Acelebrated 17th-century​ poet composes, over many years, his own eccentric version of Christian theology. Along the way he becomes blind, and so dictates the work to scribes, with additions, revisions and recopyings as it proceeds. Before he can publish his treatise the political winds shift, and it is not printed in his lifetime. After the poet’s death an attempt is made to have it printed in Amsterdam, but the Dutch publisher finds it too heretical and sends it back. A government official takes possession of the manuscript, and it sits forgotten for a century and a half until a librarian discovers it in a cupboard, still in the envelope in which it was returned from the Netherlands. When the treatise is translated from its original Latin and published, it becomes clear how strange the poet’s Christianity was. The poet is John Milton, and the work is De Doctrina Christiana.

De Doctrina Christiana matters because it mattered to Milton. You can ignore it and still enjoy his poetry, but if you become seriously interested in Milton your interest will sooner or later extend to his religious views, and once that happens, De Doctrina Christiana is a vast, detailed record of his thought on the subject he considered more important than any other. He put an enormous amount of time and effort into this treatise: the sheer bulk of the manuscript and its intricate revisions establish that. He meant to publish it, but then the Restoration of 1660 put his revolutionary party on the wrong side of history. Milton was writing for an international readership of Protestant intellectuals, a learned few who would, he hoped, be willing to hear him out. He was well aware that his treatise contained provocative material and expected pushback, which he tried to pre-empt in the preface by asserting his good intentions. ‘If I now make this account public,’ he wrote,

if (as I call God to witness) it is with brotherly and friendly feelings towards all people that I share as freely and widely as possible this my best and most precious possession, I hope that everyone will be kindly disposed towards me, and no one prejudiced or hostile, even though I shall seem to have brought to light many things which will at once be discovered to conflict with some received opinions.

De Doctrina Christiana is eccentric in content, but not in form. Although Milton was not a professional theologian and says scornful things about such people, his treatise makes a bid to join their conversation. It belongs to the sober, scholarly genre of systematic theology, and like previous works of Reformed systematics, it is structured according to the method popularised by the French logician Petrus Ramus, with material organised into subdividing branches (it is a coincidence that ramus means ‘branch’). De Doctrina Christiana divides Christian doctrine into two main parts: faith, or the knowledge of God, and charity, or the worship of God. These two parts split into further parts, which comprise the individual chapters, and the process of defining and subdividing topics repeats within each chapter. Neat and tidy, until it isn’t.

The great challenge for Reformed systematics was that scripture is not easily rendered systematic, and Protestants were supposed to rely on scripture alone. According to the core Reformation principle of sola scriptura, all points of Christian belief and worship should be grounded in the Bible, and anything non-biblical discarded. In practice, Reformed systematic theologians did not rely on scripture alone. They drew on one another’s work, and on that of their pre-Reformation precursors. They followed long-standing interpretive principles such as the Augustinian rule of faith, which held that you use the plainer parts of the Bible to explain the more obscure parts, assuming an underlying coherence. Over time, the process of collecting together scriptural passages on particular topics became conventional. Milton drew on works by the Swiss theologian Johannes Wolleb and the English puritan divine William Ames. De Doctrina Christiana follows Wolleb and Ames in dozens of ways, from its Ramist structure to many of its definitions, although Milton does not acknowledge these borrowings. This occlusion need not be thought of as dishonest. Silently borrowed language was a common feature of the genre.

In his preface, Milton describes the genesis of De Doctrina Christiana. In adolescence, he tells us, he began a programme of serious Bible study, reading both Testaments in their original tongues and arranging passages under topical headings. Eventually, finding all existing systematic works unsatisfactory, he decided to produce his own. The idea, Milton explained, was to create the most scripture-stuffed work of theology possible. Rather than relegating biblical proof texts to the margins, as others did, he intended to include them in the main text, leaving ‘as little room as possible for my own words, though they arise from the weaving together of the [biblical] passages’. The planned work wouldn’t be so much a commentary on scripture as an ordering of it. This plan does not accurately describe the work as it developed, but it is undoubtedly scripture-stuffed. About half of the text consists of Bible verses in Latin, quoted or in Milton’s own translations into Latin. The sheer bulk of scriptural evidence, Milton hoped, would lead his readers to take his heterodox conclusions seriously.

