In the beginning of the fourth book Hilary refers back to the proof furnished in the previous books, written some time ago, of the Scriptural character of his faith and of the unscriptural nature of all the heresies. Setting aside the first book, which does not correspond to this description, we find what he describes in the second and third. These form a short connected treatise, complete in itself. It is much more academic than that of which we have already spoken; it deals briefly with all the current heresies (ii. 4 ff.), but shews no sign that one of them, more than the others, was an urgent danger. There is none of the passion of conflict; Hilary is in the mood for rhetoric, and makes the most of his opportunities. He expatiates, for instance, on the greatness of his theme (ii. 5), harps almost to excess upon the fisherman to whom mysteries so great were revealed (ii. 13 ff.), dilates, after the manner of a sermon, upon the condescension and the glory manifested in the Incarnation, describes miracles with much liveliness of detail (iii. 5, 20), and ends the treatise (iii. 24-26) with a nobly eloquent statement of the paradox of wisdom which is folly and folly which is wisdom, and of faith as the only means of knowing God. The little work, though it deals professedly with certain heresies, is in the main constructive. It contains far more of positive assertion of the truth, without reference to opponents, than it does of criticism of their views. In sustained calmness of tone--it recognises the existence of honest doubt (iii. 1),--and in literary workmanship, it excels any other part of the De Trinitate and in the latter respect is certainly superior to the more conversational Homilies on the Psalms. But it suffers, in comparison with the books which follow, by a certain want of intensity; the reader feels that it was written, in one sense, for the sake of writing it, and written, in another sense, for purposes of general utility. It is not, as later portions of the work were, forged as a weapon for use in a conflict of life and death. Yet, standing as it does, at the beginning of the whole great treatise, it serves admirably as an introduction. It is clear, convincing and interesting, and its eloquent peroration carries the reader on to the central portion of the work, which begins with the fourth book. Except that the second book has lost its exordium, for the same reason that the seventh has lost its conclusion, the two books are complete as well as homogeneous. Of the date nothing definite can be said. There is no sign of any special interest in Arianism; and Hilary's leisure for a paper conflict with a dead foe like Ebionism suggests that he was writing before the strife had reached Gaul. The general tone of the two books is quite consistent with this; and we may regard it as more probable than not that they were composed before the exile; whether they were published at the time as a separate treatise, or laid on one side for a while, cannot be known; the former supposition is the more reasonable.
The remaining books, from the eighth to the twelfth, appear to have been written continuously, with a view to their forming part of the present connected whole. They were, no doubt, published separately, and they, with books iv. to vii., may well be the letters (stripped, of course, in their permanent shape of their epistolary accessories) which, Hilary feared, were obtaining no recognition from his friends in Gaul. The last five have certain references back to arguments in previous books [78] , while these do not refer forward, nor do the groups ii. iii. and iv.-vii. refer to one another. But books viii.-xii. have also internal references, and promise that a subject shall be fully treated in due course [79] . We may therefore assume that, when he began to write book viii., Hilary had already determined to make use of his previous minor works, and that he now proceeded to complete his task with constant reference to these. Evidences of exact date are here again lacking; he writes as a bishop and as an exile [80] , and under a most pressing necessity. The preface to book viii., with its description of the dangers of the time and of Hilary's sense of the duty of a bishop, seems to represent the state of mind in which he resolved to construct the present De Trinitate. It is too emphatic for a mere transition from one step in a continuous discussion to another. Regarding these last five books, then, as written continuously, with one purpose and with one theological outlook, we may fix an approximate date for them by two considerations. They shew, in books ix. and x., that he was thoroughly conscious of the increasing peril of Apollinarianism. They shew also, by their silence, that he had determined to ignore what was one of the most obvious and certainly the most offensive of the current modes of thought. There is no refutation, except implicitly, and no mention of Anomoeanism, that extreme Arianism which pronounced the Son unlike the Father [81] . This can be explained only in one way. We have seen that Hilary thinks Arianism worth attack because it is an 'inward evil;' that he does not, except in early and leisurely work such as book ii., pay any attention to heresies which were obviously outside the Church and had an organization of their own. We have seen also that the Homoeans cast out their more holiest Anomoean brethren in 359. The latter made no attempt to retrieve their position within the church; they proceeded to establish a Church of their own, which was, so they protested, the true one. It was under Jovian (a.d. 362-363) that they consecrated their own bishop for Constantinople [82] ; but the separation must have been visible for some time before that decisive step was taken. Thus, when the De Trinitate took its present form, Apollinarianism was risen above the Church's horizon and Anomoeanism was sunk below it. We cannot, therefore, put the completion of the work earlier shall the remission of Hilary's exile; we cannot, indeed, suppose that he had leisure to make it perfect except in his home. Yet the work must have been for the most part finished before its writer reached Italy on his return; and the issue or reissue of its several portions was a natural, and certainly a powerful, measure towards the end which he had at heart.
