The Statement and the Enunciation
Incorporating a Commentary on:
Barthes. “From Science to Literature.” The Rustle of Language,
trans. Richard Howard.
Barthes begins with an apparently trivial (because uncontroversial) observation: that the institution determines the nature of knowledge. His second point, hidden away as an analogy, is less obvious: the institution determines knowledge in the same way as language determines the way we think. Ferdinand de Saussure’s notion of la Langue can help here as a way of thinking about how institutions function—remember that la Langue designates the institution of similarities and differences from which classes and divisions are produced.
In particular Barthes is interested in the division between science and literature. What defines science, he says, is its status. Science is the name given by society for what counts as knowledge (whatever that happens to be). If, however, we were to define science by its content, its methods, its morality and its modes of communication, then there’d be no significant difference between science and literature.
Having identified the ways in which science and literature can be considered united, he then offers one last similarity—and this is the crucial one—that they are both discourses. In other words, if the only thing that distinguishes a science is that it is a particular kind of discourse then distinctions between kinds of discourse indicate relative values. Literature is no less a discourse than science yet each discourse has a different status from the other.
Now, given that both science and literature are discourses, we assume that they are each constituted by language. This, again, could not be controversial. However, as Barthes writes, “science and literature do not assume—do not profess—the language that constitutes them in the same way.” A scientist uses language as a tool, an instrument of communication. Indeed, the sciences of language—Theoretical and Applied Linguistics—treat their object, language, as if it was a tool or an instrument. For literature, on the other hand, language is the “very being of literature, its very world.”
Let’s now look at the earlier assumption: the difference between science and literature is their relative status. In all other respects they are the same. They are the same, too, because they are both discourses. However, each of them reveals a different attitude towards what they are. Sciences uses discourse as if it was there to function as a tool or instrument in its service in discovering and communicating its knowledge. Literature, on the other hand, acknowledges from the beginning that it, and thus all other discourses too, are constituted in and by the language it thematizes.
So Barthes can say “Literature is alone today in bearing the entire responsibility of language.” This would be because all other discourses fail to reflect on their being discourses. Only literature assumes the language that constitutes it. All other discourses dodge the question of language, and thus the question of discourse, exploiting all its possibilities without acknowledging where those possibilities derive.
What does it mean to assume the language that constitutes you as a speaking being? Here we can draw on a worthwhile distinction (it has different names and is quite old and established in some discourses): the difference between the statement and the mode of enunciation.
Emphasis on the statement would draw attention to its content, its sense and reference, whether it is true or false, and on what attributes or qualities are predicated of what subject. For instance the statement “this vase is yellow” predicates some actually existing ornament with a yellow color and can thus be tested against the actually existing ornament for truth or falsity. If the only part of language we were interested in was its function as statement then we’d all be scientists, predicating objects of all kinds with qualities, attributes and causes and declaring statements true or false or even establishing conditions according to which truth and falsity could be guaranteed.
If we shift our emphasis now onto the mode of enunciation we are less interested in what is being said and instead look at how, by saying it, the speaker is constituted institutionally in some way or another according to value and status. We are interested in the performance or the practice of speaking in this way or that—the role it plays in constituting or perpetuating a particular world of discourse. The speaker is no longer a subject with autonomous feelings and thoughts but, rather, is constituted as this or that according to a modality of discourse. Barthes calls all the different ways of addressing oneself to others “image-repertoires.”
In “From Science to Literature” Barthes is using lessons learned from the failure of structuralism to turn itself into the science of literature, in order to establish the enunciative modality of science itself. In this way he forces science (in his hands, as a structuralist) to reflect on its own conditions of being a science, and thereby taking it outside science (which we must still understand on the model of predicative and cause-effect oriented statements). Science in his hands thus becomes a kind of literature as it begins to more and more rigorously attempt to come to terms with the impossibility of grasping its own conditions of possibility, as a scientist would grasp the identifiable qualities of some object.
Metafiction—at base—would be the kind of fiction that explores the nature and status of fiction. In which case, would it strictly be fiction (which in a standard sense would be non-factual narrative)? For it would explore the possibility and nature of narrative itself. It would thus explore the difference between factual and non-factual narrative. At its most fecund it would perhaps, being reducible neither to fact nor to fiction itself, put into question the rigid dichotomy between fact and fiction. What is the possibility of metafiction? If it is possible a fiction would be thematizable within a fiction. Then it must be possible because there has never been a period without narrative that thematizes narrative and Shakespeare is not the first nor the last but perhaps one of the most persistent metafictionalizers in the history of literary forms. The issue of metafiction shares a common ground with other self-reflexive paradoxes. The concept HORSE is not a concept. Language always can be referential. But that does not mean that it is essentially. Reference (as we know it) is a possibility of language, a potential, but not a necessary fact of language. When a sentence takes itself as its referent a fact emerges (a rhetorical fact) that is not reducible to reference. This is what self-reference teaches us. HORSE can refer to a horse but not to the concept HORSE. The concept becomes an object. In the same way, the fiction about a fiction is not a fiction. Here’s the rub: if something is repeatable it can refer to itself (and then it is no longer itself). The possibility of meta-fiction would be fiction’s iterability. Now—a further implication—if there are texts that deliberately thematize fiction or narrativity, fictionality, metaphoricity—and I accept there are—then there are texts that can be read on a metafictional level even if they did not intend to be. You can you read television and cinema shows that unintentionally thematize the relationship between TV and its audience. Once you know what to look for you see nothing else. At every stage there is a self-reflexive meta-commentary to be observed. I’ve never seen a film where this does not apply. In which case, metafiction would be the rule. All fiction is metafiction. Some fictions acknowledge this. Others don’t. The place to look is at the level of address (who addresses who?). Here we return to the now classical distinction between the subject of the statement (sense, reference) and the subject of enunciation (signifier, address). How can you distinguish the two rigorously? To do so would be to enter the world of the meta-reader.