4 - A New Order of Books in the Theodosian Age

Summary

Chapter 4 traces novel scholastic practices from the realm of theology into “secular” domains during the Theodosian Age, showing how a scholarly method created to solve theological problems came to be used in legal, historical, and scientific texts of the late fourth and fifth centuries.

Tools of the Trade: Aggregation, Distillation, and Promulgation

Leontius of Jerusalem was a strict defender of the theological advances of the fourth and fifth centuries. Writing from the Judean hills in the mid-sixth century, he inherited from his elders a Chalcedonian Orthodoxy along with a manner of argumentation that focused on compiling excerpts of authoritative theological scholarship from the past. He admits that his opponents find this to be an aggravating tactic. “Here are exactly the kinds of things they’ll offer in opposition to what we’ve said: ‘Why do you, when you buzz around patristic texts like bees, harvest honey from whatever example pleases you, and continually bombard us with your buzzing about them, but fly right over others that are hostile to your purposes, darting away from them in silence?’”¹ The bee metaphor is intriguing. Leontius describes his search for certainty as dramatically aleatory, predicated on the aggregation and distillation of authorized voices from the past – a sort of sentimental antiquarianism meant to lead the careful reader through a maze of scholarly material to an ultimate truth at the path’s end.

Leontius was no innovator. In fact, his style of argumentation was already traditional by the time that he wrote in the sixth century.² The metaphor of following the example of a bee when reading and digesting information transparently invokes Lucretius’s On the Nature of Things 3.11, a perennial favorite among the Latin literary elite, evoking an eclectic pattern of reading and borrowing from sources in a variety of genres.³ More importantly, already for nearly a century and a half Christian scholars of the Nicene tradition had engaged in a practice of aggregation as foundational to the adequate demonstration of truth.

Scott Johnson recently described the move to collections of authoritative and/or discursive scholarly material in Late Antiquity, specifically in the fourth through sixth centuries, as “an aesthetic of accumulation.”⁴ Another way to describe the increasing centrality of aggregation as a precursor for valid scholarly work is to say that such scholarship takes part in “the rise of the code.” Both will suffice as a description of the strange contours of a shifting late ancient book culture. But the description of an “aesthetic” should not be confused with an explanation for the shift to aggregation as a central scholarly method. With this chapter I describe the material and intellectual framing of Theodosian Age scholarship. I also explain why aggregation came to be more than just an aesthetic choice commonly held across the Roman empire. Aggregation, in the Theodosian Age, was an expectation held by producers of technical literature about what it looks like, and means, to do rigorous and worthwhile work. These changes compounded from a shift in theological argumentation that I detailed in Chapter 3. I argue that the move to aggregation as a foundational tool did not long remain strictly within the purview of Christian theological scholarship. Rather, the rise of aggregation accounts for the rise of the code as a nexus of power and truth, as well as the shifting facets of Theodosian book culture outlined later. The Council of Nicaea blazed a path that led to the possibility of an authoritative and generative canon of scripture that yields to a tradition of interpretation, and even, ultimately, to the possibility of a text like the Theodosian Code.

Interpretation and “Patristic Commentary”

Vincent of Lérins wrote for himself two commonitoria: aides-mémoires which lay out in unadorned language the method “how and by what sure (so as to say general and common [quasi generali ac regulari]) rule I might distinguish the truth of Catholic faith from the falsehood of depraved heresy.”⁵ Surveying the field of “men eminent in sanctity and in learning” he came to the conclusion that he could detect heresy and remain pure in his own faith with reference to two resources: first, the “authority of divine law (divinae legis auctoritate)” and “second, the tradition of the Catholic community (deinde ecclesiae catholicae traditione).” Vincent insists that the latter – orthodox patrimony – is necessary to consult because the former, which he defines as “the canon of scripture,” is dangerously underdetermined.

Here perhaps someone will ask, “Since the canon of Scripture (scripturarum canon) is complete and is in itself sufficient and more for everything, what need is there to join to it the authority of the church’s understanding of it?” You see, all do not understand sacred scripture in one and the same sense on account of its very depth, but each and every person interprets its statements in a different way, such that it seems that as many opinions can be extracted from it as there are people (ut paene quot homines sunt, tot illinc sententiae erui posse videantur).⁶

By the time of Vincent’s late life literary floruit in the 430s ce, scriptural interpretation had long since faded as a central and sufficient locus for the production of theological truth. In Vincent’s estimation, as many interpretations could be extracted from scripture as there were people to perform the task, and if one’s aim was to produce authoritative knowledge it was necessary to adjoin patrimony of the Catholic tradition to any textual argument. Vincent’s contemporary Augustine, too, considered the strength of intellectual patrimony to lie precisely in its ability to account for, and exploit, the indeterminacy of scripture.⁷

A generation before Augustine, Hilary of Poitiers understood the indeterminacy of scripture as the reason that Christian theologians of his generation fundamentally changed their method. It was precisely the failure of scriptural interpretation and the proliferation of heresies in the years following the Council of Nicaea, that Hilary saw as the central, motivating factor underlying the remarkably new manner of scholastic argumentation among Christians on the eve of the Theodosian Age. He writes in his treatise Concerning the Synods that in generations past “you didn’t long for a scribe to write what you believed in your heart and professed unto salvation. As bishop you didn’t need to read the things that you held as new converts.”⁸ According to Hilary, an age of interpretative flexibility had not arisen before the upheavals of the 320s. As such, there was no need for the collection of interpretations and their distillation into creeds. He continues: “Necessity, however, introduced the custom of defining the faith and of signing on to the definition (exponi fides, et expositis subscribi)” (63). Textual interpretation failed to settle debates about the relationship between the Christian Father and his Son, and creeds were introduced to perform the task that scripture was incapable of performing.

Hilary saw the production of truth as a project involving two primary operations: first aggregation of a patrimony and then distillation of a universal statement of truth. And he understood that this new form of argumentation arose because of new concerns following the Council of Nicaea:

You perceive that the truth has been sought by many paths through the advice and opinions of different bishops, and the ground of their views has been set forth by the separate declarations inscribed in this creed. Every separate point of heretical assertion has been successfully refuted. The infinite and boundless God cannot be made comprehensible by a few words of human speech. Brevity often misleads both learner and teacher, and a concentrated discourse either causes a subject not to be understood or spoils the meaning of an argument where a thing is hinted at, and is not proved by full demonstration. The bishops fully understood this, and therefore have used for the purpose of teaching many definitions and a profusion of words, in order that the ordinary understanding might find no difficulty, but that their hearers might be saturated with the truth thus differently expressed, and that in treating of divine things these adequate and manifold definitions might leave no room for danger or obscurity.

(62)

Hilary continues admonishing his reader: “You must not be surprised, dear brethren, that so many creeds have recently been written. The frenzy of heretics makes it necessary” (63). In Hilary’s estimation, this structure of knowledge – aggregation of material followed by distillation into a creed or universal statement – is the baseline operation of any fight against heresy. As we saw earlier in the works of Vincent and Augustine, and as we will see later in the works of Ambrose and Jerome, this change was predicated on the indeterminacy of scripture exposed by the Nicene controversy, resulting in the need to join the patrimony of the Catholic tradition to the results of scriptural interpretation.

Late ancient Christian scholars knew about the shift in Christian scholastic methodology described in Chapter 3 and they actively reflected upon it in the later fourth century and the beginning of the fifth. Hilary wrote his Concerning the Synods from the Latin West at almost precisely the same time that Athanasius wrote Concerning the Decrees from the Greek East. Each offers a genealogy for the shift to the “code” format that begins with the proliferation of false interpretations – readings predicated on the interpretative art itself without recourse to the history of scholarship. In the work of Hilary and Athanasius alike we see a coherent statement of the method that would come to dominate nearly every piece of scholarship in the Theodosian Age. Both argue that truth can be found only by compiling the great diversity of opinions and distilling from that collection a universal statement which supersedes and governs the subsequent interpretation of its sources. We see, in other words, the invention of aggregation as a central scholarly tool. I turn now to the development and deployment of this tool.

Christian Aggregation

Among Christian scholars from the 350s through the Theodosian Age, aggregation was more than a method: it was an epistemic operation. What I mean by this is that aggregation was not simply one method out of many by which an argument about universal truth could be made. Rather, aggregation was the necessary precursor to any such knowledge. It was the only way to produce universal truth reliably.

For example, it is theoretically possible to produce a final statement of universal truth in theology, or in law, simply by fiat. If it were considered to be a reasonable method of accessing truth, then a final, unimpeachable statement could be handed down without supporting documentation from an accepted authority, like an emperor or a metropolitan bishop. It is simply the case, however, that no one in the Theodosian Age did so. Rather, among Christians and eventually in “secular” domains, statements of universal truth were predicated on a collation of sources and on the aggregation of previous opinions about the subject at hand.

