Start Submission

Reading: What Is a Novel?

Download

A- A+
Alt. Display

Research Article

What Is a Novel?

Author:

Alexey Aliyev

HSE University, Moscow, RU
X close

Abstract

The question ‘What is a novel?’ has received scant attention in the philosophical literature. Meanwhile, this question is important. In the light of this, in this paper, I would like to address it, suggesting a potential answer. I begin by defining what I call ‘novel in the restricted sense’ – the concept that covers all novels except the so-called nonfiction novels, graphic novels, and novels in verse. Then, drawing upon Jerrold Levinson’s approach to defining ‘art’, I provide a definition of the concept that covers nonfiction novels, graphic novels, and novels in verse. Finally, with the help of this definition and the definition of ‘novel in the restricted sense’, I formulate a definition of ‘novel’ simpliciter and defend it against potential objections.
How to Cite: Aliyev, Alexey. “What Is a Novel?”. Estetika: The European Journal of Aesthetics 58, no. 1 (2021): 19–34. DOI: http://doi.org/10.33134/eeja.215
197
Views
56
Downloads
1
Twitter
  Published on 17 Mar 2021
 Accepted on 12 Oct 2020            Submitted on 11 Jan 2020

I. Introduction

Even cursory examination of objects that fall under the expression ‘novels’ reveals that there are considerable structural-level differences between them. One of these differences concerns the word count: although a typical novel has 80–100,000 words, there are novels that are several times shorter (E. B. White’s Charlottes Web) and novels whose word count drastically exceeds the foregoing number (Mark Leach’s Marienbad My Love). Another significant structural-level difference between novels is related to their constitutive ‘matter’: although in many cases this matter is solely text (Jane Austen’s Pride and Prejudice), that is not how things are in all cases, as there are novels that, in addition to having text as their main component, also involve graphic elements (W. G. Sebald’s Vertigo) and novels that are, for the most part, graphic (Art Spiegelman’s Maus). Furthermore, a number of serious structural-level differences between novels are concerned with various aspects of the semantic content. Take, for instance, the presence of main characters: There are novels with several such characters (George R. R. Martin’s A Song of Ice and Fire), novels that have just one main character (Gary Paulsen’s Hatchet), and novels without any main characters at all (Milorad Pavić’s Dictionary of the Khazars: A Lexicon Novel). Or consider the narrative structure: In some novels, it is chronological (Stephen King’s Misery), and in some, it is nonlinear (James Joyce’s Ulysses). Even fictionality – which is often treated as a paradigmatic semantic feature of novels – does not seem to be common to all novels, as there are novels categorized as nonfictional (Rodolfo Walsh’s Operación Masacre).

In the light of the ontological diversity of novels, the attempt to define1 ‘novel’ may seem futile. Indeed, to do this, it is necessary to find a sufficiently informative (nontrivial) feature possessed by all novels and only by them. But what could such a feature be, if, as is clear from the above, the nature of novels can vary drastically with regard to both the formal and the semantic aspects? Perhaps, after all, Morris Weitz was right to regard the concept ‘novel’ as essentially undefinable…2 I think, however, that the situation with defining ‘novel’ is not as hopeless as it may appear at first. In fact, my contention is that ‘novel’ can be defined. The purpose of this essay is to explain how this can be done.

The rest of the essay is structured as follows. I begin by defining ‘novel in the restricted sense’ – the concept that covers typical, or ‘standard’, novels and, therefore, constitutes the core of the concept ‘novel’ (Section II). Next, I define ‘nonstandard novel’ – the concept that covers less common novels, in particular nonfiction novels, novels in verse, and graphic novels (Section III). Finally, with the help of the definitions of ‘novel in the restricted sense’ and ‘nonstandard novel’, I formulate a definition of ‘novel’ simpliciter (Section III), defend it against potential objections (Section IV), and review potential theoretical and practical applications of it (Section V).

II. Defining ‘Novel in the Restricted Sense’

Let us begin by defining ‘novel in the restricted sense’ (hereafter: ‘novelr’) – the concept that, as already mentioned, covers all novels except nonfiction novels, novels in verse, and graphic novels. How can this concept be defined? To answer this question, let us consider the characteristic features of a novelr.

Presumably, the most obvious of these features are that of being a verbal object (that is, an object composed, for the most part, of linguistic elements, such as words and punctuation marks) and that of being written, for the most part, in prose. It may seem that the fact that a novelr is a verbal object precludes such a novel from possessing nonlinguistic elements, such as photographs, maps, diagrams, and drawings. But this is not the case, as being a verbal object, by definition, does not require being composed solely of linguistic elements. It is also worth noting that being written, for the most part, in prose is compatible with containing nonprosaic textual elements (such as verse) – and so the fact that a novelr is written, for the most part, in prose does not imply that such a novel cannot involve such elements.

In addition to being a verbal object and being written, for the most part, in prose, a novelr has yet another feature essential to it – namely, that of having an appropriate length. Here, of course, a natural question arises: What exactly is this length? To answer this question, it is sufficient to answer two questions: (a) ‘What is the maximal length of a novelr?’ and (b) ‘What is the minimal length of such a novel?’3 Let us begin with the former question. Although many novelsr have about 80–100,000 words, there are novelsr that are much longer. Samuel Richardson’s Clarissa has over 950,000 words, Marcel Proust’s À la recherche du temps perdu exceeds a million words, Madeleine de Scudéry’s Artamène ou le Grand Cyrus contains about 2.1 million words, and Mark Leach’s Marienbad My Love has approximately 17 million words. Can a novelr be even longer than that? There seems no reason to think otherwise. It is even possible to conceive of a novelr that contains an infinite loop and, hence, is never-ending. Given this, it can be concluded that the length of a novelr does not have an upper bound.

Now, what about the lower bound? Does a novelr have it, and if so, what is it? Answering this question precisely – by specifying an exact number of words – is hardly possible. The reason for this is that any attempt to do this faces the paradox similar to the paradoxes of the Heap and the Bald Man. Suppose we find out that the shortest novelr ever written has 34,381 words (Julie Otsuka’s novelr When the Emperor Was Divine is about that long). It seems bizarre to say that anything that is just one word shorter than this novelr is not a novelr. So, a novelr can contain 34,380 words. But taking just one word out of this 34,380-word novel will not turn it into something that is not a novelr. So, it is possible for a novelr to have 34,379 words. But, again, if we remove only one word from a 34,379-word novelr, we will not turn it into a non-novelr. Thus, a novelr can have 34,378 words. But, again… We can continue this reasoning until we reach an absurd claim – that a novelr can have just one word. The absurdity of this claim suggests that there must exist a limit to how short a novelr can be. But it seems that, regardless of what we take this limit to be, we face the paradox just sketched. So, how should we proceed?

