Why Theosis is Nothing to Fear: A Patient Lesson in Logic, Participation, and Peace of Mind

From the earliest centuries of Christian theology, believers have spoken with remarkable boldness about salvation. Few statements are bolder than the one found in Irenaeus of Lyons and later given its most famous expression by Athanasius of Alexandria:

αὐτὸς γὰρ ἐνηνθρώπησεν, ἵνα ἡμεῖς θεοποιηθῶμεν

"He became human, so that we might be made godlike."

From the beginning, however, this claim was never intended as a denial of the Creator–creature distinction. It was a disciplined way of speaking about the depth of salvation rather than its confusion.

For many modern readers—especially within Protestant traditions—the sentence can nevertheless trigger immediate alarm. It sounds as though something essential has been crossed. Are we saying that human beings become divine? Does the line between Creator and creature dissolve? Has Christian theology slipped into myth, mysticism, or, some fear, even blasphemy?

Historically, such reactions have been intense. In the late nineteenth century, Adolf von Harnack recoiled from the language of deification, convinced that it represented a corruption of Christianity by Greek metaphysics or mystery religions. Much later, similar anxiety resurfaced in German reactions to the Finnish Luther research associated with Tuomo Mannermaa. The response was often emotional, even fearful. Something, it was thought, was about to collapse.

That fear is understandable. We can make no progress in understanding God’s grace and the justification of the sinner if we lose what Luther presupposed: the infinite qualitative distinction between the divine and the human. If that distinction collapses, grace ceases to be grace.

But the fear ultimately rests on a confusion—one that can be removed once we slow down and attend carefully to what is actually being claimed.

The Christian tradition never meant the crude slogan, “God became human so that humans might stop being human.” What it consistently meant was something far more careful, and far more faithful:

God became human while remaining God, so that the human might become godlike while remaining human.

The question, then, is not whether the claims of theosis are dangerous. Claims that salvation involves human beings becoming, in some sense, godlike are not in themselves dangerous. The real question is whether we understand the logic of the relations involved—and therefore why these claims do not threaten the faith.

To see why these fears arise, and why they are unnecessary, we need to pause and learn a few very simple but powerful logical distinctions.

Step One: What Is a Relation?

A relation tells us how things are connected. “Is taller than,” “is the parent of,” “is identical with,” and “depends upon” are all relations. Theology uses relations constantly, often without noticing that it is doing so.

What matters is that different relations behave differently. Once we fail to notice how a relation behaves, we begin drawing conclusions that do not actually follow.

Step Two: Four Basic Logical Features

A relation is reflexive if everything in a domain is related to itself. Identity is reflexive: I am identical to myself. Participation is not reflexive. Nothing participates in itself.

A relation is symmetric if it runs both directions. “Is married to” is symmetric. If A is married to B, then B is married to A. Participation is not symmetric. If a creature participates in God, God does not participate in the creature.

A relation is transitive if it carries across chains. “Is taller than” is transitive. Some relations behave this way; others do not.

A relation is connected if everything can be compared with everything else. A total ranking—such as the natural numbers—is connected. Many relations are not. Some things are simply incomparable. Consider the British constitution, the feeling of remorse, and the number 1729 with respect to the relation “higher than.” Nothing follows.

These features determine what follows from what. Most theological anxiety arises when one of them is silently assumed where it does not belong.

In particular, trouble arises when reflexivity, symmetry, and transitivity are combined into what logicians call an equivalence relation. Equivalence relations divide reality into mutually exclusive regions. Identity is the clearest example.

What critics of theosis often fear is precisely this: that the divine and the human are being identified—placed in the same ontological region. This is what is meant by ontological collapse, and conscientious theologians are right to reject it.

Step Three: Why Theosis Has Frightened People

Historically, fear surrounding theosis has taken four recurring forms:

• If creatures participate in God, then creatures must become divine.

• If divine attributes relate to one another, they must collapse into one another.

• If union with Christ is real, it must be an emanation of divine being.

• If justification is participatory, it must become hierarchical.

Each of these outcomes would indeed be disastrous.

But each rests on the same mistake: smuggling into the relation logical features that it does not, in fact, possess.

Step Four: Three Distinct Senses of Participation

Here is the crucial point. Participation does not name a single relation. At least three distinct senses are at work in the tradition.

Platonic participation describes the relation between a particular and what gives it form. A table participates in tableness. This relation is not reflexive, not symmetric, and not transitive. It does not rank beings or place them on a ladder. Nothing here threatens the Creator–creature distinction.