Jerome, writing in 415 ce, expresses an expectation that truth is predicated on aggregation in the strongest of terms – calling it among the “laws of commentary writing.”

Yet, while snoring in extreme dementia, he [Pelagius] failed to understand the laws of commentary writing (leges commentariorum), in which the divergent opinions of many people are cited (multae diversorum ponuntur opiniones) – sometimes leaving out their names and sometimes just mentioning them – so that it is left up to the judgement of the reader to decide which interpretation ought to be chosen as best.⁹

We likely will never know what comprises the full content of Jerome’s proposed leges commentariorum. He does gloss the term briefly in Against Rufinus 1.16, but never reflects on these “laws” in an extended manner. It is clear that he does not have in mind norms like those laid down by the contemporary theologian Tychonius, which have to do with the process of exegesis. Rather, the “laws of commentary writing” define scriptural commentary’s proper textualized form. Jerome here is concerned with the structure of a proper commentary. His chief expectation is that many diverse opinions are offered in addition to that of the author so that the discursive and commentarial tradition which forms the basis for authoritative statements of Orthodoxy can be investigated.

Jerome restated his expectation that proper argumentation cites many diverse opinions in his commentary on Isaiah. Writing in 410 ce, he claims:

In discussing these, I have briefly summarized the discourses of Africanus the chronologist, Origen, Eusebius of Caesarea, and also Clement, a priest of the Alexandrian presbyter, and Apollinarius the Laodicean. Likewise those of Hippolytus, the Hebrews, and Tertullian. I left it to the reader to choose what to select from the many views presented … In any event, if I have called the men mentioned above “teachers of the church,” they should understand: I do not approve the faith of them all (me non omnium probare fidem).¹⁰

Here, Jerome voices an expectation that aggregation is the proper method for scriptural commentary, but he also shows a concern for the eventualities brought about by this method of argumentation. While he clearly expects that scholastically valid commentaries aggregate previous opinions and display them for the benefit of the reader, he is aware that writing a commentary in the form of an aggregative code involves promulgating false opinions that are not approved or endorsed by the author of the collection.¹¹ I return to this and other concerns stemming from the centrality of aggregation in the Theodosian Age in Chapters 6 and 8. For now I simply note that fourth- and fifth-century commentators were cognizant of issues of discernment related to their chosen form of argumentation, where it may not be immediately apparent to a reader which of the many opinions presented was endorsed by the author of a work.

For Jerome, the function of a commentary is to collate interpretations and present them so that the reader, “like a good banker, can reject the money of spurious mintage” – opinions supported by poor evidence or insufficient reason.¹² As Jerome states, and as I argue later, by the time that Jerome wrote his Apology against the Books of Rufinus in the early fifth century, this was a common expectation of scholastic work even among Traditionalists and scholars working in “secular” domains.

What I have done in that and other commentaries is to develop both my own opinion and that of others, stating clearly which are Catholic and which heretical. This is the custom of commentators and the rule of exegetes (Hic est enim commentariorum mos et explanantium regula): they give at length in their exposition the various opinions and explain what is thought by themselves and by others. And this procedure is adopted not only by those who expound the holy Scriptures, but also by those who explain secular literature (sed saecularium quoque litterarum explanatores), whether in Greek or in Latin.

(3.11)

This method, which “is the custom and the rule of exegetes,” is schoolboy stuff, according to Jerome – the classical rhetorical education through which Jerome, Rufinus, and their scholarly peers were all trained necessarily included reading “Asper’s commentaries on Vergil and Sallust, those of Vulcatius on Cicero’s Orations, of Victorinus upon his Dialogues and the Comedies of Terence,” etc. (3.11). The inclusion of a variety of opinions has a long and august history, according to Jerome, even if the underlying rationale for doing so had shifted fundamentally in the preceding generation.

Jerome’s long-form attack on Rufinus was occasioned by the latter’s edition of Origen’s On First Principles, which Rufinus claims to have purged from heretical interpolations in order to return Origen’s text to its original, Orthodox state. Continuing from the previous quotation, Jerome clarifies that Rufinus’s error was not that he has included heretical opinions in his edition of Origen’s On First Principles: including such material would be perfectly in line with the task of a commentator, or that of a translator. Rather, Rufinus’s sin was cutting the heretical bits out! According to Jerome, what remains in the work, “whether good or bad, must be held to be part of the work – not of the author whom you are translating, but of you who has made the translation” (3.11). Rufinus opened himself up to the charge of heresy by failing to stick to commentarial practices accepted by Theodosian Christians and Traditionalists alike. Methodological aberration is not just gauche; it is spiritually risky.

Writing in the early 440s, Socrates “the Scholar (scholasticus)” considered aggregation to be a scholarly tool useful and necessary precisely for “searching out the truth (τὴν ἀλήθειαν ἀνιχνεύσαμεν).”¹³ He produced two editions of his Ecclesiastical History. The first was intended for a general readership. It picks up on Eusebius of Caesarea’s own Ecclesiastical History and emends errors introduced in Rufinus’s Latin translation, offering “the unadorned facts (γυμνὰ τὰ πράγματα) in order that the history might not become verbose, and weary the readers with tedious matters of detail” (2.1.5). The second edition, however, had a higher aim and a nobler audience: it was a work intended for a scholar. As such, Socrates claims that this second version used current scholarly methods, by which he meant that it was a work of aggregation: it brought together a variety of sources without alteration, because this was the only way that Socrates thought the work might be useful for another scholar.

But in the present edition such alterations and additions have been made for your sake Theodore, sacred man of God, in order that you might not be ignorant of what the emperors wrote in their own words, or of the bishop’s synodal pronouncements, where they continually refined the faith. For this reason we have inserted in this secondary compilation whatever we deemed necessary.

(2.1.6)

Socrates worked in the budding discipline of Ecclesiastical History. He had a few examples upon which to base his own contribution to the genre, but there was no fully articulated methodology of the discipline, especially where it differed from “history” as classically conceived and practiced. Eusebius self-consciously invented the genre in which Socrates worked, and claimed to bring together bits and pieces quite literally as a συγγραφεύς – one who collects facts into a narrative. Eusebius’s introduction offers that “we shall attempt through historical narration to create a body (δι᾿ ὑφηγήσεως ἱστορικῆς πειρασόμεθα σωματοποιῆσαι)” from such scattered sources as he is able to lay eyes upon.¹⁴ Now, Eusebius’s aims and his results are distinct: his work is not a coherent tapestry, in fact, but rather a messy patchwork, often presenting archival sources stitched into a narrative frame in a way that struck his readers as original and methodologically savvy.¹⁵ But his stated aim was “to create a body” from such distinct sources. As with the work of other scholars engaged in this book, Eusebius’s and Socrates’s methods, at times, appear to be identical, and in more than a few instances their results significantly align. But in this case and others, scholars working in the same discipline performed similar tasks for different reasons, with distinct aims calibrated to the intellectual culture of their generation.

Socrates claims to have produced a first edition for the masses, one that has clear methodological resonances with Eusebius’s History and following similar aims. His “second edition,” on the other hand, is no such work. Rather, this new version of the Ecclesiastical History is steeped in the scholastic trends of the Theodosian Age, involving methods and aims foreign to the early fourth century but right at home in the fifth. Furthermore, not only does Socrates diverge from Eusebius’s method in favor of a Theodosian mode of aggregation, he also faults his own sources for failing to do the same. He castigates Sabinus of Heraclea for putting together a dossier of conciliar material that is both impudent¹⁶ and – worse yet – incomplete, because it failed to bring together both heretical and orthodox material.¹⁷ According to Socrates, Sabinus’s book was scholastically useless because it was methodologically flawed: it did not aggregate.

Sozomen wrote his own Ecclesiastical History in the 440s, dedicating it to Theodosius II. He, too, echoes the notion that proper knowledge production is fundamentally based in aggregation. He had intended to “trace the course of events from the beginning” – meaning from the time of Jesus’s life – but upon reflection that such luminaries as Clement, Hegesippus, Africanus, and Eusebius had already treated such matters exhaustively, he decided rather to offer only an abridged version and to focus on events beginning with the reign of Constantine.¹⁸ Sozomen’s work integrates much of Socrates’s account and adds to it a host of documents and oral sources, especially relating to ecclesiastical affairs of the mid-fifth century, in many of which Sozomen had been personally involved. Sozomen’s approach to the production of an authoritative history wrestles both with the need to aggregate archival material in its original form and with the exigencies of the method itself; namely, if he were to bring together all the material which he surveyed, as was his original intent, the work would be too cumbersome to be useful. His response to this problem builds on the method that he learned from Eusebius, but the direction that Socrates took shows that he was intimately familiar with what was expected from him as a Theodosian scholar.