The most significant of his heterodoxies, and the one most likely to shock his contemporaries, was his antitrinitarian or Arian Christology. Milton believed in the Father, the Son and the Holy Spirit – all three are well attested in scripture – but the doctrine of the Trinity notoriously is not, and he concluded that traditional formulas such as the Nicene Creed had got it wrong. His objections were biblical and logical. Biblical, in that such terms as ‘Trinity’ and doctrines such as the eternal generation of the Son – the doctrine that the Son and the Father have coexisted for all eternity – are nowhere to be found in the Bible. Logical, in that the elements of trinitarian doctrine ran counter, as Milton saw it, to basic principles of reason. The Father and the Son cannot share an essence, for example, because essences cannot be shared: one is one and two is two, even for God. Scripture speaks of a Father and a Son; the Father must therefore come before the Son – why else would the divinely authoritative text use those terms? Father and Son cannot be co-eternal. God the Father, Milton reasons, must have begotten the Son in time, not of necessity but of his own free will. The Father then chose the Son as the agent through whom he created everything else.

The most common 17th-century term for this position was Arianism, after Bishop Arius of Alexandria, an early fourth-century churchman whose views were condemned at the First Council of Nicaea in 325. Since Arianism was a major heresy Milton didn’t use the term, but the Dutch publisher Daniel Elzevier gave it as a reason for his rejection of the manuscript after Milton’s death. Elzevier had sent it to a local reader, the theologian Philipp van Limborch, who reported years later that he had advised against its publication because ‘the strongest Arianism was to be found throughout it.’ De Doctrina Christiana contains other unusual views. It espouses mortalism – the view that the soul dies or sleeps along with the body until the last judgment. It approves polygamy, on biblical grounds. It holds that God created ex deo rather than ex nihilo. But its Arian Christology is the major issue.

Ever since De Doctrina Christiana was rediscovered in the 1820s, there have been readers of Milton to whom its heterodoxy is unwelcome. If you don’t like it, you have several options. You can deny that Milton wrote it. Or you can concede his authorship but minimise the heterodoxy of the work. Or you can concede its heterodoxy but deny it is shared by Paradise Lost and Milton’s other poems – the position taken by C.S. Lewis, among others. If you want, you can loop back and argue that this theological disparity is a reason to deny that Milton wrote De Doctrina Christiana.

Authorship denial began with Bishop Burgess of Salisbury in 1829, and had a revival in the late 20th century. The issue was thoroughly reappraised, with near universal consensus that the case for Milton’s authorship is sound. We know independently, from Milton’s early biographers, that he wrote a Latin theological treatise, and a considerable paper trail connects him to the one with his name on it found in that cupboard at the State Paper Office at Whitehall in 1823. Since the 1990s books on Milton have often included notes acknowledging an ‘authorship controversy’ around De Doctrina Christiana. Such gestures seem to me no longer necessary, not that it’s up to me.

Heterodoxy minimisers are more numerous than authorship deniers, and have a more respectable case. They point out that much of the work is standard Reformed theology; that there was no single orthodoxy among 17th-century Protestants; that Milton was not alone in expressing doubts about the foundations of trinitarian doctrine. Their main problem is that Milton rejects trinitarian doctrine in De Doctrina Christiana explicitly and at length. You could get in real trouble for doing this in Milton’s day. The last two men burned for heresy in England, Bartholomew Legate and Edward Whiteman in 1612, were antitrinitarians. In the mid-1650s, when Milton was working on De Doctrina Christiana, his contemporary John Biddle was imprisoned and interrogated by a parliamentary committee for publishing antitrinitarian views; Cromwell had Biddle exiled to the Scilly Isles to avoid worse punishment. Milton’s treatise is defensive on the subject. He precedes the section in which his antitrinitarian views are most fully spelled out – Chapters 1.5, ‘On the Son of God’, and 1.6, ‘On the Holy Spirit’ – with a secondary preface in which he exhorts readers to give him a fair hearing. In taking his position, Milton claims, he is challenging not divine authority but merely other human interpreters, and does so by following the pan-Protestant principle of sola scriptura. All good Protestants, therefore, should grant him the freedom to make his argument, and should be willing to assess it without prejudice. Then follows ‘On the Son of God’, the longest, most argumentative and most heretical chapter in the book.