I think we should admit that since the least number of words a novelr can have is (at least, epistemically) essentially vague, the minimal length of such a novel cannot be precisely identified. Yet we can characterize it imprecisely. There are objects with a word count in the 30,000-word range that are widely recognized as novelsr – for instance, Albert Camus’s The Stranger (ca. 36,500 words), E. L. Konigsburg’s From the Mixed Up Files of Mrs. Basil E. Frankweiler (ca. 31,300 words), Julie Otsuka’s When the Emperor Was Divine (ca. 34,300 words), and E. B. White’s Charlottes Web (ca. 33,500 words). At the same time, there seem to be no novelsr whose word count is in the 20,000 or some lower word range. So, it appears reasonable to adopt the view that the border that separates a novelr from a non-novelr is in the 30,000-word range.4 Surely, this view is rather imprecise – but this should not be viewed as a shortcoming. It would be a shortcoming if it were possible to precisely identify the minimal word count of a novelr, but, as already noted, such identification cannot be done: There is simply no fact of the matter with regard to the exact minimal number of words a novelr can have.

Thus, taking into account what has been said, if something is a novelr, it must be a verbal object that is written, for the most part, in prose and has at least m words, where m is a number in the 30,000-word range. The converse, however, is not true. Being an m– or m+–word verbal object written, for the most part, in prose is not sufficient for being a novel and a fortiori a novelr. Consider, for instance, historical monographs. They are typically m– or m+–word verbal objects written in prose, but they are not novels. And the same can be said about an overwhelming majority of dictionaries, collections of short stories, textbooks and philosophical treatises. So, what distinguishes novelsr from the foregoing non-novelistic works?

Here is a possible answer: Novelsr differ from non-novelistic works in that unlike the latter, novelsr were seriously intended by their authors to tell fictional stories. This answer seems promising but is not unproblematic. In order for it to be satisfactory, there must be no work that is a verbal object that (a) is written in prose, has at least m words, and was seriously intended to tell a fictional story but (b) is not a novelr. However, there can be such a work. Suppose Mary has written a 100,000-word textbook in biology. Suppose also that, as a result of some psychological aberration, when working on this textbook, she seriously intended to tell a fictional story. Now, Mary’s textbook is an m+–word verbal object that is written in prose and was seriously intended to tell a fictional story. But this textbook is not a novel and a fortiori a novelr. Or consider another situation. Suppose John has written a 100,000-word prosaic text that is completely meaningless (it is just a collection of arbitrarily chosen words). Suppose also that when working on this text, he seriously intended to tell a fictional story. Is John’s text a novelr? Prima facie, it is not. At the same time, John’s text is an m+–word verbal object that is written in prose and was seriously intended to tell a fictional story.

Given what has been said, the foregoing answer to the question ‘What distinguishes novelsr from non-novelistic works?’ cannot be accepted. However, there is a way to modify this answer so that it would be acceptable. The reason why Mary’s and John’s works are not novelsr is that these works do not, in fact, tell any fictional stories. If Mary’s and John’s works did tell such stories at least to some extent, these works could be considered novelsr. Given this, we can make the foregoing answer acceptable by adding to it the claim that the author’s intention to tell a fictional story is realized at least to some extent. Thus, the acceptable answer to the question being discussed is as follows: what distinguishes novelsr from non-novelistic works is that unlike the latter, novelsr (a) were seriously intended (hereafter: ‘intended’), by their authors, to tell fictional stories and (b) realize this intention at least to some extent.

With regard to this answer, one might ask: What exactly is meant here by ‘fictional story’? To answer this question, it is necessary to clarify the concepts ‘fiction’ and ‘story’. Due to space limitations, examining these concepts in detail is beyond the scope of this essay, and so only a rather brief clarification of them can be provided here; such a clarification, however, should be sufficient for the purposes of the current project. With regard to the concept ‘fiction’, we can adopt the view – endorsed, in one version or another, by Gregory Currie, Peter Lamarque, and Nicholas Wolterstorff,5 among others – according to which fiction is the result of a particular intention – namely, ‘an author’s intention that her audience make-believe the narrated events’.6 This view is rather intuitive. Furthermore, it correlates with the intentional component of the foregoing account of what distinguishes novelsr from non-novelistic works (recall that on this account, an author of a novelr must have the intention to tell a fictional story). It should be noted, however, that it is not necessary to understand ‘fiction’ the way it has been suggested. The reader is free to use some alternative account of fiction – such as the account put forward by John Searle or the one advanced by Kendall Walton – as long as it is compatible with the view defended in this essay.7

Let us now turn to the concept ‘story’. Prima facie, a story is a set of utterances. But, obviously, not any such set is a story. So, how can we distinguish those sets of utterances that are stories from those that are not? Presumably, the best way to do this is by appealing to the contents of the utterances. According to Paisley Livingston,8 potential proposals on the nature of these contents are as follows:

  1. ‘At least one event, or a change of state from one situation or state of affairs to another, for example, some object instantiates at least one new property.’9
  2. ‘At least two events standing in the right kind of relation to each other.’10
  3. The same as in (2), ‘with an additional, pragmatic requirement to the effect that the actual storyteller or author of the narrative effectively design the pattern of story events so as to convey some “point” or to achieve some other expressive or communicative goal’.11
  4. ‘The same as one of the above proposals, plus the additional requirement that the events include either an implicit or explicit telling of these events: in every narrative there must be at least one narrator who narrates what happens in the story.’12

Which of these proposals, if any, should we endorse? (3) is too restrictive, because a story does not have to have a goal (after all, why can’t there be a ‘pointless’ story?). At the same time, (1) is too liberal. If it is adopted, then we must agree that virtually every utterance that describes an event is a story. But this consequence is counterintuitive. Take, for instance, the utterance ‘The cat sleeps’: It describes an event, but, considered on its own, without any context, it is not a story – or so it seems. The most acceptable proposal seems to amount to the conjunction of (2) and (4). On this proposal, the contents of a story include (a) at least two events that stand in a particular relation to each other and (b) at least one (implicit or explicit) narrator, who can be either the author herself or some fictional entity. Here, of course, a natural question arises: What exactly is the relation in which the events of a story stand to each other?