I will record the events at which I happen to have been present, and concerning those which happened in our day or before our generation I learned from those having known or seen the events. Of earlier events I have sought for records among the archived laws appertaining to worship, among the records of the synods of the period, among the innovations that arose, and in the letters of emperors and clerics, of which some have been saved in imperial residences and in churches, while others are scattered, and in the possession of scholars. I considered often transcribing the whole of the texts, but on further reflection I deemed it better, on account of the cumbersomeness of the task (διὰ τὸν ὄγκον τῆς πραγματείας), to offer a synopsis of their rationale (τὴν ἐν αὐτοῖς διάνοιαν συντόμως ἀπαγγεῖλαι). However, whenever disputed issues are introduced, I will readily transcribe freely from any document that may tend toward demonstration of the truth (παραθήσομαι ταύτην εἰς ἀπόδειξιν τῆς ἀληθείας).

(1.1.13–14.)

Here we see Sozomen’s extraordinary attention to documents, his intention to offer as many and as wide-ranging views as possible, and his conviction that on disputed topics, the range of documents should be allowed to speak in their fullness. It is clear that he has in mind the particular necessity to aggregate both orthodox and heretical opinions in disputed areas, because he moves on directly to criticize the failure of partisans – lesser historians – to do just that. Heretics fail to employ proper methods of knowledge production.

In order to demonstrate the correctness of their own theological ideas, both those inclined to this side and to the other side created a dossier of such letters as favored their own heresy, omitting the ones contrary (οἱ δὲ ἐκείνοις προστιθέμενοι συναγωγὴν ἐποιήσαντο τῶν ὑπὲρ τῆς οἰκείας αἱρέσεως φερομένων ἐπιστολῶν καὶ τὰς ἐναντίας παρέλιπον) … As it is requisite to pay strictest attention to the means of eliciting truth in order to maintain historical accuracy (τὸ τῆς ἱστορίας ἀκίβδηλον), it seemed to me necessary to look extensively into any such documents of this type, according to my ability.

(1.15–16)

It is of course inevitable that such dossiers will skew to one ideological pole or another, and Sozomen’s own “collection” has a thoroughgoing partisan agenda. At stake for my argument is not whether Sozomen, Socrates, or anyone else successfully lived up to their ideals; this is a history of their ideals themselves. The acceptable form of scholastic truth production, for Socrates, Sozomen, and others in the Theodosian Age, was based in impartial aggregation.

The Proceedings of the Council of Ephesus (431)

Like Athanasius’s Concerning the Decrees and Hilary’s Concerning the Synods some seventy years earlier, the Proceedings of the Council of Ephesus (431) were constructed to make a point. The acta from July 22, 431 appear to supplement those from a month previous and, like Concerning the Decrees, constitute a dossier intended to establish a set of criteria: technical calques – not on scripture this time, but on the Nicene Creed itself – in order to secure the condemnation of Nestorius, the bishop of Constantinople. The proceedings begin with a full presentation of the Creed of Nicaea before turning to the impetus for the meeting in the first place:

However, because some pretend to profess accordingly and to agree [with the Nicene Creed], but in fact misinterpret the force of the ideas according to their own opinion and distort the truth (being sons of error and children of depravity), it has become absolutely necessary to introduce passages from the holy and orthodox Fathers that can give assurance in what way they understood [the Nicene Creed] and had the confidence to preach it, in order that, clearly, all who hold the correct and irreproachable belief may also understand and interpret and proclaim it in that way.¹⁹

The First Council of Ephesus was held in 431, and its main task was to deal with the problem of unauthorized interpretation of the Nicene Creed. Even in the run-up to the council it was clear that heretical readings of the Nicene Creed could not be defended on the basis of the text of the creed alone. Cyril’s Third Letter to Nestorius, written in late 430 immediately after a papal judgment against the latter and laying out the terms of his reconciliation, urges Nestorius to follow the “royal road” of patristic interpretation of the Nicene Creed, because his exposition of the text itself was heretically faulty. As Mark S. Smith put it: “A ‘bare’ confession of Nicaea, Cyril contended, was no longer sufficient for the authentic articulation of Nicene orthodoxy.”²⁰ In the proceedings from June 22, Cyril went so far as to suggest that his own Second Letter should be “clearly established as the authoritative and necessary lens through which the Nicene Creed must be read, and the Creed itself rather drops out of view.”²¹ The Nicene Creed itself was intended as a distillate of scripture, as a guide which dealt with the problem of scripture’s underdetermined nature. After 100 years, however, the creed had become so central to theological disputation that a new council was convened to deal with its own underdetermined nature. “The simulacrum is true.”²²

Over the course of the fourth century, and into the fifth, we see a movement from primarily scriptural citation as central to the production of truth to a form of argumentation based in the aggregation of various, sometimes competing interpretive voices. This movement has been described variously as “patristic citation”²³ or “patristic retractation.”²⁴ The change defines the fundamental shift in Orthodox theological scholarship after the Council of Nicaea, and especially in the Theodosian empire. But the move to “patristic retractation” is not merely a modern scholarly interpretation: late ancient scholars also noticed that book forms, and forms of argumentation, underwent a revolution in the years after the Council of Nicaea. Returning to Vincent’s Commonitorium with which this chapter started, we find a Theodosian scholar struggling with the change in citational forms and the relationship between a new, Theodosian scholastic methodology producing new theological truths and an ancient, and (notionally) unchanging message.

But perhaps someone will say, “Is there to be no progress in the religion of the Christian church?” There is, clearly – and substantial! For who is there who is so envious of humans, and so hateful of God, that he would try to forbid it?²⁵ But such progress must occur in such a way that it is truly progress in faith, not change! (Sed ita tamen ut vere profectus sit ille fidei, non permutatio) … For there is a great difference between the flower of youth and the maturity of old age, and yet when old they remain the very same people as they were when young, in this sense, that although the stature and carriage of individuals change, nevertheless each person’s nature is one and the same in every respect, likewise his or her character … And so it befits doctrine of the Christian religion to follow the same laws of progress (ita etiam Christianae religionis dogma sequatur has decet profectuum leges).²⁶

Here Vincent impresses upon his reader that the methods of scholarly disputation may change as theologians polish the interpretive lens to reveal new, deeper truths, but that the underlying message remains constant and universal. Hilary, too, thought that the Nicene Creed was timelessly true but that it would require ongoing support and new arguments. Writing in 359 from the East, and addressing primarily Western bishops, he offered to send an account of all the creeds between Nicaea and Sirmium in hopes that he would have their support at “councils to come (futuri synodi),” the need for which was inevitable.²⁷

Throughout the late fourth and fifth centuries, Christian scholars were preoccupied with negotiating new book forms and new styles of argumentation, and attempting to square these radical changes with an ancient tradition of interpretation that is supposed to undergird invariable, universal truths. When Christian scholars thought back on the changing forms of knowledge production in their generation and those before – from Hilary in the 350s to Vincent in the 430s – they saw the indeterminacy of Scripture as a motivating factor for the change. While the change was new to Hilary, and required both a genealogy and explanation, for Vincent the structure of knowledge in which aggregation is pivotal to the production of trustworthy knowledge was already an ingrained facet of his intellectual environment. Hilary, along with Athanasius, was an innovator in the movement to patristic rather than scriptural citation, and to aggregation as a fundamental tool. By the 430s, however, this ideology of knowledge production could be read from just about any randomly selected product of scholarship. This is to say that a particular facet of Christian book culture in the mid-fourth century came to be a generalized facet of elite book culture by the early fifth century. Or, put differently: when Christians came into a ruling elite for the first time, Christian book culture became Roman book culture.

Ambrose considered himself a poor man’s Cicero. While reading the gospel, the holy spirit confirmed to Ambrose that speaking of “duties” should not be the sole purview of philosophers (1.25). His work On Duties (De officiis) comprises the patrimony of Traditionalist and Christian learning placed within a Ciceronian literary frame, in order to offer a new scholastic production capable of superseding the classical treatise. Philosophers, he explains, devised a method of deciding between two things – those which are honorable and those which are beneficial – but Christians should not countenance an action unless it is satisfies both criteria.²⁸ Ambrose reveals the reason for his digression only thereafter: his discourse deals explicitly with the manner in which philosophers weigh the moral virtue of actions, but in truth he speaks proleptically about a proper Christian relationship with the classical tradition of which philosophers are a part – and, importantly, the value of Ambrose’s own book On Duties standing in the gap. “From now on, those who choose not to read the works of these people will be able to read ours if they so wish – those who are looking not for ornate language or verbal artistry but for the simple grace of things as they really are” (1.29). Here Ambrose argues that his and similar works by Christians replace the classical tradition not by expunging them, but by incorporating and distilling them to present a product that is fit for Christian use, placed alongside Christian materials of superseding value, if lesser artistry. This is not a repudiation of the classical tradition, but neither is it an embrace.²⁹ Rather, Ambrose argues that his and similar works collect the best of the tradition, doing the work of scholarly aggregation that is expected of him and producing a manual of practice so that readers have no need to consult those old books. He does not seem to think that this point needs belaboring, however – his methodological statement is dropped into the narrative of book one, after which he returns to a discussion of Panaetius, Aristotle, Pythagoras, and their relationship with the teachings of the biblical king David. On Duties presents an early example, from the late 380s, of a framework for scholastic production that became quite common in the Theodosian Age, which we see present from Macrobius to Servius to Martianus Capella, as detailed later.³⁰ All of these authors purposefully invoke a classical style, topic, or even particular work as a container for the aggregation of a patrimony, and for the creation of a resource of superseding – and universal – value.

