Icons and Analogy: Expanding our Language Games

Philosophy would be nowhere without words. Words allow us the clarity and rigor of a solid argument, words can be recorded and exchanged quickly, words can be invented and transformed. However, it can also happen that by favoring certain forms of expression we lose sight of the value of others. We can mistake our oversights as philosophical necessity and use this as a limitation on our discourse. But after Wittgenstein, it is imperative we recognize that language is at play even outside of our technical propositional frameworks. We must welcome all forms of meaningful expression into the conversation, even those whose rigor, if we can call it this, is not so easily pinned down. Would this not apply as well to what is outside words as such, to “language” in its broadest sense of any form of communicative practice embedded in a form of life? Could philosophy learn from the “language” of images?

Of particular significance in answering such a question is the language we use in religion. We are all aware of the danger of thinking God is immediately adaptable to our propositional language. This is why some philosophers have rallied the resources of the word to create certain strategies in language for speaking about God, notable among which is “analogy.” But there are other ways of confronting this problem as well. Jean-Luc Marion, for example, has addressed it through his work on the “icon” and “idol,” most famously in God Without Being.¹ Of course, Marion's “icon” is not strictly identifiable with the tradition of religious art, but his use of the term does remind us that the image, too, has also developed certain strategies of refusing a direct presentation of God, indeed, most notably in the icon. Both icon and analogy are thus ways of using finite expressions to indicate what is beyond expression. Their comparison merits further exploration.

This is not to say that a comparison has never been attempted. Adam Glover has linked these two traditions of religious “language” in his recent New Blackfriars article, “Creation, Icons, and the Language of Poetry.”² This article leaves us with some fruitful considerations for creation, analogy, and religious language based on metaphor and the poetic style of Vicente Huidobro. However, a close reading of the article reveals that while the author claims these innovations rest on his explication of the icon, they in fact have nothing to do with images. They stand on their own right, or at least on a broader understanding of reference that is shaped by and once again absorbed into the linguistic paradigm. This in itself takes nothing away from his conclusions about the nature of religious language, which I invite the reader to consider for herself. But it still remains to be said how the word might genuinely be illuminated by the image, rather than showing once again how an image can be understood in terms of the word. In this paper I aim to do just that, taking up these themes by asking, quite simply, how an icon can enrich our discourse on God.

I begin by pointing out where Glover first stepped aside from his professed interest in the image. By guarding ourselves against a similar move, and correcting what I believe to be a misreading of Jean-Luc Marion, I will make a beginning examination into what the tradition of icons can offer philosophy of religion.

1. Icon as Double Vision

Defining the icon is no easy task, for it quickly plunges us to the heart of theological controversy and oftentimes to ancient philosophical systems foreign to our day. Glover notes some of these difficulties as he engages with John Damascene as well as Jean-Luc Marion. How can we make a visible image of the invisible, uncircumscribable God? —If God did indeed become man, we can show the human face of Christ. How can we give honor to a material picture? —Because it is a conduit of that which it points to. But can't we also honor it in the wrong way? Each answer provokes more questions that would require a much deeper investigation, calling for the full resources of theology, and so perhaps this is why Glover finally settles on a definition which is a bit broader: icons provoke a “double vision;” they show “what they are and as something other than what they are.”³

“icons stand at the intersection between the visible and the invisible; and, as such, they invite us to look in two directions at once and hence to see two realities at once: on the one hand, created reality in all its materiality, and, on the other, created reality as mysteriously saturated by the invisible splendor of divine glory.”

It is this definition of double vision that he rescues from the fray of iconology to guide the rest of his consideration of metaphor and poetry. However, there are two critical problems with this definition as it stands.

1. a) It no longer specifies anything religious. After all, other pictures visibly present something invisible. In a portrait of my grandfather, I am drawn into seeing both what the image presents of my grandfather and the flesh and blood person of my grandfather who is not made of paint and canvas. Thus, a portrait visibly presents someone who is invisible, because absent. Oftentimes the referent of a painting is invisible by necessity, whether because I have no access to its personal presence (a portrait of a world leader), because it is long past (Monet's haystacks at a frosty dawn), or because it is imaginary (Boticelli's Venus). Even beyond explicitly referential painting, we could argue that the abstract art of Rothko or Kandinsky present to us the visibility of an affection, movement, impression, or idea, expressing through colors and form what is as such invisible to the eye, even impossible to put precisely into words.

   In short, to provoke a “double vision,” to present the invisible through and alongside its visibility, is what any painting does. If this is the case, so far we have not managed to explain why the icon has a particularly religious character. Can such a definition bring to bear the full resources the icon can offer to religious understanding?

