Title of the Paper [16 point font] © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. Argumentation and Fiction: Types of Overlaps and Their Functions GUILLERMO SIERRA-CATALÁN Department of Philosophy I University of Granada Edificio de Psicología. Campus de Cartuja, s/n. Granada 18071 Spain sierracatalan@gmail.com Abstract: Argumentation and fiction are quite different types of communi- cative phenomena. However, overlaps between them happen to be very frequent. We can both fictionalize by means of argumentation and argue by means of fiction. The main goal of this paper is to analyse the different types of overlap that may arise be- tween argumentation and fiction. In this paper, the defended hypothesis is that by considering who the “charac- ter” that is arguing is, we can get an exhaustive account of any possible overlap, as well as an explanation of the different functions that such overlaps can play. Résumé: Les arguments et les fic- tions sont des types de phénomènes de communication assez différents. Cependant, les chevauchements entre eux sont très fréquents. On peut à la fois fictionnaliser au moyen d'argu- mentations, et argumenter au moyen de fictions. L'objectif principal de cet article est d'analyser les différents types de chevauchements qui peuvent survenir entre l'argumentation et la fiction. Dans cet article, l'hypothèse défendue est qu'en considérant qui est le « personnage » qui argumente, nous pouvons obtenir un compte rendu exhaustif de tout chevauche- ment possible, ainsi qu'une explica- tion des différentes fonctions que de tels chevauchements peuvent jouer. Keywords: argumentation, fiction, speech act, pragmatics, interpretation 1. Introduction In principle, fiction and argumentation might seem rather distant things. By means of fiction, we imagine events and tell stories. In argumentation, on the contrary, we give reasons in order to justify certain points of view. Accordingly, we tend to identify fiction with imagination and description, and argumentation with reason and justification. They do not seem to have too much in common. 310 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. Consider, however, this extract from chapter 17 of Herman Mel- ville’s Moby Dick: Well, Captain Bildad,” interrupted Peleg, “what d'ye say, what lay shall we give this young man?” “Thou knowest best," was the sepulchral reply, "the seven hundred and seventy-seventh wouldn't be too much, would it?” […] “Why, blast your eyes, Bildad," cried Peleg, "thou dost not want to swindle this young man! he must have more than that.” “Seven hundred and seventy-seventh,” again said Bildad, without lifting his eyes; and then went on mumbling — “for where your treasure is, there will your heart be also.” “I am going to put him down for the three hundredth,” said Peleg, “do ye hear that, Bildad! The three hundredth lay, I say.” Bildad laid down his book, and turning solemnly towards him said, “Captain Peleg, thou hast a generous heart; but thou must consider the duty thou owest to the other owners of this ship — widows and orphans, many of them — and that if we too abun- dantly reward the labors of this young man, we may be taking the bread from those widows and those orphans. The seven hundred and seventy-seventh lay, Captain Peleg. Despite being a fictional narrative, this extract contains a quot- ed argumentation. Captain Bildad adduces certain reasons (name- ly, that he and Captain Peleg must consider the duty they owe to the other owners of the ship—which are mostly widows and or- phans—and that if they reward Ismael’s job too well, it will be unfair for the other owners of the Pequod. By adducing these reasons, Captain Bildad tries to justify (not too enthusiastically, though) the conclusion that he puts forward at the beginning of the quote, namely, that Ismael should receive the seven hundred and seventy-seventh part of the earnings of the expedition. Classical fables are typical examples of fiction functioning in an argumentative way. Think, for example, of the fable “The fox and the grapes” attributed to Aesop: A Fox one day spied a beautiful bunch of ripe grapes hanging from a vine trained along the branches of a tree. The grapes Argumentation and Fiction 311 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. seemed ready to burst with juice, and the Fox's mouth watered as he gazed longingly at them. The bunch hung from a high branch, and the Fox had to jump for it. The first time he jumped he missed it by a long way. So he walked off a short distance and took a running leap at it, only to fall short once more. Again and again he tried, but in vain. Now he sat down and looked at the grapes in disgust. "What a fool I am," he said. "Here I am wearing myself out to get a bunch of sour grapes that are not worth gaping for." And off he walked very, very scornfully. Moral: There are many who pretend to despise and belittle that which is beyond their reach. In this fable, the narrated facts are presented as reasons for the conclusion put forward in the moral. In this way, this fictional narrative is part of a piece of argumentative communication be- tween the author and the reader. These are not the only cases in which a text may present argu- mentative and fictional characteristics simultaneously. Some fictions include argumentative elements, and some argumentations include fictive ones. Yet, the overlap between fiction and argu- mentation may take different forms: Melville’s way of introducing argumentation in his fictional narrative is different from that of Aesop. The main goal of this paper is to describe and analyse the overlaps between argumentation and fiction from a pragmatic perspective focusing on speech-act theory. These connections are widespread phenomena that have recently received significant attention (Olmos 2017).⁠ In this article, I will combine structural insights (paying attention to who argues, either in the real world or in a world of fiction) with functional criteria (paying attention to why and for what this subject is actually arguing). In section 2, I will describe the characteristics of fiction, argu- mentation, fictional texts, and argumentative texts. In section 3, I will present a criterion to classify any possible relationship be- tween argumentation and narrative. This criterion focuses on the speech act of arguing and who the “character” that is arguing in a text or discourse is. By considering this criterion, we will be able to answer the question of whether the author is using fiction as a 312 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. tool for their argumentation or whether they are engaging in argu- mentation in order to build their fiction. In section 4, some struc- tural considerations will be examined. Finally, in section 5, the roles played by speech acts of fictionalizing inside argumentative texts, as well as speech acts of argumentation inside fictional texts, will be studied. Thus, the proposed classification is meant to provide tools for analysing specific overlaps between fiction and argumentation. Thereby, it also allows to explain the communicative function of each overlap. 