The Trinity in the Creeds

Around the year 318 a presbyter of the church in Alexandria named Arius fell out with his bishop, Alexander, over how to describe the Son's relation to the Father. Arius is the most caricatured figure in Christian history, and the caricature makes the whole controversy unintelligible, so start with what he actually taught. His own words survive in two letters, one to Eusebius of Nicomedia and one to Alexander, and in fragments of a work called the Thalia quoted by his opponents.

Arius held that God is one, unoriginate and alone unbegotten, and that being unbegotten is what it is to be God in the strict sense. The Son is begotten, and on Arius's account whatever is begotten has a beginning. So the Son was brought into existence by the Father's will before all ages and before time, and through him everything else was made. The Son is therefore not a creature like other creatures: he is before them all, the agent of their making, the object of the Father's love, and to be honoured far above them. But he is not the Father's own being. The slogan that got everyone's attention was that there was when he was not, and Arius said in his letter to Alexander that the Son came into existence out of nothing, or at least not out of the Father's substance.

Notice what drove that. Arius was not a sceptic looking for a way to lower Christ. He was, as Rowan Williams argued in his 1987 study of him, a conservative Alexandrian academic protecting what he took to be the church's most basic conviction: that God alone is unoriginate, and that anything sharing the divine substance by generation would divide it or make it material. His problem was how the utterly transcendent could be the Father of anyone at all. That is a serious problem and his answer was a bad one, but the problem was real and the people who defeated him had to answer it too.

The second correction is larger. There was no single Arian party. R. P. C. Hanson's history of the controversy, published in 1988, and Lewis Ayres's work on the fourth century have made it standard that 'Arianism' is largely a polemical construction of the Nicene side, and that the actual landscape had at least four positions. The Nicenes held that the Son is homoousios, of the same substance, with the Father. The Homoiousians, associated with Basil of Ancyra, held that the Son is of like substance, homoiousios, which preserved the distinction of persons while resisting what they feared was a Sabellian collapse in the Nicene word; Athanasius eventually recognised them as allies and said so in his work on the councils. The Homoians held simply that the Son is like the Father according to the scriptures, refusing substance language altogether as unbiblical; this was the position of the creeds of Rimini and Seleucia in 359 and of Constantinople in 360, and for a generation it was the official faith of the empire. The Heteroousians or Anomoians, led by Aetius and then by Eunomius, held that the Son is unlike the Father in substance, and Eunomius further claimed that the divine essence is fully knowable and is precisely unbegottenness, which is the claim Gregory of Nyssa and Basil answered in the way the first lesson described.

One more party has to be in the picture or the story is unbalanced, because the Nicenes had a problem of their own. Marcellus of Ancyra was a firm supporter of Nicaea whose account of the Godhead looked to many like a denial of any real distinction, with the Word going out for the work of salvation and returning into God at the end. He was condemned in the East and defended in the West for years. The clause in the creed of 381 that the Son's kingdom shall have no end was aimed at him. So the fourth century was not a contest between orthodoxy and one heresy. It was a long, confused search for language that would exclude both the collapse of the three into one and the reduction of the Son to a creature, conducted by people who kept suspecting each other of the opposite error.

Constantine convened the council at Nicaea in Bithynia in 325, and something over two hundred and fifty bishops attended, traditionally numbered at three hundred and eighteen. The creed it produced said that the Son is begotten of the Father, only begotten, that is, from the substance of the Father, God from God, light from light, true God from true God, begotten not made, of one substance with the Father, homoousios. It then anathematised, in explicit terms, those who say that there was when he was not, that before being begotten he was not, that he came to be out of nothing, and that he is of another hypostasis or ousia.

The word homoousios was a scandal to many who were not Arians, and the reasons should be understood rather than waved away. It is not in the Bible. It had been used by Gnostic writers. It had reportedly been rejected at a synod at Antioch in 268 in connection with the condemnation of Paul of Samosata. It could be heard as saying that Father and Son are two portions of one material stuff, or as saying that they are one and the same with no distinction, which is Sabellianism. Eusebius of Caesarea signed and then wrote to his own church explaining, somewhat uncomfortably, the sense in which he had understood it. The council chose a non-scriptural word precisely because every scriptural word had already been shown to be readable both ways, which is the honest and slightly uncomfortable lesson in it.