Neoplatonic participation introduces hierarchy. Higher realities flow into lower ones. This relation is transitive and often treated as connected. Everything lines up on a single scale. This is the metaphysical background of the “great chain of being.” If this were the logic of theosis, fear would be justified.

Constitutive participation is the relation that actually matters for theology. One reality grounds another without absorbing it. A foundation supports a house without becoming the house. A promise establishes a relationship without erasing the persons involved.

Logically, this relation is not reflexive, not symmetric, is transitive, and—crucially—not connected. There is direction without hierarchy. Dependence without ranking.

This is the logical space in which strong theological claims can be made without ontological collapse.

Step Five: Hearing Theosis Calmly

Once these distinctions are in place, the classic affirmations of theosis no longer sound reckless. They sound precise.

• Creatures participate in God without becoming divine, because participation is not identity.

• Divine attributes determine one another without collapsing into one another, because determination is not equivalence.

• Union with Christ is real without being emanative, because grounding is not a flow of substance.

• Justification is decisive without being hierarchical, because not all relations form ladders.

The fear that has haunted theosis has always been the fear of collapse. That fear dissolves when our logic is disciplined.

Properly understood, theosis is not speculative excess. It is a careful confession of grace. And grace—when spoken carefully—never abolishes what God has made.

Theory Construction in Theology

For sometime now I have had the reoccurring thought that theology employs theories in much the same way as the natural sciences. Of course, with the natural sciences, theory construction and disconfirmation is patent. We know that if theory T has as a logical consequence P, and if ~P actually turns out true empirically, then there is something wrong with theory T. (Assuming, of course, that one's inferences are in fact deducible theorems of T.) It is all rather straightforward for the scientist - - or at least it seems prima facie so.

A number of years ago I was excited by some of the similarities I found between scientific theory construction and theological construction. It was to me then rather exciting to think that somehow theology uses theories. (I confess to having a bit of the natural theologian in me in those days.) But something has happened. While it is true that I am no longer much excited by the similarities between scientific and theological theory construction, it is not because the seeming similarities have faded for me. No - - it is because it seems to me now patently obvious, and not at all surprising, that theological theory and scientific theory have the same structure. The excitement has faded because there is no longer anything creative in the thought. They just are of the same kind. Let me explain with an example that is not that of natural theology at all.

I am rereading some of the Finnish work on Luther in teaching 'Luther, Justification and Deification' in the Institute of Lutheran Theology Masters of Theology program. Among the many claims made by the Finns is that Luther employs the notion of theosis as a central motif within his theology, that his notion of justification presupposes the unio cum Christo. The way that this is argued is to take a number of themes in Luther, chart the interrelationship between these themes, and then go to the Luther texts to see if perusal experience is consistent with the theory built out of the interrelationship of these themes.

For instance, they argue that the inhabitatio Christi grounds both forensic and effective justification, that the imputational notion of justification is the divine favor or gratia, while the effective notion is the divine donum or gift. While the favor of God addresses the wrath of God, the gift of God pertains to the corruption of our natures. Just as favor of God undergirds the gift of God, so the gift of God grounds the favor of God. For Luther, justification is a unitary process that includes what is often regarded by the Reformation traditions to be sanctification. God gives Himself to His creation in love, and thus all of creation is butressed by the indwelling God. But fallen creation groans in travail for salvation. This salvation is available through the gift of divine love which is the presence of Christ in the believer grasped through faith. Thus, 'x has faith' and 'x has the presence of Christ' are materially equivalent. (I thought about claiming that they were conceptually equivalent, but I can imagine x having faith without x having the presence of Christ. How is this possible? It seems that much of Lutheran Orthodoxy was quite capable of asserting the truth of the former without asserting the truth of the latter.)

Now these general assertions could be clearly stated as propositions of a theory. One would start with some statement such as 'x has the indwelling of Christ if and only if Christ gives himself to x'. One might say then that 'for any divine property P had by D, if x has P, then x has D itself''. After such definitions, one might declare as theoretical postulates that 'for any x, if x has the presence of Christ, then x is justified', and 'for all x, if x is justified, then x is both declared righteous by divine favor, and made righteous by divine gift'. That this would be tedious work, is readily granted; that it would succeed in laying bare the structure of a class of theoretical assertions is only hoped for.

Given that a theory could be structured in which the logical and conceptual relationships between the assertions of the theory were aptly displayed, the question arises as to the applicability. Is this theory applicable to theological reality itself? Is it applicable to a class of texts written within a tradition, or written by a single author? Is it applicable to the Luther texts? Here is seems that what the theory would have to have besides the internal marks of consistency and coherency, are the external characteristics of applicability, adequacy, and fecundity. I shall treat each in turn.