According to Livingston, potential candidates for the role of this relation include the following:

  • a temporal ordering (‘Autumn came, and then the leaf fell’),
  • a causal ordering (‘The arrival of autumn caused the leaf to fall’),
  • the relation ‘x occasions y’ (‘Autumn came, and the leaf fell’),
  • a relation that results in manifesting some goal-directed activity by at least one of the events (‘The leaf decided to fall’), and
  • a relation that results in there being ‘two or more actions involving purposeful activity, with the additional requirement of a form of “closure”, meaning that there must be events that function as a coherent sequence of purposeful activity, with a beginning and resolution’13 (‘The leaf was full of life, but then autumn came, as a result of which the leaf withered and fell to the ground’).

It could be assumed that only one of these relations – or perhaps some other relation not mentioned above – is the relation in which the events of a story stand to each other. But in this case, because our intuitions about the basic nature of relations between the events of a story are vague, it would be hard, if possible at all, to provide a plausible explanation of why the selected, and not some other, relation can play the role of the relation in which the events of a story stand to each other. Fortunately, for the purposes of the project developed in this essay, we can assume that any of the foregoing relations can play the mentioned role. It may turn out, of course, that by making this assumption, we depart from some ‘objective’ sense of ‘story’. But in this case, we can adopt the view that that the sense of ‘story’ specified above is, at least in part, stipulative: It is the sense in which ‘story’ is to be understood in this essay, not in general.

In the light of what has been said, a fictional story can be characterized as a collection of utterances describing events that (a) are accompanied by an author’s intention that the audience make-believe these events, (b) stand in a particular relation to each other (for example, one of the relations listed in the previous paragraph), and (c) include at least one narrator. Having clarified the concept ‘fictional story’, let us return to the question of what distinguishes novelsr from non-novelistic works. As has been pointed out, this question can be answered as follows: What distinguishes novelsr from non-novelistic works is that unlike the latter, novelsr (a) were seriously intended by their authors to tell fictional stories and (b) realize this intention at least to some extent. Before proceeding further, some additional remarks concerning the intention that figures in the foregoing answer are worth making. First, it should be underlined that this intention is the intention to tell a fictional story, not stories. Second, the author may not intend to tell a story that is completely fictional, as this story may involve real people and objects and describe events that have actually occurred. Finally, the intention to tell a fictional story does not necessarily involve the intention to present this story to the audience as fictional.

Given what has been said, a novelr has the following characteristic features: (a) being a verbal object, (b) being written, for the most part, in prose, (c) having at least m words, where m is a number in the 30,000-word range, (d) being intended by its author(s) to tell a fictional story, and (e) realizing this intention at least to some extent. Yet having these features is still not sufficient for being a novelr. For consider screenplays and theatrical scripts. Of course, because in most cases they have fewer than 20,000 words, they are not novelsr. But nothing stops us from imagining, say, a 30,000-word screenplay or theatrical script. Furthermore, we can imagine that this screenplay/theatrical script (a) is a verbal object, (b) is written for the most part, in prose, (c) was intended by its author to tell a fictional story, and (d) realizes this intention to some extent. So, if having the mentioned features were sufficient for being a novelr, then some screenplays/theatrical scripts would be novelsr. But, surely, neither screenplays nor theatrical scripts are novels and a fortiori novelsr.

Is there a feature that can be used to distinguish novelsr from screenplays and theatrical scripts? The function of screenplays and theatrical scripts is to provide instructions or guidelines on how to create certain works of art (namely, films – in the case of screenplays; and theatrical performances – in the case of theatrical scripts). Thus, neither screenplays nor theatrical scripts are what might be called ‘self-standing’, that is, intended to serve as primary foci of appreciation, and not merely as means to create such foci. But that is not how things are in the case of novelsr. In the light of this, we can say that what distinguishes novelsr from screenplays and theatrical scripts is that unlike the latter, novelsr are self-standing (in the sense defined above).

Now, taking into account what has been said, ‘novelr’ can be defined as follows:

Novelr: For all x, x is a novelr if and only if x is a verbal object that (a) is written, for the most part, in prose, (b) has at least m words, where m is a number in the 30,000-word range, (c) was intended by its author(s) to tell a fictional story, (d) realizes this intention at least to some extent, and (e) is self-standing.

III. Extending the Definition of ‘Novelr

Clearly, the foregoing definition, although covering an overwhelming majority of novels, excludes nonstandard novels – namely, graphic novels, novels in verse, and nonfiction novels, and so cannot serve as a definition of ‘novel’ simpliciter. Can’t it be modified, however, so that it could serve as such a definition?

Let us begin to answer this question by examining nonstandard novels. Consider first novels in verse. Like novelsr, novels in verse are verbal objects. Furthermore, there is good reason to think that the minimal length of novels in verse, similar to the minimal length of novelsr, is in the 30,000-word range. Next, like novelsr, novels in verse are intended by their authors to tell a story. Moreover, like in the case of novelsr, in the case of novels in verse, the authors’ intention to tell a story is realized at least to some extent. Finally, like novelsr, novels in verse are self-standing. At the same time, novels in verse differ from novelsr in that unlike the latter, novels in verse are written in verse, not in prose. Furthermore, in the case of novelsr, the authors intend to tell a fictional story, whereas in the case of novels in verse, the authors intend to tell some (not necessarily fictional) story.

Consider now nonfiction novels. Like novelsr, nonfiction novels are verbal objects. Also, similar to novelsr, nonfiction novels have at least m words. Next, like novelsr, nonfiction novels are intended by their authors to tell a story. Furthermore, like in the case of novelsr, in the case of nonfiction novels, the authors’ intention to tell a story is realized at least to some extent. Finally, like novelsr, nonfiction novels are self-standing. At the same time, unlike novelsr, nonfiction novels are intended by their authors to tell a real story. Also, unlike novelsr, nonfiction novels can be written not only in prose but also in verse.

Finally, let us have a look at graphic novels. Like novelsr, graphic novels are intended by their authors to tell a story. Also, like in the case of novelsr, in the case of graphic novels, the authors’ intention to tell a story is realized at least to some extent. Furthermore, graphic novels are similar to novelsr with regard to how long they are: Like novelsr, graphic novels have an appropriate length. Finally, similar to novelsr, graphic novels are self-standing. At the same time, in the case of novelsr, the authors intend to tell a fictional story, whereas in the case of graphic novels, the authors intend to tell some (not necessarily fictional) story. Also, unlike novelsr, graphic novels are not verbal objects; rather, they are objects that are composed of pictures and/or words organized in a comic-strip format. Lastly, graphic novels differ with regard to the minimal word count: Unlike novelsr, graphic novels may have fewer than m words.

