Contending for Nicene Trinitarianism in an egalitarian age...

Error message

[Editors note: Prof. Steven D. Boyer's article below, first published back in 2009, clarifies the present debate over the nature and meaning of the Fatherhood of God and the Sonship of our Lord Jesus within the Trinity. Here, Dr. Boyer (Professor of Theology at Eastern University) demonstrates that the church's orthodox confession of the Trinity has, from the time of the Arian heresy, explicitly declared the order within the Trinity. Further, that this declared order (or hierarchy) is not merely analogical, nor is it limited to the Son's mediatorial work. Rather, the order must be (in some sense) ontological—and therefore eternal.

Dr. Boyer warns that the orthodox confession of the Trinity has fallen on hard times due to the egalitarian spirit of our age. He discusses the pros and cons of terminology used to discuss Trinitarian order today such as "roles," "command and obedience," and "subordination." He explains the confusion surrounding the word "ontological," pointing out that the denial of ontological order is a doctrinal error equivalent to the denial of ontological equality. Finally, Boyer makes some recommendations for word usage that may protect the order of the Trinity in this age when order and authority are despised.]

Articulating Order: Trinitarian Discourse in an Egalitarian Age1 

by Steven D. Boyer

Throughout its history, Christian orthodoxy has affirmed an understanding of the triune nature of God that includes, despite certain logical tensions, both order and equality among the divine Persons. Since most of that history played out in a social context that took hierarchy for granted and that therefore required a sturdy articulation and defense of the equality of the Persons, it sometimes appears that the tradition emphasized equality alone, and not order. But this conclusion is easily upset by a closer look at the evidence. To speak of order within the Godhead has been a commonplace ever since the patristic era, and it is often embodied especially in affirmations about the unique position of the Father in the Godhead. The Father is the “beginning of the whole divinity,” says Augustine; “the source” of Son and Spirit, says Gregory Nazianzen; the “cause of the Son”, says John of Damascus; “the principle of the Son,” says Thomas Aquinas; the “origin” of Son and Spirit, says Calvin; the “fountain of deity,” says Richard Hooker; “first in order,” says Jonathan Edwards.[1] Ordered relationships within the Trinity are as strongly affirmed by the orthodox tradition as equality is.

Yet the last two centuries have seen dramatic changes in the social context of the Western world, and many Christian theologians today work in a culture in which equality is the dominant principle. Hence, the equality of the divine Persons is easily granted in contemporary discussion, whereas the notion of order in the Trinity is often addressed with less conviction, and sometimes even with suspicion...

In this essay, I wish to elucidate this notion of order in the Trinity by, first, reviewing its place in Christian orthodoxy; second, evaluating some increasingly common trinitarian language regarding personal distinctions within the Godhead; and third, pointing out and disarming a crucial ambiguity that keeps some debates over order from making any real progress.


Let us begin with a historical point of reference, in order to see as clearly as we can what alternative readings of the trinitarian tradition are available. Consider the theological reflection that lay behind the christological debates of the Council of Nicaea. On the one side stood Arius and the Arians, with their theology of diminished deity in Christ. They insisted that, on any monotheistic reckoning, the Son of God can be “divine” only in a relative or metaphorical sense. The Son cannot be a second deity alongside of the first, or an isolated “part” of the one divine whole, or a modalistic re-presentation of the one God.[2] According to the Arians, these options exalt the Son at the expense of a coherent, biblical account of his Sonship. To be a son is necessarily to be subject to a father, and Scripture seems to require this as definitely of the “divine” Son as of human sons: “the Father is greater than I” and so on. The Arian conclusion was that the subordination of the Son to the Father seen in the New Testament reflects an eternal subordination that is rooted in the ontological difference between the literally divine Creator and the metaphorically “divine” Son who is really the first and most glorious of the creatures.

Against this conclusion stands Athanasius of Alexandria. Readily acknowledging the biblical texts that point to some sort of subordination, Athanasius employs a hermeneutic that focuses upon Scripture’s “double account of the Saviour”, an account rooted, on the one hand, in texts that affirm the unqualified deity of the Son (e.g., “the Word was God”—John 1:1) and, on the other hand, in texts that describe the Son’s humiliation for the sake of our salvation (e.g., he “emptied himself, taking the form of a servant”—Philippians 2:6-8).[3] For Athanasius, the frailty and submission and obedience of the Son in the New Testament reflect this economic subordination without at all undermining the eternal ontological equality of Father and Son.

Now, if we are interested in the relationship between order and equality in the divine life, we can easily discern two options in this debate. The first is the Arian option: the order that characterizes the Father-Son relation is such that any equality between the Persons is relativized and ruled out. The second, which we might be tempted to describe as the “orthodox” or the “Athanasian” option, suggests, by contrast, that the equality in the relation relativizes the order: order must be relegated to the economy of salvation so that the eternal equality of Father and Son may be left intact. Both of these options are logically coherent, though the second is the only “live” option for orthodox Christianity after the Council of Nicaea.

Now this second, “Athanasian” option may be a faithful rendering of many biblical texts and emphases, but it does face one rather glaring difficulty: it does not seem to be what Athanasius actually held. Or rather, it portrays one side of Athanasius without attending adequately to another side. To be sure, equality is the dominant theme of Athanasius’s writings: the burden of nearly all of his work is explicitly to overthrow the Arian error by insisting that what the Father is, the Son is also. Yet given the ubiquitous presence of Arianism in the fourth century, this emphasis is hardly surprising. The question is whether we see in Athanasius’s work another side, less prominent because less controversial in the fourth century, a side that gives us a deeper clue into Athanasius’s comprehensive understanding of the equality and order within the Godhead. I think we do.