There remains the first book, which was obviously, as Erasmus saw, the last to be composed. It is a survey of the accomplished task, beginning with that account of Hilary's spiritual birth and growth which has already been mentioned. This is a piece of writing which it is no undue praise to rank, for dignity and felicity of language, among the noblest examples of Roman eloquence. Hooker, among English authors, is the one whom it most suggests. Then there follows a brief summary of the argument of the successive books, and a prayer for the success of the work. This reads, and perhaps it was meant to read, as though it were a prayer that he might worthily execute a plan which as yet existed only in his brain; but it may also be interpreted, in the more natural sense, as a petition that his hope might not be frustrated, and that his book might appear to others what he trusted, in his own mind, that it was, true to Scripture, sound in logic, and written with that lofty gravity which befitted the greatness of his theme.
After speaking of the construction of the work, as Hilary framed it, something must be said of certain interpolations which it has suffered. The most important are those at the end of book ix. and in x. 8, which flatly contradict his teaching [83] . They are obvious intrusions, imperfectly attested by manuscript authority, and condemned by their own character. Hilary was not the writer to stultify himself and confuse his readers by so clumsy a device as that of appending a bald denial of its truth to a long and careful exposition of his characteristic doctrine. Another passage, where the scholarship seems to indicate the work of an inferior hand, is Trin. x. 40, in which there is a singular misunderstanding of the Greek Testament [84] . The writer must have known Greek, for no manuscript of the Latin Bible would have suggested his mistake, and therefore he must have written in early days. It is even possible that Hilary himself was, for once, at fault in his scholarship. Yet, at the most, the interpolations are few and, where they seriously affect the sense, are easily detected [85] . Not many authors of antiquity have escaped so lightly in this respect as Hilary.
Hilary certainly intended his work to be regarded as a whole; as a treatise Concerning the Faith, for it had grown into something more than a refutation of Arianism. He has carefully avoided, so far as the circumstances of the time and the composite character of the treatise would allow him, any allusion to names and events of temporary interest; there is, in fact, nothing more definite than a repetition of the wish expressed in the Second Epistle to Constantius, that it were possible to recur to the Baptismal formula as the authoritative statement of the Faith [86] . It is not, like the De Synodis; written with a diplomatic purpose; it is, though cast inevitably in a controversial form, a statement of permanent truths. This has involved the sacrifice of much that would have been of immediate service, and deprived the book of a great part of its value as a weapon in the conflicts of the day. But we can see, by the selection he made of a document to controvert, that Hilary's choice was deliberate. It was no recent creed, no confession to which any existing body of partisans was pledged. He chose for refutation the Epistle of Arius to Alexander, written almost forty years ago and destitute, it must have seemed, of any but an historical interest. And it was no extreme statement of the Arian position. This Epistle was far more temperate and cautious [87] ' than its alternative, Arius' letter to Eusebius. The same wide outlook as is manifest in this indifference to the interests of the moment is seen also in Hilary's silence in regard to the names of friends and foes. Marcellus, Apollinaris, Eudoxius, Acacius are a few of those whom it must have seemed that he would do well to renounce as imagined friends who brought his cause discredit, or bitter enemies to truth and its advocates. But here also he refrains; no names are mentioned except those of men whose heresies were already the commonplaces of controversy. And there is also an absolute silence concerning the feuds and alliances of the day. No notice is taken of the loyalty of living confessors or the approximation to truth of well-meaning waverers. The book contains no sign that it has any but a general object; it is, as far as possible, an impersonal refutation of error and statement of truth.
