What qualifies?

Where should the process of compilation begin? To put it otherwise: must a creed be a synodical declaration, accredited by signatures and fortified by civil and ecclesiastical sanctions? If the answer were yes, the harvest would be lean. Kinzig instead follows KellyFootnote ³ and others in quoting, as antecedents to the first official creeds, a handful of passages from Ignatius of Antioch, all of which affirm the divinity of Jesus, in most cases with his birth from Mary and the Holy Spirit, his salvific death on the cross and his resurrection (i. 172–86). Texts of similar import are extracted from Aristides, Apelles, Polycarp and then in greater abundance from Justin Martyr, who is also found to offer frequent testimony to the worship of Father, Son and Holy Spirit, sometimes in conjunction with the angels (i. 192–200). Formulations of varying length are extracted from Irenaeus, while a number of Tertullian's works enunciate the faith of the Church in solemn and rhythmic clauses, with results that are close in diction and almost coextensive in content with the future Nicene Creed (i. 222–33). No two of his summaries, however, are identical: it rather seems that, like his Greek precursors, he makes a virtue of clothing his beliefs in a somewhat different form of words on each occasion. These texts are thus not creeds in the strict ecclesiastical sense, which implies (or at least is supposed to imply) a rigidity of phrasing which prevents any surreptitious change in the substance of what is taught. Such improvised confessions did not, of course, disappear from Christian literature as soon as the Church began to promulgate official creeds; we must therefore be careful not to treat them as though they were merely preparatory. They invite comparison, not so much with the products of laboured drafting that were to emanate from councils, as with the preacher's stock of formulae which enable him to convey the same fundamental truths in sermon after sermon without exact repetition or startling innovation. This tradition of credal extemporisation (as we may call itFootnote ⁴) never ceased to influence, or to be influenced by, the composition of formal creeds, any more than the oral ministry has ceased to inform, or be informed by, the culture of the pen.

Declaratory creeds did not put an end to the free permutation of formulae: why should they, when the Gospel of Nicodemus, the English mystery plays and innumerable lives of Christ in the modern era have all been inspired by the four canonical Gospels, and even these do not have quite the same content in all editions? The term ‘declaratory’ suggests an audience, yet in fact we seldom know by what means the promulgations (as we call them) of a council held in one place were made known in other quarters of the empire. Force might be given to them by an imperial decree, like that of Constantine against Arius, but if an emperor chose to ignore them, as Constantius ii ignored the Nicene decree against the translation of bishops, they might quickly become the province of antiquarians. We know of more than one instance in which the decree of a council, confidently invoked at a later gathering, merely gave rise to a new dispute regarding its provenance or its validity. As for the Nicene Creed itself, we have no attestation of its liturgical use, and we do not know how the typical bishop treated a member of his flock whom he judged to have fallen under its anathemas. Hilary of Poitiers, in a passage adduced by Kinzig,Footnote ⁵ says that the writing of creeds was a novelty of his own time; if true, this assertion permits us to believe that the ‘improvised confessions’ of which I spoke in the previous paragraphs were variations on a fixed but unwritten formula. The same Hilary of Poitiers, however, confesses his ignorance of the Nicene Creed some thirty years after it was not only written but signed by some 250 prelates; his admission of ignorance teaches us, if nothing else, that whereas historiography is often hindered by want of knowledge on our part, it is also hindered by our knowing more than it was then usual to know.

In 341 the Dedication Council of Antioch issued a declaration (the so-called ‘second creed’) which has repeatedly been compared with the Nicene Creed on the assumption that it must be intended either to confirm or to supersede it (i. 342–4). This is not so often assumed of the creed pronounced in his own defence by Theophronius on this occasion, or of the ‘doctrinal statement’ which occurs in a letter addressed by the council to Julius of Rome (i. 354–6). By parity of reasoning, we cannot be certain that the ‘second creed’ has any reason for omitting the word homoousios except that it happened not be pertinent to their deliberations at a time when it was not being held up, even by Athanasius, as the only diagnostic of a sound faith. We seldom hear of a creed that was expressly designed to overrule an earlier one: the proscription of the term ousia at Sirmium in 357 aroused unprecedented indignation because it was itself without precedent (i. 404-8). Elsewhere the principal function of the creed was to divide the assembled bishops into signatories and non-signatories, with the consequence that the latter could expect to be banished from their sees until the next turn of fortune. The synodal creed of this era is a test for the clergy, and only when the Nicene Creed had been incorporated into the liturgy as a symbol of lay consensus was it necessary to devise new instruments, the first and best-known of which is the Chalcedonian Definition.

