], The Theodosian Code: Studies in the Imperial Law of Late Antiquity [London, 1993], 97–119) as, notably, can one or perhaps even two laws of Maximus himself: T. Honoré, Law in the Crisis of Empire 379–455 a.d.: The Theodosian Dynasty and its Quaestors (Oxford, 1998), 187–9. Notably, these laws survive without the name of the condemned attached to them, and only retrojecting dates and place of issue can help us detect their origins. Laws issued to individuals who later suffered memory sanctions likewise survive: H. Ménard, ‘La mémoire et sa condamnation d'après les codes tardifs: l'exemple de la révolte d'Héraclien en 413 apr. J.-C.’, in S. Benoist and A. Daguet-Gagey (edd.), Mémoire et histoire: les procédures de condamnation dans l'Antiquité romaine (Metz, 2007), 267–78. Notably, our letters are not texts of this category but rather documents that—particularly in the case of Coll. Av. 40—directly incriminated specific individuals in treasonous activity.

⁵⁹ In general on Rufinus, see F.X. Murphy, Rufinus of Aquileia (345–411): His Life and Works (Washington, DC, 1945); Hammond, C.P., ‘The last ten years of Rufinus’ life and the date of his move south from Aquileia’, JThS 28 (1977), 372–429Google Scholar; Fedalto, G., ‘Rufino di Concordia: elementi di una biografia’, AAAD 39 (1992), 19–44Google Scholar; D. Rohrbacher, The Historians of Late Antiquity (London, 2002), 93–107; P.R. Amidon (transl.), Rufinus of Aquileia: History of the Church (Washington, DC, 2016).

⁶⁰ Hammond (n. 59), 373.

⁶¹ Cf. Escribano Paño (n. 12), 74–6, who also explores their use by Theodoret of Cyrus.

⁶² On the court's movements between 386 and 402, see O. Seeck (ed.), Regesten der Kaiser und Päpste für die Jahre 311 bis 476 n. Chr.: Vorarbeit zu einer Prosopographie der Christlichen Kaiserzeit (Stuttgart, 1919), 268–304. On Rufinus’ sources: Duval, Y.-M., ‘Sur quelques sources latines de l’Histoire de l’Église de Rufin d'Aquilée’, Cassiodorus 3 (1997), 131–51Google Scholar.

⁶³ At least so says Socrates (Hist. eccl. 5.14.7).

⁶⁴ J.A. McGeachy Jr., ‘The editing of the Letters of Symmachus’, CPh 44 (1949), 222–9, at 223–4; A. Cameron, The Last Pagans of Rome (Oxford, 2011), 370–1.

⁶⁵ A.F. Norman, Libanius’ Autobiography and Selected Letters, 2 vols. (Cambridge, Mass., 1992), 1.42–3; N. Lenski, Failure of Empire: Valens and the Roman State in the Fourth Century a.d. (Berkeley, 2002), 6–7. In his Ep. 840, from 388, Libanius bears witness to the cloud that had hung over him in his relief at learning, through certain of his allies at court, that he had now passed beyond suspicion and could rest easy. Recent work on the gap in Libanius’ letters (e.g. L. Van Hoof, ‘Self-censorship and self-fashioning: gaps in Libanius’ letter collection’, RBPh 91 [2014], 209–29) enriches this picture rather than fundamentally contradicting it.

⁶⁶ The usurper Silvanus was, for example, driven to rebellion after being convicted on the (false) evidence of letters that fell into the hands of Constantius’ agents (Amm. Marc. 15.5.3–5). For other examples of letters getting either sender or recipient in trouble, see 19.12.4–5, 28.1.20, 28.6.26 and 29.2.25.

⁶⁷ Stancliffe (n. 30), 71 and 80–1.

⁶⁸ Sulp. Sev. Dial. 2.6; cf. 3.11 and V. Mart. 20. For the Gospel account of the woman who washed Christ's feet, which Sulpicius explicitly references, see Luke 7:36–50.

⁶⁹ Sozom. Hist. eccl. 7.13. See also Zos. 4.42.

⁷⁰ Günther (n. 6); P. Fournier and G. Le Bras, Histoire des collections canoniques en Occident depuis les Fausses décretales jusqu'au Décret de Gratien (Paris, 1931); L. Kéry, Canonical Collections of the Early Middle Ages (ca. 400–1140) (Washington, DC, 1999). The Collectio Avellana survives in two principal manuscripts now housed in the Vatican. Vat. Lat. 3787 is considered both the earliest and the best version of the text, dating from the late tenth or early eleventh century. Vat. Lat. 4961 is considered slightly younger, certainly from the eleventh century, and was formerly housed at S. Croce di Fonte Avellana in the Marche, from which the Collectio Avellana derives its name (the name can be traced back to the treatment of the Ballerini brothers in the eighteenth century who described a collection of Avellana). A number of later manuscripts from between the fifteenth and the seventeenth centuries exist but are demonstrably dependent upon Vat. Lat. 3787 and 4961: D. Jasper and H. Fuhrmann, Papal Letters in the Early Middle Ages (Washington, DC, 2001), 83–5. It is from Vat. Lat. 3787 that the German scholar Otto Günther (n. 6) made his two-part edition, published in 1895 and 1898 and still considered to be the canonical edition of the text.