Given the latter fact, as well as the abovementioned fact that graphic novels must have an appropriate length, a natural question arises: What exactly is the appropriate length of a graphic novel? Presumably, the best answer to this question is as follows: the length of a graphic novel is that of a book (under normal printing conditions). Of course, such characterization of the length of a graphic novel is rather vague. But this should not be considered a shortcoming: There is no fact of the matter with regard to even an approximate minimal numerical length of graphic novels, and so, providing a characterization of the length of a graphic novel that would be noticeably more precise than the one given above (for example, a characterization of the form ‘the length of a graphic novel is in the x-word/picture/page range’, where x is some number) is hardly possible.

Taking into account what has been said, nonstandard novels can be characterized as (1) objects that (a) are verbal, (b) have at least m words, (c) were intended by their authors to tell a story, (d) realize this intention at least to some extent, (e) are written in verse and/or were intended by their authors to tell a nonfictional story, and (f) are self-standing – or (2) objects that (g) are composed of pictures and/or words organized in a comic-strip format, (h) are book-length, (i) were intended by their authors to tell a story, (j) realize this intention at least to some extent, and (k) are self-standing. However, not only nonstandard novels can be characterized this way. Consider a typical biography. Because it is not a novel, it is not a nonstandard novel. But it falls under the first disjunct: It is a verbal object that has at least m words, was intended by its author to tell a nonfictional story (the story of someone’s life), does, in fact, tell such a story, and is self-standing. Or imagine a historical monograph that is book-length, is composed of pictures and/or words organized in a comic-strip format, was intended by its author to tell a story, does, in fact, tell a story, and is self-standing. Such a monograph is not a nonstandard novel, as it is not a novel at all. But it falls under the second disjunct.

So, what distinguishes nonstandard novels from biographies, historical monographs, and similar non-novelistic works? To answer this question, let us first consider Jerrold Levinson’s definition of ‘art’.14 On this definition, an artwork is ‘a thing (item, object, entity) that has been seriously intended for regard-as-a-work-of-art, i.e., regard in any way preexisting artworks are or were correctly regarded’,15 where ‘regard’ refers to ‘any mode of interaction with an object’.16 Thus, according to Levinson, an artwork has the property of having been seriously intended for regard in a way preexisting artworks were correctly regarded. A similar property, I think, is possessed by a nonstandard novel. Such a novel was seriously intended (hereafter: ‘intended’) by its author(s) to be regarded in a way similar to the way novelsr were correctly regarded. Thus, Alexander Pushkin intended Eugene Onegin to be regarded in a way similar to the way novelsr were correctly regarded. And an analogous intention with regard to their nonstandard novels was possessed by Truman Capote, Art Spiegelman, Will Eisner, Rodolfo Walsh, and other ‘nonstandard’ novelists.

Here, one might ask: What exactly is a way in which a nonstandard novel must be intended by its author(s) to be regarded? Before answering this question, let us first consider the way novelsr are correctly regarded. This way can be characterized as a set of regards (approaches, attitudes) that includes regards that presuppose:

  • knowledge of the historical context (provided that any such context is relevant to the story being told);
  • sensitivity to the formal structure;
  • ability to understand the story being told;
  • sensitivity to the stylistic features;
  • sensitivity to the sonic techniques (rhythm, alliteration, consonance, dissonance and so on);
  • ability to empathize with the characters;
  • willingness to read (or listen) with proper attention;
  • willingness to attend to the graphic elements (if there are any);
  • awareness of the tradition of novelr writing (that is, the tradition of writing novels such as Don Quixote, Pride and Prejudice, and War and Peace);
  • knowledge of the fact that the story being told is not intended by the author(s) to be nonfictional;
  • knowledge of the fact that the main function of a novelr is not merely to inform.

– as well as perhaps some other regards. Now, as mentioned above, a way in which a nonstandard novel must be intended by its author(s) to be regarded is similar to the foregoing way. In particular, such a way involves a considerable number of the regards of the way novelsr are correctly regarded, including the regards that presuppose at least some knowledge of the historical context (if any such context is relevant to the story), sensitivity to the formal structure, ability to understand the story, willingness to read (or listen) with proper attention, awareness of the tradition of novelr writing, and sensitivity to the stylistic features. At the same time, a way in which a nonstandard novel must be intended by its author(s) to be regarded is not necessarily identical to the way novelsr are correctly regarded. Depending on the kind of nonstandard novel, such a way (a) may not involve some of the regards of the way novelsr are correctly regarded or (b) can contain some additional regards. Thus, in the case of a nonfiction novel, the way in which such a novel must be intended by its author(s) to be regarded does not presuppose regarding this novel with awareness of the fact that the story being told is not intended by the author(s) to be nonfictional. And in the case of a graphic novel, the way in which such a novel must be intended by its author(s) to be regarded involves a regard that is not involved in the way novelsr are correctly regarded – namely, the regard that presupposes willingness to attend to the comic-strip format as well as at least some knowledge of the tradition of comic book writing.

It is important to underline that the fact that x was intended by its author to be regarded in a way similar to the way novelsr were correctly regarded does not necessarily imply that the author had a thought like ‘I want x to be regarded with at least some knowledge of the historical context, sensitivity to the formal structure, ability to understand the story, willingness to read with proper attention, knowledge of the tradition of novelr writing, and sensitivity to the stylistic features, and so on’. Of course, if the author had such a thought, then x was intended by her to be regarded in a way similar to the way novelsr were correctly regarded. But x can be considered as having been intended by x’s author to be regarded in such a way even if x’s author had a different thought – for instance, the thought ‘I want x to be regarded in a way similar to the way novelsr are correctly regarded’ or ‘I want x to be regarded like novelsr are correctly regarded’.17

Thus, a nonstandard novel was intended by its author(s) to be regarded in a way similar to the way novelsr were correctly regarded. Now, what about biographies and other non-novelistic works? Were any of them intended by their authors to be regarded in a way similar to the way novelsr were correctly regarded? There is no doubt that the answer to this question is ‘No’. So, what distinguishes nonstandard novels from biographies and other non-novelistic works is that unlike the latter, nonstandard novels were intended by their authors to be regarded in a way similar to the way novelsr were correctly regarded.