Note first Athanasius’s well-known claim that “the same things are said of the Son, which are said of the Father, except His being said to be Father.”[4] What does this exception signify? In one sense, the answer is obvious. Of course the Son is not said to be the Father. The Son is not the Father, but the Son of the Father. This distinction between Father and Son, between the nature of Fatherhood and the nature of Sonship, is solidly grounded in Scripture, especially in the Johannine notion of “begottenness”, and it is explicitly affirmed by the creeds, as when the Athanasian Creed insists that “There is one Father, not three Fathers; one Son, not three Sons,” and so on.[5] Note that the distinction includes a certain asymmetry as its fundamental feature. It is not an asymmetry that excludes mutuality, for of course just as no son can be “son” without a father (or mother), so also no father can be “father” without a son (or daughter).[6] But this mutuality is also plainly not a mutuality that excludes asymmetry, for both in biological origin and in cultural inheritance—probably in other ways, too, but these two stand out—a certain “giving” and a certain “receiving” define what it is to be a “father” and a “son”, respectively. To be distinguished throughout eternity as Father and Son is to be defined, in part, by this asymmetry. Hence, there is in the relation between the first and second Persons of the Trinity an unreciprocated and irreversible order of dependence. To get at the mutuality of the relation, we could say that the Father is dependent on the Son in the logical (and linguistic) sense that every father must have a child in order to be a father. But we must go on to say that the Son is dependent on the Father not only in order to be a son, but also in order to be the particular Son that he is, namely God the Son. As fons divinitatis, the Father constitutes that being of God in which the Son perfectly and completely shares. The only-begotten Son is fully God just as the Father is, but it is the unbegotten Father who makes the triune God to be God at all.

We see this way of thinking in Athanasius at a variety of points. Consider, for instance, the way he treats passages like John 10:18 and Matthew 28:18 in the following surprisingly forceful presentation of the distinction. According to Athanasius, Jesus

has said “was given unto Me,” and “I received,” and “were delivered to Me,” only to shew that He is not the Father, but the Father’s Word, and the Eternal Son, who because of His likeness to the Father, has eternally what He has from Him, and because He is the Son, has from the Father what He has eternally. Moreover . . . “was given” and “were delivered,” and the like, do not impair the Godhead of the Son, but rather shew Him to be truly Son . . . . For the Saviour Himself says, “As the Father hath life in Himself, so hath He given also to the Son to have life in Himself.” Now from the words “hath given,” He signifies that He is not the Father; but in saying “so,” He shows the Son’s natural likeness and propriety towards the Father. . . . Then is the Word faithful, and all things which He says that He has received, He has always, yet has from the Father; and the Father indeed not from any, but the Son from the Father.[7]

Note a few lines especially. Jesus’s intention, as Athanasius reads it, is to show precisely “that He is not the Father, but the Father’s Word”. And what is the meaning or the content of this distinction? It is this: all that the Son has, he has “from the Father”. Here is the asymmetrical order that is necessarily a part of Father-Son language. But there is no sign here that this ordered dependence is a function exclusively of the incarnation. On the contrary, it is rooted in the very nature of Sonship. Athanasius goes out of his way to explain what Jesus’s status as “the Eternal Son” involves: the likeness of Son to Father means that Jesus “has eternally” whatever he has from his Father, and the dependence of Son on Father means that Jesus “has from the Father” whatever he has eternally. Of course, this ordered relation does not entail that the Son is less than the Father: on the contrary, it demonstrates conclusively “the Son’s natural likeness and propriety towards the Father.” Here is an affirmation of equality that, rather than relativizing the order between the Father and the Son, is actually built upon that order. The Son receives “from the Father”, and receives from the Father “always”, precisely because he is the co-essential Son of the Father.

Other texts from Athanasius betray the same interest. For instance, the unity of the Godhead is rooted, for Athanasius, not simply in the divine nature abstractly considered, but in the Person of the Father. “For the Word, being Son of the One God, is referred to Him of whom also He is; so that Father and Son are two yet the Monad of the Godhead is indivisible and inseparable. And thus too we preserve One Beginning of Godhead and not two Beginnings. . . . And of this very Beginning the Word is by nature Son.”[8] Or again, consider the following lines, which embody Athanasius’s reply to the Arian argument that an eternal Son would necessarily be parallel to or coordinate with the eternal Father, and hence would be not the Father’s Son but the Father’s brother:

If we said only that He [the Son] was eternally with the Father, and not His Son, their pretended scruple would have some plausibility; but if, while we say that He is eternal, we also confess Him to be Son from the Father, how can He that is begotten be considered brother of Him who begets? And if our faith is in Father and Son, what brotherhood is there between them? And how can the Word be called brother of Him whose Word He is? . . . For the Father and the Son were not generated from some pre-existing origin, that we may account Them brothers, but the Father is the Origin of the Son and begat Him.[9]

Here again Athanasius roots his whole reply to the Arians in what he regards as the obvious, asymmetrical distinction between one who begets and one who is begotten, or between one who speaks a word and the word that is spoken. Note that he affirms not mere differentiation, but ordered or asymmetrical differentiation; and again he affirms not a merely economic order, but an immanent or eternal order, in the Godhead. He regards this element of Christian doctrine as foundational to the orthodox Christian understanding of God.