⁷¹ Jasper and Fuhrmann (n. 70), 84.

⁷² K. Blair-Dixon, ‘Memory and authority in sixth-century Rome: the Liber Pontificalis and the Collectio Avellana’, in K. Cooper and J. Hillner (edd.), Religion, Dynasty, and Patronage in Early Christian Rome, 300–900 (Cambridge, 2007), 60–5. Though the collection as a whole is believed to have been put together in the mid sixth century, shortly after the date of its latest document, the only secure terminus ante quem is the late tenth or early eleventh century, the date of our earliest manuscript. Blair-Dixon (this note), perhaps flippantly, asserts that an eleventh-century context would not be impossible.

⁷³ Günther (n. 6), 1.3–19.

⁷⁴ Lippold, A., ‘Ursinus und Damasus’, Historia 14 (1965), 105–28Google Scholar; Cristo, S., ‘Some notes on the Bonifacian-Eulalian schism’, Aevum 51 (1977), 163–7Google Scholar.

⁷⁵ Note that Coll. Av. 4 is an official letter to Siricius from Valentinian, Theodosius and Arcadius (not Maximus), congratulating the bishop on his election in the face of an attempted coup by Ursinus, and so Coll. Av. 40 could—at a stretch—be seen as in harmony with that.

⁷⁶ McLynn (n. 13).

⁷⁷ Escribano Paño (n. 12), 80–1.

⁷⁸ Ambr. Ep. extra coll. 5[11].3; as McLynn (n. 7), 58–60 points out, the association of Ursinus with the Arians is highly improbable, given Ursinus’ exemplarily orthodox credentials, but Ambrose's letter demonstrates that the accusation was being noised about.

⁷⁹ The hodgepodge nature of the collection has certainly been noted by commentators: Blair-Dixon (n. 72), 61–4.

⁸⁰ This heading is very obviously a later insertion, presumably made by the individual who put these letters together into the Collectio Avellana (since the reference to Manichaeans actually seems to pick up the topic of Coll. Av. 40).

⁸¹ Throughout the course of both letters, Maximus switches freely between singular and plural personal pronouns to refer to himself, a stylistic eccentricity that Honoré (n. 58), 187–8 posits no quaestor would exhibit, suggesting that these letters were personally composed (though of course not physically written, save for the six words following et manu imperatoris at the end of Coll. Av. 40) by Maximus himself. Certainly their content, for all its artifice, is personal and reflective of the emperor's own inner world, and I would see this hypothesis as highly probable.

⁸² Maximus’ statement that he wishes to discuss things ‘now occurring’ (nunc agi) provides the grounds to date the letter to the period of or immediately after Easter 386, that is, the first half of April 386.

⁸³ This statement is highly opaque, but it would seem to be that Maximus argues that, if he were Valentinian's enemy (which his language implies he is not; likewise §7), then Valentinian's current behaviour would be playing directly into his hands. This notion is here couched as a counterfactual hypothetical musing, but considering the strained relations between the two courts it should, I think, be read very clearly as a threat to the young emperor.

⁸⁴ The choice to make reference here to Valentinian's youth (the emperor was, in 386, fifteen years old) is a calculated one, and Maximus used their respective ages as a coercive tool in his dealings with the child emperor (cf. ‘behind the cloak of a little boy’ and ‘as a son to a father’, Ambr. Ep. 30[24].4 and 7). Ambrose too reminded Valentinian of his youth when it suited him to do so (e.g. Ep. 72[17].15).

⁸⁵ This is clearly a reference to the blockade of the church that took place at the end of March 386 (see page 4 above), and its present tense helps us to date this letter as having been written while these events were still ongoing, that is (allowing for the time it would take news to cross the Alps), sometime in early-to-mid April 386. Liebeschuetz (n. 7), 130–3 suggests that there may also have been an earlier siege in (perhaps) December 385, during which Ambrose was under semi-blockade with his congregation in his church, the Basilica Nova, by soldiers sent to arrest the bishop. However, the chronology here is so complex (thanks largely to the uncertain chronology of the three main Ambrosian sources) that the reality of the earlier siege can be legitimately questioned: Williams (n. 7), 226–39.

⁸⁶ On Palm Sunday 386 (29 March), whilst on their way to their sit-in at the Basilica Portiana, the Nicene faithful had accosted and beaten a Homoian priest, Castulus. In response the court had imposed an enormous fine of two hundred pounds of gold, to be paid within three days, upon the merchants of Milan (corpus negotiatorum), who, we must assume, were conspicuous among the crowd and also, presumably, an easy target for a fine because of both their wealth and their corporate status. Widespread arrests accompanied this action (Ambr. Ep. 76[20].6). Valentinian ultimately returned this money to the merchants (Ambr. Ep. 76[20].26).