Given what has been said, ‘nonstandard novel’ can be defined as follows:

Nonstandard Novel: For all x, x is a nonstandard novel just in case x is:

  1. a novel in verse – a verbal object that (a) is written in verse (b) has at least m words, where m is a number in the 30,000-word range, (c) was intended by its author(s) to tell a story, (d) realizes this intention at least to some extent, (e) was intended, by its author(s), to be regarded in a way similar to the way novelsr were correctly regarded, and (f) is self-standing;
  2. a nonfiction novel – a verbal object that (a) has at least m words, where m is a number in the 30,000-word range, (b) was intended by its author(s) to tell a nonfiction story, (c) realizes this intention at least to some extent, (d) was intended, by its author(s), to be regarded in a way similar to the way novelsr were correctly regarded, and (e) is self-standing; or
  3. a graphic novel – an object that (a) is book-length, (b) is composed of pictures and/or words that are organized in a comic-strip format, (c) was intended by its author(s) to tell a story, (d) realizes this intention at least to some extent, (e) was intended by its author(s) to be regarded in a way similar to the way novelsr were correctly regarded, and (f) is self-standing.

We are now in a position to answer the question posed at the beginning of this section – namely, ‘Can’t the definition of “novelr” be modified so that it could serve as a definition of “novel” simpliciter?’ The definition of ‘novelr’ can indeed be modified this way. To do this, the definiens of ‘nonstandard novel’ should be added, as a disjunct, to the definiens of the definition of ‘novelr’. The result of this modification amounts to the following definition:

Novel: For all x, x is a novel if and only if x is a novelr or a nonstandard novel (where ‘novelr’ and ‘nonstandard novel’ are to be understood according to the corresponding definitions provided above).

Before proceeding further, it is worth making two remarks concerning the foregoing definition (hereafter: ‘DN’). First, there is good reason to hold that DN reflects the actual structure of the concept expressed by the word ‘novel’. One of our intuitions about novels is that some entities called ‘novels’ are doubtless novels, whereas other such entities do not seem entirely like novels. Consider novelsr: Pride and Prejudice, War and Peace, Moby-Dick, and so on. There is no doubt that each of them is a novel. The same, however, cannot be said about nonstandard novels. Take, for instance, Eugene Onegin. Although it is categorized as a novel, it does not seem like a novel (‘How can it be a novel, given that it is written in verse? It’s a poem!’). Or take Maus. The consensus is that it is a novel. But, intuitively, it is not (‘How can it be a novel, given that it has a comic-strip format? It’s a comic book!’). What has been said about Eugene Onegin and Maus can be said about other nonstandard novels as well. Thus, from an intuitive viewpoint, novelsr are clearly novels, whereas the status of nonstandard novels qua novels is dubious. In the light of this, there is good reason to hold that the concept expressed by the word ‘novel’ is composed of two sub-concepts: (a) the (core) concept that covers novels like Pride and Prejudice, War and Peace, Moby-Dick and (b) the (peripheral) concept that covers novels of a less traditional sort, such as Eugene Onegin and Maus. But if this is so, then a definition of ‘novel’ reflects the structure of the concept ‘novel’ just in case the definiens of this definition is composed of two disjuncts that express, respectively, the concept that covers novels such as Pride and Prejudice, War and Peace, Moby-Dick and the concept that covers novels of a less traditional sort, such as Eugene Onegin and Maus. Meanwhile, the definiens of DN is, in fact, composed of these disjuncts.

Second, DN has certain historical implications with regard to the novelistic tradition. One of these implications is that the first novel came into existence in or before the first century CE, long before the time when a considerable number of first paradigmatic novels were written (the seventeenth and eighteenth centuries). Consider Callirhoe – one of the so-called ‘ancient Greek novels’ that was written in the first century CE. It is a verbal object that is written in prose, has more than 30,000 words, tells a fictional story, and, clearly, was intended by its author, Chariton of Aphrodisias, to tell such a story. So, according to DN, Callirhoe is a novel. As a result, if no entity created before the first century falls under DN, then DN implies that the first novel was written in the first century; if, on the other hand, there is an entity that was created before the first century and falls under DN, then DN implies that the first novel was written before the first century.

Another historical implication of DN is that nonstandard novels – nonfiction novels, novels in verse, and graphic novels – were not created before the tradition of novelr writing came into existence. As mentioned above, to regard x in a way similar to the way novelsr were correctly regarded, it is necessary to regard x with at least some awareness of the tradition of novelr writing. Suppose now that this tradition does not exist. Then x cannot be regarded in a way similar to the way novelsr were correctly regarded – and, hence, cannot be intended to be regarded in such a way.18 Meanwhile, according to DN, having been intended to be regarded in a way similar to the way novelsr were correctly regarded is a necessary property of a nonstandard novel. Thus, if DN is true, then the creation of the first nonstandard novel took place after the tradition of novelr writing came into existence.

IV. Some Objections

Let us now examine some potential objections to DN.

Objection 1. DN involves concepts that either are not defined at all or require further clarification (in particular, ‘story’, ‘fiction’, ‘verse’, ‘book-length’, and ‘comic-strip format’). But if this is so, then DN is not sufficiently informative.

Response. This objection assumes that to be sufficiently informative, DN must define ‘story’, ‘fiction’, ‘verse’, ‘book-length’, and ‘comic-strip format’. Why think this assumption is true? One possible explanation is that any sufficiently informative definition must define all the concepts it involves. However, this explanation is unsatisfactory. Consider the following definition of ‘water’: for all x, x is water just in case x has the molecular structure H2O. This definition is doubtless acceptable and, hence, is sufficiently informative. However, it does not provide the definitions of ‘molecular structure’, ‘H’, and ‘O’. Furthermore, if a definition defines all the concepts it involves, then there must be infinite chains of (noncircular) definitions. But such chains do not exist, as some concepts are basic and, hence, indefinable. Thus, there is no definition that defines all the concepts it involves. So, if the explanation being discussed is correct, no definition is sufficiently informative, which is, of course, absurd.

Another potential way to explain why the assumption being discussed should be regarded as true is to say that a sufficiently informative definition must define all definable concepts – that is, concepts that can be defined using solely basic concepts – and the concepts ‘story’, ‘fiction’, ‘verse’, ‘book-length’, and ‘comic-strip format’ are definable. But this explanation also fails. Consider, once again, the definition of ‘water’ – ‘for all x, x is water just in case x has the molecular structure H2O’. As already mentioned, it is sufficiently informative. But it does not define ‘H2O’, ‘molecular structure’, and other definable concepts it contains.