2. Argumentation and fiction: speech-acts and texts For the goals of this research, argumentation is most fruitfully described from a linguistic-pragmatic perspective. Specifically, I will adopt Bermejo-Luque’s (2011) definition of argumentation as a second order speech-act complex, composed of the constative speech-act of adducing (i.e., the reason) and the constative speech- act of concluding (i.e., the conclusion) (Bermejo-Luque 2011, pp. 60-62). The illocutionary force of this speech-act complex amounts to an attempt to show that the conclusion is correct, and its characteristic perlocutionary effect invites an inference to the conclusion on the grounds of the adduced reasons. If, by means of their speech act of arguing, the speaker succeeds in showing that the conclusion is correct—which involves the accomplishment of certain semantic conditions of correctness as well as certain prag- matic conditions about what a good act “of showing” for a speech act is (Bermejo-Luque 2011, pp. 186-194)—then they justify the conclusion. This model focuses on the communicative nature of argumen- tation, but it also offers an analysis of several related concepts. Specifically, arguments are defined as mere representations of the syntactic and semantic properties of the inferences underlying argumentation or inner reasoning. Thus, the relation between the concepts of argument, argumentation, inference, and reasoning is described in the following way: reasoning (mental processes) and argumentation (communicative processes) consist of inferences that establish the acceptance of a belief or a claim on the grounds Argumentation and Fiction 313 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. of some other beliefs or claims. The syntactic and semantic prop- erties of these inferences are represented by means of arguments. Since fiction consists of particular cases of narratives, these should be described first. Gerald Prince (1982) proposed an in- sightful definition of narrative as “the representation of at least two real or fictive events or situations in a time sequence, neither of which presupposes or entails the other” (1982, p. 4). However, as Garrido Domínguez (1993) pointed out, mixing genres while producing narratives is not such an uncommon practice— especially in fictions produced from the beginning of the 20th century onwards. Garrido Domínguez also highlights that dra- matic, lyrical, and argumentative elements usually coexist in nar- ratives alongside the strictly narrative elements. The relationship between these elements is normally so strong that “no attempt at isolation can succeed without attacking the essence of this kind of story” (Garrido Domínguez 1993, p. 2; my translation). Prince’s (1982) aforementioned definition of narrative will be taken as a starting point. I will add to it that the representation is made by means of certain speech acts that constitute, in Gérard Genette’s words, the “narrating of the story” (Genette 1980, p. 27). Taking into account this definition of narrative, let us then describe the characteristics of fiction. Consider the following declarative sentence from the fable “The fox and the grapes”: A Fox one day spied a beautiful bunch of ripe grapes hanging from a vine trained along the branches of a tree. Consider that a speaker recites this sentence to a five-year-old child who is disposed to believe whatever they are told. In this situation, the child will conventionally trust the speaker, that is, they will conventionally consider the speaker to be serious, not a joker. Therefore, the child will believe the speaker is asserting. However, consider that a well-informed adult reads this sentence in its context as a part of an Aesop’s fable. In this case, the adult will not conventionally assume that, when Aesop wrote his fable, he was trying to inform his readers of any particular event1— 1 Or putting us in a position to be informed. This more refined account of assertion, which aligns with Williamson’s (1996) position as explored by 314 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. which can intuitively be considered the standard general purpose of any asserter. This is because Aesop was not asserting but fic- tionalizing. García-Carpintero (2016) understands fictionalizing as uttering a sentence with the communicative intention of putting an ad- dressee in a position to imagine the content of the proposition conveyed by the sentence. Following Currie (1990), García- Carpintero states that fictionalizing can be considered a speech act. In this way, although the external (syntactic and semantic) features of Aesop’s text could look like the those of assertions, the inten- tion with which it was uttered differs. Or, in terms of speech-act theory, the same locutionary act (external form of the utterance) will count as a fiction or not depending on the illocutionary force with which it was uttered. Taking this definition of the speech act of fictionalizing as a basis, I define a fiction text as one whose main function is to fic- tionalize a certain sequence of events. Thus, at least one of the representations of events that were alluded to by Prince in his definition should be made by means of a speech-act of fictionaliz- ing. In turn, an argumentative text would be a text whose main function is to show that a certain conclusion (target-claim) is correct. Consequently, it should contain at least an argumentation speech act. 3. Arguing by means of fiction and fictionalizing by means of argumentation As pointed out in section 1, overlaps between fiction and argumen- tation are abound. In order to better analyse the communicative functions of these overlaps for both fiction and argumentation, it seems useful to classify the different situations in which they may take place. Overall, there are two main types of overlap: either people fictionalize in order to argue or they argue in order to fictionalize. As we are going to see, these two main types of overlap can be Carpintero (2016), allows certain limited situations to be considered as asser- tions, for example, when the audience already knows the subject about which the asserter is trying to inform. Argumentation and Fiction 315 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. classified considering the question of who the (real or fictional) author of an argumentative text is. Such method for classifying overlaps between argumentation and fiction is also meant to pro- vide an account of the different functions that these overlaps may play. In this way, our proposed classification is based on the anal- ysis of the roles played by argumentative speech acts in fictional texts, on the one hand, and by fictionalizing speech-acts in argu- mentative texts, on the other hand. In order to illustrate these main types of overlap, we can start by considering the already quoted fable “The Fox and the Grapes.” By means of this fable, Aesop is trying to justify the conclusion that he states in the moral: “There are many who pretend to des- pise and belittle that which is beyond their reach.” In order to achieve this goal, Aesop adduces a fictional narrative (the fable) that is meant to show that the conclusion is correct, that is, to justify that conclusion. Thus, we can say that Aesop is fictionaliz- ing—based on what Carpintero (2016) proposed, this seems clear: Aesop is putting his audience in a position to imagine