Then the thing everyone forgets. Nicaea did not settle the matter. What followed was fifty-six years of councils, rival creeds, imperial intervention, depositions and exiles, and for much of that period the Nicene party was the losing side. Athanasius, who became bishop of Alexandria in 328, was driven out five times and spent something like seventeen years in exile. Jerome, looking back at the aftermath of Rimini in 359, wrote the line that has stuck: the whole world groaned and marvelled to find itself Arian. Anyone who imagines that a council settles a doctrine by voting should sit with that sentence.

Part of what broke the deadlock was a terminological repair. At Nicaea, ousia and hypostasis had been treated as synonyms, which is why the anathema condemns anyone saying the Son is of another hypostasis. But the Greek East had come to want a word for what the three are severally, distinct from the word for what they are jointly. In 362 a council at Alexandria, with Athanasius presiding, issued the document known as the Tome to the Antiochenes, which examined two rival groups and found that those who said one hypostasis and those who said three meant the same thing by different words, and ruled that both formulas could be used orthodoxly. It is one of the most quietly important documents in Christian history, and its method, of asking what a formula is being used to deny before condemning it, is worth more than most of what is said about ecumenism.

The council of Constantinople in 381, convened by the emperor Theodosius, is where the creed most churches recite on Sundays comes from, and two facts about it deserve stating. Only about a hundred and fifty bishops attended and none from the West, so the council that gave the church its most universal creed was not at the time a universal gathering. And the text we call the Niceno-Constantinopolitan Creed first appears in the acts of the council of Chalcedon in 451, which is why scholars including J. N. D. Kelly have long debated exactly how it relates to what was done in 381. Its authority in the church does not rest on settling that question, but a student should not be told a tidier story than the evidence supports.

What 381 added is mostly the third article. The Nicene creed of 325 had trailed off with 'and in the Holy Spirit', which was no longer adequate, because a party the Nicenes called the Pneumatomachi, the Spirit-fighters, were affirming the Son's deity while denying the Spirit's. The expanded article says that the Spirit is the Lord and the giver of life, who proceeds from the Father, who with the Father and the Son together is worshipped and glorified, who spoke by the prophets. Note what it does not say. It does not call the Spirit homoousios. Basil of Caesarea, in his treatise On the Holy Spirit, had deliberately argued from the church's worship rather than from the contested term: the Spirit is glorified with the Father and the Son in the church's own doxology, and what is worshipped is God. The creed follows that strategy. It is an early and striking case of doctrine being settled from the practice of prayer.

The settlement in the East is owed above all to three Cappadocians: Basil of Caesarea, his friend Gregory of Nazianzus, and his brother Gregory of Nyssa, with their sister Macrina in the background of the family's theology. Their formula is one ousia in three hypostaseis. Ousia is what God is, the one divine being; hypostasis is each of the three who is that being, distinctly and really. The word hypostasis, which in ordinary Greek meant something like underlying reality or substance, was pressed into a new technical job, and everyone knew it was being pressed.

What distinguishes the three, on this account, is nothing in the divine nature, since they share it wholly, but the relations of origin alone. Gregory of Nazianzus states it compactly in his fifth theological oration: the Father is unbegotten, the Son is begotten, and the Spirit proceeds. Those are not degrees of divinity or differences of power; they are modes of existing, tropos hyparxeos, and every attribute of God belongs equally and wholly to each. Notice the structure this produces: the Father is the source, the arche, from whom the Son is eternally begotten and the Spirit eternally proceeds, and this monarchy of the Father is what secures that there is one God rather than three. The unity is located in the Father as fountain rather than in an abstract shared essence.