Should this too be regarded as a creed, as it certainly is a conciliar resolution on a theological issue which could not have been settled otherwise? Since Kinzig does not treat it as such, we may guess that he sees it rather as a commentary of the kind that, according to his introduction, he has purposely excluded to spare us the addition of another volume. The prospect is all the more formidable because it is clear that one would not suffice: what is not so clear, however, is the criterion by which he has distinguished a creed from a commentary. He includes, for example, an excerpt from John Chrysostom, in which the bare affirmation of belief in the resurrection is accompanied by a gloss extending to four lines in his own style which has no parallel in other formularies. A later and longer document is a catena of propositions from the Tome of Leo, never designed (so far as we know) for subscription or recitation, and showing little resemblance to a creed except in its opening declaration, ‘I believe’. The Chalcedonian document lacks this exordium; on the other hand, it is justified in its own words as a necessary appendix to the Nicene Creed, innovation on which had been prohibited in 431 by the Council of Ephesus. To deflect the charge of temerity, the Chalcedonian Fathers recited not only the Nicene Creed but the Constantinopolitan creed of 381. It will be useful to set down some reflections on the genesis and content of both formulae, which might have been pre-empted had the Chalcedonian Definition been reproduced here in full.

What is the Nicene Creed?

In the long section entitled ‘The Council of Nicaea’, the Greek text of the Creed is preceded by an invaluable compilation of documents which throw light (or purport to do so) on the origin and proceedings of the council. In these there is much to ponder – above all in the assertion of Sozomen,Footnote ⁶ writing in the fifth century, that in deference to friends who urged him not to profane the mysteries, he has given only the gist of the creed, together with the more momentous of the anathemas. His reticence is even more complete than that of Cyril of Jerusalem, whose preaching allows us to reconstruct the formula that he would not recite in full. In Sozomen's time, however, when almost everyone who could read professed Christianity, we have all the more reason to ask how he could hope to conceal a document which Eusebius had published in a letter to his congregation a few months after the council, which Athanasius had then transcribed in a work addressed to heretics, and which Socrates, a contemporary of Sozomen, reproduced in full with no show of misgiving. Sozomen is a witness to the reception of the creed, though not to its genesis or to any disputed particulars of its content: if Kinzig's enterprise includes reception, one would have hoped for more commentary, but if it does not, one wonders what this unannotated text is designed to add to our understanding.

To the wording of the creed itself Kinzig lists sixteen witnesses (i. 291–2), with indications of date but no discussion of the sources on which they rely, and (more regrettably) no segregation of those who include and those who omit the anathema on the word ktiston, or ‘created’. As Maurice Wiles pointed out some twenty-five years ago (and he was not the first to notice it), the majority of the texts which include the anathema are dependent on Athanasius, who is ostensibly taking it word for word from Eusebius of Caesarea;Footnote ⁷ yet the transcript of the same letter in Theodoret (a writer of unblemished orthodoxy in his teaching on the Trinity) agrees with Athanasius in every clause but this.Footnote ⁸ It is also omitted by Basil of Caesarea and by Cyril of Alexandria, an unshakeable pillar of orthodoxy who repeatedly appeals to Athanasius as a touchstone of the true faith. The question is not trivial, for it was Athanasius who convinced the world, in the teeth of the ‘Homoiousians’ and the champions of Origen, that any attribution of creaturely status to the Son was a confession of Arianism. Wiles insinuates that Athanasius has fathered his own theology on the council: I have proposed the more charitable view that he was acquainted, not with the final text, but only with the draft submitted to Constantine by Alexander and Hosius, in which (if we can for once believe Philostorgius),Footnote ⁹ the use of the adjective ktiston was indeed anathematised. Whatever the truth may be, it behoves all scholars to remember that the word which is forbidden in the main body of the Creed is not ‘created’ but ‘made’ (poiêthenta), which was not applied by Arius to the Son.

