Ruth Ronen. 6 avril 2006

Possible worlds beyond the truth principle

Is there another possible world in which unicorns exist? Does Hamlet exist in the drama written by Shakespeare? Is there a difference between attributing existence to an entity like Frankenstein and attributing it to monstrosity? These are questions which the interdisciplinary discourse between philosophy and literary theory, that is to say philosophy's interest in fiction, has in many ways trivialized. In my book from 1994 the aim was to locate this triviliazation. Granted we accept the discourse of possible worlds, the idea that fictional worlds are very much like possible worlds, as the latter are understood by logicians and philosophers of language, it is difficult to restitute what is different in the literary. Can the literary study benefit from this interdisciplinary standardization? Alain Badiou, in his Petit manuel d'inesthétique, claims that philosophy can dwell with art only when philosophy succeeds to explain the singularity and immanence of artistic truth, that is the singularity, the fact that art conveys a truth that belongs to it absolutely, and immanence, the fact that truth is internal to the artistic effect of works of art. Can possible worlds talk stand up to these conditions and say something that refers to the singularity and immanence of the literary? I would like to approach this problem through the issue of naming and its relation to fictional worlds and I will attempt to show something crucial about the way truth, a truth about fiction, is constituted through the mechanism of naming. I will refer to the problem of naming as developed by Saul Kripke in his lectures given after "Naming and Necessity". Kripke is the first philosopher to say that it is possible to decide whether a proposition is true or false regarding Sherlock Holmes, for instance, even if Holmes did not exist.

In 1973, Saul Kripke delivered “the John Locke lectures” at Oxford, which are in many respects a continuation of Naming and Necessity, and deal with the subjects of fictional names and perceptual error. In these lectures, Kripke wants to reconcile 1) the fact that with ordinary language we can talk about Sherlock Holmes and say whether a proposition is true or false, with 2) the fact that 'Holmes' appears to have no reference. Kripke locates the problem of ordinary language practice and that of constructing worlds in the question of naming and his main idea is that names carry no descriptive content. Kripke's lectures constitute an attack on the descriptivist theory of reference with respect to proper names, according to which a name refers to an object by virtue of the name being associated with a description that the object in turn satisfies.

Concerning the question of proper names, the orthodox view holds that names function by virtue of the predicates associated with the name, predicates that can uniquely pick one object that satisfies them. For instance, Moses as an entity satisfies the attribute: “he who led his people out of Egypt”, but this carries no commitment to Moses' existence. According to Gottlob Frege and Bertrand Russell (whom Kripke opposes), this is an aberrant situation, an aberrant of instantiation without existence. We refer to something that does not exist. For them, the reference is "determined" as whatever fits the description, so if Moses in another possible world does not lead the people out of Egypt, this 'Moses' is somebody else. Whereas for Kripke, the notion of rigid designation is crucial. We baptize an object by picking it out, and speakers who wish to preserve the reference will use this name by passing it from link to link. The name will thus spread and the reference be preserved. In this way, a name functions as a rigid designator: it refers to the named object in every possible world in which the object appears. Myths may arise about an object and the reference will still works even if the myth is false. For Kripke, it is a mistake to identify epistemology with necessity. Even if it is dubitable that Moses existed and did what he did, it does not mean the reference does not work.

Now, rigid designation works just as well in a story-world. 'Hamlet soliloquizes', if true in the story about Hamlet, should prevent us from saying that Hamlet does not exist. Kripke thus holds that “Hamlet existed” is true as Hamlet existed in the story, whether or not it is just a story. Could we say then that everything to be found in stories exist? Not for Kripke, who considers that there may be fictional characters within a fiction. In Hamlet, for instance, the play within the play stages fictional characters. As for the question of the existence of the ghost, that is to say: did Hamlet really see the ghost? It could be debated by literary critics. If, for instance, we agree that he imagined that there was a ghost, then, even though Hamlet existed, the ghost of his father did not. Kripke talks of rigid designation, as for him reference works the same in science and in myths. The same language mechanism is in question. It is not more or less strange to talk about Hamlet than to talk about unicorns. (“Unicorns” is not a proper name, but a common name which yet involves a similar problematic). He writes “no pagan gods would exist if paganism didn't exist”. All names in fiction really refer. Even though there is a distinction between Hamlet and Napoleon, there is no difference in naming them. The naming of fictional things does not pose an ontological problem.