Now the Latin problem, which caused real and avoidable damage. Latin had taken substantia as the natural rendering of hypostasis, since both words mean, literally, what stands under. So when Greeks said three hypostaseis, Latin ears heard three substances, which is tritheism. Jerome wrote to Pope Damasus in some distress about being asked in Syria to confess three hypostases, complaining that the whole school of secular letters knows hypostasis as nothing but ousia. The Latin word that did the work instead was persona, which Tertullian had already used around the year 200 when he wrote of one substance and three persons and gave the West the word trinitas. But persona had its own liability: it meant a mask worn in the theatre and a party in a lawsuit, so Greeks could hear it as three roles played by one actor, which is Sabellianism. Each side's best word sounded to the other like the heresy each most feared. A student who grasps that will be slower to assume that a formula he dislikes means what it sounds like it means in his own idiom.

Augustine's De Trinitate, written over roughly twenty years from the late 390s, is the most influential single book on this subject in the Western tradition, and it is more careful than its reputation. He is frank that the vocabulary is inadequate: at book five he says the three are called persons not in order to say something but in order not to be silent. He works from the unity of God and asks what in creation could serve as an image of a threefold God, and finds his best material where scripture told him to look, in the human mind made in God's image. In book nine he offers the mind, its knowledge of itself and its love of itself; in book ten, memory, understanding and will. These are the psychological analogies. Two things must be said about them. Augustine does not offer them as proofs or as explanations; he offers them as traces, and he dismantles each one himself before moving on. And the trajectory of the whole work is towards the conclusion that the image is found not in the mind's structure as such but in the mind remembering, understanding and loving God, which is a considerably less abstract conclusion than the analogies alone suggest.

The Eastern reservations about this are longstanding and they deserve to be heard in their own voice rather than paraphrased into a Western frame. The charge is that beginning from the one essence and then looking for threeness inside a single mind produces a God who is one subject with three faculties, and that this is modalism with better manners. Vladimir Lossky argued that the Western order of treatment, essence first and persons second, makes the persons into relations within a prior divine substance rather than three who genuinely are. John Zizioulas, in Being as Communion in 1985, developed the positive Eastern alternative: being is communion, personhood is prior to substance, and the Father's free begetting and spirating is what constitutes the divine being, so that God's unity is personal rather than essential. On this account the West secured the oneness at the cost of the threeness, and the practical consequence is a Western piety in which the Trinity is an article to be believed rather than the shape of the Christian's actual relation to God.

There is now a substantial scholarly qualification to that story and honesty requires reporting it. The contrast itself, Greeks starting from the persons and Latins from the essence, was given its classic form by the French Jesuit Theodore de Regnon in 1892, and it has been heavily criticised in the last thirty years. Michel Rene Barnes's article reconsidering de Regnon, published in 1995, argued that the paradigm was read back into the Fathers rather than found in them. Lewis Ayres's Augustine and the Trinity in 2010 argued in detail that Augustine is a pro-Nicene theologian of essentially the same kind as the Cappadocians and that the standard account of his method is a distortion. Sarah Coakley has argued that Gregory of Nyssa is not the social trinitarian modern readers want him to be. It should be added that this correction has been welcomed by some Orthodox theologians too, and that Lossky's sharply drawn contrast is not the only Orthodox reading available. The Orthodox rejoinder to all of it is that the correction is historiographical and leaves the dogmatic question untouched: whether the Father's monarchy is in fact preserved by a given account of the Godhead is a question that would still have to be answered if de Regnon had never written a word, and the East's answer to it has not changed. So there are two things to hold: the Eastern theological objection is a real and continuing one, made by serious people for reasons internal to their tradition, and the historical narrative in which it is usually framed is now widely judged to be too neat.

One term belongs to both traditions and is regularly misused. Perichoresis names the mutual indwelling of the three: each is wholly in the others, so that there are not three separate centres standing side by side. Its scriptural basis is John 14:10-11, the Father in the Son and the Son in the Father, and John 17:21. It was used first of the two natures in Christ, by Gregory of Nazianzus and then Maximus the Confessor, and John of Damascus in the eighth century applied it systematically to the Trinity. The Latins rendered it circumincessio, with a variant circuminsessio, and the pair of Latin words preserves two shades, moving around one another and resting within one another. Now the misuse. Perichoresis is very often glossed as divine dance, and the etymology is wrong: the word comes from chorein, to make room or to contain, not from choreuo, to dance. The dance image may be a fine piece of preaching, but it should not be presented as the meaning of the Greek, and a student who repeats it in front of someone who knows will lose the room.