Thus, neither explanation is satisfactory. Meanwhile, there seems no other potentially satisfactory way to substantiate the foregoing assumption of Objection 1. So, this assumption can be rejected.

Objection 2. Suppose there is an object O that (a) is verbal, (b) is written in prose, (c) was intended by its author to tell a fictional story, (d) does, in fact, tell such a story, and (e) is self-standing. Suppose next that O has 27,700 words. Is it a novelr? An answer to this question depends on whether O’s word count is in the 30,000-word range. Is it, in fact, in this range? Intuition cannot help us answer this question. It is not obvious that the number of words O has – 27,700 – is not in the 30,000-word range; likewise, it is not obvious that this number is in this range. At the same time, the answer to the foregoing question cannot be found using some principle that precisely determines the lower bound of the 30,000-word range, as, given the essential vagueness of this bound, no such principle exists. In the light of this, there is good reason to hold that using DN, it is impossible to establish whether O is a novelr. Meanwhile, if a definition cannot be used to establish whether an object falls under the concept being defined, then according to this definition, it is indeterminate whether this object falls under this concept. Thus, DN entails that for some objects, their status as novels is indeterminate. However, in fact, any entity is either clearly a novel or clearly a non-novel. So, DN either excludes from the extension of ‘novel’ some novels or fails to exclude from this extension some objects that are not novels.

Response. The foregoing objection assumes that no entities are indeterminate qua novels. But this assumption is false. In fact, besides entities that are clearly novels and entities that are clearly non-novels, there are also borderline cases – entities that are neither clearly novels nor clearly non-novels.19

Objection 3. If DN is true, then, contrary to standard practice, at least some novellas are to be categorized as novels. Take, for instance, Ian McEwan’s On Chesil Beach. It is normally categorized as a novella, not a novel (it has been categorized as such by its author as well as a number of critics).20 However, according to DN, it is a novel, because it is a verbal object that is written in prose, has more than 30,000 words, was intended by its author to tell a fictional story, does, in fact, tell such a story and is self-standing.21

Response. This objection can be successful only if it is assumed that something cannot be simultaneously categorized as both a novel and a novella. However, a novella is sometimes characterized as a short novel and a fortiori a novel (see, for instance, the definitions of ‘novella’ provided by the Cambridge Dictionary and the Collins Dictionary). If this characterization is acceptable – and there seems no real reason think otherwise – then the foregoing assumption is ungrounded.

Furthermore, Objection 3 is problematic even if we grant that the concepts ‘novel’ and ‘novella’ are mutually exclusive. This objection substantiates the thesis that On Chesil Beach is a novella with the help of the fact that it has been categorized as a novella by its author and some critics. However, a similar, and equally powerful, substantiation can be given in favour of the thesis being rejected – that On Chesil Beach is a novel. The author of this work categorizes it not just as a novella but also as a novel, as is evidenced, for example, by the fact that he calls it a novel in his interview to The Telegraph.22 Furthermore, a number of critics categorize it as a novel, not a novella.23 Thus, Objection 3 fails to provide sufficient reason to accept one of its premises – that On Chesil Beach is a novella and not a novel.

Objection 4. Consider the Iliad. It is an m+–word verbal object that is written in verse, was intended by its author – Homer – to tell a story, and does, in fact, tell a story. Furthermore, most likely, Homer intended it to be regarded in a way similar to the way novelsr were correctly regarded. Thus, the Iliad seems to fall under the definition of ‘novel in verse’ and, hence, under DN. But the Iliad is not a novel.

Response. According to this objection, the Iliad satisfies all the conditions of the definition of ‘novel in verse’. But does the Iliad, in fact, satisfy all of them? It is doubtless an m+–word verbal object that is written in verse, was intended by its author to tell a story, and does, in fact, tell a story. But was it intended by its author, Homer, to be regarded in a way similar to the way novelsr were correctly regarded? Given what has been said in Section III, in order for the Iliad to have been intended by Homer to be regarded in such a way, he had to intend it to be regarded in a way that included the regard presupposing at least some awareness of the tradition of novelr writing. But he could not have intended the Iliad to be regarded in a way that included this regard – because the tradition of novelr writing did not exist at that time. Thus, the Iliad was not intended by Homer to be regarded in a way similar to the way novelsr were correctly regarded. But if this is so, then, contrary to what Objection 4 assumes, the Iliad does not satisfy all the conditions of the definition of ‘novel in verse’.

Objection 5. According to DN, some works written before the seventeenth century might be novels. However, the consensus is that the novel qua a genre – and, hence, works of this genre – did not come into existence until the seventeenth century.

Response. This objection assumes that there is a generally accepted view as to when the novel came into existence. Is this assumption true? There are a considerable number of theorists who believe that the first European novel was written in the seventeenth century. But there is no consensus as to when the first novel simpliciter was written. Some theorists argue that it was written in the seventeenth century (Cervantes’s Don Quixote). Others claim that it was written in the eleventh century (Murasaki Shikibu’s The Tale of Genji). There are also theorists arguing that the creation of the first novel dates back to the first three centuries CE (the so-called ‘ancient Greek novels’: Chariton’s Callirhoe, Achilles Tatius’ Leucippe and Clitophon, Longus’ Daphnis and Chloe, Xenophon’s Ephesian Tale, and Heliodorus of Emesa’s Aethiopica.) Thus, the foregoing assumption of Objection 5 is false.

Objection 6. Consider the Harry Potter series of novels. It is composed of seven books, each of which is an m+–word verbal object written, for the most part, in prose. Meanwhile, anything that is composed of m+–word verbal objects written, for the most part, in prose is an m+–word verbal object written, for the most part, in prose. So, the Harry Potter series is such an object. Furthermore, this series was intended by its author, J. K. Rowling, to tell a fictional story – the story about Harry Potter – and this intention was successfully realized. Finally, the Harry Potter series is a self-standing object. Thus, according to DN, the Harry Potter series of novels is a novel. But this series is not a novel; it is a collection of novels.

Response. A core assumption of the foregoing objection is that the Harry Potter series of novels is not a novel. But this assumption is far from obvious and can, in fact, be rejected. Indeed, the Harry Potter series is not normally categorized as a novel. Yet there seems no real reason against categorizing this series as such. Furthermore, it is common for some analogous series to be categorized as novels. Consider, for instance, Marcel Proust’s À la recherche du temps perdu. Like the Harry Potter series, it is a series of novels. At the same time, it is categorized as a novel.