It appears, then, that Athanasius’s account of equality and order in the Godhead is not quite so simple as the two options we were examining a moment ago. There is a third possibility. And we are now in a position to recognize the assumption that both of those earlier options shared, namely, the assumption that an eternal or permanent order of dependence necessarily entails ontological attenuation in the dependent party. If this assumption is legitimate, then the relation of the Son to the Father may be characterized by either equality or order, but not by both. Yet Athanasius insists upon both—and derives both from the nature of Sonship itself.[10]

Nor is Athanasius alone in this judgment. Our discussion here could easily be extended to show that the vast majority of the Christian tradition views intra-trinitarian order in much the same way. The Son eternally comes from and is eternally dependent upon the Father, yet in a manner that in no way entails the Son’s being less than or inferior to the Father. To connect dependence to inferiority is in fact to accept an axiom of Neoplatonism that the fourth-century fathers who knew Neoplatonism best went out of their way to reject. In words that could be extended to describe fourth-century orthodoxy as a whole, Anthony Meredith says that Basil of Caesarea, in his writings against the Arian Eunomius, recognizes “the place of taxis or order within the Trinity” and simultaneously

refuses to follow Eunomius in inferring from this order a lessening of essential being and Godhead. Though the Son comes from the Father he is not therefore any less than the Father. In holding this Basil set himself against what can be called a tenet of Platonism, namely that the cause is superior to the effect.[11]

And by rejecting this “tenet of Platonism”, the fathers paved the way for a full-blooded trinitarian tradition that speaks over and over not of equality or order, but of equality and order. We might even say, equality because of order, for Father and Son are completely equal, yet the Father is the asymmetrical source of equality, as he is also the source of everything in the Godhead. Thomas Aquinas refers to this as an “equality with movement” in it, such that “the Son receives from the Father, this, namely that He is equal to the Father, and not conversely.”[12] Gregory of Nazianzus had earlier made the same point: “From him [the Father] flows both the Equality and the Being of the Equals.”[13]


With this historical insistence on intra-divine order in mind, we can begin to detect an instructive significance in a debate currently raging among Protestant evangelicals, especially in the United States and Australia, over how to articulate this order appropriately. Conversation about trinitarian order initially surfaced among evangelicals as part of their on-going dialogue about gender, in which so-called “complementarians” affirmed the value of certain forms of biblically mandated hierarchy in the relation between men and women in the Church and in the family, while “egalitarians” rejected such hierarchical ordering as both unbiblical and intrinsically demeaning to women.[14] This discussion about the legitimacy of hierarchy found representatives of both positions appealing for support to the orthodox doctrine of the Trinity, with its paradigmatic account of order and equality; and this appeal from both sides has thus developed into a vigorous debate in its own right.

I do not intend to focus on the details of the evangelical discussion, but from it have emerged several new and somewhat controversial ways of articulating the relation between order and equality in the Godhead, and I would like to consider three of these provocative locutions. For convenience, we may consider all three in a single locale, namely, in the Systematic Theology of American “complementarian” Wayne Grudem. Grudem expressly affirms order in the Trinity in terms of an “eternal equality in being but subordination in role”, such that “the role of commanding, directing and sending is appropriate to the position of the Father,” while “the role of obeying, going as the Father sends, and revealing God to us is appropriate” to the Son.[15]

Following Grudem’s lead, let us consider the validity of referring to the ordered life of the Godhead in terms of (1) distinct “roles” played by the divine Persons, (2) “command” and “obedience” within the Trinity, and (3) the “subordination” of the Son to the Father.

Distinction of “roles” 

Speaking of the Persons of the Trinity in terms of distinct “roles” sounds a bit like the (pseudo-) “Athanasian” option we discussed above, as if the ordered relationship between Father and Son were primarily a function of the incarnation, or at least of the economy of salvation. Yet many (like Grudem) employ this terminology precisely to give an account of the eternal order in the Trinity without compromising the essential oneness of all three Persons. Since all three possess completely the undivided divine nature, it cannot be difference in nature that distinguishes Father and Son—and therefore it must be difference in role.[16]

This reasoning makes sense, I think, but it has weaknesses. Its chief weakness—and it is a major one—is its significant departure from Nicene and post-Nicene usage. Nicaea’s homoousios was interpreted by its promulgators and by later Christians as an affirmation not of identical substances in the Godhead, substances that might in principle take part in different activities or play different roles, but of identity of substance. In other words, the substance of the Son is not only exactly like that of the Father; the substance of the Son is that of the Father. The “oneness of substance” is a numerical oneness. Furthermore, it is a oneness that is especially expressed in the common activity or the common energy that all three Persons indivisibly produce. Thus, John of Damascus summarizes the Christian teaching in this way:

The community and unity [of the Godhead] are observed in fact, through the co-eternity of the subsistences, and through their having the same essence and energy and will and concord of mind, and then being identical in authority and power and goodness—I do not say similar but identical—and then movement by one impulse. For there is one essence, one goodness, one power, one will, one energy, one authority, one and the same, I repeat, not three resembling each other. But the three subsistences have one and same movement. For each one of them is related as closely to the other as to itself.[17]

The intimacy of this union gives rise in virtually all Christian traditions to the doctrine variously known as perichoresis (Greek) or circumincession (Latin) or coinherence.  The three Persons of the Trinity eternally and incomprehensibly “indwell” one another—“I am in the Father, and the Father in me,” says Jesus (John 14:11)—so that every act of one involves the activity of the others, not as three partners each contributing one part of the total achievement, but as three subjects of one undivided activity. This oneness of activity directly resists any suggestion that Father, Son, and Spirit have different roles to play throughout eternity. Any hard and fast differentiation of role would seem in fact to deny the very thing that historically has accounted for the unity of the Persons of the Godhead.[18]

Yet there may be differentiations that are not “hard and fast”. Both Scripture and the church fathers seem to suggest a sort of non-technical differentiation of roles when they commonly associate the work of creation with the Father, or the work of redemption with the Son, or the work of illumination or empowerment with the Spirit, or again, when they distinguish between, say, the Father as “sender” and the Son as “sent” (the very distinction that Grudem also builds upon in the quotation above). These are not absolute, water-tight “compartments of responsibility” that run afoul of the affirmation of perichoresis. They are simply helpful ways of thinking about how God has made himself known.[19]