Theory T would be applicable to a class of texts C if and only if it were not disconfirmed by any particular assertions found within C. Theory T would be adequate to a class of texts C if there were not assertions of C that T could not in principle handle. Theory T would be fecund with respect to C if it generated a continuing program of fruitful and creative insights concerning the relationship of T to other theories.

What is different between scientific theory construction and this theological theory construction is what Heidegger called the Befragte, that which is asked questions on the basis of the theory. In natural science theory construction, nature is the Befragte; in theological theory construction it is most often a class of texts that are questioned. To find out what view Luther held, one must be content to advance theoretical models, some of which are contradicted by the texts and some of which are not, some which fit nicely into other overarching theories, and some which do not. Just as we cannot know the Ding an Sich in nature, but must model nature and build a sustainable "take" on nature given our experience with it, so too in theology, we cannot know the mens auctoritatis (mind of the author), but content ourselves with sustainable "reads" on the basis of the Luther texts themselves. Moreover, just as traditional scientific theory must not be easily discarded in favor of newer scientific theory, but generally regarded as authoritative unless directly contradicted by new empirical evidence, so too should newer theological theory not supplant traditional readings unless there is a compelling reason to do so - - a reason arising from a straightforward experience with the Luther texts themselves.

So I find these days in theology that some statements are "more theoretical" and some "less theoretical" on the basis of whether the first are "further removed" from the primary literary experience of reading the text. So too are some theological terms "more concrete" or "less concrete" as to how they cash on the basis of the particular texts. Accordingly, oftentimes the most "theoretical" of terms are those that are presupposed everywhere in our perusal experience without being asserted directly very many times at all! It is of this latter nature, it seems, that the Finnish Lutheran notion of divinization participates.

Luther: Justification and/or Deification?

In 1988 I attended my first Luther Congress. We met in Oslo, Norway. While there I met a young Finn named Risto Saarinen who gave me a copy of the book Thesarus Lutheri. Later I was given a copy of Luther und Theosis and I began to read.

In the early 1990s, I became quite interested in whether or not Luther was a theologian of theosis (or deification). I remember standing up at the Lutheran gathering at the American Academy of Religion one year, and talking about the new Finnish research. It was new and exciting research in America. At the next Luther Congress in St. Paul in 1993, I was in Mannermaa's seminar. I found him to be an immensely likable man, someone who was willing to question his own research, someone who would genuinely dialogue. I got to know some of the other young scholars in Mannermaa's group. They were intensely interested in theological conversation.

When working through Simo Peura's Mehr als ein Mensch, my uneasiness with the way that the Finns were reading Luther grew. It seemed to me that so much of the thesis of deification depended upon a rather small group of passages, and these mostly from the early Luther. Moreover, as I read a bit more of Augustine (and those that know Augustine), it seemed to me that deification imagery was palpable in the Augustinian tradition. I concluded that in order to show that Luther was a theologian of deification, one would have thereby to establish that he was using the imagery of deification differently from how it was employed by theologians who have generally been thought to uphold justification, not deification, as their central salvific notion.

As I considered the historical question of Luther's adherence to deification, I quickly determined that I would need to know what deification is if I were to be able to determine whether Luther held to it. I looked at the question of what deification is both semantically and ontologically. Firstly, I wanted to know the identity conditions of 'deification' so that the term could be properly applied. Secondly, I wanted to know what state of affairs would make true the claim that deification was present.

My contribution to the Mannermaa Festschrift in 1997 considered the ontology of deification. What claim could we be making about the divine/human reltionship when asserting that person p is deified? While the essay was itself speculative and inconclusive, the exercise was useful to me, for I found how little textual evidence there was to adjudicate among senses of 'deification', and I discerned that some notions of justification were not entirely unlike some notions of deification. In fact, the more I thought about it, the more it seemed like the boundary between 'justification' and 'deification' was becoming porous. What began as a seemingly firm distinction dissolved upon deeper reflection.

Beginning in two weeks, I will present a course entitled 'Luther: Justification and/or Deification'. The course will, as the title suggests, try to get clear on the claim that Luther is a theologian of theosis, by getting clear about what state of affairs would make true a statement about the deification of a person. Accordingly, we shall start in the course by understanding justification in the tradition generally, and the late medieval options on justification. After this we shall read some of what the Finns claim about deification. Looking specifically at the Luther texts, we shall try to answer this question: Was Luther, as Mannermaa has suggested, a theologian of theosis? Please visit the Institute of Lutheran Theology web page at www.ilt.org for details. The course is in the new ILT Masters of Theology program. This degree is designed for those wishing to study theology beyond the M. Div. level. All are welcome. Any takers?