The question I hope to answer is how does Derrida interpret this responsibility. "In other words," how is the essay "Living On" both about and not about Shelly's Triumph of Life. In order to investigate this question, I looked at the recommended reading in Paul De Man's Rhetoric of Romanticism. While I was reading, I was often perplexed at how this related whatsoever to Derrida's essay. But then I realized that in order to get De Man's points, I really needed to read Shelly's poem (at least the finished version). Of course, according to Derrida in "Borderlines," I would also have had to read all of Blanchot's stories, and his essays including The Step Not Beyond and The Infinite Conversation, and Rousseau, and his own Glas, and parts of Dissemination, and "Ousia and Gramme" (in Margins of Philosophy), and probably more Shelly, and Spenser, and Dante, and and and and and. . . . .perhaps the entire Western Canon?
Of course, this is an impossible task, relating to the "impossible wager" of merely treating Shelly's Triumph of Life (see note on pg 102).
So I contented myself with Shelly, de Man, and Derrida (as well as the Blanchot stories). The best way I found into this question is through de Man. de Man points to the infinite regressive structure (and the re-citative structure) of Shelly's poem as characteristic of all its tropes. The poem is a series of questions that are never answered. When Rousseau asks what is his origin, de Man writes,
[Rousseau] is granted a vision of the same spectacle that prompted the poet narrator's questioning in the first place; we have to imagine the same sequence of events repeating themselves for Shelly, for Rousseau, and for whomever Rousseau chose to question in his turn as Shelly questioned him. The structure of the text is not one of question and answer, but of a question whose meaning, as question, is effaced from the moment it is asked. (98)It is this effacing and "infinite regress" structure that I think "legitimizes" (if such a thing can be said) Derrida's forays into Blanchot's texts, such that they become the focus of the essay. Derrida puts into question the very possibility of remaining within the text of Shelly, affirming that there may be a way in which the two texts "read" each other. This sort of comparative literary analysis, which I am very familiar with, seems to legitimated by many things, but Derrida suggests that it is by the fact of mere language. For instance, in the discussion in "Borderlines" of Blanchot's use of "perfect rose," Derrida muses on the many "roses" in Shelly's Poem (see pg 167 of the essay). Or the connection between Shelly's use of "Brain turning to sand" and the imagery of the "pulse scattering like sand" in Blanchot. Although this would not be "serious" it still seems like there is a connection--an affection at a "distance." This is precisely the wager.
The structural connection I see between Blanchot and Shelly stems from the de Man quote above. This idea of infinite regress, of a repetitive structure that repeats the event within the story. In this sense, perhaps we can ask if Shelly's poem is a recit [?] in Blanchot's sense (Derrida writes: "the Triumph of Life [. . .] belongs in many ways to the category of the recit" pg 112). Rousseau repeats the *same* questions that the narrator of the poem asks and no one is given an answer.
This recitative structure is present also in Blanchot's story, the Madness of the Day, which was initially titled "un recit?" (with a question mark). The narrator begins the story demanded by some forceful men, "I am not learned. I am not ignorant," which is how the story we as readers read begins as well. It is impossible to tell which came "first." Derrida writes, "it is impossible to say which one cites the other and above all which one forms the border of the other. each includes the other comprehends the other, which is to say that neither comprehends the other" (126).
In the same way, Derrida opens his essay with the question of citing: "In other words on living? This time it sounds to you more surely like a citation" (103). On level, yes, we have already seen this "in other words," but was it in the same context? Can we say that this is a citation of the "first" "in other words on living"? Thus his essay tries to perform the conflict and weird space of "invagination" that he describes. The recit of Madness of the Day has the structure of "the recit of deconstruction in deconstruction" (127). In de Man's language, it is the de-figuring of figuration.
In a way, though, since Madness of the Day re-cites its story verbatim (or is it?), Death Sentence is a better analogy to the Triumph of Life (but can I say analogy?). In Death Sentence, which really should be left in the French: "L'arret de mort." The structure of the l'arret de mort is two "separate" recit's within the same 'title' of "l'arret de mort."
Derrida will ultimately argue that even though we cannot show how these two 'recit's' are connected--indeed, that they may even have entirely different narrators, he thinks that the connection lies in the two women: "they--the two women, the two voiceless voices, tele-phone one another: come. And the relationship, the connection, between the two recits would be telegraphic in nature" (187). The word "telegraphic" can be translated as "that which writes at a distance." Somehow, the language that the two women use connect them--they connect them through the narrator as a kind of "medium."
I keep trying to work out how Derrida arrives at this "mad hypothesis." On one level, we want to connect the two seemingly independent and separate recits, because we want to believe there is a reason for them united under one title: L'arret de mort. We assume the narrators are the same and that these are two different women in his life. Blanchot's language does suggest a connection between these two women, particularly describing them as "statues" or almost like a statue. Derrida picks up this image and runs with it, claiming that "each woman lives off and dies of the other [. . .] each woman preserving the other's narrator" (184).
I suspect that this notion of "preserving" or "monumentalizing" or "statue"-esque themes are why the narrator is so initially disturbed at Nathalie's decision to make a death mask and a cast of her hands. Derrida asks (and qualifies): "Should we say that he gave her the idea of or the desire for the death mask, as he had wished to embalm the other woman, in order to preserve both of them, to keep them alive-and-dead, living on? Yes and no" (189). It is through him that the woman finds the card, but she is the one who finds it, so we cannot simply ascribe agency to the narrator or the woman, exclusively.
This theme of "monumentalizing" brings me back to the de Man reading of Shelly, which it is quite possible Derrida read or knew about (or perhaps these two texts "love" each other): "Is the status of a text like the status of a statue?" (de Man 95). de Man also writes that in order to deal with Shelly (and the Romantics) we "bury them [. . .] to bury them in their own texts made into epitaphs and monumental graves. They have been made into statues for the benefit of future archeologists" (121). How might we understand de Man's reading of Shelly in terms of Derrida and Blanchot?
To me, Derrida's "mad hypothesis" about the two women calling one another to "come" (and, I may add, I cannot but think the erotic connotations of two women telling each other at a distance to "come"), is also the possibility that the two texts, L'arret de mort and Shelly's Triumph of Life call each other, love each other, can be "superimposed" on one another and made to "live on" in a way, perhaps, that does not merely "monumentalize" and bury both Shelly and Blanchot.
I say what must not be said: for example, that a text can stand in a relationship of transference (primarily in the psychoanalytical sense) to another text. And, since Freud reminds us that the relationship of transference is a 'love' relationship, stress the point: one text loves another" (165).