Objection 7. If DN is true, then a nonstandard novel must be intended, by its author(s), to be regarded in a way similar to the way novelsr are correctly regarded. However, there is no real reason to hold that, say, In Cold Blood was, in fact, intended by its author, Truman Capote, to be regarded in such a way. So, if DN is true, then In Cold Blood is not a novel, which is, of course, false.

Response. This objection is based on the idea that there is no real reason to hold that In Cold Blood was, in fact, intended by Capote to be regarded in a way similar to the way novelsr were correctly regarded. But this idea is highly questionable. According to Capote, ‘a nonfiction novel […] employ[s] all the techniques of fictional art but [is] nevertheless immaculately factual’.24 Now, there is little doubt that by ‘fictional art’ here, he means primarily the art of the traditional novel, or novelr. So, it can be said that according to Capote, a nonfiction novel is to be regarded qua an entity that possesses the relevant features of a novelr except the feature of having been intended by the author(s) to tell a fictional story. But if this is the case, then it is reasonable to suppose that in his view, a nonfiction novel should be regarded in a way similar to the way novelsr are correctly regarded. Meanwhile, if this supposition is true, then, given the fact that In Cold Blood is explicitly characterized by Capote as a nonfiction novel,25 there is, in fact, a good reason to think that he intended In Cold Blood to be regarded in a way similar to the way novelsr were correctly regarded.

Here, one could ask: Can’t Objection 7 be made successful by replacing In Cold Blood with some other nonstandard novel – in particular, some nonstandard novel that was not intended by its author to be regarded in a way similar to the way novelsr were correctly regarded? Having examined a considerable number of relevant sources, I have not found any evidence that can be used to show that there is an entity that (a) is clearly a nonstandard novel but (b) was not intended by its author to be regarded in a way similar to the way novelsr were correctly regarded. At the same time, in an overwhelming majority of cases, there is evidence suggesting that the author of a nonstandard novel did, in fact, intend this novel to be regarded in such a way.26 Taking this into account, the foregoing question, I think, should be answered in the negative.

V. Conclusion

So, none of the objections discussed in Section IV stands up to criticism, and there seem to be no other potentially successful objections to DN. In the light of this, there is good reason to hold that DN is sufficiently informative and covers all and only those entities that actually fall under the concept being defined. Meanwhile, if this is so, then DN can be considered a satisfactory definition of ‘novel’.

In closing, it is worth addressing the question about the value of DN. Unlike analogous projects concerning a number of some other literary art concepts (in particular, the concepts ‘literature’, ‘poetry’, and ‘comics’), the project of defining ‘novel’ has received scant attention in the philosophical literature.27 In the light of this, it may seem reasonable to conclude that DN – as well as any other definition of ‘novel’ – is of little theoretical and practical interest. Is this conclusion justified?

The concept ‘novel’ plays a key role in a number of projects in aesthetics, for instance, the project of elucidating the ontology of literature as well as various projects concerned with questions about literary value and interpretation. This concept, however, is often left without clarification – as if it were completely transparent. Yet the level of its pretheoretic transparency is insufficient: Surely, we have some intuitive understanding of it, but this understanding is rather imprecise and, as a result, theoretically lacking. Because of this, it is not always clear what exactly the foregoing projects examine when they examine what they call ‘novels’. Resolving this problem is possible with the help of DN: It can be used to explicate the concept ‘novel’ and thereby make the subject matter of these projects less vague.

Another consideration in favour of the usefulness of DN concerns the fact that it can be incorporated into some ongoing philosophical projects, thereby contributing to their realization. To illustrate, consider Dominic McIver Lopes’s project of defining ‘art’ using the concepts of art kinds.28 On this project, ‘x is a work of art = x is a work of K, where K is an art’29 – and so to define ‘art’, it is sufficient to define the art kind concepts: ‘music’, ‘theater’, ‘literature’, ‘dance’, ‘painting’, and so on. DN can help us define (at least) one of these concepts – ‘literature’. Thus, similar to how ‘art’ is treated, by Lopes, as a disjunction of the art kind concepts, ‘literature’ can be treated as a disjunction of the literary kind concepts: ‘poetry’, ‘novel’, ‘short story’, ‘novella’, and so on. Using DN, we can define one of these concepts – ‘novel’ – and, as a result, get closer to defining ‘literature’.

The foregoing considerations imply that DN has particular applications in philosophy. But it can also be applied, in similar ways, in other academic disciplines, such as literary theory and history of literature. Furthermore, DN is useful in the nontheoretical context as well. Many of us are interested in knowing why some texts are categorized as ‘novels’, whereas other texts are not, and DN can, at least partially, help us satisfy this interest.

In the light of what has been said, there is good reason to answer the question posed above in the negative: The lack of interest in the philosophical project of defining ‘novel’ does not mean that this project is not worth pursuing. Rather, this lack of interest should be viewed as a perhaps surprising but definitely undesirable omission – which, I hope, this essay, by providing a definition of ‘novel’, can help to eliminate.

Notes

1A definition of x (where x is some concept) can be characterized as whatever explicates the meaning of x by specifying a set of conditions that (a) are satisfied by all entities that actually fall under x and only by such entities and (b) are sufficiently informative, in particular, (1) are not enumerative (that is, do not amount to a disjunctive list of objects that fall under x), (2) do not involve a vicious circle (that is, do not explicitly or implicitly contain x), and (3) do not contain meaningless concepts or concepts whose meanings cannot be understood. 

2See Morris Weitz, ‘The Role of Theory in Aesthetics’, Journal of Aesthetics and Art Criticism 15 (1956): 31–32. 

3Here and in what follows, a novel’s having x words is understood as (a) the possession of x words by this novel – if it is written in English, or (b) the possession of x words by its correct English translation – if it is written in a language other than English. 

4That said, however, to be completely justifiable, the view that the border that separates a novelr from a non-novelr is in the 30,000-word range – as well as any other view that specifies this border in numerical terms – requires a more serious substantiation than the one given above – a substantiation based on the results of a comprehensive empirical (primarily, sociological) investigation into the issue of what novels are considered the shortest ones and how many words they have. Because, so far as I am aware, no such investigation has been performed, the view being discussed should be treated as defeasibly true and, hence, as open in principle to revision. 

5See Gregory Currie, The Nature of Fiction (Cambridge: Cambridge University Press, 1990); Peter Lamarque, Fictional Points of View (Ithaca, NY: Cornell University Press, 1996); Nicholas Wolterstorff, Works and Worlds of Art (Oxford: Oxford University Press, 1980). 