Indeed, the deeply personal character of every divine activity—i.e., the fact that every activity is the activity of a Person, not the activity of some undifferentiated divine substance—seems to require some distinctive mode of expression, however limited, to describe the interpersonal aspects of the divine life. For instance, it is certainly the case that God loves the world with one love, a love that is fully the Father’s because it originates in the Father and that is every bit as fully the Son’s because the Son is the Son of the Father. No differentiation between the Father’s love and the Son’s love is necessary here: it is one, perichoretic reality. Yet when we speak of the Father’s love for the Son and of the Son’s love for the Father, some differentiation does seem to be in order. For the Father cannot love the Son as his Son in the same way that the Son loves the Father as his Father. The asymmetry of the relation entails an asymmetry of personal involvement within the relation. Calvin recognizes this distinctive interpersonal dimension when he affirms that “the Son, regarded as God, and without reference to person, is . . . of himself”, but then immediately adds that, “regarded as Son, he is of the Father.”[20] In other words, one always speaks in one way regarding God simply as God, or with respect to creation, but in another way when divine interpersonal concerns are to the fore. In the technical terminology of the 15th century Council of Florence,[21] everything in the divine Persons is one unless there is “opposition of relationship”, i.e., unless the Persons are related to one another in opposite or asymmetrical ways. But where there is “opposition of relationship”, the oneness necessarily takes on a more nuanced character. In this case, it is fully appropriate to recognize an ordered, interpersonal differentiation that characterizes the perichoretic oneness of every divine attribute and act.

Nevertheless, this differentiation is a delicate thing. And so, paradoxically, one problem that “roles” language faces is its clarity. It provides an unambiguous account of the distinction between the divine Persons that is so crisp and obvious that it obscures the real depth and mystery of the perichoretic relationship. It suggests that the distinction between divine Persons is not really so difficult to explain, that we can talk about the relation between Father and Son in much the same way we talk about any other relation: they are equal parties with different responsibilities. This is a rather dramatic oversimplification, and one that could lead to a kind of implicit Arianism if “roles” language is not supplemented by other metaphors for intra-divine life.

Yet at the same time, if its limitations are acknowledged, “roles” language does make a legitimate point about inner differentiation in the divine life—and in fact, a point that unrestricted emphasis on co-equality is likely to obscure. “Roles” language is not so simple and unambiguous as its advocates may suppose, but in an egalitarian world the greater danger may lie in the opposite direction, namely, in the untroubled supposition that the co-equality of the divine Persons is plain, simple, and unambiguous. This supposition is deeply flawed. Language of “equality” in modern democratic contexts is inevitably laden with problematic individualistic implications that are themselves rooted in assumptions about the inevitability of competition and the ever-present danger of oppression or abuse—implications and assumptions that stand far outside the world of orthodox trinitarian discourse.[22] In this context, the oversimple affirmation of intra-divine equality may need to be challenged even more than an oversimple affirmation of order. Thus, “roles” language, even with all its weaknesses, may give expression to that aspect of trinitarian life that contemporary observers are most in danger of forgetting. In my judgment, the fact that it stands in dialectical tension with the perichoretic unity of the Godhead does not invalidate “roles” language—so long as no other, deeper problem presents itself.

“Command” and “obedience” 

One can follow much the same logic in trying to recognize what is appropriate and what is dangerous in descriptions of the ordered relation between the Father and the Son in terms of “command” and “obedience”. Once again, the terms point to a legitimate distinction, while they also suffer from an excessive clarity that is rather blind to the triune mystery. But there are at least three aspects of this language that might make us even more hesitant to apply it to trinitarian relations. For in relationships of command and obedience, (1) two independent wills are clearly involved; (2) the natural, unconstrained preference of the “obeying” will is inappropriately overridden; and (3) the obedient party therefore seems to be inferior to the commanding party in some way, whether in power or in knowledge or in rank. Let us take these hesitations one at a time.

First, two wills are involved in command and obedience, whereas we noted above that the Persons of the Trinity share one undivided will. Yes. But we noted also certain distinctions in that undivided will, and this complicates matters. In principle, command and obedience might be thought of as the endpoints of a single, shared volition, in spite of the fact that two independent wills are typically involved in the “sharing”. In other words, to “obey” is to accept as one’s own a volition that originates in an other, viz., in one who “commands”. Now this is exactly the relation that obtains between the Father and the Son, except that the Son accepts the will of “an other” without thereby having “another” will. The will that the Father and the Son share is, as we have seen, numerically one will, but it remains a personal will—that is, the will of a Person. Originally it is the personal will of the Father, and by begetting it is the personal will of the Son. As the Father’s will it must include some sort of distinctive “volitional perspective” with respect to the Son, and as the Son’s will it must include some sort of “volitional perspective” with respect to the Father. In neither case should this “perspective” be construed as a disposition of the will toward an external object, for the Father and the Son are not external to one another. Instead, the “perspective” is a mode of having the will that is distinctive to each particular Person. The Father “has” the will, or wills as he wills, in a manner that reflects his initiative as begetter of the Son, and the Son “has” the will, or wills what he wills, in a manner that reflects his responsiveness as begotten of the Father. Without some sort of differentiation at this interpersonal level, we inevitably end up confusing the divine Persons. To avoid that confusion, a cautious analogical use of “command” and “obedience” might prove quite useful.

What about the second hesitation? Doesn’t “obedience” suggest an inappropriate constraint, or even a coercive imposition, on the Son? This inference is frequently drawn,[23] but I cannot see how the charge can be made to stick. It may be natural for a democratic culture to suppose that every decision I make should be simply and autonomously my own, with no compelling influence arising from any source, but if orthodox Christianity is right to believe that an eternal order of dependence characterizes the relationship between the Father and the Son, then the insistence on autonomy cannot be justified on Christian principles. According to the Athanasian vision of the Trinity, the Son’s will is not one whit less the Son’s for having originated with the Father. Nor is it less fully or less happily the Son’s. The Son is the Son of the Father, and so he is pleased, from all eternity, to accept the Father’s will as his own. It might not be very precise to say that he accepts it “voluntarily”, for this implies a second, independent will that could, in principle, not will what the Father wills. But it would be equally misleading to say that he accepts it “involuntarily”—as if he preferred to go his own way, but found that this was not an option. On the contrary, it is precisely the numerical oneness of the divine will that guarantees that the Son’s receptive willing of that will is no more “forced” or “coerced” than the Father’s originative willing of it.