the content of certain propositions that have been adduced about the events in which the Fox was involved. But Aesop is not fictionalizing for fictionalizing’s sake, he is clearly doing it in order to try to justify the moral of his fable. In turn, the extract from Moby Dick presented above exempli- fies a different way of overlapping fiction and argumentation. In this case, Herman Melville is fictionalizing a story by means of presenting argumentation that, in the fictive world of Moby Dick, is performed by Captain Bildad. In principle, these are the two main ways in which argumenta- tion and fiction can overlap. Accordingly, the author of a text or discourse with such overlap can be, overall, either arguing or fictionalizing. In either case, a speech act is performed by means of another one: Aesop argues through the fictionalization in the fable. Henceforth, he is arguing indirectly through fiction. In the same way, Melville is fictionalizing indirectly through argumenta- tion when he presents Captain Bildad’s plea. According to Searle (1975), “a sentence that contains the illocu- tionary force indicators for one kind of illocutionary act can be uttered to perform, in addition, another type of illocutionary act” 316 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. (p. 168). For instance, uttering the question of whether the hearer wishes me to leave “Do you want me to leave?” can constitute an offer to leave (p. 181). In this vein, he defines the indirect speech act as one that is performed by way of performing another one. He labels the speech act that is performed literally as the “secondary” speech act, while he labels the speech act that is performed by means of the secondary speech act the “primary” speech act (p. 170). Searle highlights two fundamental questions in relation to indirect speech acts: how is it possible for the addresser of an indirect speech-act to mean the primary speech-act by uttering the secondary one (whose literal meaning is different)? And how is it possible for the addressee to understand the primary speech-act when they are only receiving the secondary one, attached to its literal meaning? The answer to these questions is advanced by Searle through the tools needed to explain an indirect speech act: a theory of speech acts, general principles of cooperative conversa- tion, mutually shared information between the addresser and the addressee, and the addressee’s ability to properly make inferences (p. 169). In the case this paper deals with, some variations should be made to these resources. Even though I admit that a theory of speech acts is essential in order to explain the indirect speech acts that arise between fiction and argumentation, as well as the mutu- ally shared information between the parts of the speech act and the addressee’s ability to infer, the reference to the general principles of conversation does not stand. The communicative situation that arises when someone is reading or listening to a fiction story differs from that of a conversation. As van Dijk (1980) stresses, it is difficult to say that reading a literary text generates any direct social relationship between the parts as “[a] literary text in general does not put a reader under obligation, does not necessarily direct the reader to a form of (social) action as advice, orders or requests do”2 (p. 8). However, there are specific social and institutional 2 Despite its major interest, the discussion about the relations and differences between the notions of ‘literature’ and ‘fiction’ lies beyond the purposes of this paper. I will trust the commonly accepted assumption that (almost) every fiction is of literary nature. Argumentation and Fiction 317 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. properties of literary texts that can contribute in adapting Searle’s tools for understanding an indirect speech act. As van Dijk recalls, “in our culture literature is typically produced by those speakers who have a specific, institutionalized role, namely as ‘au- thors.’ Similarly, literature is typically ‘public’ and ‘published,’ having a group as ‘hearer,’ being discussed, commented upon, and possibly canonized” (p. 13). Taking this observation into account, we can substitute the requirement for general principles of cooper- ative conversation that Searle made. Instead of this, I suggest that, in order for an addresser and an addressee to properly use an indi- rect speech act between fiction and argumentation, they should have the conscience of the literary (and fictional) communication process they are involved in. Specifically, the addressee should be aware of what they are reading, whether it be literature, something about fictive facts, or something that may mean more than it seems. This is because, as we have brought up and as van Dijk also pointed out, “literature may also […] be taken as an assertion, warning, congratulation, etc., depending on both the meaning of the text and the structure of the context […] This phenomenon may be explained in terms of the notion of an indirect speech act” (p. 10). According to this account of indirect speech acts, in “The Fox and the Grapes,” Aesop would be using the speech act of fictional- izing (telling the story) in order to perform an argumentative speech act (justifying his conclusion established in the moral by adducing the facts narrated through the story). Arguing would have been his primary speech-act, whereas fictionalizing would merely be the secondary and literal one. So, we can say that Aesop was performing an argumentation through the utterance of his fictional text. The interpretation of the fable depends on the condi- tions that have been presented, referring to particular readers. In a similar way, we could say that the primary speech act that Melville performed when presenting Captain Bildad’s argumenta- tion is not actually an act of argumentation.3 On the contrary, what Melville was doing was fictionalizing; he was putting us in a 3 That is, quoting, which constitutes his secondary speech act. Melville is neither arguing, despite what it might seem, nor is he conventionally trying to show any claim to be correct. 318 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. position to imagine Captain Bildad while mumbling, reading, confirming profits. This introduction of Captain Bildad as a parsi- monious and reflexive man is masterfully accomplished through the representation of his argumentation. Consequently, we can say that Melville was fictionalizing through the representation of this argumentation. Presenting argumentation, as Melville did in the already ana- lysed fragment, is not the only way of using it in order to fictional- ize. Argumentation can also be performed in fictional texts in order to argue. But, when argumentation appears explicitly in a fictional text, it is always performed (uttered) by either a character or the narrator. Consider, then, argumentation that is explicitly presented in a fictional text. In the world of fiction, it is performed by a character, but it might seem plausible that the views ex- pressed through the argumentation coincide with the author’s own. How could we determine, as readers, if the author is performing the argumentation shielded by his character or merely representing it? I will try to elucidate an answer to this question in the follow- ing sections. 