Kinzig's sixteen witnesses do not include the Armenian Church, but he includes the Greek translation of the creed which this Church submitted to Proclus of Constantinople after the Council of Chalcedon (ii. 84–5); as he remarks, ‘it deviates from [the Nicene formula] in certain respects’ – most strikingly, in replacing ‘from the substance of the Father’ by ‘substance from substance, begotten substance’. Here again the question of reception is forced upon us, for the Armenians may not have been alone in tendering an augmented text in the belief that it was the original. Kinzig gives us to understand (i. 292, 513) that the formula which was recited at the fifth session of Chalcedon as the creed of the 318 Fathers at Nicaea was in substance identical with the creed which they go on to attribute to the 150 at Constantinople in 381; the latter, as he reminds us, is the first full recitation of the text which is now known to most Churches as the Nicene Creed (iv. 511). But had the learned clergy of the fifth century already ceased to distinguish the creed of 325 from its Constantinopolitan successor? If that were the case, it is hard to account for the reproduction of both in the Chalcedonian Definition, for the stated purpose of this is to show that, while a council has no right to innovate on the substance of the Nicene Creed, it may lawfully canonise new formulations to preserve the intention of those who framed it against a heresy which they had not foreseen. Moreover, as Richard Price and Michael Gaddis have remarked in their translation of the Acts of Chalcedon,Footnote ¹⁰ certain manuscripts give an augmented version of the Nicene Creed which is not yet identical with that of 381. A work such as the present one, which will undoubtedly (and deservedly) supersede all other textbooks, is in danger of misleading students if it fails even to give them notice of this controversy.

What then of the second creed recited at Chalcedon? The present consensus (for which the principal evidence is the lack of any evidence against it) is that it was indeed promulgated at Constantinople in 381 as a new redaction of the Nicene Creed, but following precedents for the expansion of this in Epiphanius and Cyril of Jerusalem.Footnote ¹¹ The chief additions were articles endorsing the pneumatology of Basil the Great, rejecting the eschatology of Marcellus and reaffirming the Virgin Birth against the putative docetism of Apollinarius. The consensus further holds (and Kinzig agrees) that there is no evidence for knowledge, let alone for approbation, of this new shibboleth before it was quoted in the Chalcedonian Definition of 451. This purported argument from silence, however, ignores the explicit statement of Nestorius that the Fathers at Nicaea had declared Christ to be the son of the Virgin MaryFootnote ¹² – an error which would be explicable if he mistook the version current in his adopted city for the original creed of 325. Cyril's exact transcription of the Nicene Creed of 325 in his third letter to Nestorius would not, on this hypothesis, have been a superfluous insult but a proof that Alexandria had the advantage both in piety and in learning. The same theory will explain why Leo, in his Tome to Flavian of 449, asserts that the ‘creed’ describes the saviour both as ‘God from God’ and as son of the Virgin, although the first of these clauses is lacking in the Apostles’ Creed and the second in that of Nicaea.Footnote ¹³ Both are found in the creed which is supposed to have been proclaimed in 381 in Constantinople, and it is surely more probable that by ‘creed’ the pontiff meant one document rather than a conflation of two. If this is so, his appeal to the creed of 325 in his second Tome was a silent recantation of an error which was all the more heinous in one whose office forbade him to grant authority to the council of 381.

Leo might also have come upon the same combination of tenets in a Latin version of the Nicene Creed which was endorsed in 393 at the Council of Hippo (i. 299) – the same occasion on which the African Church proclaimed the whole of the Septuagint canonical. This is one of forty-six renderings, collated with remarkable diligence (i. 296–335). By contrast, as we have noted, Greek witnesses to the Creed of 325 are simply listed without quotation. The reasons for this asymmetry are easily divined: the Greek texts are regarded as variants of a single archetype, while the Latin texts are all equally original, none being a mere derivative or transcription of another. While, therefore, their discrepancies wither the hopes of anyone hoping to establish the most primitive form of the Greek, they are all of independent interest to the student of reception; once again, however, we might ask why a similar case could not be made for reproducing all the Greek versions, or at least as many as would be necessary to illustrate all the divergences in the tradition. A counsel of perfection would recommend also the inclusion of a Latin translation of one of the letters of Arius in Marius Victorinus,Footnote ¹⁴ for this epitome of the rejected doctrine will have shaped and assisted the understanding of the Nicene norm.