But why should the Son be always, so to speak, “on the receiving end” of the divine will? Doesn’t this suggest, according to our third hesitation, that the Son is somehow inferior to the Father? Ironically, this question, which arises so readily in our egalitarian culture, is exactly the same question asked by the self-consciously hierarchical culture of the fourth century. The assumed answer is also the same, but we have already seen the orthodox response. There is no ground, according to Athanasius and Basil and others, for accepting Platonism’s axiom that order entails inferiority. On the contrary, the Son is everything that the Father is, except for being Father. Shocking as it may be to either fourth century or twenty-first century sensibilities, to obey is as God-like as to command.[24] Glad submission is intrinsic to the divine nature itself, and in this respect, provocative language of “command” and “obedience” regarding the Trinity may point us, once again, to the very aspect of intra-divine order that contemporary culture is most likely to overlook. This is risky language, of course, for the Father does not “command” the Son in the same way that one creature commands another, or in the way that Father, Son, and Holy Spirit perichoretically “command” all of creation. Yet with appropriate care, this phrasing seems to me able to make an illuminating contribution to a doctrine of the Trinity that includes both ordered distinction and perichoretic union.[25]


And what about the term that is perhaps more provocative and controversial than any of the others, the term “subordination” itself? It is almost impossible to hear this word in a theological context without being reminded of Arianism, which Maurice Wiles refers to as the “archetypal heresy”.[26] This association might be reason enough for Christians to avoid the term, but not all have done so: theologians like Charles Hodge and A. H. Strong and historians like J. N. D. Kelly and Phillip Schaff have had occasion to use “subordination” to describe trinitarian relations that are construed in an explicitly non-Arian fashion.[27] Nevertheless, should the term be used? Or is it objectionable in itself, in all contexts, because of its implicit connection to the Arian outlook even where Arianism is explicitly rejected?

I would argue that it is not intrinsically objectionable. If we are right to detect order in the Trinity, then there is nothing obviously or necessarily wrong with describing that order using a spatial metaphor that involves height, that is, “ordering under” or “sub-ordering”. One correspondent suggested to me that “under” always implies an Arian ontology, and so we should say instead that, in the order of the Trinity, the Son comes after the Father. But surely the temporal language of “after” is every bit as misleading as the spatial language of “under”; moreover, “after” was explicitly condemned by the Nicene fathers because it was Arius’s literal teaching! Again, it is sometimes objected that “subordination” implies Arian ranks or degrees of deity because the term is used so often in hierarchical contexts (the military, the church, etc.). I reply that these commonplace hierarchies do not include any ranking according to “essence”: the humanity of the five-star general is no different from that of the buck private. Instead, here is an established asymmetrical order of initiative and response among ontological equals—exactly what the doctrine of the Trinity, mutatis mutandis, also demands. Another correspondent observed that the hierarchical ordering of general over private is compatible with essential equality only because the private could, in principle, become a general. But I cannot see why this condition should be granted. Indeed, if the orthodox account of order in the Trinity is correct, then the condition should not be granted, for the Son is the equal of the Father not because he might one day become the Father, but precisely because he is eternally and unchangeably the Son of the Father.

Of course, I am not suggesting that the relation between the Father and the Son in the Trinity is exactly like the relation between a general and a private in the armed forces. Indeed, the parallel is likely to be misleading if not properly understood. But the same can be said of any parallel, and hence of any language that we employ to describe the Trinity—including language that is far more palatable, and therefore less subject to scrutiny, in an egalitarian climate.  In such a climate, to speak provocatively of “subordination” in the Trinity can be enriching and instructive, so long as the meaning of the term is carefully and conscientiously circumscribed.


But circumscribed how? This is a crucial matter, I think, for one of the great problems in discussion of trinitarian order is here. It is not at all uncommon to hear the claim that, while there may be diversity or order within the Godhead, any form of ontological subordination is ruled out by Nicaea: there is differentiation, but no difference in being among the divine Persons. Ironically, both sides in the evangelical debate noted above use this language, and so do a variety of other thinkers. To take but one example, Thomas Torrance maintains that “inner trinitarian order is not to be understood in an ontologically differentiated way, for it does not apply to the Being or the Deity of the divine Persons which each individually and all together have absolutely in common, but only to the mysterious ‘disposition or economy’ which they have among themselves within the unity of the Godhead.”[28]

Yet I submit that we need to ask what exactly is meant by an “ontological” order or by differences in “being”. For both ancient and contemporary trinitarian discourse recognizes a very definite “ontological finality” in the divine Persons. There is no “being”, no ontos, that is deeper than or behind these Persons. As Augustine noted, the three divine subsistences are not separate from or posterior to the one divine substance like three rings that are made out of the one substance gold. No, the personal subsistences are the divine being, for (to quote Augustine yet again), “it is the same thing with God to be (esse) as to subsist (subsistere).”[29] Indeed, the genius of the Cappadocian Fathers, and especially of Basil of Caesarea, may be located precisely in their development of a terminology for the divine subsistences that was explicitly ontological in character.[30]