The creed of 381 says that the Holy Spirit proceeds from the Father. Most Western churches recite it saying that the Spirit proceeds from the Father and the Son. The Latin for 'and the Son' is filioque, and that single word is the oldest formal doctrinal division between the Eastern and Western halves of the church. It is still in force. Give it the care it deserves, because a great many students have opinions about it and very few know the history.

The history first, and it is not disputed by either side. The word is not in the creed of 381. It appears in Spain, in the acts of the Third Council of Toledo in 589, at the conversion of the Visigothic kingdom from Arianism to Catholic faith, and whatever the precise state of the Toledo manuscripts, the usage is firmly established in Spain during the seventh century. From Spain it spread into Gaul, and Charlemagne's court promoted it vigorously; a synod at Aachen in 809 affirmed it, and the Frankish theologians pressed Rome to adopt it. Pope Leo the Third refused. He accepted the doctrine that the Spirit proceeds from the Father and the Son, and he declined to alter the text of the creed, and according to the papal biographies he had the creed without the addition engraved on silver shields set up in Saint Peter's. Rome finally admitted the word into its liturgical creed only around 1014, under Benedict the Eighth, at the imperial coronation of Henry the Second. So for roughly two centuries the papacy itself held the position the East still holds about the text.

The Eastern objection has two parts and the first is not about the doctrine at all. It is that the creed is the common property of the whole church, given by councils, and that the council of Ephesus in 431 forbade the production of any other creed; a portion of the church therefore had no authority to add a word to it unilaterally, whatever the word meant. Photius, patriarch of Constantinople, made both arguments in the ninth century and his treatise on the Spirit's mystagogy is the classic Eastern statement. The theological part is this. In the Cappadocian settlement the Father alone is the source, the sole cause of the Godhead, and that monarchy is what keeps the three from being three gods. To say that the Spirit proceeds from the Father and the Son either gives the Godhead two sources, which divides it, or makes the common divine essence the source, which dissolves the persons; Photius pressed the dilemma in exactly that form. Lossky in the twentieth century added a further charge, that the filioque subordinates the Spirit to the Son and issues in a Western Christianity in which the Spirit's freedom is subordinated to the institution of the church. And on the exegesis, the East notes that John 15:26 uses its verb of procession with the Father alone, in the very sentence that also says the Son sends, so the distinction between eternal origin and temporal mission is drawn by the text itself.

The Western case is equally substantive. It begins from the scriptural pattern: the Spirit is called the Spirit of the Son in Galatians 4:6 and the Spirit of Christ in Romans 8:9, the risen Christ breathes the Spirit on the disciples in John 20:22, and in John 16:14-15 the Spirit takes what belongs to the Son, which belongs to him because all that the Father has is his. Augustine's formulation in book fifteen of De Trinitate is careful and is not the crude version usually reported: the Spirit proceeds principally, principaliter, from the Father, and also from the Son because the Father has given the Son that the Spirit should proceed from him, so that the Father remains the origin of the origin. Aquinas then supplies the systematic argument: if the persons are distinguished from one another by nothing except relations of origin, and if the Spirit stands in no relation of origin to the Son, then there is nothing to distinguish the Spirit from the Son, and the doctrine loses its own principle of distinction. The council of Florence in 1439 accordingly defined the procession as from the Father and the Son as from one principle and by a single spiration, which is precisely the denial that there are two sources.

The Eastern reply to that is not a refusal to engage. Greek theology has its own long-standing formula for the Son's part in the Spirit's being, from the Father through the Son, dia tou Huiou, used by Gregory of Nyssa, by Cyril of Alexandria and by John of Damascus, and many Orthodox theologians regard it as expressing everything that is true in the Western concern while keeping the Father as sole cause. On Aquinas's argument they answer that relations of origin already distinguish the Spirit sufficiently, since the Spirit proceeds and the Son is begotten, and that these are different modes of coming from the same Father; needing a relation between Son and Spirit assumes a Latin framework rather than proving it. And on Florence they observe that a formula agreed under political pressure, with the Byzantine empire seeking military aid, was repudiated by the Eastern churches when the delegation came home, and that Mark of Ephesus refused to sign it at the time.

