New Critical Approaches to the Short Stories of Ernest Hemingway

by Jackson J Benson

  Because of the author’s restrictions against reprinting “A Very Short Story” as a whole in any work other than a volume made up exclusively of his own work, the full text of the story has not been included here. The reader is requested to consult the text of “A Very Short Story” in Hemingway’s In Our Time or The Short Stories of Ernest Hemingway, New York: Charles Scribner’s Sons (The Scribner Library), before reading this essay. My apologies for the inconvenience.

  In any fictional text, then, we can discern certain features that are of the story: reports on actions, mentions of times and places, and the like. We can also find elements that are of the discourse: evaluations, reflections, language that suggests an authorial or at least narratorial presence who is addressing a reader or narratee with a persuasive aim in mind. When we are told that someone “smiled cruelly,” we can detect more of story in the verb and more of discourse in the adverb. Some fictional texts, those of D. H. Lawrence for example, are highly discursive. To read a Lawrence story is to enter into a personal relationship with someone who resembles the writer of Lawrence’s private correspondence. In contrast, Hemingway often seems to have made a strong effort to eliminate discourse altogether—an effort that is apparent in “A Very Short Story.”

  The distinction between story and discourse is closely related to another with which it is sometimes confused, and that is the distinction between the récit and diégésis of a narrative. In this case we are meant to distinguish between the whole text of a narration as a text on the one hand and the events narrated as events on the other. We can take over the Greek term, diegesis, for the system of characters and events and simply anglicize the other term as recital, or just refer to the “text” when we mean the words and the “diegesis” for what they encourage us to create as a fiction.

  The text itself may be analyzed into components of story and discourse, but it may also be considered in relation to the diegesis. One of the primary qualities of those texts we understand as fiction is that they generate a diegetic order that has an astonishing independence from its text. To put it simply, once a story is told it can be recreated in a recognizable way by a totally new set of words—in another language, for instance—or in another medium altogether. The implications of this for analysis are profound. Let us explore some of them.

  A fictional diegesis draws its nourishment not simply from the words of its text but from its immediate culture and its literary tradition. The magical words “once upon a time” in English set in motion a machine of considerable momentum which can hardly be turned off without the equally magical “they lived happily ever after” or some near equivalent. The diegetic processes of “realistic” narrative are no less insistent. “A Very Short Story,” by its location in Hemingway’s larger text (In Our Time) and a few key words (Padua, carried, searchlights, duty, operating, front, armistice), allows us to supply the crucial notions of military hospital, nurse, soldier, and World War I that the diegesis requires.

  This process is so crucial that we should perhaps stop and explore its implications. The words on the page are not the story. The text is not the diegesis. The story is constructed by the reader from the words on the page by an inferential process—a skill that can be developed. The reader’s role is in a sense creative—without it no story exists—but it is also constrained by rules of inference that set limits to the legitimacy of the reader’s constructions. Any interpretive dispute may be properly brought back to the “words on the page,” of course, but these words never speak their own meaning. The essence of writing, as opposed to speech, is that the reader speaks the written words, the words that the writer has abandoned. A keen sense of this situation motivates the various sorts of “envoi” that writers supplied for their books in the early days of printing. They felt that their books were mute and would be spoken by others.

  In reading a narrative, then, we translate a text into a diegesis according to the codes we have internalized. This is simply the narrative version of the normal reading process. As E. D. Hirsch has recently reminded us (in the Philosophy of Composition [Chicago, 1977], 122–23), for almost a century research in reading (Binet and Henri in 1894, Fillenbaum in 1966, Sachs in 1967, Johnson-Laird in 1970, Levelt and Kampen in 1975, and Brewer in 1975—specific citations can be found in Hirsch) has shown us that memory stores not the words of texts but their concepts, not the signifiers but the signifieds. When we read a narrative text, then, we process it as a diegesis. If we retell the story, it will be in our own words. To the extent that the distinction between poetry and fiction is a useful one, it is based on the notion of poetry as monumental, fixed in the words of the text and therefore untranslatable; while fiction has proved highly translatable because its essence is not in its language but in its diegetic structure. As fiction approaches the condition of poetry, its precise words become more important; as poetry moves toward narrative, its specific language decreases in importance.

  In reading fiction, then, we actually translate from the text to a diegesis, substituting narrative units (characters, scenes, events, and so on) for verbal units (nouns, adjectives, phrases, clauses, etc.). And we perform other changes as well. We organize the material we receive so as to make it memorable, which means that we systematize it as much as possible. In the diegetic system we construct, time flows at a uniform rate; events occur in chronological order; people and places have the qualities expected of them—unless the text specifies otherwise. A writer may relocate the Eiffel Tower to Chicago, but unless we are told this we will assume that a scene below that tower takes place in Paris—a Paris equipped with all the other items accorded it in our cultural paradigm.