Then how does rigid-designation guarantee the designation of fictional characters? There is only one entity, a unique link of name to entity. Kripke refers again to the example of unicorns. If, in Africa, a species was to be found with the properties of unicorns, or, alternatively, medieval culture ran into a unicorn in the forest and discovered that it does not have one horn but another tiny one hidden under its mane, medieval people would continue to see the magic animal as before and the species found in Africa would be called by some other name. The uniqueness of reference must be kept and it is bound by a context. Mythological unicorns are not affected by zoological discoveries. This is how language behaves. If so, one cannot say that the animals found, if similar to unicorns, are instances of the species unicorn. One must acknowledge the connection with the myth in which unicorns were founded. This is the necessity that attaches the animal to its name. Now, given that there are no unicorns, are there other circumstances under which they could exist? Not for Kripke. Names are expected to fix entities. In deciding whether a proposition concerning unicorns is true or not we are carried to the medieval moment of baptizing, with which the practice of language operates beyond the principle of truth. For Kripke, the notion of rigid-designation traverses the question of truth as generally conceived. The difference between truth and falsity has changed its grounds.

In order to contextualize these thoughts about naming and to explore their literary significance, I would like to consider two examples that do not undermine philosophical propositions but rather make a productive use of them. With both examples the relevant distinction is not that between the fictional and the actual as they traverse the line dividing fiction from the outside world. In both examples the name constitutes a context of its own within which reference is guaranteed beyond the question of truth. 1. Rembrandt The first example concerns Rembrandt and the constitution of the self through his act of painting, a thesis forwarded by Svetlana Alpers, a Dutch art historian, according to whom Rembrandt uses his name as a way to create an image of his self. His real artistic revelation had to do with his practice, both: - in the studio - and in the art market. Within the studio, Rembrandt imposed his image of self by changing things in his models, according to his own conception. As for his relation to the art market, he changed the mechanism of how the market determines the value of an artwork. For example, both he and Rubens had apprentice workshops working for them, but the difference was that Rubens did not sign his works. According to Alpers, trying to say which painting was painted by Rubens and which by his assistants is futile and irrelevant. Whereas for Rembrandt, there is a whole research going on to find out which works he painted and which ones were the work of his assistants. The question of identity is here decided by Rembrandt's own practice within the art world. With Rembrandt, the identity of the painting-self became relevant to the way his art is looked at. This is precisely what S. Alpers attempts to account for, this moment where Rembrandt creates an unprecedented possibility of the observer looking at an artwork and saying "this is a Rembrandt!", that is what his innovation constituted. From now on, the name is attached to the work. The artwork is given the name of the painting-self. In terms of Kripke's position we can say that "this is a Rembrandt!" refers to a moment of baptizing by the self.

Rembrandt did several things to add up to the value of his work, without attempting to comply with the taste of the public, which is an exception in the art of his time. Rembrandt, although a great expert in the art of repetitive reproduction, the art of etching, marked these works as unique by inserting minor changes from one etching to the next. The same is to be found in his self-portraits (he did more than seventy of them), a practice through which he created an image of his self. For Alpers, “Rembrandt was an entrepreneur of the self”, had a real interest in creating images of the self. For Rembrandt, the self-portraits were not a way to reproduce a given identity; on the contrary, the accumulation of self-portraits questions the possibility of representing the self. Rembrandt managed to create some value for self-imaging beyond representation, and was in this sense a radical artist. But Rembrandt also used the market to the same purpose. The name Rembrandt was associated with this investment: there is no pre-given self or entity attached to the name, but it has become a baptizing moment in which “a Rembrandt” has been born. The name thus refers to more than the person of the artist. It is the artist, “a Rembrandt”.

2. Joyce according to Lacan. My second example concerns Joyce's Ulysses, and Lacan's reading of it in his Seminar 23. Lacan makes the claim that Joyce did not have an image of his own body; he lacked an imaginary ability to consolidate and totalize his self, his ego. In A Portrait of the artist as a young man, Stephen himself experiences the failure to grasp his body. It is striking in some scenes of the novel, for instance when Stephen is beaten up by school-mates and he doesn't feel like his body belongs to him. According to Lacan, the paternal function did not work for Joyce, and writing served him as support of the self. This compensation through writing is described by Lacan as the attaining of a name. Joyce used to say that critics would have to decipher his work for more than 300 years. He replaced the name of his father who did not function as father to him with his own name of the artist (and not of his person). He turned his name into something more than himself, by making it into a common noun: "This is a Joycean writing!" .