4. A structural insight: who is actually arguing? As it was outlined before, there are a number of ways in which an argumentative speech act can take place within a fictional text. Some of them correspond to the first type we have considered: when producing them, the author is arguing by means of fiction. Others fit better in the second main type of overlap we have con- sidered: the author is fictionalizing by means of argumentation. In both cases, argumentation is being performed although in quite different ways. In this section, I propose to distinguish the types of overlap between fiction and argumentation that may take place in a text or discourse by considering who the subject of a fictional narrative that is arguing is. In principle, there are three possibilities: the arguer may be one of the characters of the story, the narrator (who is also a fictional character as long as they tell the story as if they know it), or the real-world author of the story. This distinction of subjects in a narrative is long-haul in literature analysis: the char- Argumentation and Fiction 319 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. acters and the narrator are “beings” from the world of the narrated story, while the author is the real-world “being” who writes the story. This can be summarized in the words of Roland Barthes: “[…] both narrator and characters are essentially ‘paper beings’. The living author of a narrative can in no way be mistaken for the narrator of that narrative […]” (Barthes 1975, p. 261). Thus, three types of overlap can be considered. The first structural type (henceforth, type S1) occurs in situations in which the author of the text presents argumentation performed by one of the charac- ters of the story. Such argumentation is usually quoted in the text. Captain Bildad’s argumentation in the extract from Moby Dick cited above would be an example of an S1 overlap. There is also an overlap between fiction and argumentation when a fictional narrative text includes argumentative speech acts performed by the narrator of the story. This happens, for example, in the novel Lolita, by Vladimir Nabokov. In this fiction, the nar- rator is presented as defending himself in court. He bases his defence in the narration of the story. Henceforth, the narrator argues by adducing the facts narrated in the novel in order to try to justify his innocence: “Ladies and gentlemen of the jury, exhibit number one is what the seraphs, the misinformed, simple, noble- winged seraphs, envied. Look at this tangle of thorns” (p. 11), he states at the beginning, thereby introducing the narration of the facts involving Lolita. This is type S2 of overlap.4 Finally, the whole fictional narrative text may constitute an instance of argumentation if it is adduced as a reason for a further claim. We should remember that argumentative speech acts were described as second order speech-act complexes, consisting of a speech act of adducing and a speech act of concluding (Bermejo- Luque 2011). ⁠ On the other hand, the notion of fictional narrative texts held in this paper, and adapted from Prince (1982)⁠, is the representation of certain events, made by fictionalizing speech- acts. Consequently, a narrative text can be part of an instance of 4 Although this is an example of argumentation performed by the narrator of a story (S2), it might be a limit case. The argumentation that the narrator presents is a way of introducing his narration of the facts (of which he is one of the main characters) rather than “pure” argumentation performed by a narrator who is not one of the characters of the story. 320 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. argumentation if the speech acts that constitute the fictional narra- tive text play the role of a speech act of adducing, making of it a reason for some conclusion by the real author of the text. The paradigmatic example of this situation is a classic fable, where the story is presented as a reason that justifies the conclusion settled explicitly in the moral. This is type S3 of overlap. Considering fictions and argumentations as speech acts and fictional and argumentative texts as sets of speech acts whose main communicative goal is either telling a story or justifying a claim allows for the possibility of analysing each of these structur- al types as follows: In type S1 situations, one of the characters of the story is per- forming an argumentative speech act. Given this, what actually appears in the text is merely the representation of an argumenta- tive speech act. In type S2 situations, the narrator of the story is performing an argumentative speech act, along with the rest of speech acts that make up her fictional narration.5 In cases S1 and S2, the real- world author of the text is representing argumentation in order to create their fictional text. As it will be argued in section 5, the functions of such representations can be either rhetorical or a matter of inducing the reader to make certain inferences. Contrastingly, in type S3 situations, the author is adducing the set of speech acts that constitutes the fiction text as a reason, with the conventional illocutionary force of arguing, that is, of trying to justify a certain conclusion that may be explicitly stated in the text or may be left implicit. As mentioned above, type S3 is clearly illustrated by classic fables, with their explicit morals. In these cases, the author of the text is arguing by means of fictionalizing. Importantly, in some cases, a text might fit properly in two of these categories allowing some kind of crossing between structural types. For instance, as the author’s voice and the narrator’s voice may at times coincide, types S2 and S3 may overlap. An example of this situation is the interpretation of Lolita as a defence—made 5 All types of narrators fit in this wide category from omniscient narrators (whose voices are often mistakenly identified with the real authors’ voices) to narrators who are also characters of the stories they narrate (like Humbert in Lolita); all of them can be fictionally arguing, in the sense of type S2. Argumentation and Fiction 321 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. by Nabokov—of a pedophile. Structurally, it is the narrator and not the author who is making this defence. Despite this, it might be reasonable to think that the author is just interposing a narrator between himself and the argumentation in order to argue covertly. This phenomenon results in a tricky situation as it creates a jump between the two main types we are considering; is Nabokov argu- ing by means of his fiction or is he fictionalizing by means of representing an instance of argumentation?6 Based on the func- tions performed by these overlaps between argumentation and fiction will provide effective tools for answering this question as I will explain in section 5. In sum, by considering the classic distinction between subjects in literature (author, narrator, and character), we get an exhaustive classification of types of overlaps between argumentation and fiction. Type S1 corresponds to situations in which a character is arguing, and their argumentation is part of the narrated story. Type S2 takes place when the narrator of the story is arguing, and her argumentation is part of the narration of the story. Finally, type S3 is associated with situations in which the author of the story is actually arguing by adducing the story as a reason, and their argu- mentation takes place in the real world. On the other hand, mixes between these types of overlap may exist due to the fact that the subjects of narratives may mix in turn in some ways, leading to a concoction of identities. This phenom- enon yields a variety of expressive resources that fictional narra- tive authors—especially literature writers—frequently exploit. The classification proposed in this paper aims to clarify this variety of resources. By doing this, this classification also attempts to high- light the relevance of an adequate conception of argumentation for the broad field of literary studies. 