  Places and other entities with recognizable proper names (Napoleon, Waterloo, Broadway) enter the diegesis coded by culture. The events reported in a narrative text, however, will be stored in accordance with a syntactic code based on a chronological structure. The text may present the events that compose a story in any order, plunging in medias res or following through from beginning to end, but the diegesis always seeks to arrange them in chronological sequence. The text may expand a minute into pages or cram years into a single sentence for its own ends, but the minutes and years remain minutes and years of diegetic time all the same. In short, the text may discuss what it chooses, but once a diegesis is set in motion no text can ever completely control it. “How many children had Lady Macbeth?” is not simply the query of a naive interpreter but the expression of a normal diegetic impulse. Where authors and texts delight in equivocation, the reader needs certainty and closure to complete the diegetic processing of textual materials. From this conflict of interests comes a tension that many modern writers exploit.

  The semiotician takes the reader’s diegetic impulse and establishes it as a principle of structuration. The logic of diegetic structure provides a norm, a benchmark for the study of textual strategies, enabling us to explore the dialogue between text and diegesis, looking for points of stress, where the text changes its ways in order to control the diegetic material for its own ends. The keys to both affect and intention may be found at these points. Does the text return obsessively to one episode of diegetic history? Does it disturb diegetic order to tell about something important to its own discursive ends? Does it omit something that diegetic inertia deems important? Does it change its viewpoint on diegetic events? Does it conceal things? Does it force evaluations through the rhetoric of its discourse? The calm inertia of diegetic process, moved by the weight of culture and tradition and the needs of memory itself, offers a stable background for the mapping of textual strategies. And our most aesthetically ambitious texts will be those that find it most necessary to put their own stamp on the diegetic process.

  Hemingway’s “A Very Short Story” presents itself as exceptionally reticent. The familiar Hemingway style, which Gerard Genette has called “behaviorist,” seems to efface itself, to offer us a pure diegesis. Boy meets girl—a “cute meet,” as they used to say in Hollywood—they fall in love, become lo
vers, plan to marry, but the vicissitudes of war separate them, and finally forces that are too strong for them bring about their defeat. This is the story, is it not: a quasi-naturalistic slice of life that begins almost like a fairy tale (“Once upon a time in another country . . .”)—and ends with the negation of the fairy-tale formula (“and they lived unhappily ever after”)—a negation that proclaims the text’s realistic or naturalistic status? But there is already a tension here, between the open form of the slice of life and the neat closure of the fairy tale, which emerges most clearly if we compare the progress of diegetic time with the movement of the text. We can do this in a crude way by mapping the hours, days, and weeks of diegetic time against the paragraphs of the text. The slowest paragraphs are the first: one night; and the third: one trip to the Duomo. The fastest are the fourth: his time at the front; the sixth: Luz’s time in Pordenone; and the seventh or last: which carries Luz to the point of infinity with the word “never.” The narrative thus increases its speed throughout and achieves its effect of culmination by the use of the infinite terms in the last paragraph. The text might easily have contented itself with recounting the fact that the major did not marry Luz in the spring, but it feels obliged to add “or any other time,” just as it is obliged to use the word “never” in the next sentence. Something punitive is going on here as the discourse seems to be revenging itself upon the character. Why?

  Before trying to answer that question, we would do well to consider some other features of the text/diegesis relationship. From the first paragraph on it is noticeable that one of the two main characters in the diegesis has a name in the text while the other is always referred to by a pronoun. Why should this be? The answer emerges when we correlate this detail with other features of the text/diegesis relationship. The text, as we have observed, is reticent, as if it, too, does not want to “blab about anything during the silly, talky time.” But it is more reticent about some things than others. In the first paragraph the male character is introduced in the first sentence. Luz appears in the fifth. When she sits on the bed, we are told “she was cool and fresh in the hot night.” Why this information about her temperature? She is the nurse, after all, and he the patient. In fact it is not important about how she feels at all, but about how she appears to him. The text is completely reticent about how he feels himself, though the implication is that he finds her coolness attractive. How he seems to her or how she feels about him are not considered relevant. This is a selective reticence. Our vision is subjectively with him (as the personal pronoun implies), while Luz is seen more objectively (as the proper name implies). The final implication of paragraph one is that they make love right then and there. But the reticent text makes the reader responsible for closing that little gap in the diegesis.

  This matter of the point of view taken by the text can be established more clearly with the use of a sort of litmus test developed by Roland Barthes. If we rewrite the text substituting the first-person pronoun for the third, we can tell whether or not we are dealing with what Barthes calls a “personal system,” a covert, first-person narration (see “Introduction to the Structural Analysis of Narratives,” in Image-Music-Text, 112). In the case of “A Very Short Story,” where we have two third-person characters of apparently equal consequence, we must rewrite the story twice to find out what we need to know. Actually, the issue is settled conclusively after the first two paragraphs, which are all I will present here:

  The first two paragraphs of “A Very Short Story” rewritten—“he” transposed to “I”:

  One hot evening in Padua they carried me up onto the roof and I could look out over the top of the town. There were chimney swifts in the sky. After a while it got dark and the searchlights came out. The others went down and took the bottle with them. Luz and I could hear them below on the balcony. Luz sat on the bed. She was cool and fresh in the hot night.