Of course, one might wish to argue that the order between the Persons is not really an “ontological” order since the distinctions in the Godhead are grounded not in separate entities or “beings”, but only in the relations of origin between the Persons. This emphasis on relations is very appropriate, and it is universally recognized, but it does not help matters. For the phrase “relations of origin” denotes the character of the Father and the Son respectively as “unbegotten” and “begotten”, terms that cannot refer to some temporal starting point, as Arius tried to refer them, but that must refer to who the Father and the Son eternally are, to their “being”—or else they refer to nothing at all. As Colin Gunton concisely points out, “A relation is first of all to be conceived as the way by which persons are mutually constituted, made what they are.”[31] If relations of origin really do “constitute” the Father and the Son, then relations of origin are what these Persons are: the relations are themselves “ontological”. Even Torrance, concerned as he is to resist “ontological” differentiation, seems ready to agree: “The relations between [the Persons] are just as substantial as what they unchangeably are in themselves and by themselves. Thus the Father is Father precisely in his indivisible ontic relation to the Son and the Spirit, and the Son and the Spirit are what whey are as Son and Spirit precisely in their indivisible ontic relations to the Father and to one another.”[32] To admit asymmetrical “relations of origin” is therefore to admit some kind of “ontological” asymmetry as well, for, in Torrance’s own pithy phrase, “‘Person’ is an onto-relational concept.”[33]

So it begins to appear that “ontological” language about God may reasonably refer either to the one substance that the three Persons share or to the three Persons who share the one substance. For when we speak of the divine Persons, we really are speaking “ontologically”—i.e., we are speaking, in a certain sense, about nothing less than the divine “being”. And therefore any ordered distinction among the divine Persons may legitimately be referred to as an “ontological” order or a distinction of “being”. Now, to admit this usage is immediately to be embroiled in paradox, for if both ousia and hypostasis are ontological categories, then one cannot help noticing a certain awkwardness in the orthodox declaration that while the Son is the same ousia as the Father, the hypostases are related to one another asymmetrically. Yet this is precisely the orthodox claim. Again, we can hardly blame Kevin Giles when he complains about the “contradicting assertions” of those who “hold at one and the same time that the three persons of the Trinity are ‘one in being’ and that there are ‘differences in being’ between the three divine persons.”[34] Yet this is just the sort of odd juxtaposition that a vague English word like being will invite, for there is no difference in the “being” qua ousia of the Persons, but there is a very definite difference in the “being” qua hypostasis of the Persons. Worst of all, we seem to be driven ineluctably to the paradoxical conclusion that the Son both is and is not “ontologically” subordinate to the Father! Of course, those who did not buy my argument a few paragraphs ago might not want to say anything positive about a doctrine that bears the label “subordination”, but the point is that, in some sense, a genuinely ontological order seems to be inescapably present in the divine life. To err with respect to ontological equality leads to a heretical ascription of diminished deity to the Son, but to err with respect to ontological order leads just as surely to a heretical confusion of the Father and the Son.

There is no space here to develop this argument very far, but we ought to note one of its important implications. It turns out that the genuinely ontological character of the divine distinctions invites a second and more critical look at the language of “roles”. Note the contrast that is made in the not uncommon claim that the Son is “identical to the Father in being but subordinate in role”. This way of speaking explicitly suggests that what the Son “is” at the deepest level—what it is for the Son to “be”—is really no different from what the Father “is”, from what it is for the Father to “be”. The (sub-) ordered distinctness describes only what the Father and the Son “do”, only the roles they play; it explicitly does not describe what they “are”. One must look deeper, behind or beneath the roles, and therefore beneath the Persons, to find the solitary, self-identical “being” of God. Now this way of putting the matter seems to me very clearly to abandon the ultimacy of the Persons of the Trinity, which is, in effect, to de-personalize the Godhead and simultaneously to lapse into modalism.[35] In fact, this is precisely what we see when, for instance, The New Schaff-Herzog Encyclopedia of Religious Knowledge explains that the three divine Persons “are absolutely equal in virtue of the identity of their divine essence” and hence that the order in the Trinity “relates only to ‘order of subsistence’ not to being; it is merely logical, not real.”[36] Any time “order of subsistence” is contrasted to “being”—as it surely is when the order is defined solely in terms of “roles”—the conclusion that this order is “merely logical, not real” cannot be far behind. And Thomas Aquinas expresses a prophetically accurate judgment when he notes that, “if no real paternity or filiation existed in God, it would follow that God is not really Father or Son, but only in our manner of understanding; and this is the Sabellian heresy.”[37]

Now, if this entire line of argument about “trinitarian ontology” is correct, then it seems that Christian theology needs to find a new and more nuanced way to speak about the equality and order that characterize the being of God. I take it that those, like Reformed theologian Robert Letham, who affirm an “order of subsistence” that is different from an “ontological order” are trying to make just such a technical distinction.[38] But my argument is that we should avoid restricting the term “ontological” to one side of the distinction, for it legitimately applies to both. As an alternative, we might follow Phillip Schaff in his distinction between “a subordinationism of essence (ousia) and a subordinationism of hypostasis, of order and dignity”, the former denied by the Nicene fathers, the latter affirmed.[39] Or for simplicity’s sake, I would prefer to stipulate a distinction between “substantial subordination”, by which we would designate the Arian error, and “hypostatic subordination” (or, for those who still respond viscerally to “subordination” language of any kind, “hypostatic order”), by which we would mean an orthodox, non-Arian asymmetry. I do not find this option to be particularly elegant, for we are being forced to make an arbitrary distinction between adjectives that are etymologically synonymous. But since both kinds of order—both the Arian kind that is denied and the orthodox kind that is affirmed—are genuinely “ontological”, a certain arbitrariness might be unavoidable.