6 A crossing between types S1 and S3 will lead to the same situation as it involves a controversy about whether the author is arguing or just representing an instance of argumentation. 322 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. 5. A functional insight: why—and what for—would anyone present argumentation in any of these ways? Going back to Lolita, we can read it in two different ways. We might think that Nabokov produced his novel in order to present an apology for free sexual intercourse between adults and children. Or we could think that he just presents argumentation, fictionally performed by his narrator, Humbert Humbert, in order to convey and make more vivid this character’s personality. In other words, should we interpret Lolita as argumentation performed by Nabokov in order to fictionalize his novel or as fiction that Nabokov uses to argue in favour of his cause—namely, an excuse for hebephilia? In the former case, the story would amount to fictional argumentation by Humbert (S2), whereas, in the second case, it would amount to real argumentation by Nabo- kov (S3). The key point in analysing this dichotomy is to clearly distinguish who is performing the speech-act of arguing. In the world of fiction, the whole story amounts to argumentation per- formed by Humbert. He uses this story as a reason to justify his innocence in front of a jury. Yet, even though the argumentation is performed by Humbert in the world of the story, the speech act is actually put forward by Nabokov in the real world. But with which illocutionary force? Different interpretations arise: Nabokov could be trying to show that the alleged target claim “hebephilia is justi- fiable” is correct. In this case, we could read Lolita as if Nabokov were just arguing, shielding his illocutionary force behind Hum- bert’s. Or we could think that Nabokov was trying to achieve some rhetorical effects as regards the characterization of Humbert. In this case, we could read Lolita as if Nabokov were using his act of presenting argumentation with the illocutionary force of fictional- izing—that is, we can identify here an indirect speech act. In this way, the choice between the two interpretations involves finding an answer for the following question: was Nabokov using argu- mentation as a part of an indirect speech act or not? Michael Bamberg (1997) provides an insightful tool for ad- dressing this problem. He distinguishes three levels in which a story is operative (for any function, including argumentative): the story world, the interactional situation, and a wider discursive context. Based on this, it is clear that the facts narrated in Lolita Argumentation and Fiction 323 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. function as argumentation in the story world; Humbert is undoubt- edly addressing the court and trying to show his innocence claim to be correct. But, does Lolita function in an argumentative way at the other levels? Before answering this question, I should distinguish three types of inferences that a piece of discourse or text may invite the ad- dressee to make. As stated in section 2, overlap between fiction and argumentation takes place as the addresser performs an indi- rect speech act. Searle (1975, p. 170) pointed out that the interpre- tation of a secondary speech act requires, among other conditions, an ability on the part of the addressee to infer. These interpretative inferences can be prompted by several types of input, among which we can find both fictional and argumentative speech acts as we are going to see. According to Sperber and Wilson (2005), there are two differ- ent meanings involved in any communicative act. First, the sen- tence meaning is the conventional meaning that is decoded by applying the associated grammatical code. Second, the speaker’s meaning is the pragmatic meaning decoded (or inferred) through contextual conditions, non-verbal communicative elements, as well as the shared knowledge between the addresser and the ad- dressee. Searle (1975, p. 170) highlighted the importance of the addressee taking into account the background information they share with the addresser, which allows them to properly detect an indirect speech act. Let us, then, call those inferences made by the addressee in order to get the speaker’s meaning of an utterance7 contextual inferences. An example of a contextual inference prompted by the already quoted fable “The Fox and the Grapes” would be “in classic fables personified animals are typical [shared knowledge between ad- dresser and addressee]; therefore, Aesop is not presenting the ability of the fox to speak and reason as something extraordinary.” On the other hand, if the author does not provide enough details about some features of a story, the reader will normally feel curi- ous about them and thus will be led to wonder about certain relat- 7 The topic of context-based inferences has been widely studied—from multiple perspectives—in relation to semantics, pragmatics, and its contextual dependence. See, for example, Sperber and Wilson (2005). 324 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. ed questions. In order to try to answer these questions, the reader will have to make some inferences. Let us call those inferences made by the addressee in order to reconstruct beyond the speaker’s meaning explanatory inferences. Such inferences are indeed be- yond the speaker’s meaning because the author does not really provide clues or inputs for the reader to make specific inferential jumps but only blank spaces to be filled in with a certain degree of freedom.8 Explanatory inferences have been described in literature from various perspectives. As a part of a more general framework on reception theory, Wolfgang Iser, in The act of reading (1979), defended the claim that every narrative text presents a set of “emp- ty spaces” that has to be completed by each reader on the basis of their cultural, historical, and personal features. A clear example of this phenomenon can be found in “The Dinosaur” by Augusto Monterroso, which was, for some time, the shortest story in the history of literature: When s/he woke up, the dinosaur was still there. This is a paradigmatic source of potential explanatory infer- ences. “Who woke up?” “Where did they wake up?” “Why was a dinosaur there?” These are