The Western reply in turn is that the East's own sources are less uniform than the argument requires. Maximus the Confessor, in a letter to Marinus in the seventh century, defended the Latin usage to Greek readers on the ground that the Latins do not make the Son a cause of the Spirit but use their language to express the Spirit's consubstantial coming through him, and that letter is claimed by both sides and cited by both to this day. Latin theologians also press the point that if the temporal missions reveal the eternal relations at all, as the East agrees they do in the case of the Son's generation, it is hard to see why the Spirit's being sent by the Son reveals nothing about his eternal origin. And the canonical objection, the West says, cuts both ways: the East has also developed its doctrine, most obviously in the fourteenth-century councils on the essence and energies, without the West's consent.

Where it stands now is worth knowing precisely, because it has moved. The Pontifical Council for Promoting Christian Unity published a clarification in 1995 which distinguished the Greek verb used at John 15:26, which the East reserves for origination from the Father alone, from the wider Latin word processio, and stated that the Father is the sole origin of both the Son and the Spirit. The North American Orthodox and Catholic theological consultation issued a joint statement in Washington in October 2003, titled as a question, whether the filioque is a church-dividing issue, and closed it with recommendations to both communions. Some are addressed to both sides equally: that Orthodox and Catholics refrain in future from labelling the other side's tradition on the procession of the Spirit as heretical, and that both distinguish the divinity and hypostatic identity of the Spirit, which is received dogma in both churches, from the manner of the Spirit's origin, which the statement says still awaits full and final ecumenical resolution. Two are addressed to the Catholic Church alone: that it use the original Greek text of the creed of 381 alone in making translations for catechetical and liturgical use, and that it declare the condemnation issued at the Second Council of Lyons in 1274, against those who presume to deny that the Holy Spirit proceeds eternally from the Father and the Son, to be no longer applicable. The Lambeth Conferences of 1978 and 1988 recommended that Anglican provinces drop the word from the creed, and several have. Old Catholic churches removed it. Recent popes have recited the creed in Greek without it at ecumenical liturgies. None of that has ended the division, the word remains in the ordinary Latin and vernacular texts used by most Western Christians every Sunday, and the Orthodox churches have not accepted it. A student should be able to say what each side holds, why each holds it, what has changed and what has not, and should resist the temptation to announce that it was all a misunderstanding. Serious people on both sides think something is at stake.

Eternal generation is the doctrine that the Son is begotten of the Father eternally, without beginning, without division and without the Father being prior in time. The creed's own gloss is exact: begotten of the Father before all ages, God from God, begotten not made. The distinction between begetting and making is the hinge. A maker produces something other than himself; a begetter produces one of the same nature. The point of saying eternal is to strip from begetting everything that in creatures implies before and after.

The scriptural basis is the adjective monogenes in John 1:14, 1:18, 3:16, 3:18 and 1 John 4:9, together with John 5:26, where the Father who has life in himself has granted the Son to have life in himself, and the language of Hebrews 1:3 about radiance and exact representation. There is a genuine lexical argument here that students should know. Older translations rendered monogenes 'only begotten', deriving it from the verb to beget; twentieth-century lexicography, following an argument Dale Moody published in the Journal of Biblical Literature in 1953 and now standard in the major lexicons, derived it instead from genos, kind, giving 'only, unique, one of a kind', which is why the BSB reads 'one and only Son' and footnotes 'Or only begotten or unique' at John 3:16. More recently Charles Lee Irons, in a chapter of a 2017 volume on retrieving eternal generation edited by Fred Sanders and Scott Swain, has argued the lexical case for 'only begotten' back onto the table. The doctrine does not stand or fall with the word, but a teacher should know that the word is argued about and should not build a sermon on a lexicon entry he has not checked.