What is the nature of the possible worlds that art makes in which proper names can change their value in this manner? Joyce's name is made into a master signifier. The name of the artist replaces that of the father. No father controls meaning, a fact in the light of which we can re-read the "Circe" chapter in Ulysses. This drama of identities, in which Bloom changes his self-image incessantly, in which we are told that his father changed his name - attests to the decline of the name of the father and to the way writing stands for the missing key. Writing constitutes a self for the artist.

Debate:

Daniel Ferrer: Are these John Locke lectures published?

Ruth Ronen: Not as far as I know.

Daniel Ferrer: What you say about Rembrandt's use of painting to establish the characteristics of an empty name is true enough, but wouldn't it be also true of many previous artists? It is more obvious in the case of a compulsive self-portrait painter like Rembrandt and his historical moment is an important turning point in this respect. But, according to Vasari, even a medieval artist like Giotto would make himself known to the Pope by his way of drawing a circle, a zero, the most impersonal of shapes.

Ruth Ronen: Rembrandt did not invent anything, but with him, the moment of baptizing is very important. What makes 'a Rembrandt', what makes 'a Giotto', is not something we can characterize, and yet the name functions in a way that activates this language mechanism.

Françoise Lavocat: Saul Kripke's theory concerning possible worlds as naming is not interested in fiction. You underlined this important point several times in your book. You refer now to a completely new direction of Kripke's theory which allows us to apply it to fictional beings. Could you clarify this crucial shift in Kripke's thought ?

Ruth Ronen: One couldn't really speak of a shift, as the fact of excluding fiction in Naming and Necessity was not really motivated. He didn't change his view radically.

Otto Pfersmann : Thank you for this very stimulating talk triggering a lot of questions. First, you seem to be saying that Kripke is concerned with ordinary language. But it is far from sure that he does not just want to show that language may precisely not be working without a certain normative frame. Reference is not just an empirical question. Second, you raised the important problem of the value of a name. Is its object just the person Rembrandt or Joyce? Or has the very term “Rembrandt” or “Joyce” a second value and a second function: to epitomise art and a certain way of life. In the examples you draw on, name-terms do not refer to the self or person. “Joyce” refers to Joyce and to something else as well, i.e. a concept of a certain way of conceiving literature and life, just as “Rembrandt” is not only the name of a certain person, but a concept of a certain way of conceiving the visual arts. But becoming a concept, we find again what was previously expelled from the Krikpean picture of names, namely their descriptive content. Rembrandt could have been someone else, perhaps even someone who had never painted, but the reference of “Rembrandt”, if used as in the work of Alpers, cannot be anything else but the conceptual content which she precisely introduces. Hence, we have lots of techniques allowing us the use of name terms for definite descriptions, even though it may be correct that no definite description may be able to fix the reference of “Rembrandt” intended to refer to a person. In other words, “Rembrandt” is an ambiguous term, referring both to Rembrandt – whoever he may have been – and to an artist placing himself at the reflexive centre of visual art, etc

Ruth Ronen: I didn't say that Kripke was talking about ordinary language. He is a Kantian in the sense that he is interested in the things that decide the necessary conditions under which language can function. His theory is demonstrated through ordinary language, but it is not about it. There is a whole debate between Kripke and Putnam, about what necessity is, its status. His position regarding the place of necessity in logic is tested through ordinary language practice. Now, about the second part, I wanted to show that a Rembrandt can function without any content value assigned to it. Kripke's rigid designation works: it's not the way he paints that names Rembrandt's production as 'a Rembrandt'. The name Rembrandt comes to function because it is an empty name, it does not carry any descriptive value of any kind. What Joyce means is that it is impossible to pour content in the name Joyce, and yet it functions in our culture, we know whom we are talking about.

Anne Duprat: There is also the question of the value of the name in economical terms; the name Rembrandt is related to the art market.

Ruth Ronen: True. Rembrandt had a habit of painting in order to pay people back. When he had a debt, he started a painting but didn't finish it, so the person for whom he painted would ask him to finish it, and he would bid an additional sum to finish the work up. He thus created an exchange around his work to create a value. Although he was interested in money, it was also a way of making the art market into a symbolic functioning system. He never painted what was expected. He always manipulated the relation to the end-product and to the person for whom it was done, thus tying together the market issue as money and a symbolic system.