examples of questions that anyone could ask themselves when reading the text. As pointed out before, when hypothesizing answers to these questions, the reader is mak- ing inferences. However, these inferences arise without any specif- ic contextual input from Monterroso; he just artfully leaves the right spaces throughout the story to be filled by each reader. Finally, the third type of inference we consider can be exempli- fied by the previously quoted fragment (in the Introduction of this paper) from Moby Dick. As with any story, this extract has the typical effect of inducing its readers to make inferences. First, it invites them to make contextual inferences by presenting a scene and some characters. In this case, an occidental-educated reader acquainted with Melville’s work and in particular, with the context 8 The writer Antonio Muñoz Molina addresses this phenomenon in an informal, non-academic but masterful way: “when you write, you must choose one specific detail, two at most […] And then, the reader will add everything else.” (Muñoz Molina 2015; my translation) Argumentation and Fiction 325 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. of Moby Dick, shares some knowledge with the author. For in- stance, when Captain Bildad mumbles “where moth and rust do corrupt, but lay,” the reader is expected to understand that Bildad, an old Quaker administrator, is going to pay more attention to his Bible than to Ismael’s economic welfare. In this way, contextual inferences are prompted. Explanatory inferences can also take place. Although this is the most subjective type of inference, we may still find some exam- ples when we hypothesize answers to questions arising from the text, such as: “Why was Captain Peleg trying to get Ismael a better wage?” In turn, because this text involves argumentation by the charac- ters, it also has the potential to invite the reader to make the infer- ences that the very argumentation invites of its fictional address- ees. Captain Peleg, thou hast a generous heart; but thou must consider the duty thou owest to the other owners of this ship — widows and orphans, many of them — and that if we too abundantly re- ward the labors of this young man, we may be taking the bread from those widows and those orphans. The seven hundred and seventy-seventh lay, Captain Peleg. On the basis of this argumentation, an explanatory inference may lead the reader to believe that Captain Peleg was a stingy person. However, such belief is not stated as the conclusion of the argumentation performed by Bildad himself, that is, “Ismael should perceive the seven hundred and seventy-seventh lay.” A reader who comes to believe the latter because of Captain Bildad’s argumentation would be making an argumentative-input inference. When a reader makes this last type of inference, they are coming to believe both the reason and the conclusion of the argumentation quoted, neither of which have to be (conventionally) part of the speaker’s meaning. To sum up, three types of inferences have been distinguished in this section: contextual, explanatory, and argumentative. Except for the latter type, the rest can be prompted by a variety of inputs, including fictionalizing and argumentative speech acts. Let us then analyse their roles in relation to the different types of overlap 326 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. between argumentation and fiction that have been described above. 5.1. The role of fictionalizing speech acts in argumentative texts On the basis of the previous definition of an argumentative text as one whose main goal is to show that a target claim is correct, this section discusses the roles played by fictionalizing speech acts within such texts. That is, I am going to deal with the third type of overlap, namely, when real-world authors fictionalize in order to actually argue (S3). Given that argumentation is modelled as a (second order and complex) speech-act, it presents, along with its illocutionary force (considered an attempt to show that some conclusion is correct), a perlocutionary force, which is a function of the perlocutionary effect that it is meant to produce on its addressee. Specifically, in the case of the speech-act of arguing, this perlocutionary force is a matter of the extent to which the addressee can induce an act of arguing to make the same inference expressed in the act of arguing itself. In order to accomplish this perlocutionary effect, the speaker must make the addressee believe the reason they are adducing and the inference claim that leads to the conclusion from the reason (Bermejo-Luque 2011, pp. 122-128). Thus, a main function that using fiction for arguing can accom- plish is to convey dynamism and vividness to both the reason and the inference-claim and, as a consequence, to increase the perlocu- tionary force of the argumentation. How is this so? The presence of fictional speech-acts provides details or examples that can act as inputs prompting contextual inferences. Also, as explained above, “blank spaces” can prompt the reader to make explanatory infer- ences. These types of inferences sustain both the reason and the inference claim by providing them with a web of interrelated inferences departing from the adduced narrative. 5.2. The role of argumentative speech acts in fictional texts Let us now consider the possible functions of S1 and S2 types of overlap. As we have seen, these types of overlap appear when the author fictionalizes by means of argumentation. Argumentative Argumentation and Fiction 327 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. speech acts inside fictional texts work as narrative resources. They are just one kind among all those that an author can exploit in order to build their fiction. By introducing argumentative speech acts in fiction texts, an author may either have non-argumentative intentions or try to invite the reader to make argumentative-input inferences. As regards the first function, presenting argumentations as made by any of the characters may help to fictionalize their evilness, weak- ness, ridiculousness, eccentricity, etc. The following fragment of The Magic Mountain by Thomas Mann is an example of this usage: On the contrary, Naphta hastened to say, disease was very human indeed. For to be man was to be ailing. Man was essentially ailing, his state of unhealthiness was what made him man. There were those who wanted to make him “healthy,” to make him “go back to nature,” when, the truth was, he never had been “natural.” All the propaganda carried on to-day by the prophets of nature, the experiments in regeneration, the uncooked food, fresh-air cures, sun-bathing, and so on, the whole Rousseauian paraphernalia, had as its goal nothing but the dehumanization, the animalizing of man. They talked of “humanity”, of nobility — but it was the spir- it alone that distinguished man, as a creature largely divorced from nature, largely opposed to her in feeling, from all other forms of organic life (p. 465). The conventional illocutionary intention with which Thomas Mann introduces Naphta’s argumentation is not to convince the reader of what Naphta is defending, but to give them clues about Naphta’s personality, namely, that he was cynical, morbid, com- plicated, etc. In this way, the quoted argumentation becomes part of Naphta’s context. The reader can make contextual inferences on its basis, which may allow them to elucidate the author’s usage of an indirect speech act; an instance argumentation is being present- ed with the illocutionary force of fictionalizing (the primary and non-literal speech act). Bamberg’s levels can shed some light on this case; similarly to Lolita, as mentioned above, the fragment by Thomas Mann func- tions argumentatively in the story world, but, I dare say only there. 