The agreement in doctrine between De Doctrina Christiana and Paradise Lost is a large and technical subject. It’s an apples-to-oranges comparison, first of all. Milton’s Latin theological treatise and his English epic poem on the Fall cover different ground, and any common ground is dealt with at different length and in different ways. It’s easy to find disparities; harder to find are cases where epic and treatise are plainly at odds. Paradise Lost is less explicitly Arian, partly because a treatise of systematics is bound to be more explicit about doctrine than a narrative poem. It may also be that Milton was cautious enough, in a vernacular poem published during his lifetime, to avoid bringing too much attention to his most controversial positions. Paradise Lost does not show God the Father creating the Son. But it does show the Father summoning the angels to a high assembly at which he announces that ‘This day I have begot whom I declare/My only Son,’ whom he decrees vicegerent ruler. This decree causes Satan to rebel. Paradise Lost shows the Son of God sweeping the rebel angels out of heaven; after that, it shows the Son, on the Father’s instructions, creating the world. Far from co-existing in eternal unity, the Father and Son have distinct characters and roles. The Father commands from his invisible throne; the Son, with exemplary obedience, executes the Father’s commands. They are, as William Empson put it, ‘about as unidentical as a terrier and a camel’.

In the century and a half between the publication of Paradise Lost and the rediscovery of De Doctrina Christiana, a minority of readers expressed doubts about the poem’s Christology; most did not. The moral, I think, is not that Paradise Lost is orthodox, but that it is very hard to read a work of Christian literature without unconsciously importing assumptions from the version of Christianity you know best (the same is true a fortiori of reading the Bible). Readers of Paradise Lost in the 18th century were mostly trinitarian Christians of one stripe or another, Dissenters included, and brought these assumptions to Milton’s epic. Paradise Lost thus came to be admired as a wholesome Christian poem by many conventional believers who abhorred Milton’s politics and would have abhorred his theology had they recognised it. That is the reason De Doctrina Christiana caused a stir: it forced a rethink. After its publication in 1825, readers like the young Macaulay professed themselves unsurprised:

Some of the heterodox opinions which he avows seem to have excited considerable amazement; particularly his Arianism, and his notions on the subject of polygamy. Yet we can scarcely conceive that any person could have read the Paradise Lost without suspecting him of the former; nor do we think that any reader, acquainted with the history of his life, ought to be much startled at the latter.

Macaulay’s take can be discounted as hindsight (he wasn’t fair about Milton’s private life either). But Milton’s heterodoxy has always been evident to some, with or without De Doctrina Christiana. When I arrived at the Catholic University of America I was introduced to a senior colleague, a distinguished philosopher, who asked me what I was working on, and when I answered ‘Milton’ followed up: ‘Interesting poem, Paradise Lost. Christologically interesting. Basically Arian, isn’t it?’

For the past two centuries, most readers of De Doctrina Christiana have come to it from Milton’s poetry. This has resulted in its being read as a gloss on the poems – a practice explicit in the title of Maurice Kelley’s still valuable 1941 study, This Great Argument: A Study of Milton’s ‘De Doctrina Christiana’ as a Gloss upon ‘Paradise Lost’. On this approach, the treatise functions as a repository or archive of Miltonic religious beliefs. The poems raise questions about which doctrines Milton held; you turn to De Doctrina Christiana to find the answers.

Jason Kerr​ wants to move away from that approach. Milton’s Theological Process aims to read De Doctrina Christiana not as an archive of Miltonic positions but ‘as a literary work in its own right’. We have gone wrong, Kerr claims, in reducing it to its doctrinal content. In doing so we have neglected its rhetorical dimensions, its evolution over time and its unfinished condition. We should understand it as a work in which Milton was ‘still figuring things out’, as an ‘artefact of a theological process’. ‘This process appears,’ he goes on, ‘in the materiality of the manuscript, which bears abundant traces of both revision and expansion. These material traces correspond to a shift in the treatise’s literary form as it moves (in places) further and further away from its own stated formal intentions … towards new generic territory marked by rhetorical modes of expression at odds with the Ramist project.’