1. b) It no longer specifies an image as such. A second problem arises. This function of double vision, to show “what they are” and “something other than what they are,” may be true of images, but not exclusively. It might in fact be a general definition of reference. This is what a word does, which is why Glover is able to link the definition so quickly linked to metaphor. It is also what a sign does, and a symbol, each in their different way. There is indeed a precedent for taking “icon” in a way that is not exclusively related to its pictorial function, and recognizing that “visibility” in this sense is not so much the range of phenomena visible to the human eye but rather a close approximation for the word “intelligibility.” And Glover settles on this definition after an interpretation of Marion's analysis of icons, which likewise uses “icon” with a broader, conceptual meaning. If our goal is simply to speak broadly how we as finite beings can think of a God who surpasses any limits, the generality is perfectly acceptable.

   But might the icon also have something to teach us specifically as a picture? If our goal is to do justice to the uniqueness of the icon as a specifically religious image, we will have to alter this definition in both of these respects.

2. What is Unique to the Icon as Religious?

As we have seen, to say that the icon provokes a double vision, of the visible and invisible, is not enough to qualify the icon as religious, since there are many different senses in which something can present the invisible.

A first and obvious solution to this problem would be to appeal to the specific content of this reference. If royal portraits are those depicting royal persons, and abstract paintings are those presenting abstract affections or ideas, then holy images would be those depicting holy persons. An icon would be a “religious double vision.”

But are these different kinds of “double vision” really on the same plane? These other invisibilities portrayed in art differ in kind, from persons to things, but these subjects all by nature have a “visible” (or in some way “intelligible”) expression, and it is this visible expression we meaningfully encounter both in person and in a painting. Persons and things are visible in the literal sense. A goddess may never be actually perceived, but she can be imagined, and therefore portrayed visibly, as for example a Greek woman with a war helmet, spear, and owl. An idea or affection may not be visible like a face is, but artists can still render these abstract experiences visible through shape, color, and texture, at least to some extent. In short, the invisibility here is an absence that could be made present according to some mode of perception, memory, or imagination. Visibility (in its broad sense) belongs to them by nature. Is this really the case with God? Is this same kind of “visibility” appropriate to God by nature?

We have reason to doubt that God's invisibility would be only a simple matter of his absence, like the haystack and the queen. Nor is God's invisibility due to a lack of artistic effort to imagine and render him, like one renders anthropomorphic goddesses and abstract ideas. At least according to the mystical tradition, an intense experience of God's presence is not described as visible clarity but blinding light or deepest darkness. This is not so much due to God's absence, but our inability to take in God's presence. This invisibility of God then is at a different level than any other visible, for it cannot be simply made visible to us because of the limitations of our nature.

This difficulty need not stop us from our attempts to depict God, at least if we follow the lead of language. Assuming pure equivocity between God and creatures, requiring a purely negative discourse which has abandoned all hope of reaching God, would be difficult to distinguish from atheism.⁴ Most philosophers of religion would argue that there is some kind of positive expression that is possible which would avoid this error. But an appropriate expression about God must be qualified. It is different than the other kinds of expressions, a way of thinking which holds each concept or image that comes to us, alongside a negation or surpassing of it: God is like this, but also exceeds this infinitely: “God is good” and “God infinitely surpasses any goodness we know.” By preserving both moves at once, holding loosely, we commit ourselves only to being religious nomads, dwelling here only provisionally until we are called to a different image which is more like, if infinitely unlike, the invisible God. This general practice of language which says and unsays is seen in various ways from thinkers from Dionysius and Aquinas to the present, and philosophers often call it the language of analogy. Thus, the strategy of analogy would advise us that if we want any hope of speaking of God, we must present not just any double visibility, but one that allows itself to be infinitely transgressed in order to be faithful to the vertical invisibility which we are trying to show. An icon by this definition would be a pointer that directs us beyond any definitive place for our gaze.

But how do we apply this logic of analogy to a painting? It is one thing to talk about a word which both expresses a meaning and infinitely distances itself from it. Words are flexible, abstract. Paintings can perhaps have a form of abstraction as well, but this is not the case in an icon, where we are faced with the image of a face. Can we commit ourselves to living out the provisional hold of analogy here? This would mean something like claiming, “This image is like the face of God,” while holding, “The face of God infinitely surpasses the face we see.” But it doesn't quite make sense to apply this logic here. The practice of analogous language operates according to general predicates with a very broad application: Justice, Love, Mercy, Goodness. But an image is not one among other predicates belonging to one person. The image shows the person, the subject of whom we are predicating and removing our predictions.