The doctrine has had Protestant wobbles. Calvin affirmed eternal generation and also insisted, in the Institutes, that the Son is autotheos, God of himself, with respect to his essence, while being from the Father with respect to his person; the formulation was attacked in his own lifetime and afterwards as an innovation that weakened the Father's priority, and defenders answer that it protects the Son from being a lesser God. In the twentieth century some evangelical writers proposed setting the doctrine aside as speculative, and the best-known case is Wayne Grudem, whose Systematic Theology in 1994 did not merely judge the exegesis weak but drew the conclusion out loud. In the sixth appendix of that first edition he wrote that it would seem more helpful if the language of the eternal begetting of the Son, also called the eternal generation of the Son, were not retained in any modern theological formulations, and he argued that the phrases should not stand in the Nicene Creed either. In the second edition, published in 2020, he reversed himself and affirmed eternal generation. That is a rare and creditable thing in print and it is worth telling students about, because it shows what changing one's mind in public looks like, and because the distance between the two editions is the measure of how seriously the argument of the last decade has been taken.

The recent argument is about something adjacent, and it is not settled. Its defenders call it the eternal relations of authority and submission, or eternal functional subordination, and its best-known advocates are Wayne Grudem and Bruce Ware, and it is Ware's own 2005 book on the Father, Son and Holy Spirit that sets out the case at length. The claim is that the three are equal in essence and eternally distinguished by a relation in which the Father has authority and the Son joyfully submits, and that this is not subordination of being but of role. The texts appealed to are the Son's language of doing the Father's will and being sent, John 5:19, 6:38 and 8:28-29, the Father being greater at John 14:28, the ordering of 1 Corinthians 11:3, and the subjection of the Son at 1 Corinthians 15:28, which is placed after the end of the redemptive work and so, they argue, cannot be merely a temporary economic arrangement.

The critics are numerous and their central argument is technical rather than rhetorical. Kevin Giles, in The Trinity and Subordinationism in 2002 and Jesus and the Father in 2006, and Millard Erickson, in Who's Tampering with the Trinity? in 2009, argued that the position revives the subordinationism Nicaea excluded under a new label. The argument sharpened in the summer of 2016, when Liam Goligher and Carl Trueman published criticisms and were joined by patristic scholars including Michel Rene Barnes and Lewis Ayres and by theologians including Fred Sanders, Scott Swain and Michael Bird. The decisive objection runs through the sixth ecumenical council of 681, which defined that Christ has two wills because will belongs to nature rather than to person, which is why the two natures give two wills. If will follows nature, then the one divine nature has one will, and there is nothing there for the Son to submit to the Father with; to have an eternal submission of the Son's will to the Father's you need two wills in God, and two wills means two natures, which means two gods. A second objection is historical: the tradition does speak of an order, a taxis, and of the Father as the source, but it locates the distinction in relations of origin rather than in authority, and it insists the persons act inseparably.

The defenders answer that they affirm one divine essence and reject subordination of being, that they are describing relations between persons and not a division of nature, and that the tradition's own language of the Father sending and the Son being sent already implies an eternal order that is more than a courtesy of vocabulary. They also point out that the critics must explain what an eternal taxis amounts to if it carries no content at all. It is worth adding two things that moved during the dispute: Grudem's affirmation of eternal generation in 2020 removed one of the criticisms that had been most pressed against him, and several writers on that side have modified how they state the position. It is also worth saying plainly that the argument has been entangled from the start with the argument about men and women, because the analogy between the Son's submission to the Father and a wife's to a husband has been drawn by advocates and attacked by critics, and the entanglement has made both discussions worse. A student should be able to separate them, and should notice that complementarians are found on both sides of the trinitarian question, which shows the two do not stand or fall together.

So end the lesson where the creeds end. The church said, and has not unsaid, that there is one God; that the Father, the Son and the Holy Spirit are each wholly that God; that they are really distinguished from one another and not three names for one; and that the distinction lies in the relations of origin rather than in anything that would make one more God than another. Everything else in this lesson is argument about how to say that without saying something else by accident. The creeds are short because everything they leave out is what people were dying over.