By this Kerr means, first of all, that Milton set out to produce something neat and tidy and found that he couldn’t. This is plausible, in a sense that you’ll grasp if you’ve ever outlined a piece and then found, once you started writing, that the crystalline logic of the outline did not survive first contact with the enemy. Ramism is essentially a technique of outlining. Since Petrus Ramus was among the many French Protestants killed in the St Bartholomew’s Day massacre of 1572, his status as a Huguenot martyr contributed to his method’s vogue among Protestant intellectuals, but there is nothing intrinsically Protestant about Ramism. It organises content – any content – into a certain logical form. Some chapters of De Doctrina Christiana stick closely to the Ramist format. Others spill outside it, as Milton departs from his stated goal – maximal scripture, minimal commentary – in favour of arguing his more controversial positions.

From De Doctrina Christiana’s uneven adherence to Ramist form, Kerr draws a broader conclusion, that ‘the treatise’s shift towards rhetoric indicates a crisis in its theology and methodology.’ As the work proceeded, Milton’s own commentary increasingly took over. As this change came about, Kerr claims, Milton reconceived the work of theology itself, moving from an emphasis on ‘external scripture’, or the biblical text, to ‘internal scripture’, or its human interpretation. According to the Ramist plan, interpretation was barely necessary: organise the Bible in the right logical form, line up your proof texts and the Word of God will speak for itself. But Milton came to realise by the late 1650s that this way of doing theology was inadequate. His treatise was getting more discursive and less scriptural, and so he abandoned it in favour of the new literary-theological project he had taken up around the same time: Paradise Lost.

These are interesting claims, because they amount to a new explanation for Milton stopping work on De Doctrina Christiana around 1660, long before his death in 1674. The usual assumption is that he abandoned the work because of the Restoration: with the return of the Stuart monarchy in 1660 came the return of the established English Church, and as a well-known opponent of both institutions, Milton was in no position to publish a heterodox treatise in England. He might have pursued publication anonymously and abroad, but that too would have carried risks. After a decade spent writing on behalf of the revolutionary cause, Milton had newly exposed himself in April 1660, on the eve of the Restoration, by publishing the second edition of a fierce republican tract entitled The Ready and Easy Way to Establish a Free Commonwealth. When the king returned in May Milton went into hiding. He was arrested and briefly imprisoned that autumn in the Tower. The experience might well have left him with a desire to avoid further trouble. He was ageing, blind and living in London – a sitting duck if the authorities reconsidered their forbearance towards him. His biographers Gordon Campbell and Thomas Corns point out that after the Restoration those associated with the Interregnum were liable to have their papers seized. Milton may have felt it too dangerous to keep the manuscript of De Doctrina Christiana in his possession, and sent it elsewhere for the manuscript’s safety and his own.

Kerr does not deny that the Restoration removed the possibility of the work being published. He proposes, however, that Milton abandoned it not only because he had to, but because ‘the increasing literariness making its way into the manuscript meant that the genre as he had initially conceived it was becoming less and less apt for the work he needed it to do.’ Kerr is certainly right that some parts of De Doctrina Christiana are less tightly structured than others. He may well be right that the work loosened up, becoming more diffuse, rhetorical and argumentative (this is what he means by ‘increasing literariness’). I am less convinced that Milton experienced this loosening up as a crisis of method that led him to abandon the work.

Unless something turns up, external evidence isn’t much help to Kerr’s argument. Our best source for Milton’s intentions regarding his treatise is its preface, although we don’t know when it was written or revised. It would work best for Kerr’s thesis if the preface had been written at an early stage. It could then be read as an initial design that Milton failed to accomplish, and this failure could then be construed as a reason for his abandoning the work. But it’s no less likely – likelier, if I had to guess – that the preface was written near the end of the 1650s, when Milton had largely completed De Doctrina Christiana and was getting ready to publish it. That is the easiest inference from the ‘if I now make this account public’ passage quoted above. That the preface’s stated goal – maximal scripture, minimal commentary – does not fully match what he wrote in the work itself need not indicate that it was written earlier; authorial prefaces and introductions misdescribe their books all the time, at whatever stage they’re written. Perhaps Milton was trying to present his treatise as tidier than it was. Perhaps he wanted to highlight its distinctiveness: my treatise is more rigorously scriptural than any other. Perhaps he emphasised its closeness to the Bible because he knew how unwelcome some of its conclusions would be and was aiming to pre-empt objections. Or perhaps he thought his description accurate enough; after all, it fits large portions of the manuscript.

