On which edition of the early church historians—principally Eusebius, Socrates Scholasticus, and Sozomen—did Marvell rely when composing the Short Historical Essay, touching General Councils, Creeds, and Imposition in Religion that forms part of Mr. Smirke (1676)? Since Marvell frequently cites the historians in conjunction, and moves between them freely, it seems likely that he was resorting to one of the many collections that conveniently brought them all together. The answer that currently holds the field is the one given by Annabel Patterson in her edition of Mr. Smirke: that is to say, that

While, as we shall see, it is certainly the case that Marvell was working with Latin rather than Greek versions of the histories, my argument is that he was reliant not on a Counter-Reformation compilation of them but, as might perhaps have been expected, one with a more obviously Reformed provenance. What follows is a process of winnowing out the various contenders by reference to a set of differentiae until we end up with the book that was undoubtedly on Marvell’s desk as he composed the Essay.

Perhaps the most important of the church historians for Marvell’s purposes in the Essay was Eusebius. But this immediately raises the question of which of the four available Latin translations of Eusebius he was following. The first, by Rufinus of Aquileia (340–411 CE), collapsed Eusebius’s ten books into nine and was followed by a two-book continuation of his own which took the story down to the death of Theodosius in 395CE. This was the version frequently printed by the house of Froben in Basel as the lead item in their composite volume, Autores Historiae Ecclesiasticae (1523, 1528, 1533, 1535, 1539, and 1544) and by Galeoto Pratensi in Paris in 1541 under the same title. In all these editions, the nine books of Eusebius and the two by Rufinus are numbered consecutively, one to eleven. But when Marvell quotes from Rufinus, his reference—“saith Ruffin L. 1. c. 2.”—shows that he was following a version in which the Rufinus continuation was printed as a separate item in a standalone two-book format.2 If what we are after is a single volume in which Marvell could find almost all of his historical materials, then this in itself is sufficient grounds for excluding the seven volumes printed between 1523 and 1544 from consideration.

Furthermore, at one point in the Essay Marvell quotes from the original tenth book of Eusebius, of which Rufinus had incorporated only a few fragments in his ninth book. Since the passage is a crucial piece of evidence to which I shall return, it is here quoted at length:

It should be noted, first of all, that the citation is a mistake or a misprint; the text of Constantine’s rescript to Chrestus, bishop of Syracuse, explaining why the emperor had found it necessary to summon a council at Arles in 314CE to settle the controversy over the Donatists, actually comes at the end of chapter five of Eusebius’s tenth book, and is followed immediately by chapter six. Either way, this brings into play the three translations of Eusebius in ten books that Marvell could in principle have consulted.

As we have seen, the first of these, by John Christopherson (d. 1568), was originally published in 1569.4 However, this edition did not include the Rufinus continuation, something that is also true of the translation with annotations by Henricus Valesius (1603–76) that was published in Paris in 1659 and reprinted in Mainz in 1672. Admittedly, this is not in itself sufficient grounds for dismissing Christopherson since his translations were reprinted in Paris in 1571 in a single volume entitled Historiae Ecclesiasticae Scriptores Graeci that did include Rufinus’s continuation. What definitively rules out Christopherson (as well as Valesius) is Marvell’s emphatic reference to “the Catholick Heresy” in the passage quoted above. Patterson was puzzled by this because it was “not clear why Marvell chooses ‘heresy’ as the term here, unless for irony’s sake. Christopherson renders it ‘catholica Ecclesia[e] opinione,’ and Hanmer translates it as ‘opinion.’”5 It is true that if Marvell did have access to the 1612 Geneva edition, which printed Christopherson’s Latin and the original Greek in parallel, then he might have seen that the latter actually read αἱρέσεως —the genitive of αἵρεσις (hairesis)—and on that basis decided to substitute “heresy” for Christopherson’s “opinione.”6 (The same goes for the parallel text of Valesius, where the relevant phrase is rendered as “Ecclesiae Catholicae sententia.”7) But this is not what happened. Instead, Marvell was simply following the earlier translation of Eusebius by the Reformed theologian Wolfgang Musculus (1497–1563), first published by Froben in Basel in 1549 as the lead item in a collection now entitled Ecclesiasticae Historiae Autores and subsequently reprinted in 1554, 1557, and 1562. Musculus’s more faithful rendition of the Greek, “haeresi catholica,” is seized on by Marvell as telling evidence for his claim that originally heresy was a neutral term that only acquired its negative significance as a result of factional disputes within the early Church and, above all, in consequence of proceedings at the Council of Nicaea.8

Marvell’s reliance on Musculus’s translation of Eusebius’s history, together with the fact that the 1549, 1554, 1557 and 1562 Froben volumes also include the Rufinus continuation, might be thought conclusive. However, this is not so. Significant complications arise when Marvell cites another work by Eusebius; namely, his life of Constantine (De vita Constantini). While this work too was translated by Musculus, neither the Latin cues nor the references supplied by Marvell match his version. For example, Marvell points out that Constantine took action against the Novatians by publishing

The Latin here matches Christopherson’s translation rather than that of Musculus, which reads “Agnoscite nunc per hanc legis constitutionem, o Nouatiani.”10 We also need to take into account the peculiarities of the way in which the Musculus translation is presented. Each of the books is preceded by a list of the chapter headings, though these are numbered only in the case of books 2 and 3. Their efficacy as a means of navigating the work is, however, greatly limited by the fact that in the body of the text the successive chapters are distinguished only by indents or blank spaces, making it difficult for a reader to be sure of the number of any given chapter as they read it. By contrast, the Christopherson text presents the chapters neatly separated by rules, headings, and numbers. Nor do the chapter numbers themselves tally; the Latin quotation above occurs in what is chapter 62 in Christopherson but 65 in Musculus.