If not directly following the form of the language of analogy, perhaps we might be able to continue a kind of metaphoric deferral with some icons of the Divine, which present the visibility of God based on scriptural metaphors: the Trinity symbolized as the three angels visiting Abraham, the Father symbolized by a ray of light or a hand reaching from heaven, the Spirit as dove, the pre-incarnate Son as Holy Wisdom. These images show persons or visible figures, but are not meant as direct portrayals, which seems to allow both a visibility and an invisibility.

But in the majority of cases, this will not help us, for the majority of icons are boldly direct. Icons usually give a representation of a very specific person, and lest there be any doubt about this, this person's name is inscribed alongside their painted face. It would not then be a matter of attributing qualities and deferring them, nor understanding a more abstract idea symbolized by a human form. The image insists on being seen as a particular person. Worse, the icon does not even shy away from giving us the image of one who bears the name, in its Greek abbreviation of IC XC: it is Jesus, the human son born in history of Mary, and the Christ, the Messiah, the anointed one promised by God to redeem his people.⁵ And lest we think that historical human name and a traditional Messianic role lets us off safety from making claims about the infinite God, the icon immediately pins us down with a third name: “ho on,” the Septuagint translation of the “Divine Name” (rather, the mistranslation of the Hebrew's refusal to give a name) in Exodus (Ex 3:14), rendered in English as “I am.” This image is of a man, but it is at the same time claimed as the very God who revealed himself to Moses.

Ultimately, the icon denies us the infinite deferral of analogous language, for it leaves us with a directly visible reference to a person who is directly claimed to be God. How, then, do we avoid univocity in our images? Does a religious image simply come down to an image with religious content, in the end, parallel to imagined content, historical content, and so forth? How can we preserve the icon's “vertical” character when it claims to give us directly the visibility and name of God, although both are said to infinitely surpass human understanding of it? Does not the blatant visibility of the image claim too much knowledge, and offer too little for us to recognize it as transcending our understanding?

A second problem presents itself. If we try to define the religious character of the image merely by its content, we miss a very significant and important fact: the icon is a very particular kind of religious image. That is, not all images with religious content are icons. How can we distinguish the images with religious content that are icons from the images with religious content that are not icons?

Both of these problems emerge when we try to model our discussion of images as religious referents after the discussion of religious language. Thus far the word simply fails as a guideline for us; it is not a fitting way to describe the kind of reference of an icon.

2. What is Unique to the Icon as an Image?

These problems may not be unsolvable as they stand. But perhaps we might make more progress with a second starting point. Rather than beginning with defining the specifically religious character of the icon, we might begin by addressing what kind of image the icon is.

An icon in the strictest sense is usually understood to be a particular genre of religious art emerging from Byzantine culture and propagating through Eastern Christianity across history. The content alone is not unique to the icon, for images of Christ, the saints, and biblical events are found in many other traditions of religious art, including Western art. But we must also recognize that from the Renaissance onward, Western paintings have closely paralleled nonreligious art, and are appreciated for their aesthetic value as well as for their commemorative and didactic function, serving as the “Gospel for the illiterate.” An icon, on the other hand, while it may display similar content as Western images and may also serve commemorative or didactic purposes, is not primarily created to be an aesthetic masterpiece or even primarily for knowledge. This is not to say the icon shuns beauty or intelligible content, but that the icon's primary goal is beyond this. We begin to see this when we consider its particular stylistic choices.

One of the most central features of the icon's style, which is directly related to its use, is the direct, face-to-face orientation of the figures depicted with the viewer. In its most basic form, the icon is an image of Christ facing us. To face someone opens up the possibility of communication or personal encounter. The face can be turned more or less to the side while still maintaining eye contact, but this link is broken when the face reaches full profile, turned away from any possible encounter, a gesture we repeat when refusing to engage with someone. (This is why icons reserve the profile for the figures of demons or the wicked, to show us first that the very nature of evil is a closure of encounter, and second, to spare us from a personal relation with someone so dangerous to us.) In fact, many of the other artistic conventions of the icon are also oriented toward this goal. This gaze is also carried by the use of inverse perspective, where lines converge not in a point off in the distance of the painting, but leading out of the painting to converge at the heart of the viewer. The line of the icon is strong yet fluid, resulting not in a static symmetry but a dynamic tension that infuses energy into its geometrical composition. This character of line and perspective opens onto a shallow field within the painting itself. The line, along with the color and other conventions of the iconic style, thus pushes the image forward into the plane of the viewer rather than pulling the viewer back into its receding depths.⁶