The long chapter ‘On the Son of God’ looms large in Kerr’s argument, as it does in the manuscript itself. As Kerr observes, it includes a vestigial Ramist skeleton: a list of divine attributes, adapted from Milton’s precursors Wolleb and Ames. In Wolleb, these attributes – omniscience, supreme goodness, supreme glory and so on – are predicated as belonging to the triune God. Milton, by contrast, argues that they belong to the Father alone, with the Son subordinate to the Father. Much of the chapter’s argument is devoted to saying what the Son is not. It looks as though Milton began his work on this topic in conventional fashion, starting with divine attributes, as his more orthodox precursors had done, then wrote his way into heterodoxy.

But if Milton argued himself into heterodoxy, did he also argue himself out of systematics? Kerr writes that the chapter ‘shows the genre channelling Milton’s literary energies in a not particularly productive way’. Kerr is not the first to find it a bit of a slog. I’m not sure, however, that Milton would have agreed. As he dictated material to his scribes, he must have realised that the chapter was moving away from his ideal of ordered Bible verses with minimal commentary. So be it; this was where his Spirit-led inquiry had taken him. The orthodox trinitarian view would have to be opposed; given the importance and sensitivity of the subject, the argumentation was necessary. He was correcting a misunderstanding deeply embedded in Christian thought, one that the light of the Reformation had yet to reach. Careful philological work would be required on his part, and open-mindedness on the part of his readers. The task included an acknowledgment that there are questions scripture doesn’t allow us to answer: the way Christ’s mortal and divine natures fit together, for example. Difficult and obscure though the subject was, Milton believed he could do at least as well as any previous theologian who had approached it. As was his wont, he persevered. Where Kerr sees a crisis of method, I draw a less interesting conclusion: Chapter 1.5 shows that messy topics get messy, and the Son of God in Christian theology is a messy topic if ever there was one.

Kerr’s​ book is excellent on Milton’s non-institutional understanding of the church. He was an anticlerical, anti-hierarchical, individualistic godly Protestant. He belonged to no parish, sect or gathered church. He drew no distinction between laity and clergy. He opposed the public maintenance of ministers. He insisted on the need to think for yourself in religious matters. His DIY approach to doctrine was matched by his DIY approach to worship: he believed that you didn’t need a liturgy, or a church building, or a priest. Yet he had a notion of ecclesial community, if a loose one. As Kerr points out, Milton saw theology as a collective endeavour. Every believer has the right and obligation to work out the truths of Christian religion for themselves, but they should refine and test these beliefs in dialogue with fellow believers. No institutional authority should intervene to settle disagreements. Religious truth will be approached gradually by the broad community of believers – the church – who work it out together, through the study of scripture, free discussion and the guidance of the Spirit. This dialogue among the faithful, Milton thought, was the means by which the unfinished Reformation would progress. Its leaders would be free-thinking learned lay theologians like Milton himself. The process would require from its participants a willingness to listen to one another; it would require from the state a commitment to pan-Protestant toleration. The point of writing a theology treatise was to contribute to this dialogue. As Kerr puts it, ‘Milton undertook the inordinate labour of compiling the treatise not because he wanted to express his own theological views but because he wanted to contribute to the ecclesial process of discovering and giving voice to a truth larger than himself.’ Replace ‘not … but’ with ‘both … and’ and this sentence gets it right.

Kerr presents an appealing picture of Milton’s open-ended, communal approach to theology, one in which all inquirers – all Protestant inquirers, at least – count as fellow labourers in the vineyard, working together towards a truth that none will possess alone. I would add that this communitarian stance is one you take when trying to clear space for ideas you know are likely to be met with hostility. Milton often found himself in the position of writing from the fringe towards the mainstream of godly Protestant opinion, attempting to gain his readers’ trust before they threw his book across the room. It’s easy for modern intellectuals to miss this side of Milton, because we are usually incentivised to play up the novelty of our own ideas. Milton’s incentives pulled in the opposite direction: he held some alarming views and wanted to put them across as non-alarmingly as possible. Many of his rhetorical choices become intelligible in this light.