328 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. There are no reasons to believe that Mann aimed to justify the points of view defended by Naphta to his readers, especially given the fact that a significant part of The Magic Mountain is based on the unsolved discussions between Naphta and Settembrini, whose opinions frequently diverge. These facts leave aside any possibility of understanding the previous fragment of The Magic Mountain as functioning argumentatively at Bamberg levels 2 or 3. On the other hand, at times, the distance between the author and the narrator or any of his/her characters can vanish, so that the former can express him or herself directly through the text. In these cases, since argumentative speech acts are both represented and performed, they keep their typical perlocutionary force. Thus, they serve as a device to try to induce the addressee to make the same inferences made by the character in her speech-act: what we have called argumentative-input inferences. 5.3. An application: Lolita The fruitfulness of the functional analysis that our proposed classi- fication of overlaps between argumentation and narration enables becomes particularly apparent in literary analysis and criticism. So, let me end by offering an example of an application of the tools developed in this paper.9 The connection between the question “who is arguing?” and the analysis of the functions that a speech-act of arguing may play within a fictional narrative allows for an answer to the following question already propounded in section 4: how can we distinguish whether argumentative speech-acts are represented or performed within the frame of a fictional text? We can give a first answer by considering the illocutionary force that is conventionally attributed to the author of the speech 9 Through its history, literary criticism has dealt with aesthetic, ethical, moral, and structural features, but it has not paid enough attention to the argumentative character of narrative. In relation to this, there are some senses or some aspects of meaning that may emerge from literary texts that cannot be gathered nor analysed without studying their relationships to argumentation. My contention is that in order to reveal the whole meaning of certain literary works, we must also analyse the relationships between narrative and argumentation from this functional perspective. Argumentation and Fiction 329 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. act of arguing (real or fictive). I think, however, that the best option is thinking in terms of perlocutionary forces, rather than illocutionary ones, that is, which rhetorical function is the speech act of arguing actually performing? Let me explain this point by means of the following extract from Chapter 13 of Vladimir Nab- okov’s Lolita: As she strained to chuck the core of her abolished apple into the fender, her young weight, her shameless innocent shanks and round bottom, shifted in my tense, tortured, surreptitiously labor- ing lap; and all of a sudden a mysterious change came over my senses. I entered a plane of being where nothing mattered, save the infusion of joy brewed within my body. What had begun as a delicious distention of my innermost roots became a glowing tin- gle which now had reached that state of absolute security, confi- dence and reliance not found elsewhere in conscious life [...] The implied sun pulsated in the supplied poplars; we were fantas- tically and divinely alone; I watched her, rosy, gold-dusted, be- yond the veil of my controlled delight, unaware of it, alien to it, and the sun was on her lips, and her lips were apparently still forming the words of the Carmen-barmen ditty that no longer reached my consciousness. Everything was now ready. The nerves of pleasure had been laid bare. The corpuscles of Krause were en- tering the phase of frenzy. The least pressure would suffice to set all paradise loose. I had ceased to be Humbert the Hound, the sad- eyed degenerate cur clasping the boot that would presently kick him away. I was above the tribulations of ridicule [...] In this fragment, Nabokov is playing with the author-narrator (S2-S3) crossing.10 Through Humbert’s narration of how Lolita moved on his lap as she threw the apple, how his feelings were evolving while that was happening, the contemplation of a dreamy girl in an intimal situation, etc., we may get the impression that Nabokov himself is trying to justify Humbert’s own conclusion: “I was above the tribulations of ridicule” (notice that, on this ac- count, Nabokov’s claim would not be that Humbert felt that way, but that he actually was above ridicule). On this interpretation, 10 This particular crossing can vary from absolute ambivalence to a subtle distance. 330 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. Nabokov would be aiming at convincing the reader that Humbert’s feelings are justified in this way (1). Contrastingly, a second inter- pretation would be that Nabokov aims to create an ambivalence between the closeness that may arise between Humbert and the reader as they follow his reasonings and experiences and the rejec- tion that they may provoke (2). This distinction could not be made by exclusively relying on an account of illocutions since both interpretations correspond to type S2 overlapped with type S3. Instead, it requires taking into ac- count functional criteria: the interpretation (1) is a case where the author is providing certain information at the disposal of the read- er, which they can use to infer conclusions about Humbert. This interpretation corresponds to a standard (non-indirect) speech-act of arguing. Nabokov is not presenting Humbert’s argumentation for anything but presenting his own. That is, Nabokov is merely arguing shielded behind Humbert. Otherwise, interpretation (2) is a case where the rhetorical effect of presenting the argumentation enhances the rhetorical force of the fragment. This interpretation corresponds to a case of an indirect speech act. Nabokov is presenting Humbert’s argumen- tation in order to fictionalize, which turns out to be his primary and literal speech act. Bamberg’s levels can be useful in order to analyse these situa- tions. Which are the clues that allow us to discover Nabokov’s intentions? Is Nabokov merely positioning his narrator as an argu- er and then making the story function argumentatively in the story world? Is he making Lolita argumentatively operative in relation to the interactional situation that is delivered to the readers (every reading process for any reader)? Or in a wider discursive context? The fact that Lolita functions argumentatively at Level 1 is clear. On Level 2, I do not think there is any contextual