This was the deliberate purpose of Hilary, and he had certainly counted its cost in immediate popularity and success. For though, as we have seen, the work did produce, as it deserved, a considerable effect at the time of its publication, it has remained ever since, in spite of all its merits, in a certain obscurity. There can be no doubt that this is largely due to the Mezentian union with such a document as Arius' Epistle to Alexander of the decisively important section of the De Trinitate. The books in which that Epistle is controverted were those of vital interest for the age; and the method which Hilary's plan constrained him to adopt was such as to invite younger theologians to compete with him. Future generations could not be satisfied with his presentation of the case. And again, his plan of refuting the Arian document point by point [88] , contrasting as it does with the free course of his thought in the earlier and later books, tends to repel the reader. The fourth book proves from certain texts that the Son is God; the fifth from the same texts that He is true God. Hence this part of the treatise is pervaded by a certain monotony; a cumulative impression is produced by our being led forward again and again along successive lines of argument to the same point, beyond which we make no progress till the last proof is stated. The work is admirably and convincingly done, but we are glad to hear the last of the Epistle of Arius to Alexander, and accompany Hilary in a less embarrassed enquiry.
Yet the whole work has defects of its own. It is burdened with much repetition; subjects, especially, which have been treated in books ii. and iii. are discussed again at great length in later books [89] . The frequent stress laid upon the infinity of God, the limitations of human speech and knowledge, the consequent incompleteness of the argument from analogy, the humility necessary when dealing with infinities apparently opposed [90] , though it adds to the solemnity of the writer's tone and was doubtless necessary when the work was published in parts, becomes somewhat tedious in the course of a continuous reading. And something must here be said of the peculiarities of style. We saw that in places, as for instance in the beginning of the De Trinitate, Hilary can rise to a singularly lofty eloquence. This eloquence is not merely the unstudied utterance of an earnest faith, but the expression given to it by one whom natural talent and careful training had made a master of literary form. Yet, since his training was that of an age whose standard of taste was far from classical purity, much that must have seemed to him and to his contemporaries to be admirably effective can excite no admiration now. He prays, at the end of the first book, that his diction may be worthy of his theme, and doubtless his effort was as sincere as his prayer. Had there been less effort, there would certainly, in the judgment of a modern reader, have been more success. But he could not foresee the future, and ingenious affectations such as occur at the end of book viii. § 1, impietati insolenti, et insolentiæ vaniloquæ, et vaniloquio seducenti, with the jingle of rhymes which follows, are too frequent for our taste in his pages [91] . Sometimes we find purple patches which remind us of the rhetoric of Apuleius [92] ; sometimes an excessive display of symmetry and antithesis, which suggests to us St. Cyprian at his worst. Yet Cyprian had the excuse that all his writings are short occasional papers written for immediate effect; neither he, nor any Latin Christian before Hilary, had ventured to construct a great treatise of theology, intended to influence future ages as well as the present. Another excessive development of rhetoric is the abuse of apostrophe, which Hilary sometimes rides almost to death, as in his addresses to the Fisherman, St. John, in the second book [93] . These blemishes, however, do not seriously affect his intelligibility. He has earned, in this as in greater matters, an unhappy reputation for obscurity, which he has, to a certain extent, deserved. His other writings, even the Commentary on St. Matthew, are free from the involved language which sometimes makes the De Trinitate hard to understand, and often hard to read with pleasure. When Hilary was appealing to the Emperor, or addressing his own flock, as in the Homilies on the Psalms, he has command of a style which is always clear, stately on occasion, never weak or bald; in these cases he resisted, or did not feel, the temptation to use the resources of his rhetoric. These, unfortunately, had for their result the production of sentences which are often marvels of grammatical contortion and elliptical ingenuity. Yet such sentences, though numerous, are of few and uniform types. In Hilary's case, as in that of Tertullian, familiarity makes the reader so accustomed to them that he instinctively expects their recurrence; and, at their worst, they are never actual breaches of the laws of the language. A translator can hardly be an impartial judge in this matter, for constantly, in passages where the sense is perfectly clear, the ingenuity with which words and constructions are arranged makes it almost impossible to render their meaning in idiomatic terms. One can translate him out of Latin, but not into English. In this he resembles one of the many styles of St. Augustine. There are passages in the De Trinitate, for instance viii. 27, 28, which it would seem that Augustine had deliberately imitated; a course natural enough in the case of one who was deeply indebted to his predecessor's thought, and must have looked with reverence upon the great pioneer of systematic theology in the Latin tongue. But this involution of style, irritating as it sometimes is, has the compensating advantage that it keeps the reader constantly on the alert. He cannot skim these pages in the comfortable delusion that he is following the course of thought without an effort.