6David Davies, ‘Fiction’, in The Routledge Companion to Aesthetics, ed. Berys Gaut and Dominic McIver Lopes (New York: Routledge, 2001), 265. 

7See John Searle, ‘The Logical Status of Fictional Discourse’, New Literary History 6 (1975): 319–32; Kendall Walton, Mimesis as Make-Believe: On the Foundations of the Representational Arts (Cambridge, MA: Harvard University Press, 1990). 

8See Paisley Livingston, ‘Narrative’, in Gaut and Lopes, Routledge Companion to Aesthetics, 275–84. 

9Ibid., 276–77. 

10Ibid., 277. 

11Ibid. 

12Ibid. 

13Ibid. 

14See Jerrold Levinson, ‘Refining Art Historically’, The Journal of Aesthetics and Art Criticism 47 (1989): 21–33. 

15Ibid., 21. 

16Ibid., 31. Examples of regards are a regard with close attention to form, a regard with openness to emotional suggestion, and a regard with awareness of symbolism. 

17Comp. Levinson, ‘Refining Art Historically’ (in particular, his idea of extrinsic and intrinsic modes of artmaking). 

18The expression ‘can’ here is used in the probabilistic (not the absolute) sense. 

19The same, by the way, can be said about other kinds of entities. Consider, for instance, artworks. Surely, there are entities that are clearly artworks (the Mona Lisa, David, Moby-Dick) and entities that are clearly nonart (humans, trees, planets). But there are also entities that are neither clearly artworks nor clearly nonart – for example, computer games. 

20Among these critics are, for example, Toby Clements and Alan Taylor. See Toby Clements, ‘Ian McEwan Is Lucky to Be Allowed to Publish Novellas’, Telegraph, 15 October 2012; Alan Taylor, ‘Let’s Not Talk about Sex’, Sunday Herald, 8 April 2007. 

21Thanks to Paisley Livingston for pointing out this potential objection to DN. 

22See Jake Kerridge, ‘Ian McEwan Interview: “Pornography Has Warped Our Teenagers”’, Telegraph, 17 May 2018. 

23Among these critics are, for example, Tim Adams and Natasha Walter. See Tim Adams, ‘On Chesil Beach by Ian McEwan – Review’, Guardian, 25 March 2007; Natasha Walter, ‘Young Love, Old Angst’, Guardian, 31 March 2007. 

24George Plimpton, ‘The Story Behind a Nonfiction Novel’, New York Times, 16 January 1966. 

25See ibid. 

26The most common piece of such evidence is the fact that the author explicitly characterized his or her work as a novel. 

27This is not to say, of course, that the concept ‘novel’ has not been examined at all; for possible analyses of this concept, see Walter Benjamin, The Storyteller, trans. Sam Dolbear, Esther Leslie, and Sebastian Truskolaski (London: Verso, 2016); Georg Lukács, The Theory of the Novel, trans. Anna Bostock (Cambridge, MA: MIT Press, 1974); Friedrich Schlegel, Literary Notebooks 1797–1801 (London: Athlone, 1957); and Ian Watt, The Rise of the Novel (Berkeley: University of California Press, 1967). The point is that the definitional project – of the kind common within the analytic philosophical tradition – has largely been ignored. 

28See Dominic McIver Lopes, Beyond Art (Oxford: Oxford University Press, 2014). 

29Ibid., 14. 

Acknowledgements

Thanks to Jerrold Levinson, Robbie Kubala, Anna Christina Ribeiro, and Paisley Livingston for a number of excellent suggestions that have led to substantial improvements.

Competing Interests

The author has no competing interests to declare.

References

  1. Adams, Tim. ‘On Chesil Beach by Ian McEwan – Review.’ Guardian, 25 March 2007. 

  2. Benjamin, Walter. The Storyteller. Translated by Sam Dolbear, Esther Leslie, and Sebastian Truskolaski. London: Verso, 2016. 

  3. Clements, Toby. ‘Ian McEwan Is Lucky to Be Allowed to Publish Novellas.’ Telegraph, 15 October 2012. 

  4. Currie, Gregory. The Nature of Fiction. Cambridge: Cambridge University Press, 1990. 

  5. Davies, David. ‘Fiction.’ In Gaut and Lopes, Routledge Companion to Aesthetics, 263–74. 

  6. Gaut, Berys, and Dominic McIver Lopes, eds. The Routledge Companion to Aesthetics. New York: Routledge, 2001. 

  7. Kerridge, Jake. ‘Ian McEwan Interview: “Pornography Has Warped Our Teenagers”.’ Telegraph, 17 May 2018. 

  8. Lamarque, Peter. Fictional Points of View. Ithaca, NY: Cornell University Press, 1996. 

  9. Levinson, Jerrold. ‘Refining Art Historically.’ The Journal of Aesthetics and Art Criticism 47 (1989): 21–33. DOI: https://doi.org/10.2307/431990 

  10. Livingston, Paisley. ‘Narrative.’ In Gaut and Lopes, Routledge Companion to Aesthetics, 275–84. 

  11. Lopes, Dominic McIver. Beyond Art. Oxford: Oxford University Press, 2014. 

  12. Lukács, Georg. The Theory of the Novel. Translated by Anna Bostock. Cambridge, MA: MIT Press, 1974. 

  13. Plimpton, George. ‘The Story Behind a Nonfiction Novel.’ New York Times, 16 January 1966. 

  14. Schlegel, Friedrich. Literary Notebooks 1797–1801. London: Athlone, 1957. 

  15. Searle, John. ‘The Logical Status of Fictional Discourse.’ New Literary History 6 (1975): 319–32. DOI: https://doi.org/10.2307/468422 

  16. Taylor, Alan. ‘Let’s Not Talk about Sex.’ Sunday Herald, 8 April 2007. 

  17. Walter, Natasha. ‘Young Love, Old Angst.’ Guardian, 31 March 2007. 

  18. Walton, Kendall. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press, 1990. 

  19. Watt, Ian. The Rise of the Novel. Berkeley: University of California Press, 1967. 

  20. Weitz, Morris. ‘The Role of Theory in Aesthetics.’ Journal of Aesthetics and Art Criticism 15 (1956): 27–35. DOI: https://doi.org/10.1111/1540_6245.jaac15.1.0027 

  21. Wolterstorff, Nicholas. Works and Worlds of Art. Oxford: Oxford University Press, 1980. 

comments powered by Disqus