Aggregation beyond Theology

Christian theologians were not the only scholars who saw aggregation as central to accurate knowledge of the world. In fact, we see the extraordinary interimplication of scholastic domains in the Theodosian Age by looking at the way that aggregation underlies scholastic productions across the ideological spectrum. A scholarly method that gave aggregation pride of place is visible everywhere from legal compilations to miscellanies during the Theodosian Age.

I want to be perfectly clear: I am not claiming that any of these methods are fundamentally new. Rather, my claim is that aggregation, distillation, and promulgation took a central position in scholarship during the Theodosian Age, and the centrality of that position is novel. These methods became the scholastic lingua franca – the most available and widely used tools for answering questions of knowledge production and governance. This does not mean that everyone, everywhere, in all domains used them exclusively. Of course, there are polemicists from the period who did not use the methods, and some Christian disputation did not involve these methods, either. I do not argue that there are no detractors from the method: later I detail a number of them. Nor is my argument that the theory presented here can neatly explain every bit of Theodosian scholastic production. What I describe here is a trend, visible across the ideological spectrum and in different scholastic domains, that appeared in a wide variety of sources at about the same time. My explanation for this trend is that a set of scholarly practices contrived for theological disputation became generalized and central during the late fourth through the mid-fifth century as a result of Nicene Christian dominance. Other explanations may be possible, and I encourage other proposals be made. The sheer magnitude of the change in scholarly method across disciplines in this moment demands explanation. This book offers one.

I begin my discussion of the particularly Theodosian nature of aggregation outside of theological scholarship with Macrobius’s Saturnalia. It is reasonably certain that Macrobius himself was a Christian, or at least that he was not an outspoken Traditionalist in the vein of his text’s principle characters: Praetextatus, Symmachus, and Flavianus.³¹ Nevertheless, he worked in the “secular” domain of Miscellany, a genre with an august history, and was keenly interested in what Alan Cameron has called “pre-Crisis paganism … more the paganism of Vergil than the paganism of Symmachus and Praetextatus.”³² In this sense he presents an interesting case of an apparent Christian explicitly using a method developed for Christian theological disputation in a nontheological scholarly domain, and with the aim of exploring themes of a long-lost Traditionalist past.

Macrobius composed his work of staggering antiquarian learning sometime in the 430s, reflecting a dramatic setting in the 380s.³³ But his text does not stand alone: its exemplar, and in many instances its direct source, is Aulus Gellius’s Attic Nights. Gellius pioneered the form of Roman miscellany during the Antonine dynasty, and some 200 years later Macrobius picked up his rhetoric and even the specific wording of Gellius’s preface. Opening to a random page in either Attic Nights or Saturnalia reveals remarkably similar material in a similar form: extracts, culled from a wide variety of sources, placed together under (what intend to be) useful headings. But, as I argue later, the reason that each scholar took up his task could not be more precisely at odds. Leontius of Jerusalem is not the only scholar to envision the search for truth as an apiary endeavor. Macrobius, too, introduced his work with an exhortation: “We ought to imitate bees, if I can put it that way: wandering about, sampling the flowers, they arrange whatever they’ve gathered, distributing it among the honeycomb’s cells, and by blending in the peculiar quality of their own spirit they transform the diverse kinds of nectar into a single taste.”³⁴

Saturnalia is presented within a consciously literary frame as dialogue between three great Traditionalist thinkers of the late fourth century – contemporaries of the likes of Ambrose and Jerome, and men of great wealth and imperial rank. The content of the book, however, is a series of extracts, things “initially noted down in a jumble” that were collated in order under headings useful for a reader in order that they “might come together in a coherent, organic whole” (Pref.3). Macrobius’s aim was to take the raw material of previous scholarship and collate it into an authoritative whole for his son. It is worth quoting him here at length; his justification for undertaking the project at hand should sound strikingly familiar:

We should draw upon all our sources with the aim of making a unity (ex omnibus colligamus unde unum fiat), just as one number results from a sum of individual numbers. Let this be the mind’s goal: to conceal its sources of support and to display only what it has made of them (omnia quibus est adiutus abscondat, ipsum tantum ostendat quod effecit) … You know how a chorus consists of many people’s voices, and yet they all produce a single sound. One voice is high-pitched, another low, another in the middle, men are joined by women, a pipe is added to the mix: individual voices disappear while the voices of all are revealed, and the disparate tones produce a harmony. That is my goal for the present work: it comprises many different disciplines, many lessons, examples drawn from many periods, but brought together into a harmonious whole (in unum conspirata).

(Pref. 8–10)

Macrobius uses poetic and playful language to express the same sentiment over and again in rapid succession, impressing on his reader that the ultimate aim of his seven books of miscellany was to allow for a single, universal truth to proceed from the raw, aggregated material of an antiquarian’s selection. The metaphor of diverse instruments which produce a single sound is not Macrobius’s own: it appears throughout classical literature, from Pseudo-Aristotle’s On the Cosmos (396b), to Plutarch’s Moralia (96e), and even in Philo’s Life of Moses (2.256–257), as noted by Robert Kaster.³⁵ Even in his choice of learned metaphors Macrobius innovates within tradition, as none of his forebears use the analogy in the same way: Pseudo-Aristotle invokes it to explain the consonance of the universe, even though it includes materials of different types, Plutarch speaks on the nature of friendship, and Philo the nature of worship. Macrobius uses a traditional metaphor to explain the nature of his own work, and what his son, as an idealized reader, is intended to hear amid the cacophony. All of Macrobius’s extracts combine to express “a harmony” and “a single flavor (unius saporis)” with a single underlying truth.³⁶

Nevertheless, Macrobius was no Heroditean. He hoped that as a result of his aggregative method his reader could find, eventually, the unius saporis of truth which underlies them all, but both in his explicit methodological statement and in source critical analysis Macrobius comes across as a rather faithful copyist of his sources. As argued at length by Alan Cameron, “Macrobius himself never lays claim to any originality, and where we are in a position to check, he did indeed follow single sources closely for long stretches. Notoriously, he follows Gellius so closely that in many places the Macrobian text has the authority of a manuscript of Gellius.”³⁷ The question has been asked before, whether speeches in Macrobius’s narrative reflect the opinions of the historical character who speaks in the text, that of Macrobius’s source, or Macrobius’s own personal view.³⁸ The debate would be rendered less fraught if the broader intellectual culture of the Theodosian Age were taken into account, including both theological and secular works. Cameron wouldn’t have had to go to such great lengths to show, rightly, that Macrobius did not necessarily share the opinion of his sources.³⁹ When placed next to Jerome, Ambrose, Hilary, or even the Theodosian Code, as argued later, it becomes apparent that the contemporary scholastic norm was to include material with which you disagreed, or to which you did not at least wholly assent – not ‘the norm’ as in normal, but positively normative.