The prism of history

Readers will find themselves pondering the relation between the history of reception and the history of promulgation once again when they peruse the remarkably copious, though avowedly selective, inventory of testimonia to the meaning of the term symbolum (i. 68–144), applied in both Greek and Latin to the Nicene Creed and subsequent formularies. The vast majority of relevant passages (all but one, in fact [i. 144]) are culled from Latin sources, while the earlier citations from Clement and Basil (i. 64, 65, 68) do not have the sense of ‘creed’. We are assured on p. 144 that sumbolon tês pisteos, ‘symbol of the faith’, is a frequent locution in Cyril of Alexandria; but why would it not be sufficient to say that symbolum is a common term for the creed in Latin writers from the time of the Council of Arles in 314? The catalogue extends to authors of the eighth century, none of whose opinions (if they were ever new) would be any guide to the reasons for the initial use of the word. As documents of reception they bear witness, not to any evolution in the use and interpretation of the creeds, but to a perpetual cacophony. Isidore of Seville, for instance, admits three senses – a token, a collection and a watchword – while Theodulf of Orleans, treating the first two with approbation as though they were synonyms, confesses that in his own time (the early ninth century), a symbolum was more commonly defined as an abridgment of scriptural teaching (i. 124–5). The last is not mentioned in Kinzig's introduction, where he enumerates eight senses in which a creed was understood to be a sumbolon: as a token of mutual recognition, a token of acquaintance with the content of Christian teaching, a token of orthodox faith, a summary of the said faith, a reminder and confession of this, a contract with other believers, a contract with God and a sign or symbol (not, it would seem, an abbreviation) of holy Scripture (i. 6). What place should be assigned then to the adversaries of Theodulf, either in a genealogy of the creeds or in a history of their diffusion in the West?

The data offered by Kinzig raise a methodological point of some importance. If, as it seems, the first Latin use of symbolum to designate a creed antedates the first Greek usage (here ascribed to Cyril of Alexandria at i. 144), should we assume that priority in fact belongs to the Latin Church, notwithstanding the provenance of the word itself? That Greek should follow Latin was not impossible in antiquity: the Trinitarian use of prosopon, attested first (and uniquely before Nicaea) in Hippolytus of Rome, is best explained as a calque on the term persona, which Tertullian could already assume to be a known term to his readers. But here we have a Latin word at the fountainhead, and one for which Greek had already been obliged to adopt prosopon as a translation in legal parlance; for Latin to adopt the Greek noun sumbolon in a sense that was not yet attested in Greek, and for the Greeks to borrow this usage a century later, would be far more anomalous. Kinzig seems inclined to hold that the Greeks were the first to give sumbolon the sense of ‘creed’, and most would agree that this is a case in which chance remains can be supplemented by rational conjecture. In 1999 he himself, together with Markus Vinzent, took a different view with regard to the Apostle's Creed, a text that is publicly recited only in western Churches, yet attested for the first time (and the only time in Greek) as a statement of orthodoxy presented by Marcellus of Ancyra to Julius of Rome in 341.Footnote ¹⁵ It was half a century later that Rufinus of Aquileia transcribed and commented on the formula which he professed to have received by oral instruction; this testimony reinforces Kinzig's own contention that no written creeds were employed in the West before the late fourth century. Nevertheless, the prevalent opinion up to 1999 was that the Romans would not have adopted as their own creed the confession of a suspect bishop whose faith they already had some means of testing, and that the absence of Greek congeners proves that Marcellus was purposely adopting the creed which he knew to be in use at Rome. From his footnote at Faith in formulae, i. 12 it would seem that Kinzig now accepts this reasoning, and with it (no doubt) the methodological principle that the rubble of ancient times is not a jigsaw from which everything has survived except the picture on the box.