⁸⁷ This and the ‘new edicts’ of a moment earlier are reference to Valentinian's law of 23 January granting freedom of assembly to the Homoians and threatening with death penalty anyone who attempted to oppose this: Cod. Theod. 16.1.4.

⁸⁸ Maximus’ argument here appears to be that the very idea of having to convince Valentinian that it is unwise to harm the Church of God is so absurd that, were this the only issue, he would not even attempt a defence, but that (as his argument develops) Valentinian is flouting not only divine law but human law as well, and is overturning positions legislated for by his imperial predecessors, not least his own father.

⁸⁹ The word here is sacramentum, which is difficult to interpret in this context, but given the list of regions Maximus delineates, the thread of his argument, and the haec fides of the following clause, it would seem clear that he is referring to Nicene orthodoxy, the belief in the consubstantiality of Father and Son, as opposed to the Arian subordination of Son to Father.

⁹⁰ This phrase (cuius etiam in hac parte principatus est) may, more than any other, explain the rationale for this letter's inclusion in the Collectio. Though it is certainly ambiguous, the etiam would seem to indicate that Maximus is declaring Rome's primacy not only over Valentinian's territory but over his own as well.

⁹¹ The list of territories given here is significant, comprising the regions ruled by Maximus and Valentinian. The territories are notably not an enumeration of the totality of the Roman world, for the regions ruled by Theodosius are omitted and these comments ought therefore to be seen again in the context of Maximus’ ambitions to gain total control of the western empire. Given that Maximus also stresses that he and the bishop of Rome, whose premier position he underscores, are in concord with one another, we can see that the emperor was using assertions of his own orthodoxy to advance a political position vis-à-vis the extent of his own secular authority.

⁹² Here again Maximus is obscure. Chadwick (n. 4), 118 reads in this a general assertion by Maximus that Gothic invasions (which troubled Illyricum in the 380s, though only tangentially, and which bore an obvious Arian connection) are the result of divine vengeance. This seems too obtuse. A more specific reading is that Maximus was referring to Valens of Mursa, a prominent Homoian bishop and confidant of Emperor Constantius, who would seem to fit the description that follows of a ‘former’ (quondam) ‘originator’ (auctor) of error, and who was infamous as a bulwark of Arianism, however poorly this label might fit Valens’ own beliefs; see especially Heil (n. 8), 85–115; cf. M. Dunn, Belief and Religion in Barbarian Europe c. 350–700 (London, 2013), 34–5. Valens’ reputation was an enduring one, and both he and, by association, Mursa could certainly be used as metonyms for ‘Arianism’ (cf. Ambr. de Spiritu sancto 3.10.59; Sulp. Sev. Chron. 2.36, 2.38–40, 2.44). Whatever the specific thinking behind the choice of Mursa, the general point, that Illyricum was viewed as a hotbed of Arianism, is clear: Dunn (this note), 37.

⁹³ Ambrose used exactly the same tactic when negotiating with the younger Valentinian, citing the father's highly tolerant stance on religion as a precedent to order the son to keep his nose out (Ep. 75[21].2, 5).

⁹⁴ Ambrose told his sister that Valentinian's secretary had actually denounced him as a tyrannus (Ambr. Ep. 76[20].23).

⁹⁵ nullo certe maiore genere curam meam circa clementiam tuam probare †te posse; this passage is very problematic and Günther (n. 6) marks it with an obelus to indicate that it is plainly corrupt. I have tried as best as I can to render the sense of the original here, but Günther's notes offer several possible emendations: probare tibi possem, probare recte possem, and probari credo posse.

⁹⁶ ab ipso statim salutari fonte conscenderim: here Maximus tells us that his accession was linked with his baptism, an unprecedented display of Christian kingship (see page 14 above).

⁹⁷ Maximus’ letter, though oblique and somewhat purple in its language, is nevertheless businesslike: he has three points he wishes to address with Siricius, each bulleted by a ‘moreover’ (ceterum).

⁹⁸ McLynn (n. 13) suggests that, far from meekly promising to further the bishop of Rome's interests here, Maximus was actually seeking to usurp his authority by transferring this matter to the jurisdiction of the bishops of Gaul. My own reading sees a much more cooperative Maximus.

⁹⁹ In a very indirect way, Maximus is here deferring to ecclesiastical law and to the expertise of the clergy of Gaul, saying that he will allow the leaders of the Church, who are the people most versed in the Church's own law (‘those people … who know [these things]’).

¹⁰⁰ For my comments on this phrase, see above, n. 46.

¹⁰¹ Priscillian was clearly denounced as a Manichaean, and had been so labelled under Gratian as well. Discovery of his own writings in the nineteenth century has shown that this charge was ill-founded (Lieu [n. 4], 148–50).

¹⁰² Maximus presumably included with this letter the official records of Priscillian's trial (the term he uses here is gesta, which is synonymous with the more common acta). Evidently, Maximus felt that these records would tell a story plain enough that he needed to add little to it.

¹⁰³ Here Maximus, as I argue above (see page 400), clearly defends the execution of the Priscillianists by reference to the horrifying depth of their heresy, something that Maximus suggests he cannot even bear to set down in writing.