From this closer look at the unique qualities of the visible image, with all of the devices of the icon aimed at carrying the gaze of the person depicted into the viewer's space, we are better able to appreciate what Jean-Luc Marion means when he speaks of the icon as a “counter-gaze.”⁷ I find that even before I choose to look at the icon, I have already been under the gaze of the holy one depicted there. Let us not forget that a gaze, a personal presence, is a shocking thing, experienced in a fundamentally different way than the presence of objects. We all feel the sudden pulse of energy upon discovering we have not, in fact, been alone in a room. For we can glance over the inanimate objects in the room with a certain degree of mastery and control, but when we come across the gaze of another person, we find someone measuring us up right back. We wince to stare into another's eyes too long; it's too powerful, like looking into the sun. Only lovers can bear it. The gaze has been spoken of by philosophers before, but in Marion's words, the other's gaze inverts the subjective “I” who masters objects into the dative “me,” the one who receives a gaze, the one who is addressed.⁸ Thus when the icon gazes on me, I become not the aesthetic subject viewing an art object, but one subjected to the gaze of the holy, one receiving the address of the divine.

But we must be careful not to turn this image into a magical object, or to submit God to our power to command his appearing. The painted gaze is not guaranteed, but remains ambiguous. I am not forced to see it as the gaze of the divine, but may treat it as a mythical aesthetic depiction parallel to a statue of Athena. Worse, I may also corrupt it into an idol. The idol and icon are not two fixed material objects, but, as Marion says, “two manners of being for beings.”⁹ Rather than open us to the gaze of God, the idol serves as a mirror of our own desire. It shows us the kind of divinity that most impresses us, the kind of divinity we are precisely looking for. In other words, it is still determined by the expectations of the one who sees the icon, and its worth is determined by their fulfillment. As soon as any icon is taken as definitive, as soon as it is collectively or personally taken as an unsurpassable moment of who God is, it is turned into an idol. Even Andrei Rublev's inspired icon of the Trinity can be turned into an idol if we take it as absolute.¹⁰ No matter how beautiful, authentic, or religiously “correct” the content of the painting, nothing guarantees it will remain an icon. Any God worthy of the name wildly surpasses the images of our expectations, in both delightful and terrifying ways.

But we must go a further step, if we wish to make use of Marion's insight. At least as Marion sees it, idolatry is even more sinister than we realize. It is not only a question of the holy image, but plays back into our language and concepts as well. As long as our approach to God is anchored from the point of view the subject, governed by the limited conditions of possibility for our finite, anthropocentric view, we have not left idolatry. Merely to point out that our concepts are always inadequate is not to find a way beyond them. The full reach of our finite gaze plus our projection beyond it still remains firmly trapped in the horizontal plane of human capacity. Addressing the divine is not calculus. There is no limit of approach to guide our discourse like a mathematical n + 1. A finite plus a finite is a finite. The best we can do is come up with larger expanses of finitude, and then despair of reaching for the infinite which lies on a completely different plane beyond computation.

Thus, at least according to Marion, the icon is not what points infinitely beyond itself. Whatever points for me is still trapped in the realm of the limits of my own understanding, no matter how I add, subtract, or defer, no matter how many metaphors I link up, no matter whether I paint in Byzantine style or Cubist.¹¹ The only way to be free of this, the only way to have an icon is to completely abandon this system of reference. In the icon, I give up my own initiative and open myself to the gaze of God who reaches out to me from beyond my hopelessly finite capacities to speak, refer, or paint. I am not so much a subject who looks at the icon, but someone subjected to its gaze.

Marion's account is not limited to images alone, but begins from the basic visible features of the icon to suggest his conclusions. He has used the image, not the word, to find an alternative to an anthropocentric system of reference. But this sense of religious understanding is not limited to the image, and indeed we can see that while Marion begins from the visual (countergaze), he elsewhere includes the verbal (call) to express the same basic point: that God takes the initiative.¹²

If Marion uses both word and image, others have put forward similar ideas using only words. Take the recent example of Stephen Mulhall. In The Great Riddle, Mulhall has described the question of religious language using Cora Diamond's discussion of riddles.¹³ In the riddle of the sphinx, for example (“What has four legs in the morning, two legs at noon, and three legs in the evening?”), it is not just that the answer is unknown, but that the question itself is not understood before one hears the answer. It is with the arrival of answer that the question's terms begin to be clarified, or which leads very easily and simply to the answer (that the times of day mean a human life, that “legs” includes the old man's cane as well as the child's hands). The riddle thus functions as a suspended meaning whose fulfillment can only come with the answer, which does not come through any of our devising or negating, but through a solution from beyond our initial frame. Similarly, religious language is suspended until we receive the answer that God would give.