Comparison with Aulus Gellius’s Attic Nights magnifies Macrobius’s novel aims in assuming Gellius’s format and rewriting his preface, and clarifies the conceptual overlap between Saturnalia and other Theodosian scholastic productions. Like Macrobius, Gellius wrote a preface explaining the form and goals of his work. While it is clear that Macrobius knew Attic Nights and patterned his Saturnalia on it, each author’s rationale for aggregation could not contrast more starkly. Unlike Macrobius’s miscellany, which aims to access ultimate truth through carefully chosen excerpts from past authorities, Gellius claims that his books are intended “not so much to instruct as to give a hint, and that content with my, so to speak, pointing out of the path, they may afterwards follow up those subjects, if they so desire, with the aid either of books or of teachers.”⁴⁰ As Joseph Howley argues, “juxtaposition and open-endedness [are] typical of the [Attic Nights]’s functioning as a book not of answers, but of questions; though founded on scholarly research, its literary and interactive mode is not encyclopedic but protreptic, often demanding its reader finish the work it has begun.”⁴¹

Macrobius used Attic Nights as a source, and he goes to great lengths to mimic Gellius’s preface.⁴² But Macrobius repudiates the other aims of Gellius’s work – including, most importantly, his rationale for undertaking such a production in the first place. Both authors produced a miscellany by compiling diverse material from a variety of sources, but only the Theodosian scholar considered the format to be conducive to an aim beyond the virtues of miscellany itself. Macrobius’s aim is truth, made visible through the scholarly process of aggregation, distillation, and systematic presentation, and he rewrote Gellius’s preface to stress these aims. On the other hand, Gellius warns his reader, “I have not made an excessively deep and obscure investigation of the intricacies of these questions, but I have presented the first fruits, so to say, and a kind of foretaste of the liberal arts.”⁴³

Macrobius’s contemporary Martianus Capella wrote his nine books, The Marriage of Philology and Mercury, in the early fifth century, and quickly gained notoriety as an encyclopedist.⁴⁴ He was notable as a Neoplatonist and his book is, in the estimation of Alan Cameron, “a treasure house of pagan lore” placed within the literary frame of a Greco-Roman novel, in the vein of works by Petronius or Apuleius.⁴⁵ Martianus’s is an aggregative work of diverse learning placed within the narrative frame of a wedding feast. His first two books concern attraction (broadly conceived) and detail the narrative by which Apollo played matchmaker between Mercury and Philology, the latter of whom was adopted as one of the gods. During the nuptial celebration Apollo brings forward each of the seven liberal arts, and Martianus devotes a book to each: Grammar, Dialectic, Rhetoric, Geometry, Arithmetic, Astronomy, and Harmony. By way of conclusion, Martianus calls his nine books a miscillo, and claims the true author, Satire, “has intermixed (immiscuit)” materials of all sorts. “Our garralous Satire has heaped learned doctrines upon unlearned, and crammed sacred matters into secular.”⁴⁶

In the words of Beatrice Bakhouche, the work “borrows from every previous literary genre. The grandiose meets the comic.”⁴⁷ It is indeed true that Martianus’s novel mixes tales together, but the parallels that Bakhouche suggests in attempting to understand the Martianus’s generic frame – Petronius and Apuleius, above all – do not include the seven books of technical learning explaining, inter alia, which words in Latin have an A as an ending in the nominative or the number of miles between the Arsia and Drina rivers, to choose two examples (quite literally) at random. Martianus’s work is fundamentally aggregative, in distinction from earlier examples of the novelistic genre. If Martianus had worked solely within a traditional genre, is hard to imagine that such an apology as he offers at the end of his nine books would be necessary.

There is another difference between the generic features of The Marriage and the apparent aim of other novels, such as those of Petronius and Apuleius which have most often been used to contextualize the work: Martianus presents his own work as more than either a satire or an encyclopedia – and, as Jason König and Greg Woolf rightly note, “[e]ncyclopaedism was never a genre within classical antiquity” to begin with.⁴⁸ Martianus’s work is a true “encyclopedia,” invoking explicitly the enkuklios paideia (grammar, dialectic, rhetoric, geometry, arithmetic, astronomy, and music), placing him in an intellectual tradition boasting the likes of Cato, Varro, and Celsus,⁴⁹ and suggesting that his aim, again like those of his forebears and similar to Macrobius’s production, was something like universal knowledge – an aim that can be squarely situated in the intellectual culture of the Theodosian moment, but which is found nowhere in Martianus’s preceding stylistic exemplars. And, like other writers of the Theodosian Age, he writes an aggregative work that is precisely not a manual of practice. Like his contemporary Macrobius and Gellius before, Martianus wrote a book of aggregative learning for the benefit of his son, who stands in for a general reader. His fatherly message, again in the words of Bakhouche, is this: “The liberal arts make sense only if they allow us to account for the world, to render the cosmos intelligible.”⁵⁰ Exploration of the seven subjects in this aggregative format is the first step toward universal and ultimately divine knowledge.⁵¹ Here again we see a Theodosian author taking up the framework of an august genre and reshaping it, inserting heaps of antiquarian learning and, ultimately, appropriating an old format with the aim of universality and knowledge of the divine. In this sense, the relationship between Martianus and the work of Apuleius and Petronius is similar to the relationship between Macrobius and his exemplar in Aulus Gellius: both attended to traditional topoi of the genre in which they work, while adding new features rooted in encyclopedic learning and explicitly claiming to create a resource capable of leading the discerning reader to universal truth. Further, both Macrobius and Martianus offer a rationale for their innovation within a literary tradition – apologies for a perhaps startling format which diverged from classical examples of the genre, but which hewed instead to the methodological trend of their day.

Aggregation, distillation, and systematic presentation are the central aim of another great scholastic production by Macrobius and Martianus’s contemporaries. The compilers of the Theodosian Code, however, had even loftier aims. In 429 ce, emperors Theodosius II and Valentinian III tasked seven men of high imperial rank to aggregate, distill, and systematically present the tradition of legal pronouncements and the tradition of legal scholarship from the time of Constantine’s conversion through their present day.⁵² Unlike Macrobius, their aim was not just truth, nor did they intend merely to create a resource for jurists strewn across an empire that stretched some 3,000 kilometers, though this was certainly a feature of the finished product. The imperial constitution calling for the creation of the Theodosian Code makes clear that the initial intention of this collocation and distillation of the patrimony of classical law was the production of a magisterium vitae – a comprehensive “guide to life.”

In an Appendix I discuss the peculiarly Christian usage of the term magisterium that frames the aims of the second code as ordered in CTh 1.1.5; I hope that the influence of Christian language and conceptual frameworks on the Theodosian Code is clear. Here I want to focus on the motivation for the Theodosian Code as initially ordered in 429, and on its method. The project was intended to comprise two steps: the creation of a scholarly resource, and the distillation of that resource into a universal statement. The first step was to use the method of aggregation to create a resource for scholars, or in its words “more industrious people (diligentiores).” The second step was to distill that scholarly resource into a “guide to life (magisterium vitae).” The first step was envisioned as follows:

Although it may be simpler and more in accordance with law to omit those constitutions which were invalidated by later constitutions and to set forth only those which must be valid, let us recognize that this code and the previous ones were composed for more industrious people, to whose scholarly efforts (scholasticae intentioni) it is granted to know those laws also which have been consigned to silence and have passed into disuse (illa, quae mandata silentio in desuetudinem abierunt).⁵³

Here the Eastern and Western emperors suggest that they could have ordered the Theodosian Code to include only statutes which had not been superseded by later legislation, but that the aims of the text require a different method. Namely, given that the text is intended as a legal repository for “more industrious people” engaged in scholarly work (scholasticae intentioni), the aim of creating a scholastic resource dictates the aggregation of both valid and invalidated laws. It is perhaps worth dwelling on this fact for just a moment. If the Theodosian Code were meant simply as a manual of legal practice or a resource for working jurists, then there would be little reason to include invalidated laws. As William Turpin put it: “One of the oddest things about the Roman law codes is that their contents could be inconsistent or out of date. This is most obvious in the case of Theodosian Code, which is more or less open about it.”⁵⁴ The technical nature of the document defined expectations regarding its form because it claims to be intended as a resource for scholars. The document itself was novel – never before had a Roman law code been created as a universally valid statement of legal praxis, as I discuss in the Appendix – and the method prescribed, by law, to create this novel resource was the same method that contemporary scholars used in other disciplines. Scholars across the disciplinary spectrum expected that a technical resource or scholastic production should rightly aggregate all relevant sources, their validity notwithstanding. The Theodosian Code aimed not simply to offer an authoritative statement of what law is, though that was certainly part of the project, but also to codify a discursive and commentarial tradition into a clear statement of that tradition’s past; how the tradition got from one place to another, from the old laws “consigned to silence” to the statutes that superseded them.⁵⁵

The second step envisioned by the Theodosian Code project was never undertaken. The same men tasked with creating the scholarly resource were intended to distill that work into a universally valid “guide to life (magisterium vitae).” It is not obvious that a legal codification could possibly serve such a noble goal; it is, at least, a strange choice of genre. Caroline Humfress argues persuasively that “the Codex Theodosianus does not lay down the law; instead it provides its elite, specialist readers with the tools – epistemic and material – to produce their own ‘valid’ legal knowledge as defined by and through the Codex itself.”⁵⁶ As I explore in the Appendix, the term magisterium vitae only makes sense if the term “magisterium” is understood with its peculiarly Christian meaning, as a moral exemplar in the guise of the Christian saints. In light of this I would modify Humfress’s statement only slightly. The Codex as it comes down to us provides specialists with tools to produce their own valid legal knowledge, but the Codex as intended was meant to give all people the tools to produce their own valid knowledge in any domain of life. The idea that aggregation could serve as a method to produce a magisterium vitae would seem utterly foreign to someone like Aulus Gellius, whose aim in aggregation was simply “a kind of foretaste of the liberal arts.”⁵⁷ Much more proximate is the work of Macrobius or Socrates, who saw in aggregation the possibility of universal truth.