Another example is when Marvell observes that

While the book number is not given, the Latin here matches Christopherson’s translation at Book 2, chapter 70 (of 71); by contrast, Musculus reads “Reddite itaque mihi serenos dies ac noctes a curis liberas” at what would be chapter 66 of the 67 that are listed.12

At this juncture, it might seem that the thesis that Marvell was working with a single volume to hand should be discarded on the grounds that he was apparently moving to and fro between Christopherson and Musculus. However, this difficulty is obviated if we turn to the compilation prepared by the Swiss Protestant divine Johann Jakob Grynaeus (1540–1617): Eusebii Pamphili, Ruffini, Socratis, Theodoriti, Sozomeni, Theodori, Evagrii et Dorothei Ecclesiastica Historia (Basel, 1570; reprinted in 1587). Adopting a markedly ecumenical approach, Grynaeus prints (lightly) revised and corrected versions of, on the one hand, the Musculus translation of Eusebius’s history, and, on the other, Christopherson’s translation of the life of Constantine as well as his translation of Sozomen’s Historiae Ecclesiasticae. As against that, when it came to Theodoret’s history, he turned to the translation by the Protestant humanist Joachim Camerarius (1500–74). We can be sure that Marvell was familiar with the last of these since in the Essay he quotes from Camerarius’s translation of the conspectus of heresies that Theodoret had appended to his history.13 It is true that Camerarius had first published the work separately under the title Theodoriti Episcopi Cyrensis Rerum Ecclesiasticarum in Basel in 1536 before it was incorporated in Froben’s Autores Historiae Ecclesiasticae (1539 and 1544) and Ecclesiasticae Historiae Autores (1549, 1554, 1557, and 1562). Nevertheless, the fact that Grynaeus included Camerarius’s Theodoret as well as Musculus’s version of Eusebius, Rufinus’s continuation of Eusebius, and Christopherson’s translation of Eusebius’s life of Constantine, means that virtually all of the patristic materials that Marvell required were available in either the 1570 or the 1587 Ecclesiastica Historia.

To place the issue beyond doubt, we can turn for the final time to Marvell’s quotation from Constantine’s rescript to Chrestus:

The parenthetical aside in roman type sounds quintessentially Marvellian as he lingers over the much-contested term “heresy.” However, it is in fact an expansion of a marginal gloss by Grynaeus: “Haeresis in bonam partem sumpta” (“Heresy taken in its good sense”).14 Nor should we be at all surprised by this. Such economy of effort and opportunism is absolutely consistent with what we know of Marvell’s working methods, whether early or late.

This being the case, two further issues arise. One is the question of how Marvell got access to a copy of this relatively esoteric text. It has been established that when researching The Rehearsal Transpros’d (1672) and The Rehearsall Transpros’d: The Second Part (1673) Marvell largely depended on the good offices of the Earl of Anglesey, the owner of probably the largest private library of the time.15 But while the sale catalogue of Anglesey’s library lists a copy of Eusebius “Et Aliorum Historia Ecclesiast. Græcé – Paris Rob. Stephani – 1544” this was not a volume of Latin translations but the editio princeps printed by the Estienne press in Garamond’s newly designed Greek (“grecs du roi”) typeface.16 However, there is some evidence that Marvell may also have had recourse to the extensive collection of books belonging to John Owen, the nonconformist divine who is known to have handled the proofs of The Rehearsal Transpros’d.17 When Owen’s library was sold in 1684, it included a copy of the 1587 edition of Grynaeus.18

The other question concerns the confessional significance—if any—of adopting Grynaeus’s edition. It is true that Christopherson had been the Master of Marvell’s Cambridge college, Trinity. More to the point, however, may have been the fact that Christopherson had played “a major role in enforcing the Marian regime’s heresy legislation.”19 One of the salient features of this legislation—as Marvell might have known by virtue of sitting on the Commons committee for the abolition of de haeretico comburendo, the writ that authorized the burning of heretics—was that it left Mary Tudor’s bishops free to burn heretics on their own initiative without even needing to secure the writ.20 Was one of the attractions of the Grynaeus volume therefore precisely that it did include Christopherson’s translation of the life of Constantine, from which, by a nice irony, Marvell was able to cite materials that were damaging to the reputation of the Council of Nicaea that had done so much to weaponize the notion of heresy in the first place?21