Daniel Ferrer: Zizek suggests an interesting corrective to Kripke's theory: rigid designation is not the result of a primal baptism, but of a retrospective baptism. This appears very clearly in the cases of Rembrandt and Joyce as you describe them.

Régis Salado: You said that one of Joyce's ambitions was to turn his name into a common name. It is difficult to grasp the idea of a common name because a common name includes conceptual content. In this regard, we know the figure of speech called antonomasia: turning a proper name into a common noun.

Ruth Ronen: The turning of the proper name to a common name is something Lacan says about Joyce. A common name is a name of a concept, but Lacan's intention, as I try to figure it out, is to show that the name is made into a signifier. It points at something peculiar to this name without articulating what this something is.

Régis Salado : Joyce started by signing Stephen Dedalus. It was his pen name, a path towards becoming Joyce, the artist.

Ruth Ronen: There is a whole development in his work. It culminates in Ulysses. One should notice the anarchistic way in which names are used in Ulysses.

Régis Salado : for Finnegan's Wake, he considered having James Stephen who was born on the same day to finish writing the book. It is interesting to notice that the first name is that of Joyce, and the family name the first name of Stephen Dedalus. For Joyce, the question of name and creation were deeply linked.

Ruth Ronen: I am not an expert in Joyce. I became interested in him because of Lacan and have written a chapter about him in my recent book, Aesthetics of Anxiety in the context of my discussion of the notion of 'genius'. A genius creates outside the law, and yet becomes a model for others. Kant talks about the exemplary originality of the genius. The question of names can enter in this context of the way the genius' name will be used, the way his/her name functions in relation to art. Joyce's use of names in novels is something particular that cannot be articulated in terms of a law or a universal procedure. His particular use of names definitely cannot be interpreted in terms of meaning. The laws of language fail when we encounter Joyce.

Otto Pfersmann: Kripke's point, we take it, consists in refuting the descriptive theory of names. You said the name is empty: it refers to something, but we cannot assign it a meaning. The problem is: can we eliminate all properties in establishing a causal chain by virtue of which a name refers to an object? This is difficult to achieve without any reference to precisely what makes a causal chain being a causal chain, i.e. properties of things. Concerning literature, I wonder whether we are not in an error of focus. If literary reference (however it may really work) were empty, there would not be much to be said about Hamlet or Don Juan. Instead, there is a difference between emptiness and indeterminacy: indeterminacy can be dealt with in terms of the theory of interpretation. This is what literary theorists do all the time, and they try to say something relevant in terms of properties of the persons-in-the-fiction. So are you endorsing the position according to which reference is exhausted by naming? Or instead of emptiness, there would be some degree of indeterminacy, and we could say relevant things.

Ruth Ronen: The fact that “names are empty” does not go against literary criticism. Yet I wouldn't solve the problem of naming through the alley of indeterminacy. Joyce says: you'll read my work 300 years and not be able to say anything conclusive about my writing, and yet we can say of his work: “this is Joyce”. His name is imbued with real presence. His name is determinate and yet irreducible to interpretation.

Françoise Lavocat: In Possible Worlds and Literary Theory, you define a literary world as subjected to “a unique ontic position” (relative to worlds of other ontic determinations), or subjected to one “modality.” Is naming, in your view, one of the components of this “ontic position”? How do you connect naming and modality?

Ruth Ronen: Naming oversteps differences between modalities. "What is a Rembrandt?" is a question that oversteps this: 'Rembrandt' is not an artistic, nor an actual name, it is both. Naming subsumes differences of modalities, because for a name to function it takes the same necessity of baptizing the conditions its application.

Sophie Rabau: In the case of the name of an author, what matters is not so much the difference but the principle of unity. Saying Joyce allows us to take a principle of unity, no matter what the meaning of the name is. Ex: Homer: common name, proper name, trope... It can even be found in the plural : des homères, until at one point the name of Homer could not be used, it had become impossible to use it.

Ruth Ronen: In a way, it enables us to re-think unity (biographical, personal unity...) in relation to the use of names. Does the use of names prove the coherence of the person? Bloom goes through a transformation. He changes identity and yet remains Bloom. What, then, does the name refer to? He is even split into two personas at the end, and still the uniqueness is retained. In psychoanalysis, the subject has a particularity that can be pointed out only with a signifier and nothing else.

