An Earlier Edition of Eusebius’s Church History?

 For more than a century scholars have played with the idea that Eusebius’s Ecclesiastical History once circulated in a shorter, earlier form. Harnack and Laqueur imagined a seven-book edition, finished before the Great Persecution (around 303), with the last three books added only later. Lawlor pushed the first edition to eight books. Barnes revived this line in the 1970s, pointing to the stylistic and structural difference between Books 1–7 and 8–10 and suggesting multiple redactions.

More recent work has been much less enthusiastic. Valerio Neri essentially dismantles the notion that Books 1–7 represent a self-contained “first edition,” and Cassin–Debié–Perrin argue that all ten books belong to a single compositional project, brought into final form around 324/325. Even Andrei, who still allows a long compositional process, dates the first complete ten-book edition only as late as 313, which already undercuts the older picture of a neat, pre-Diocletian prototype. On this view, what we see in the textual tradition is not a short vs. long History but one work that Eusebius kept revising, especially toward the end.

Barnes’s own collation actually points in this direction. There are places where material moves around or appears only in some witnesses. The little doxology at 9.11.8 (the “hymnus”) stands after Book IX in some manuscripts (B, D, etc.) but at the head of Book X in others. That looks exactly like the sort of seam where one edition may have ended and a later redaction smoothed the join. Likewise, the imperial letters at 10.5–7 survive only in a subset of Greek manuscripts; other codices, and later Latin versions, simply omit them. And the Syriac translation preserves readings that look like a later, “cleaned up” edition where references to Crispus have been quietly removed after 326. Cassin and others stress that such phenomena are better explained as authorial reworking and scribal updating than as completely independent interpolations. What emerges is not two different Histories but one Historia Ecclesiastica with a history of growth.

There is also no patristic smoking gun for a shorter nine-book (or seven-book) text. Later church historians treat Eusebius’s History as ten books. Socrates and Sozomen simply presuppose it as the standard narrative from Christ down to Constantine, and the ancient versions line up with that. The Syriac church history is particularly important: the Syriac manuscript is centuries older than any surviving Greek codex, and yet it is clearly a translation of a ten-book exemplar. The early evidence, in other words, knows Eusebius in ten books, with local variation in details, not a floating nine-book “proto-Eusebius” that only later got topped up.

Rufinus’s Latin version, and what he says about it, is crucial here. In his preface he frankly admits that he mutilated Book X. In the Greek exemplar before him, the tenth book contained very little straightforward “history” and a great deal of epideictic material – panegyrics of bishops, set-piece praise of Constantine, and so on – which he judged “superfluous” and useless “for knowledge of the facts.” So he took the historical fragments from Eusebius’s Book X, tacked them onto Book IX, and made that (in Latin) the formal end of Eusebius’s History. He then composed his own Books X–XI, moving the story from Constantine’s victory and the end of persecution down to the death of Theodosius.

Modern scholars who have compared the Greek and Latin confirm that his “omissions” were drastic. Humphries is right that in effect Rufinus almost entirely omits the Greek Book X. McKitterick’s larger judgment is also on point: Rufinus wants a leaner, more doctrinally “safe” and politically balanced narrative. The emperor is still important, but now deliberately subordinated to the church; triumphant imperial rhetoric is toned down, and offensive material (like Constantinopolitan panegyrics) is simply cut. That does not prove that the panegyric sections were later, non-Eusebian additions. On the contrary, they are present in our Greek witnesses, and the simplest explanation is that they formed part of Eusebius’s own final edition which Rufinus disliked and suppressed.

Rufinus also tells us how he understood his role: he translated Eusebius in nine books, then added two more books “like the two little fish appended to the five loaves.” Everything from Nicaea through the Arian crisis, Athanasius, the councils under Constantius, the Jovian–Valentinian–Valens sequence and so on is the Latin continuer speaking. For him, the “real” Eusebius ends where everyone else in the orthodox tradition also puts the terminus: with the defeat of Licinius and the “peace of the Church” under Constantine as sole emperor. That testimony is strong negative evidence against any genuinely longer Eusebian edition extending into the rule of Constantine’s sons.

And yet there is one tantalizing piece of contrary evidence: Philostorgius.

Photius, in his notice on Eusebius (cod. 27), sums up the Ecclesiastical History in a neat sentence: Eusebius’s history in ten books begins from the advent of Christ, describes in detail the times of the tyrants, and “comes down to the reign of Constantine the Great,” narrating what befell the churches under and through him. That is standard, and matches the Greek text as we have it.

But in the epitome of Philostorgius, an Arian historian of the early fifth century, there is a different note. Philostorgius, Photius reports, praises Eusebius’s historical work while criticizing his theology, and “joins with the others in bearing witness that he brought his history down to the accession of the sons of Constantine the Great.” That sounds, at first glance, exactly like the “longer edition” one might expect from an Arianizing Eusebius: a continuation into the succession crisis of 337, the bloody purge, and the subsequent ascendancy of Constantius II.