Whether arbitrary in terminology or not, some such distinction is sorely needed. The fourth century met a similar need by making its own nearly arbitrary distinction between the virtually synonymous Greek terms ousia and hypostasis. That “linguistic settlement” was not without its confusions: when the East recommended the distinction, Western Christians found themselves puzzled and troubled, for they translated hypo-stasis quite literally into Latin and came up with sub-stantia or “substance”—and thus a term for divine threeness in Greek was equivalent to a term for divine oneness in Latin![40] J. N. D. Kelly perceptively remarks, “Every alert reader must have noticed, and been astonished by, the extent to which theological divisions at this time [mid-4th century] were created and kept alive by the use of different and mutually confusing theological terms.”[41] If we are to escape such “mutual confusion” in our day, then theologians will need to devote careful attention to the precise relation between order and equality in the impeccably orthodox trinitarianism of someone like Athanasius, and they will need to provide a creative and appropriately subtle articulation of that relation—and especially of divine order—for an age that can no longer take the goodness of “order” for granted.


[1] Augustine, On the Trinity IV, 29; Gregory Nazianzen, Oration XL (On Holy Baptism), 43; John of Damascus, The Orthodox Faith I, 8; Aquinas, Summa Theologica I, 33, 1; Calvin, Institutes I, xiii, 20; Hooker, Polity, v. liv. (cited in footnote 118 of Augustine’s On the Trinity IV); Edwards, Ms sermon on John 15:10 (cited in Amy Plantinga Pauw, “The Supreme Harmony of All”: The Trinitarian Theology of Jonathan Edwards (Grand Rapids: Eerdmans, 2002), 106). Here and elsewhere in this essay, patristic references not otherwise noted are taken from the respective volumes in The Nicene and Post-Nicene Fathers of the Christian Church, ed. Philip Schaff and H. Wace (Grand Rapids: Eerdmans, 1971).

[2] See the Arian Confession quoted in Athanasius’s Of Synods 16.

[3] Athanasius, Against the Arians III, xxvi, 29.

[4] Against the Arians III, xxiii, 4 (italics added).

[5] See Philip Schaff, ed., The Greek and Latin Creeds, vol. 2 in The Creeds of Christendom, 6th ed., rev. David S. Schaff (Grand Rapids: Baker, 1998; repr. of 1931 ed. published by Harper and Row), 66-68.

[6] See, for instance, the argument by David S. Cunningham, These Three Are One: The Practice of Trinitarian Theology (Malden, MA: Blackwell, 1998), 112.

[7] Against the Arians III, xxvii, 36.

[8] Against the Arians IV, 1.                             

[9] Against the Arians I, v, 14.                                     

[10] See the account of Athanasius given by Alvyn Pettersen in his Athanasius (Harrisburg, PA: Morehouse Publishing, 1995), especially p. 176.

[11] Anthony Meredith, The Cappadocians (Crestwood, NY: St. Vladimir’s Seminary Press, 1995), 106. Similarly, T. F. Torrance acknowledges that the Cappadocian view of the Father as the Origin and Cause of the Son and Spirit involves “a unique notion of cause, as comprising and continuous with its effects” (The Trinitarian Faith: The Evangelical Theology of the Ancient Catholic Church (Edinburgh: T & T Clark, 1995), 237). For an alternative view of the Cappadocians, which sees them as still unwittingly captive to Neoplatonism and which sees Athanasius alone as escaping from an implicitly Arian view of intra-divine order, see Thomas G. Weinandy, The Father’s Spirit of Sonship: Reconceiving the Trinity (Edinburgh: T&T Clark, 1995), 10-13; see also the fuller discussion by Torrance, pp. 236-247, 302-323.

[12] Summa Theologica I, 42, 1. Here and elsewhere, I am quoting from the translation of the Fathers of the English Dominican Province, 2nd and rev. ed., available online at

[13] Oration XL (On Holy Baptism), 43.

[14] These two positions are fairly and polemically represented by two evangelical organizations, the Council on Biblical Manhood and Womanhood (the “complementarian” perspective) and Christians for Biblical Equality (the “egalitarian” stance).  Both organizations offer vast supplies of bibliographical resources for those who want to pursue the gender question: see, respectively, and

[15] Wayne Grudem, Systematic Theology: An Introduction to Biblical Doctrine (Grand Rapids: Zondervan, 1994), 250-251.

[16] This logic is made explicit in, for instance, Bruce Ware’s significantly sub-titled book, Father, Son, and Holy Spirit: Relationships, Roles, and Relevance (Wheaton, IL: Crossway, 2005), pp. 43, 69, and 103.

[17] The Orthodox Faith I, 8 (italics added).

[18] This point is strongly and repeatedly emphasized by the most prominent evangelical critic of “roles” language, Australian scholar and egalitarian Kevin Giles. See especially his The Trinity and Subordinationism: The Doctrine of God and the Contemporary Gender Debate (Downers Grove, IL: InterVarsity, 2002) and, more recently, Jesus and the Father: Modern Evangelicals Reinvent the Doctrine of the Trinity (Grand Rapids: Zondervan, 2006).

[19] Much Roman Catholic theology, following Thomas Aquinas, has formalized this rough-and-ready differentiation in its doctrine of “appropriations”. See Summa Theologica I, 39, 7-8. See also the account of the origins and development of the doctrine in Lewis Ayres, Nicaea and Its Legacy: An Approach to Fourth-Century Trinitarian Theology (New York: Oxford, 2005), 296-300.

[20] Institutes I, xiii, 25 (italics added). Here and elsewhere, I am quoting from Henry Beveridge’s 1845 translation (Grand Rapids: Eerdmans, 1972 repr.).

[21] See the summary in the Catechism of the Catholic Church, 2nd ed. (United States Catholic Conference, 1997), 67 (paragraph 255).

[22] For an account of the diversity of assumptions that may underlie affirmations of “equality”, see Elizabeth S. Anderson, “What is the Point of Equality?” Ethics 109 (January 1999): 287-337. For a more popular account of the problem, see C. S. Lewis’s address “Membership” in The Weight of Glory and Other Addresses, rev. and expanded ed., ed. by Walter Hooper (New York: Macmillan, 1980), 106-120.