The two-step process envisioned by the Theodosian Code is precisely the two-step process that we saw as early as Athanasius’s Concerning the Decrees, and which had become de rigueur throughout the landscape of scholarly production in the eighty years since. The first step, for both Athanasius and the Theodosian Code, is aggregation of previous scholarly material, regardless of its validity or authority. It is this scholastic operation that Macrobius explicitly claims to be performing in his Saturnalia. The second step is distillation of that material into a work of universal truth. In the case of Athanasius this distillate was the Nicene Creed, while the Theodosian Code envisioned a magisterium vitae: after “men … of singular trustworthiness [and] most brilliant genius” had “exclude[d] every contradiction,” the laws were meant to be consulted as a “guide to life” in the same way that Christians were urged to pattern themselves after the lives of the saints (1.1.5). The content and material of each of these projects is unique, but their aims and methods proceed from a coherent set of scholastic practices.

It was only the Theodosian Age when this structure of knowledge, born of theological controversy, first found its way into secular domains. The example of Oribasius helps to make this clear. Oribasius was a contemporary of Athanasius, and a doctor and a medical historian working from the court of Julian, serving as the emperor’s personal physician about two decades before Theodosius I ascended to the purple. One of his tasks was similar to Athanasius’s: bringing together scholastic patrimony within his own discipline of medicine. Unlike Martianus Capella, Macrobius, and the compilers of the Theodosian Code, however, Oribasius chose a method wholly different from that of Athanasius. His Medical Compilations compiles a scholastic patrimony, but it is not aggregative in a full sense. That is, his reason for bringing together patrimony is radically different from that of Theodosian Age scholars.

In the introduction to his Medical Compilations, Oribasius relates that the emperor Julian instructed him to compile a corpus of epitomes from the works of Galen alone, and thereafter to strike out on a more expansive second project. The emperor ordered “that I should search for and collect the principal writings of all the best physicians and everything that pertains to the entire medical profession,”⁵⁸ and he claims to be “zealously determined to carry out this task [of compilation], as far as [he is] able,” because such a dossier would be “extremely useful, when people who are reading it readily discover that which in each case of efficacious for those who are in need” (1.pref.2.). Oribasius created a scholarly tool at the behest of the emperor, bringing together previous sources of scholarship into a dossier meant to help the future practitioner of the medical arts. The language that he used around compilation, and even some of the compiler’s peculiar self-referential phrases, mirror strongly the method of Sozomen detailed earlier, and Oribasius’s method resembles that used by the compilers of another imperially ordered scholarly production: the Theodosian Code.⁵⁹ But Oribasius’s aim in creating a compilation could not diverge more radically. The doctor intended to create a scholarly resource but he had no expectation that a good argument, or worthwhile scholarly knowledge, required bringing valid and invalid sources together. (Such a method would be no use to the invalid, in search of a cure.) He continues: “Thinking it superfluous and altogether absurd to include in the work the same things multiple times, both from the authors of the best treatises and of those who treated the subject without a similar degree of accuracy, I will gather together only the works of the better sources” (1.pref.3). For Oribasius, scholarship is a work of curation more than a work of compilation. His stated intent is to create a resource for other doctors that separates the wheat from the chaff. We see here no evidence that the aggregation expected from Christian theological arguments in the mid-fourth century had found its way into the methodological presumptions of a medical scholar writing from the imperial court of the last “pagan” emperor. As I argued earlier, this separation did not last long.

Traditionalist Rejection

Three texts from the Theodosian Age appear to reject aggregation as a scholarly method: the Res Gestae of Ammianus Marcellinus, the Historia Augusta, and Proclus’s Ten Questions Concerning Providence.⁶⁰ Each author’s discussion of aggregation as a method is fraught, and none offer such an unimpeachable statement of rejection as one would hope. My contention, offered with due reservation, is that each of these authors seems to know of aggregation as a scholarly method, and that each rejects it in their own way. My argument is that we can see the prevalence of aggregation, distillation, and promulgation as a scholastic linga franca through pointed rejections of the methods from both chronological ends of the Theodosian Age. By militating against aggregation, each of these sources underscore the prevalence of a widespread scholarly method requiring that raw material remain part of the final scholarly product.

Ammianus completed his administrative and political history of the Roman empire in the early 390s, initially intending to write only twenty-five books covering the period from the accession of Nerva in 96 ce to the death of Valens in 378, continuing on where Tacitus’s own Histories left off and in a similar style. He chose not to write about the most recent events “partly to avoid the dangers which are often connected with the truth, and partly to escape unreasonable critics of the work which I am composing.”⁶¹ These “unreasonable critics (intempestivos),” in Ammianus’s estimation:

[C]ry out as if wronged if one has failed to mention what an emperor said at table, or left out the reason why the common soldiers were brought before the standards for confinement, or because in an ample account of regions he ought not to have been silent about some insignificant forts; also because the names of all who came together to pay their respects to the city-praetor were not given, and many similar matters, which are not in accordance with the principles of history (praeceptis historiae dissonantia).

(26.1.1)

The historian’s exasperation at what he is being asked to do is palpable. There are two ways to read Ammianus’s concern about criticisms that he is loath to incur by leaving out what he deems to be trivial details.⁶² Ronald Syme and Guy Sabbah each offer a traditional understanding, reading Ammianus’s statement at the beginning of book 26 as nothing more than a rejection of the idea that minor details are anything more than trivialities, while his own aim was to illuminate the character and actions of major players in the imperial orbit.⁶³ Here, Ammianus defends the Tacitean historiographical method’s factual remit, rejecting biography – the writing of “Lives” – which had long been a viable vehicle for the writing of imperial history, invoked most famously by Suetonius and more recently by Eusebius and the author(s) of the Historia Augusta. Ammianus complains that his critics require him to record heaps of details, which “are not in accordance with the principles of history (praeceptis historiae); for [history] is wont to detail the highlights of events (discurrere per negotiorum celsitudines assuetae)” (26.1). A passage in the Historia Augusta’s Life of Opilius Macrinus makes a similar point:

Nonetheless, we shall bring forward what we have discovered in various historical works – and they shall be facts that are worthy to be related (ea quidem quae memoratu digna erunt). For there is no man who has not done something or other every day of his life; it is the business of the biographer, however, to relate only those events that are worth the knowing.⁶⁴

Read together, Ammianus and the Life of Opilius Macrinus simply make statements about the proper writing of history, against their predecessors within the genre of historiography. But such a reading of Ammianus Marcellinus fails to account for the historian’s own explanation for his methodological choices. Suetonius and other historical biographers may be implied in Ammianus’s critique – I am persuaded that Ammianus considered himself to be a continuator of the Tacitean project, and therefore sticks to Tacitean methodology, more or less. But this is not what Ammianus says. Rather, he complains that “unreasonable critics (intempestivos)” in his own day will inevitably accuse him of failing to write a proper history precisely by virtue of sticking to an older tradition of historiography, which focuses on “the highlights of events.” Ammianus’s method is traditional, but his exasperation is timely. The intellectual context which makes sense of his concern, that people would criticize him for failing to include what he deems “trivial details,” is the intellectual context of the Theodosian Age in which aggregation was the most immediately available and widely used tool for scholastic productions, and especially historical accounts. It would be irresponsible to read Ammianus as reacting solely to trends within historiography without attending to the wider intellectual and scholastic climate in which he worked – the same failure that causes scholars of Roman law to trace every innovation in legal ideology to a wholly internal process of juristic evolution. Ammianus lived in a society and interacted with scholars working in other disciplines. When placed in the intellectual climate of the Theodosian Age of which he was but a small part, his reaction against aggregative methodologies appears in a new light. That he complains about an intellectual culture which “cries out as if wronged” when he fails to aggregate is only further evidence that what he responds to is a contemporary scholastic trend. It may be easier to ignore this fact, and one can read Ammianus profitably without placing him beside contemporaneous scholars involved in different fields. But doing so renders his text less rich, and outbursts like the methodological winging at the beginning of book 26 less rational.

An obsession with aggregation, in Ammianus’s estimation, is an affectation of the inscitia vulgari: the ignorance of the masses, who require such trivial details (26.2). On the other hand, “Julius Capitolinus,” the putative author of the Life of Opilius Macrinus collected in the Historia Augusta, has a specific polemical target in mind. “Capitolinus” argues that the form of history writing that Ammianus calls “vulgar” is a method undertaken by a historian named Junius Cordus, who is otherwise unknown outside of this citation in the Historia Augusta:

He openly declared that he would search out the most trivial details (minima), as though, in dealing with a Trajan, a Pius, or a Marcus, it should be known how often he went out walking, when he varied his diet, and when he changed his clothes, whom he advanced in public life and at what time. By searching out all this sort of thing and recording it, he filled his books with gossip, whereas either nothing at all should be said of petty matters or certainly very little, and then only when light can thereby be thrown on character. It is character (mores), of course, that we really want to know, but only to a certain extent, that from this the rest may be inferred.