The editor's task

Annotation is always too occasional for some, too officious for others. The editorial policy in these volumes is to refrain from it, except for the provision of an ample bibliography at the beginning of each new section. For the most part nothing is lost, because the reproduction of primary texts on such a scale provides us with information which would usually be gleaned only (if at all) from the small print of a commentary. Thus, if we are sufficiently erudite to frame the question, ‘When did Christ's descent to the underworld become an article of the Apostles’ Creed?’, it will not take long to discover first attestations in two texts of the early sixth century, one certainly written by, and the other attributed to, Caesarius of Arles (ii. 259–63). On the other hand, for those who were not aware that this is a question to be asked, some indication of the significance of these texts would have conferred an important piece of knowledge at little expense of space. Caesarius is also thought by some to have had the chief hand in the composition of the Athanasian Creed, which, despite this sobriquet, is clearly of western provenance, and, as Kinzig observes, is anything but a typical creed in form (iv. 1). Nevertheless we may feel that its position in the western Church, where it functioned for many centuries as a third creed, is not adequately acknowledged in the fourth volume, where translations into Greek and Hebrew eke out the Latin text and the long bibliography to fill up a modest quota of nine pages (iv. 1–9). A little lengthening of the introduction would have permitted some incipient speculation on its authorship, or at least some canvassing of its roots in Ambrose and Augustine, which are unlikely to be denied.

The most necessary parentheses to any creed are biblical references, which Kinzig supplies with frugal precision. In one instance, however, I feel the need of a supplement not only to his own but to all previous annotation. At volume i. 428 he reproduces the creed which certain Fathers, according to Basil, had accepted in the simplicity of their hearts without perceiving that it lent colour to the blasphemies of Eunomius by assigning first place to the Father, from whom are all things, the second to the Son, from whom are all things, and the third to the Spirit, unto whom are all things.Footnote ¹⁶ Kinzig rightly notes the juxtaposition at 1 Corinthians viii.4 of the ‘one God, from whom are all things’, and the ‘one Lord Jesus Christ, through whom are all things’; as Basil observes, however, the only text in which the three prepositions occur together is Romans xi.36, where God (or ‘the Lord’ is the common subject.Footnote ¹⁷ If this ramification of biblical syntax is a creed, we must surely say as much of passages in early Christian writers which attach different verbs to each participant in the divine economy. Thus Irenaeus writes that first the Spirit prepares humanity in the Son, the Son then leads us to the Father, and the Father bestows the gift of incorruption (Against heresies 4.20.5). In Origen, that which is energised by the Father and ministered by the Son is substantiated by the Spirit (Commentary on John ii.10.77); when Gregory of Nyssa asserts that all divine acts commence with the Father, are energised through the Son and are consummated by the Spirit,Footnote ¹⁸ he adapts their extemporised triads to the vindication of Nicene orthodoxy. Chronology excludes this private letter from Kinzig's remit, but Gregory's imitation of Origen's dictum suggests that for him it encapsulated the received faith of the Church.

But of course the greater the enterprise, the more diverse will be the proposals for its execution. Scholars may ruminate freely on what they might have done otherwise without belittling the excellence of Kinzig's scholarship or the utility of this work for students of early Christianity. In scope there is nothing to match it, and it makes many texts available for the first time to Anglophone readers in translations which are invariably accurate and lucid. We shall still go to Bindley for full texts (with translation and notes) of Leo's Tome and the letters of Cyril, to Heurtley for a convenient edition (without translation) of the commentary of Rufinus on the Apostles’ Creed,Footnote ¹⁹ and to Kelly for an account of the ferment of creed-making in the fourth century. None of these authorities, however, tells us anything of the reappropriation and revision of these formulae in Byzantium, Britain, Spain or the Frankish territories, but for which the Church would not have survived the fall of Rome and Rome herself would have perished without an echo. Whether or not it is true that ‘creeds are what make Christians tick’, as Kinzig asserts in his introduction, they combine with preaching, hymnody and Scripture to shape the practice and belief of the Christian laity. The clergy composed, the laity recited, and so long as this antiphony remained unbroken, ‘Christendom’ was more than a figure of speech.