gaisavoir still2 copy

“We can say all we want about what we see, but what we see is never lodged in what we say.”

Jean-Luc Godard, Le Gai savoir, 1969

Here we confront an even more stark distinction than the earlier comments on seeing, reading and writing had suggested. Now, in marking the difference between seeing and saying, it is not a matter of choosing sides or deciding upon the proper sequence. According to this remark, there appears to be such a rift between seeing and saying that nothing can guarantee a connection between them.

That is not how things usually appear. As we navigate the world around us, the split between seeing and saying is hard to notice because we carry out a loose and apparently adequate translation between them all the time. We often look around our everyday settings with a kind of vague confidence that we could talk about anything on view if we really wanted to. Even if we assume, as a kind of metaphysical prejudice, that saying is fundamentally more true or real or comprehensive than seeing (or vice versa), we usually switch between them as if they were overlapping or complementary ways of dealing with “the same thing.”

Only in certain charged moments— like trying to describe what happened during a soccer match or a bank robbery—do we admit the inadequacy of words that had been nagging us all along. Or we can be tested the other way around—say, during a game of Charades or a tourist adventure— when we are forced to make an imperfect picture of what we could have spoken quite easily.  When words or images—or the ability to toggle between them—fail us, we typically “internalize” the flaw by blaming our eyes, our memory, our talent, or our vocabulary.

Some might say that seeing and saying seem dissociated only because we have not yet learned how to integrate them. Description—the exact notation of the visible—is thought to be a fundamental verbal art, ready to be developed with great precision. Inspection—careful scrutiny of a visible scene and its objects—is likewise thought to be an eminently teachable skill. Maybe each faculty can be trained to such a high degree that they converge into a kind of all-purpose expressive insight. Such a goal seems possible if we assume that both faculties ultimately share the “same” basic thought-process, an otherwise invisible and unspoken coordination of perception and articulation whereby we make sense of what our senses tell us. Indeed, we might even hope for a virtuous circle: the better we can say what we see, the better we can see whatever we able to say. And so on, until all things are seen, spoken, and thereby known.

Contemporary theory would suggest that we have to jettison this whole model. There is no “mind’s eye” where seeing and saying can be reconciled. You might even say—to anticipate the argument a little—that seeing and saying do not belong to the same person, let alone the same world.

gaiavoir still3 copy

Let’s return to the original statement: “We can say all we want about what we see, but what we see is never lodged in what we say.” This line is spoken in Godard’s film by Patricia, who is instructing her friend Emile on the basic principles of “joyful wisdom.” In fact, this is an exact quotation from Michel Foucault’s book Les Mots et les choses [The Order of Things], first published in 1966. It appears in the middle of his celebrated explication of the painting “Las Meninas” by Velàzquez. (This is not the first time Godard has quoted someone writing on Velàzquez, but that’s another story.)

Foucault is explaining why a verbal description of the painting, in particular an inventory of the people represented there, would be unhelpful in an effort to understand what it makes visible. “[The] relation of language to painting is an infinite relation,” he writes, not simply because words are inadequate to convey the richness of the visible, but because “the space where one speaks” and “the space where one looks” operate differently. Here is the passage where Godard found his phrase:

“[It] is in vain that we say what we see; what we see never resides in what we say. And it is in vain that we attempt to show, by the use of images, metaphors, or similes, what we are saying; the space where they achieve their splendour is not that deployed by our eyes but that defined by the sequential elements of syntax.” [Foucault, The Order of Things, 9]

Foucault suggests that we should take the “incompatibility” of language and vision as a starting-point rather than an obstacle. His own reading of Velàzquez proceeds to demonstrate how the visible spectacle of the painting is structured by several invisible relationships. On one hand, the painting can never quite anchor (though it tries very hard) its relationship to the painter and the spectator, who stand outside and look; on the other hand, it can never quite fulfill its relationship to the royal figure whose power it was supposed to make manifest. Either way, no matter how full the spectacle may be, it can never come right out and say what it wants to say. The painting does indeed “show” us quite a bit, and it does indeed “say” quite a lot, but there is no actual “place” where all this showing and saying happens: there’s nobody there, not Velàzquez, not the king, not us.

We should remember that Foucault was writing about a particular painting in order to begin his examination of “classical representation.” Like Godard, Gilles Deleuze wants to draw a more general conclusion from Foucault’s argument.

Here is Deleuze’s commentary on Foucault:

“There is a disjunction between speaking and seeing, between the visible and the articulable: ‘what we see never lies is what we say,’ and vice versa. The conjunction is impossible for two reasons: the statement has its own correlative object and is not a proposition designating a state of things or a visible object, as logic would have it; but neither is the visible a mute meaning, a signified of power to be realized in language, as phenomenology would have it. The archive, the audiovisual is disjunctive.” [Deleuze, Foucault, 64]

Notice how Deleuze has shifted the terrain from “seeing and saying” (which are too easily confined to individual subjects) and “images and words” (which are too easily mistaken for objective raw material) to “the visible and the articulable.” For Foucault and Deleuze, these new terms are dynamic, constitutive elements of knowledge, historically specific and variable, rather than inert and idiosyncratic by-products of anybody, anywhere, anytime. As Deleuze elaborates: 

“As long as we stick to things and words we can believe that we are speaking of what we see, that we see what we are speaking of, and that the two are linked: in this way we remain at the level of an empirical exercise. But as soon as we open up words and things, as soon as we discover statements and visibilities, words and sight are raised to a higher exercise that is a priori, so that each reaches its own unique limit which separates it from the other, a visible element that can only be seen, and articulable element that can only be spoken. And yet the unique limit that separates each one is also the common limit that links one to the other, a limit with two irregular faces, a blind word and a mute vision. Foucault is uniquely akin to contemporary film.” [ibid, 65]

Having pushed the opposition this far—where “statements” and “visibilities” break free from the particular subjects who utter or witness them—Deleuze wants to ensure that we don’t resort to any of the usual compromises or reconciliations. For Deleuze, we cannot be content to say that “the visible” and “the articulable” are forever disconnected. Nor should we be satisifed to observe that, in fact, “the articulable” has prevailed in Western culture over the past 2000 years or so. And finally, we cannot climb into some higher faculty of reason where all previous divisions are authoritatively judged. We have to deal with the disjunction between “the visible” and “the articulable” on their own terms, even as they point to a third dimension they both traverse.

Foucault and Deleuze have their own way of conceptualizing this third space. Without following them very far down their path, we can note that they describe it in terms of “power” or rather “power relations.” As Deleuze puts it: “Seeing and Speaking are always already completely caught up within power relations which they presuppose and actualize” (ibid, 82). This is a very useful formulation, although it leaves all the work yet to be done: for every act of making-visible or making-articulable, we have to examine what powers are in play, how relationships might be altered, and what remains unseen and unsaid.

Godard’s path takes a somewhat different route. No sooner has Patricia finished telling Emile about the difference between what we see and what we say than she offers an alternative statement—as we will see next.


N.B. Godard quotes Foucault word for word. I think the unnamed subtitle translator of Le Gai savoir captures the phrase better than Alan Sheridan-Smith’s version in The Order of Things. Note, however, that the film’s subtitle is missing a word in the first frame above.