  Luz stayed on night duty for three months. They were glad to let her. When they operated on me she prepared me for the operating table; and we had a joke about friend or enema. I went under the anaesthetic holding tight on to myself so I would not blab about anything during the silly, talky time. After I got on crutches I used to take the temperatures so Luz would not have to get up from the bed. There were only a few patients, and they all knew about it. They all liked Luz. As I walked back along the halls I thought of Luz in my bed.

  The same paragraphs—“Luz” transposed to “I”:

  One hot evening in Padua they carried him up onto the roof and he could look out over the top of the town. There were chimney swifts in the sky. After a while it got dark and the searchlights came out. The others went down and took the bottles with them. He and I could hear them below on the balcony. I sat on the bed. I was cool and fresh in the hot night.

  I stayed on night duty for three months. They were glad to let me. When they operated on him I prepared him for the operating table; and we had a joke about friend or enema. He went under the anaesthetic holding tight on to himself so he would not blab about anything during the silly, talky time. After he got on crutches he used to take the temperatures so I would not have to get up from the bed. There were only a few patients, and they all knew about it. They all liked me. As he walked back along the halls he thought of me in his bed.

  “He” transposes to “I” perfectly, but “Luz” does not. In the second rewriting the first person itself enters the discourse with a shocking abruptness, since the earlier sentences seem to have been from the male patient’s point of view. The stress becomes greater in the last sentence of the first paragraph, which has been constructed to indicate how she appeared to him, not how she seemed to herself. But the last two sentences of the second paragraph in the second rewriting are even more ludicrous, with the first-person narrator informing us of how well liked she was and finally describing his thoughts about her. In this rewriting there is simply too great a tension between the angle of vision and the person of the voice. The discourse loses its coherence. But the first rewriting is completely coherent because in it voice and vision coincide. It is really his narrative all the way. The third-person narration of the original text is a disguise, a mask of pseudo-objectivity worn by the text for its own rhetorical purposes.

  The discourse of this text, as I have suggested, is marked by its reticence, but this reticence of the text is contrasted with a certain amount of talkativeness in the diegesis. He, of course, doesn’t want to “blab,” but they want “every one to know about” their relationship. Implication: she is the one who wants the news spread. There is absolutely no direct discourse in the text, but there are two paragraphs devoted to letters and one to recounting a quarrel. Here, too, we find reticence juxtaposed to talkativeness. Luz writes many letters to him while he is at the front. But the text does not say whether he wrote any to her. Hers are clearly repetitive and hyperbolic. The style of the discourse becomes unusually paratactic—even for Hemingway—whenever her letters are presented. “They were all about the hospital, and how much she loved him and how it was impossible to get along without him and how terrible it was missing him at night” (my italics). The repetitive “hows,” the hyperbolic “impossible” and “terrible,” and all the “ands” suggest an unfortunate prose style even without direct quotation. Above all, they indicate an ominous lack of reticence.

  The quarrel is not represented in the text, but the “agreement” that causes it is summarized for us, at least in part. It takes the form of a series of conditions that he must fulfill in order to be rewarded with Luz’s hand in marriage. Curiously, the conditions are represented not only as things it is “understood” that he will and will not do but also as things he wants and does not want to do. He does not “want to see his friends or any one in the States. Only to get a job and be married.” It is not difficult to imagine a man being willing to avoid his friends, to work, and to stay sober to please a woman, but it is hard to imagine any human being who does not “want to see his friends or anyone.” Not want to? Not any
one? The text seems to be reporting on the diegesis in a most curious way here. This is not simply reticence but irony. There is a strong implication that he is being coerced, pushed too far, even having his masculinity abused. If there are any conditions laid upon Luz, we do not hear of them.

  Finally, the final letter arrives. In reporting it the text clearly allows Luz’s prose to shine through once again, complete with repetition of the horrible phrase about the “boy and girl” quality of their relationship and the splendidly hyperbolic cacophony of “expected, absolutely unexpectedly.” Her behavior belies her words. Her true awfulness, amply suggested earlier by the reticent text, blazes forth here as her hideous discourse perfectly complements her treacherous behavior.

  But how did he behave while she was discovering the glories of Latin love? Nihil dixit. The text maintains what we can now clearly see as a specifically manly reticence. Did he drink? Did he see his friends? Or anyone? Did he want to? We know not. We do know, however, of his vehicular indiscretion in Lincoln Park and its result. The text is too generous and manly to say so, of course, but we know that this, too, is Luz’s fault. She wounded him in the heart and, “a short time after” this salesgirl got him in an even more vital place. The discourse leaves them both unhappy, but it clearly makes Luz the agent of the unhappiness.

  And what does it make him? Why, the patient, of course. He is always being carried about, given enemas, operated on, sent to the front, sent home, not wanting anything, reading letters. He is wounded at the beginning and wounded at the end. The ail-American victim: polite, reticent, and just waiting for an accident to happen to him. Who is to blame if his accidents keep taking the form of women? Who indeed? Whose discourse is this, whose story, whose diegesis, whose world? It is Papa’s, of course, for Hemingway taught a whole generation of male readers to prepare for a world where men may be your friends but women are surely the enema.