At this point it matters how we read the other continuators. Gelasius of Caesarea, Cyril of Jerusalem’s nephew, wrote a history that almost everyone agrees was conceived as a continuation of Eusebius. His work is lost, but enough fragments survive (and have been reconstructed by Wallraff, Stutz, Marinides, etc.) to see its contour. In his own preface he says that his task is to narrate “the things that happened after Eusebius and the things that Eusebius did not record.” He then re-tells the end of the tetrarchy, Constantine’s origins, the Milvian Bridge and Nicaea, and carries the story on through the Arian battles into the reign of Valens. In other words, if you bind together Eusebius plus Gelasius you get exactly what Philostorgius says Eusebius produced: a continuous dossier from Christ down to (at least) Constantine’s sons and the later fourth century.

Now put Rufinus back into the picture. His Latin Church History presents itself as “Eusebius in nine books” plus two books of continuation. The material he adds – the dramatic stories about Nicaea, the petitions burned by Constantine, the Paphnutius and Spyridon anecdotes, Athanasius’s childhood games and his prayer against Arius in Constantinople, the lurid death-in-the-latrine scene – overlaps in a striking way with the Gelasian fragments. Both Rufinus and the Gelasius tradition are clearly drawing on a post-Eusebian dossier of pro-Nicene Constantine and post-Constantine lore. Both treat that dossier as continuation, not as part of Eusebius’s own text. When they talk about Eusebius, they say what Socrates also says: his History ends with Constantine’s victory and the settlement of the Church under his sole rule.

Seen from that angle, Philostorgius’s remark starts to look less like hard evidence for a lost Arian long edition of Eusebius, and more like a blurred attribution. An Arian historian in the early fifth century could easily be reading an “Eusebius plus continuation” codex and casually refer to the whole package as “Eusebius’s history.” His own narrative unmistakably depends on sources beyond the ten Greek books we possess; at points he even contradicts the orthodox succession histories (for instance by erasing Paul of Constantinople and making Eusebius of Nicomedia Constantine’s immediate choice for the new capital). Those divergences are better explained by an Arianized continuation tradition than by postulating that the original Eusebius wrote a very different political endgame which later disappeared without a trace in the orthodox stream.

The Decretum Gelasianum sits in the background of all this as a kind of verdict on Eusebius from the Latin, late-antique side. The famous “not to be received” list includes all sorts of obviously apocryphal gospels, acts, and revelations, but when it reaches historical works the tone softens. Eusebius is not simply dumped in a pile of condemned heretical books. The decree says that “the chronicle of Eusebius of Caesarea and the books of his church history,” although he “fell flat in the first book of his narration” and later wrote “one book in praise and excuse of Origen the schismatic,” are nevertheless valuable for their exemplary stories and “are not said by us to be refused.” That is a grudging endorsement, not a rejection. From this same dossier we can see how later Latin churchmen read him: as somewhat compromised doctrinally, too friendly to Origen, yet indispensable as a collector of facts.

Putting all this together, the picture that emerges is something like this. Eusebius wrote and repeatedly revised a ten-book Ecclesiastical History that he brought into final form in the 320s. The end-point of the Greek text, as transmitted in the mainstream tradition, is Constantine’s victory over Licinius and the ensuing peace of the Church; Book X in its original form was heavily panegyrical and was disliked enough by some later readers (notably Rufinus) to be mutilated. In the later fourth century, pro-Nicene historians in Palestine and the Latin West began composing continuations: Gelasius of Caesarea in Greek, Rufinus in Latin. Their material overlaps, and both clearly present themselves as “after Eusebius,” not as alternative editions of him. Arian historians like Philostorgius, reading a continuous dossier that ran from Christ to the Constantinian succession, could easily slide into calling the whole thing “Eusebius’s history” and push the terminus down to the accession of the sons.

If so, then the “earlier edition” question changes its focus. Instead of hunting for a lost seven- or eight-book prototype, or an Arian long edition that conveniently vanished everywhere except in Philostorgius’s memory, it makes more sense to think in terms of an evolving ten-book Historia Ecclesiastica that spawned continuations very quickly. The real critical task is to map the seams inside Books IX–X, to understand how Eusebius himself reworked his last volume as Constantine’s regime changed, and then to distinguish that authorial process from the quite diverse continuation traditions (Gelasius, Rufinus, the later Greek church historians) which filled in the story beyond him.

And that, in turn, has consequences for how we read Eusebius’s theology and politics. If Book X, with all its triumphalist Constantinian color, belongs to his own mature composition, and if the “anti-Arian” or pro-Nicene framing of the post-Constantinian period belongs chiefly to his continuators, then any attempt to reconstruct a more modest, non-imperial Eusebius on the basis of a hypothetical short edition becomes very hard to sustain. The evidence we actually have points instead to one historian who kept editing his ten-book narrative in light of Constantine’s changing fortunes, and then to a cluster of later writers – some Nicene, some Arian – who took his work as their starting point and carried the story forward in very different directions.