(Life of Macrinus 1.3–5)

Again, in his preface the author of the Life of Opilius Macrinus invokes a debate familiar to anyone working within the ancient discipline of history. The author seems to suggest that his opponent, Junius Cordus, styled himself as a latter-day Plutarch. In his famous introduction to the Life of Alexander, Plutarch claims precisely to focus on small moments, seemingly trivial details, because he aims to understand the character (ἦθος) of his subjects and “a slight thing like a phrase or a jest often makes a greater revelation of character than battles where thousands fall.”⁶⁵ The issue, for Julius Capitolinus, is that a biographer should include only such trivial details when they shed light on the mores – character – of the subject, as Plutarch did. Cordus’s error, according to Capitolinus, was searching out trivial details in the hope of being able to illuminate the character of his subjects therewith, but failing to do so. Cordus has created a dossier of information that is beside the point.

It is possible that these concerns of Ammianus and the author of the Historia Augusta are simply aesthetic – they’d prefer not to read trivialities or burden their own literary inventories with useless knick-knacks. It is certain that Ammianus’s attack on the ignorant masses and Capitolinus’s attack on Cordus are expressed within the bounds of long-held scholarly discourse about what proper historiography comprises. But the vitriol and the specificity of their polemic suggest a more precise aim in decrying aggregation of material as an acceptable format for historiography. The fact that during the period when each of these texts were written just such a format was in vogue in the circles of elite Roman historiographers suggests that Ammianus and Capitolinus were reacting against a wider culture in which the expectation of aggregation was a central facet of the dominant scholastic method. It was perhaps not lost on outspoken Traditionalists, either, that the trend was embraced widely by the most visible and vitriolic sect of Christians.

Further, the Historia Augusta claims to be a composite work, and does not witness a singular historiographical method. A very different method is evidenced in Life of the Deified Aurelian. The preface to this biography discusses precisely the type of aggregation to which a wide variety of Theodosian Age scholars would happily accede:

“And yet, if I am not mistaken, we possess the written journal of that great man and also his wars recorded in detail in the manner of a history, and these I should like you to procure and set forth in order, adding thereto all that pertains to his life (additis quae ad vitam pertinent). All these things you may learn in your zeal for research from the linen books, for he gave instructions that in these all that he did each day should be written down. I will arrange, moreover, that the Ulpian Library shall provide you with the linen books themselves. It would be my wish that you write a work on Aurelian, representing him, to the best of your ability, just as he really was.”

I have carried out these instructions, my dear Ulpianus, I have procured the Greek books and laid my hands on all that I needed, and from these sources I have gathered together into one little book all that was worthy of mention. I hope that you think kindly of my work, and, if you are not content therewith, to study the Greeks and even to demand the linen books themselves, which the Ulpian Library will furnish you whenever you desire.

(Life of Aurelian 1.6–10)

For Flavius Vopiscus, the putative author of this Life of the Deified Aurelian, the validity of his “little book” was based on its status as a distillation of a great mass of material. The authority of the distillate, furthermore, did not depend on blind acceptance of the author’s account, but it was underwritten by the ongoing availability of the raw archival material. Aggregation is the necessary precursor to valid knowledge. It is telling that Vopiscus felt compelled to stress this point, and that his method was consonant with the dominant scholastic framework of his age.

In his study of prefaces in the Historia Augusta, Daniël den Hengst concluded that “the dominant impression after reading through the [Historia Augusta] from beginning to end is one of bewildering variety.”⁶⁶ Not only does the text comprise an eclectic mix of styles and details, but it claims to be the work of six authors, each with his own methodology. Within the Historia Augusta as a whole we find some historians, like Vopiscus, who claim a quintessentially Theodosian form of aggregation as central to their work, placed side by side with the work of others such as Capitolinus, who reject such an operation out of hand – apparently with specific polemical targets in view. Whether the Historia Augusta is the work of one author or six, with one underlying generative framework or many, one thing is clear: the effect of the compilation as a literary product is comfortably at home in Theodosian book culture, with all its variety of voices, opinions, aggregation of the admirable with the censurable, compilation of documents and archival material, etc. What we see in the sum total of the Historia Augusta as a literary product is a form of historiography that is explicitly disclaimed at some points within the text itself and embraced elsewhere. We see the type of historiography that Ammianus despised as “vulgar,” and rejected as beneath the dignity of his project. In these literary products of the 390s we see reflected exactly an elite discussion about, and perhaps embrace of, the scholastic method of aggregation and distillation that is positively endorsed by the likes of Hilary, Jerome, and the compilers of the Theodosian Code. The Historia Augusta itself is aggregative. It is interesting that we see this clearly in a text that is, in the words of Arnaldo Momigliano, “a first-class document of the reformed paganism of the fourth century.”⁶⁷ It is interesting, in other words, that explicit rejection of aggregation as a valid scholarly method is most often found in the writing of Roman Traditionalists.

A thread of Traditionalist resistance runs the length of the Theodosian dynasty. Proclus was a Neoplatonic philosopher, a practicing lawyer, and one of the few outspoken Traditionalists in the orbit of the court of Theodosius II. He is known mostly for copious commentaries on Plato, but it is his Ten Questions Concerning Providence where he betrays most clearly the wider intellectual environment in which he was writing, in the period around the death of Theodosius II (c. 450 ce). He begins his text with an apology:

Let us, then, interrogate ourselves, if that is all right, and raise problems in the secrecy of our soul and thus attempt to exercise ourselves in solving these problems. It makes no difference whether we discuss what has been said by previous thinkers or not (sive igitur dicta a prioribus, sive non, pertractemus, differentes nihil). For as long as we say what corresponds to our own view, we may seem to say and write these views as our own.⁶⁸

Proclus appears to respond to an expectation of aggregation – precisely that he should “discuss what has been said by previous thinkers” on the topic or providence as a part of his own argument, and his own search for philosophical truth. As in the cases we have already covered, there are two ways to read this comment by Proclus. In their commentary on this passage, Jan Opsomer and Carlos Steel suggest that here he offers nothing more than “a kind of apology for having copied [Plutarch’s] text almost shamelessly.”⁶⁹ Perhaps this is an explanation for this comment, which forms the last piece of Proclus’s preface. But ancient writers in general, and Proclus in particular, express little compunction about copying from their predecessors. Additionally, the passage doesn’t discuss culpability for copying one author in particular, but disclaims a requirement to “discuss what has been said by previous thinkers” writ large. A more proximate explanation for the comment is that he is responding to typical scholarly practice at the time. Like the Historia Augusta, we may reasonably read this passage as a response to typical practice of the dominant scholastic culture into which Proclus speaks: Proclus knows that others will expect him precisely to “discuss what has been said by previous thinkers,” and he retorts not only that he won’t be doing so, but that an argument structured as such would be beside the point because, as he continues, “after all, we all have ‘common Hermes’ as our leader (communem Mercurium ducem habentes), the same who is said to place in every soul the untaught preconceptions of the common notions.”

During the Theodosian Age, scholars from across the disciplinary spectrum engaged with a set of scholastic practices which were dominant and visible across the ideological and disciplinary spectrum and, importantly, which included an expectation of aggregation. Some scholars embraced these new practices while others rejected them. Even in their apparent rejection, Proclus and Ammianus Marcellinus speak to the expectation of aggregation that permeated their scholarly environment, while the Historia Augusta embraces the method as a whole, even while one of its “authors” dissents.

Post-Theodosian Collection, or the Shift to Florilegia

While the centrality of aggregation as a scholastic tool continued intact through the extent of the Theodosian Age, the post-Theodosian West saw yet another shift: this time away from the notionally dispassionate aggregation of competing voices to the collection of sources with explicit polemical and ideological aims. To reiterate: aggregative scholastic productions are never truly nonpolemical. Every source that I’ve engaged has subtle and overt polemical aims which shape the selection and presentation of sources. Aggregation is never dispassionate in actuality – how could it be? Nevertheless, in sources where aggregation is discussed as a scholarly practice by Theodosian writers, overwhelmingly they claim that their method comprises dispassionate selection; the good with the bad, the orthodox with the heretical, good law alongside that which it supervened. The scholarly ideal was that reliable knowledge could be produced through the unbiased collection of sources on all sides of an intellectual debate, followed by their distillation.