feature associat- ed with the interactional situation that could lead to the fulfilment of this function. I do not think that anyone attributes the intention of convincing particular readers of Humbert’s innocence to Nabo- kov. There is just a writer telling a story to a reader. It is a rather murky and disturbing story, but so is Dracula, and nobody thinks Bram Stoker was actually a vampire. It is on Level 3 where I think Nabokov plays masterfully. How is Nabokov positioning himself Argumentation and Fiction 331 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. through Lolita? Someone supporting interpretation (1) would answer this question by saying that he is positioning himself as an old man holding some well-known misogynist views who is actu- ally justifying the behaviour of a hebephile. On the contrary, someone supporting interpretation (2) might say that a teacher and butterfly lover who has written a novel about a trial and a love story—a rather murky and disturbing love story. I dare to say that, in a wide discursive context, there is still controversy on the matter of how Nabokov described himself to the world through Lolita. The plausibility of these two interpretations loads the text with a singular richness. As a consequence, controversy around Lolita’s “actual meaning” has taken place since its first publication in 1955. Hans-Georg Gadamer’s hermeneutics provides a theoretical ground to explain this phenomenon: “Not just occasionally but always, the meaning of a text goes beyond its author” (Gadamer 1975, p. 264). Authorial intentions do not fit in Gadamer’s views: “[…] the fusion of horizons […] does not allow the interpreter to speak of an original meaning of the work without acknowledging that, in understanding it, the interpreter's own meaning enters in as well” (p. 578). Through an argumentative analysis of certain narra- tives—like the analysed extract from Lolita—different interpreta- tions can be identified and, according to Gadamer, none of them should be considered to be the “actual” one. The only thing that can be said in this respect is related to the mild adaptation, pre- sented in section 3, of Searle’s conditions for understanding an indirect speech act on the basis of van Dijk’s remarks. According to that adaptation, in order to choose interpretation (2), the reader of Lolita should (of course, neither in an academic nor specific way) have: (a) certain knowledge of speech-act theory, (b) knowledge of background information in common with Nabokov (historical, cultural, etc.), (c) the ability to infer and (d) an aware- ness that they are reading a literary text, a story about fictive facts, and a story that may mean more than its literal meaning. The choice between interpretations, between assuming that Nabokov was arguing or using an argumentation to do something else (in a wide discursive context, not in any particular interactional situa- tion) is a cultural matter, and it depends on the context and read- 332 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. ers. And, as I have already pointed out, it is a source of narrative and argumentative richness. 6. Conclusion I have pointed out that overlaps between narratives and argumen- tation are a widespread and significant phenomenon in literature. The goal of this paper was to provide tools for analysing types of overlap between fiction and argumentation as well as the roles they may play. Literary analysis can make profit of this, as these tools highlight an element, that is, argumentation, that has not been quite taken into account in the field. In order to do so, a distinctive criterion for the analysis of these relationships has been presented. It is based on the study of the action the author is actually performing, whether within the course of argumentation, or when producing a fictional narrative. The obvious cases refer to authors that argue when producing argu- mentation as well as authors that fictionalize when producing a fictional narrative. In these cases, no overlap arises. But, as it has been shown, authors tend to produce argumentation in order to fictionalize, and they also tend to present fictional narratives in order to argue. This relation has been shown to be based on the notion of an ‘indirect speech act.’ A first structural insight has complemented the definition of this distinctive criterion. It is based on the shape that an argumentative speech act can take in a narrative text. Then, an analysis of the functions that the speech acts of fic- tionalizing can perform in argumentative texts or that argumenta- tive speech acts can perform in narrative texts has been presented. This study of the functions has been shown to be of importance in relation to the interpretation of fictional texts. Acknowledgments I wish to thank Prof. Lilian Bermejo Luque for her support and guidance. The work presented has been financed by the research project FFI2016-79317-P of the Spanish Ministry of Science and Innovation. Argumentation and Fiction 333 © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. References Aesop. The Fox and the Grapes. In Library of Congress. URL accessed 23 August 2021: http://read.gov/aesop/005.html Bamberg, Michael. 1997. Positioning between structure and perfor- mance. Journal of Narrative and Life History 7(1–4): 335–342. Barthes, Roland. 1975. An introduction to the structural analysis of narrative. New literary history 6(2): 237-272. Bermejo-Luque, Lilian. 2011. Giving reasons. A linguistic-pragmatic approach to argumentation theory. Dordrecht: Springer. Currie, Gregory. 1990. The nature of fiction. Cambridge: Cambridge University Press. Dijk, Teun A. van 1980. The pragmatics of literary communication. In On text and context, eds. E.Forastieri-Braschi, G. Guinness, and H. Lopez-Morales, 3-16. Rio Piedras, Puerto Rico: Editorial Universitar- ia. Gadamer, Hans-Georg. 1975. Truth and method (2 [1989]). London: Sheed and Ward Ltd. García-Carpintero, Manuel. 2016. Relatar lo ocurrido como invención: Una introducción a la filosofía de la ficción contemporánea. Madrid: Cátedra. Garrido Dominguez, Antonio. 1993. El texto narrativo. Madrid: Síntesis. Genette, Gérard. 1980. Narrative discourse. Ithaca, New York: Cornell University Press. Iser, Wolfgang. 1979. The act of reading: A theory of aesthetic response. Journal of Aesthetics and Art Criticism 38(1). Mann, Thomas. 1927. The magic mountain. London: Secker and War- burg. Melville, Herman. 1851. Moby Dick. New York: Harper and Brothers. Monterroso, Augusto. 1971. The black sheep and other fables. New York: Doubleday and Co. Muñoz Molina, Antonio. 2015. Imprescindibles: Antonio Muñoz Molina, el oficio de escritor. Televisión Es. Olmos, Paula, ed. 2017. Narration as argument. Dordrecht: Springer. Prince, Gerald. 1982. Narratology, The form and function of narrative. Berlin: Mouton. Searle, John R. 1975. Indirect speech acts. Syntax and semantics 3: 59– 82. Sperber, Dan and Deirdre Wilson. 2005. Pragmatics. In Handbook of philosophy of language, eds. F. Jackson and M. Smith, 468-501. Ox- ford: Oxford University Press. 334 Sierra-Catalán © Guillermo Sierra-Catalán. Informal Logic, Vol. 41, No. 3 (2021), pp. 309–334. Williamson, Timothy. 1996. Knowing and asserting. The Philosophical Review 105(4): 489-523.