[23] See the common (albeit often implicit) inference in Giles, Jesus and the Father, 59, 121, 127, 142, 238, etc.

[24] Karl Barth points to this same “most offensive fact” in the modern context when he investigates the common assumption that “there is necessarily something unworthy of God and incompatible with His being as God in supposing that there is in God a first and a second, an above and a below.” Against this idea, Barth asks a series of provocative questions: “Has there really to be something mean in God for Him to be the second, below? Does subordination in God necessarily involve an inferiority, and therefore a deprivation, a lack? Why not rather a particular being in the glory of the one equal Godhead, in whose inner order there is also, in fact, this dimension, the direction downwards, which has its own dignity? Why should not our way of finding a lesser dignity and significance in what takes the second and subordinate place . . . need to be corrected in the light of the homoousia of the modes of divine being?” (Church Dogmatics, Volume IV: The Doctrine of Reconciliation, Part 1, tr. G. W. Bromiley, eds. G. W. Bromiley and T. F. Torrance (London: T&T Clark, 1956), 201-202). See also the similar conclusion, even expressed in terms of “obedience”(!), reached by John Thompson in Modern Trinitarian Perspectives (New York: Oxford, 1994): “The obedience of the incarnate Son reflects that of the Son in the eternal life of God. . . . In going this lowly way of obedience to death Jesus is not following a capricious or arbitrary way but one which God the Father has chosen for him and which has its basis in the very being of God in the relation of Son to Father” (48).

[25] I draw the same conclusion, for the same reasons, regarding expressions like “the Father eternally has authority over the Son” or “the Son eternally submits to the Father”. These idioms are easily misunderstood (and I must confess that Ware’s untroubled descriptions of plain, easy-to-grasp “lines of authority” within the Trinity leave me uneasy—see, for instance, Father, Son, and Holy Spirit, p. 95), but I think they can be helpful when set in the proper context.

[26] Maurice Wiles, Archetypal Heresy: Arianism through the Centuries (Oxford: Clarendon Press, 1996).

[27] See Charles Hodge, Systematic Theology (Grand Rapids: Eerdmans, 1993 repr.), 1:462, and elsewhere; A. H. Strong, Systematic Theology (Valley Forge, PA: Judson, 1907), 342; J. N. D. Kelly, Early Christian Doctrines, rev. ed. (San Francisco: HarperSanFrancisco, 1978), 263; and Phillip Schaff, History of the Christian Church (Grand Rapids: Eerdmans, 1958), 3: sec. 130. For other, more recent instances of this usage, see the list compiled by Giles in The Trinity and Subordinationism, p. 23 note 8.

[28] Thomas F. Torrance, The Christian Doctrine of God: One Being Three Persons (Edinburgh: T&T Clark, 1996), 176 (italics added). Of course, the subtitle of this book tells the same tale.

[29] On the Trinity VII, 9.

[30] See John D. Zizioulas, Being as Communion: Studies in Personhood and the Church (Crestwood, NY: St. Vladimir’s Seminary Press, 1985), 37. See also the very careful explication of Zizioulas’s outlook in Aristotle Papanikolaou, “Divine Energies or Divine Personhood: Vladimir Lossky and John Zizioulas on Conceiving the Transcendent and Immanent God,” Modern Theology 19, no. 3 (July 2003): 367-368.

[31] Colin E. Gunton, The Promise of Trinitarian Theology (Edinburgh: T&T Clark, 1991), 11.

[32] Torrance, The Trinitarian Faith, 321.

[33] Torrance, The Christian Doctrine of God, 157 (italics added).

[34] The Trinity and Subordinationism, 81. The complaint is raised about the conclusions found in the Sydney Anglican Diocesan Doctrine Commission Report, “The Doctrine of the Trinity and Its Bearing on the Relationship of Men and Women,” (1999), Year Book of the Diocese of Sydney (Sydney, NSW: Diocesan Registry, 2000), 538-550. The document is reprinted in its entirety in The Trinity and Subordinationism, 122-137.

[35] This problem is related to the one pointed out by Zizioulas in his defense of the Eastern Christian affirmation—quite shocking to many Western ears—of the “monarchy of the Father” within the Godhead. See Being as Communion, especially pp. 27-100. For critical discussion, see Alan J. Torrance, Persons in Communion: An Essay on Trinitarian Description and Human Participation (Edinburgh: T&T Clark, 1996), 290-295. My argument here, however, is not that the Eastern approach to the Trinity (if Zizioulas’s really is the Eastern approach!) is superior to the Western. One can, I think, hesitate over ascribing “monarchy” exclusively to the Father and still recognize the flagrant modalism toward which all non-Arian “roles” language tends.

[36] The New Schaff-Herzog Encyclopedia of Religious Knowledge (Grand Rapids: Baker, 1949-50), 20-21 (italics added).

[37] Summa Theologica I, 28, 1 (italics added).

[38] See Robert Letham, “The Man-Woman Debate: Theological Comment”, Westminster Theological Journal 52 (1990): 65-78. See also the extended discussion in his more recent The Holy Trinity: In Scripture, History, Theology, and Worship (Phillipsburg, NJ: P & R Publishing, 2004), especially pp. 175-181 and 389-404.

[39] History of the Christian Church, 3: sec. 130.

[40] Robert Jenson’s account of the “linguistic settlement” in Greek, and the trouble that resulted among Latin theologians, is instructive. See The Triune Identity: God According to the Gospel (Philadelphia: Fortress, 1982), 103ff.

[41] Early Christian Doctrines, 253.

  • 1. This article originally appeared in Pro Ecclesia 18, no. 3 (Summer 2009): 255-272. It is reprinted here by kind permission of Steven D. Boyer, Ph.D..
Dr. Steven Boyer

Dr. Boyer is Professor of Theology at Eastern University.