As a scholarly ideal, dispassionate aggregation did not last. This distinction, between aggregation as a scholastic practice and collection of sources for admittedly polemical aims, is what separates the Theodosian order of books from what came after. The florilegia that so define the literary output of the post-Theodosian West are not intended for “more industrious types,” as the Theodosian Code might phrase it: collections of raw material from which truth may be distilled. Rather, they were intended as the distillation of truth itself. Whereas for a Theodosian scholar such as Ambrose or Macrobius claims of universality necessarily involved the aggregation of a wide variety of conflicting material, the “century of florilegia” that began in the Ostrogothic period involved scholastic production of a very different type. Leontius of Jerusalem, with whom this chapter began, inherited a focus on aggregation from his Nicene and Chalcedonian forebears. But he wrote nearly a century after the end of the Theodosian dynasty, a context which alone suggests that his opponents might have been correct when they accused him of sampling tendentiously from the patrimony of the tradition rather than presenting a wide range of opinions as befits the scholar in search of universal truth. As is the case with Macrobius and Gellius, who perform a similar operation of aggregation for divergent purposes, Theodosian scholars aggregating scholastic patrimony produced texts that look like later florilegia materially, with radically divergent stated intentions.

There was a durable shift in the material form of Christian scholarship toward end of the Theodosian Age, and especially the controversy around and leading up to the Council of Chalcedon in 451. The rather abrupt arrival on the scene of what Eduard Schwartz dubbed “curated collections” began with the so-called Collectio Novariensis de re Eutychis.⁷⁰ Pope Leo’s chancery collected materials from Roman archives in order to create a dossier that marked a new phase in Christian scholastic production, extended analysis of which is beyond the scope of this book.⁷¹ Other examples of the trend include Leo’s own collection of letters known now as the Leonis Papae I epistularum collectiones,⁷² and the more famous Collectio Avellana, a lacunose mid-sixth-century collection of imperial and ecclesiastical letters and documents ranging from the year 368 to 553 which offers a “a unique perspective on the history of the early sixth century” through selective presentation of documents relating to the papacy, the Ostrogothic court in Ravenna, and the Roman court in Constantinople.⁷³ The dossier that Athanasius proposed, produced, and appended to his Concerning the Decrees, for instance, looks materially like these later “curated collections,” and of course Athanasius collected and curated the material. But pre and early Theodosian productions have stated intentions fundamentally distinct from the catenae, “curated collections,” and florilegia that came to dominate in the post-Theodosian Age. The Theodosian collections were notionally dispassionate.

This shift did not go unnoticed during the Theodosian Age, either. Writing for the Eastern court of Theodosius II in the early 440s, Sozomen recounts that his own access to the truth of matters surrounding “the dogmas of Arius and subsequent proposals” were obscured by purposeful failure to aggregate conflicting materials.⁷⁴ He complains, “in order to demonstrate the orthodoxy of their own dogmas, the partisans of each sect respectively formed a collection of epistles that favored their own heresy, omitting all hostile documents!” (1.1.15.8–1.1.16.1). Sozomen explicitly groups such texts together as a class, what I have been calling “curated collections,” and complains in the first chapter of his Ecclesiastical History that he was forced to busy himself with analysis of these faulty dossiers nevertheless, in order that he might have at least some access to the truth. “Still, as it is requisite, in order to maintain historical accuracy, to pay the strictest attention to the means of eliciting truth, I felt myself bound to examine all writings of this sort according to my ability” (1.1.16.2–4). The shift to curated collections was a lamentable trend, according to Sozomen. And again, it was a trend with a center of gravity in theological productions.

A shift to curated collections in support of one creed or another was of course not inevitable, but this historical movement was perhaps overdetermined already by the shift from scriptural interpretation to a focus on credal language. By the middle of the fifth century, even the creeds which were intended to be a distillation of scripture and its interpretative key had undergone the same transformation that scripture underwent in the aftermath of the Council of Nicaea: they became hermeneutically impotent. As discussed earlier, by the convening of the Council of Ephesus in 431, creeds had become so central that new scholastic productions were necessary which aggregated not the patristic disputation that led to the dogmatic affirmations held in the creed, but rather compilations which attested the history of credal interpretation itself. In a supplement to the Proceedings of the Council of Ephesus condemning Nestorius, the Nicene Creed is presented and followed not by general theological disputation, as in Athanasius, but by credal disputation, and extracts from the Orthodox fathers detailing how it is that they interpreted the creed itself. In the words of Richard Price:

Here, finally, the appeal to the Fathers moves to centre stage. Taking together the acts of 22 June and this supplement of 22 July, we may conclude that the First Council of Ephesus achieved its main work, the condemnation of Nestorius, not by theological ratiocination, but by establishing the criterion of orthodoxy, namely the Nicene Creed as interpreted by the great Fathers of the fourth century and definitively by Cyril of Alexandria.⁷⁵

And, while the appeal to the fathers and the creation of scholarly resources to support particular readings of the Nicene Creed was effectively accomplished by the creation of this dossier, it was also formally legislated in canon seven of the Council of Ephesus, which declares that “no person may propose or even write or support a faith diverging from that created together with the holy spirit by the holy fathers assembled in Nicaea.”⁷⁶

Conclusion

The assumption of aggregation, distillation, and promulgation as central scholarly tools spread from the wake of the Council of Nicaea, first among Christians and eventually across the entire spectrum of Theodosian Age scholarly production. Evidence of a dominant scholastic mode, inflected by Christian theological debates, is visible from the works of Athanasius to those of Proclus. Even when scholars reject the basic methodology of aggregation and distillation, they witness its presence as a form, and perhaps the dominant form, of scholastic production.

Interconnections between disciplines are visible in every facet and corner of the great Theodosian Age scholastic productions: from theology to law to historiography, medicine, and miscellany. One glance at the pioneering work of Raban von Haehling detailing the “[r]eligious persuasion of high office holders in the Roman Empire” shows that the accession of Theodosius I was the turning point, at which Nicene Christians came to power as a ruling elite for the first time.⁷⁷ Even given the rightful criticisms of Barnes and Salzman regarding specifics of von Haehling’s prosopography, the trend is clear: Christians, and specifically Nicene Christians, came to majority power only in the late fourth century (Figure 1).⁷⁸

Figure 1. Religious identification of Western senate high-office holders at time of highest office.

Chart data from Salzman, The Making of a Christian Aristocracy, 228.

Armed with a set of scholastic practices whose development I treated in Chapter 3, Nicene Christians came to power during the Theodosian Age, and brought their peculiar structure of knowledge with them. One’s approach to proper knowledge production may be context specific to a certain extent, but in broad strokes it remains intact as individuals code switch between domains of knowledge and modes of interaction – already forty years ago, Ramsay MacMullen demonstrated that by the late fourth century a significant convergence in religious vocabulary and ideology had taken place between senatorial Christians such as Ambrose and Traditionalists such as Symmachus.⁷⁹ As Nicene Christianity proliferated through the ranks of the elite, we see a simultaneous shift in the way that arguments are made within domains of knowledge production that do no obvious theological work. I offer this analysis as a novel way of tracing what it means for elite Roman society to “become Christian.” Macrobius was a scholar working in a traditional discipline, and a rather lonely one at that, far removed from the bustle of theological disputation undertaken by broad swathes of his contemporaries. Yet his reformulation of the antiquarian format adopted from Gellius and his redeployment of the format with new intellectual aims points to the new and predominantly Christian scholastic environment in which he lived and to which he spoke. Each of the examples in this chapter could be multiplied, and each speaks to a coherent shift in practice among works of Roman scholarship in the years after Nicene Christians first came to hold significant political power. The widespread assumption of a mode of scholastic production which began as a set of theological practices was not a one-off event, and the shift did not occur in the same way in all domains. But a gradual change is a change nevertheless; in fits and starts, the acorn does eventually become an oak tree.

The assumption of aggregation and distillation as central methods in knowledge production led to downstream changes in the way that scholars approached books: what they thought books did, how works of scholarship ought to look, how they were best encountered, and the manner in which readers assessed their contents. These changes were not merely intellectual, confined to discussion in statements of purpose and programmatic methodological musings. We can see the effect of shifting scholarly practices in the pages of fifth-century manuscripts and in the innovations in style and format meant to deal with the fact that scholarship in the Theodosian Age looked different than it had before. A new order of books is visible beyond the methodological statements and intellectual productions of Theodosian writers; it is visible perhaps most clearly in the pages presented to Theodosian readers. I turn now to trace the effect of shifting scholarly practices in manuscripts of the Theodosian Age. This project has an epigraph: “New readers of course make new texts, and their new meanings are a function of their new forms.”⁸⁰ Chapter 4 has profiled the new readers; Chapter 5 begins to investigate their new texts.