Thus, for Jean-Luc Marion, as well as for Stephen Mulhall, even our maximum finite capacity to engage with God falls too short. The religious icon is not the negation or suspension of this condition, as if its visibility directly led us to God, but the entry into an entirely different “logic.” It shows us by the personal gaze already upon us that whatever we are seeking is not of our own devising, but can only be received from elsewhere. The “icon” is religious because it does not depend on my own capacities, because through it God speaks to me. It is God alone, and not my own capacities of imaging or knowing, however creative, that reveals God to me.

3. Icon as Image: Educating the Word

Thus the word and the image have the capacity to reach a similar point, whether we start from the countergaze of the icon or the fulfilled prophecies found in the Scriptures. Playing the one off the other has helped us to find our way through many tangles of philosophy of religion.

However, even if we can set word and image on a mostly parallel plane, let us take one further step. Beyond what we are given in Mulhall and to some extent Marion, I argue that the icon has something unique to contribute as an image, something beyond what other forms of religious expression in poetry, riddles, analogy, or metaphors can offer.

Namely, what is significant about the icon is not that it is a divine pointer, although it may be that. Nor is it limited to the question of who has the initiative, although this is critical, allowing us to escape the closure of finitude by an impossible rupture from outside of us. Rather, we must recognize that the unique feature of the icon over the word is the manner it gives us to touch upon the holy in the image. Marion moves so quickly from image to a more general sense of “icon” that he seems to undermine this point.¹⁴

That is, in the icon, we encounter the holy face to face, and the initiative of the holy invites us not to defer our words in the play of analogy and metaphors, nor to await a word that will someday be fulfilled, but to join here and now in personal communion. The icon is not one-way mirror or word from ages ago, but a face, the sustained presence of a listening gaze. The icon does not give us the direct and full presence of God, of course, but only an image; yet this image holds out to us an invitation to place ourselves before the holy here and now.

We do not have to respond to this invitation. We can turn our own gaze away in profile so we do not see, or we can fixate on the image–sign as an object of human production, a historical artifact, a stage of art history, or a beautiful image. Nothing guarantees such a communion is true, and not mere indulgence of our imagination and desire. The icon is no proof that God exists at all. But icons are created and placed there in such a way as to be seen, for the purpose of calling us into this communion. And the practice of prayer before the icons is to decide to take up the risk and enter into it precisely as an invitation. By looking at this image we are invited to remember that the holy we address is not a “what” but a “who,” and a “who” that invites us into a personal encounter of gazes more intimate than any verbal play.

This face-to-face communion is precisely the link that cannot come from verbal metaphor alone. Thus the image, in this case, can indeed help teach us a different way of seeing religious expression, including language. Words, too, can be words of address, naming, speaking, all of which enter into the practice of prayer before the icon. But words seem to be in danger of slipping away from this relational, communal address. For one can still think of the word as a deferred address, waiting over the ages for the final promise made long ago. One can speak about someone in his absence, but in the icon we are put before him face-to-face, and must recognize that whatever we say, even if we speak about him in the third person, is said to his hearing. It informs us that we are never really alone, away from the absolute, but are always in the listening presence. We are not embarking on a search for a God who is only waiting to be cast into light through our metaphors, but a God who casts the light on us. Nor is it even that we are waiting for God to complete the riddles that we begin—someday. Ultimately, the icon bids us acknowledge that before we even begin to speak or look, we are already known. And what we say is only the beginning of our response to a call that precedes us.

While the icon helps us understand the role of God's initiative, the use of the icon can tell us something we wordy philosophers have a tendency to forget: that talk about God is not first and foremost an intellectual exercise, but a communion we enter, and our words are not first of all intellectual objectification or even endless deferral, but responses of love and communion. Perhaps we also need thoughts that are detached and critical, so that we can better enter in to the truth, just like some representational images take us into another world to better enter our own. We cannot deny that icons do that too, help us better enter into the imagining of what has happened. But ultimately what makes the icon unique is the prayer it presents. It is not a magical talisman or a representational removal. Unlike even analogy or reference, it is the invitation to a personal communal relationship. In short, the icon tells us that talk about God is a prayer, or should be. A discourse about God that does not see itself as a prayerful response to God's first call is indeed blind.