Sophie Rabau: But precisely, Bloom's name changes all the times in Ulysses.

Otto Pfersmann: You cannot escape the real name (this is the same in all possible worlds).

Sophie Rabau: : Joyce points to a danger, the name changes when the context changes. His name could then change according to the different parts of his work. Pessoa for instance did it. Stendhal did too. This is the opposite of Kripke's theory in practice.

Ruth Ronen: No, precisely. That was my point about the particularity of the literary. The creator/writer's name functions as a common noun; the name points at what is real about creation, without articulating the content of creation or the characteristics of the creator. 'This is Joyce!' no matter how the context changes, and regardless of how his different works are deciphered.

Sophie Rabau: thanks to rigid designation!

John Pier: If naming the author can be taken as a means of designating his work or the unity of a given work or even the multiple identities projected by that author, then it would seem that the discussion is now bordering on Booth's much disputed notion of “implied author.” Rejected outright by some and radically re-theorized by others, the implied author is generally understood as the author's “second self” who is responsible for the work as a whole, either the image the author projects of himself or that image as reconstructed by the reader. Is it possible that naming in Kripke's sense might shed light on this problem?

Ruth Ronen: Booth's author is a fictional concept, of an imaginary author. In any case there is a difference between designation and unification. The fact that 'Rembrandt' designates all occurrences of certain artworks does not mean that it works as a unifying mechanism, which Booth's concept clearly does.

Françoise Lavocat: Do you connect the theory of names with transfictionality? There are characters that have the same name in different books. For instance, Don Juan. What is the status of these beings? When they are the 2nd, the 3rd, or the hundredth to have the same name?

Ruth Ronen: Baptizing is a 'thick' procedure, in the sense that it is not a mechanical thing that you take for granted. Don Juan can be said to refer to everything our culture said about him but the point about the designation by the name in the way I described is that the artist accedes to his name. References are subsequently made possible. Names function by using ordinary language mechanisms but it is clear that at a certain context 'Don Juan' can bread the chain of designation (as the case of Unicorns exemplified).

Otto Pfersmann: But this leads to a problem: then there is a world that conflates all Don Juans. Whereas Tirso de Molina's Don Juan is different from Da Ponte's, even though there are some family resemblances.

Ruth Ronen: You are resorting to the theory of content which Kripke opposes. According to you, these would be different entities because they are characterized differently?

Otto Pfersmann: Yes.

Françoise Lavocat: So, for you, there is no transfictionality! I disagree.

Otto Pfersmann: In literary analysis, one would be conflating them. But precisely, Don Juan is not Don Giovanni.

Ruth Ronen: Then, for you, there is no rigid designation? Each refers to a different entity?

Otto Pfersmann: There may be rigid designation, but not to the same entities. The same sounds do not make the same name.

Ruth Ronen: For Kripke, fiction is a problem for philosophers. There is no bottom line with regard to the question of naming. Kripke's theory is useful for literary production outside literary interpretation. In my book, I explain how interdisciplinarity can lead to a rigidity of a new kind. Concepts are taken and made adjustable or shaped for literary theorists. We should not aim at solving fiction transdisciplinarily. The question is to know whether it is productive to use Kripke's theory about naming. Well, sometimes it is.

Otto Pfersmann: What do we mean by “productive”? What are the criteria? To what account is Kripke's theory accurate? It tries to eliminate all conceptual traces of content.

Ruth Ronen: Kripke shows that designation, and a successful one, does not depend on conceptual content. “Moses” does refer successfully even if the Bible is wrong. How is that possible?

Anne Duprat: Is there a shift between your statements of 1996 about the use of P.W. literary theory and your attitude towards the problem today?

Ruth Ronen: There has been a shift I can only now appreciate as being an ethical shift. In my book from 1994 I thought the philosophy of possible worlds can enlighten literary theory. But the problem is by getting tuned to the philosophical subtlety of possible worlds talk we can lose what is particular to art. Maybe analytical philosophy is not equipped (“equipped” is not the right word) to the task of grasping this particularity as it tends to commensurate the question of art and literature with certain other things. Alain Badiou has a very different view as he tries to say something philosophical about the truth of art, to grasp its immanence.