This week at The Junkyard, we're hosting a symposium on Kathleen Stock's recent book:  Only Imagine: Fiction, Interpretation, and Imagination (Oxford University Press, 2017).  See here for Kathleen's introduction.  Additional commentaries and replies will run each day this week.

* * *

Commentary from Tobias Klauk: Stock on Fiction and Imagining

Kathleen Stock has written an engaging and thoroughly interesting book, which was a joy to read and to discuss. In this commentary, I shall concentrate upon chapter five on the notion of fiction. “A fiction,” Stock writes, “should be understood as a collection of utterances of a certain kind: utterances intended by their author to produce conjoined F-imagining in readers or hearers, in the reflexive way indicative of communication according to Grice.” (149-150) These utterances are characterized by three conditions, of which only the first shall concern us here: The author must intend that her “utterance of x should cause some particular F-imagining in her intended audience B.” (149) Finally, “F-imagining exhibits the following basic features […]. It is propositional and ‘quasi-factual’ […]: it involves thinking that p. It is potentially conjunctive in the sense I have also indicated. And its content can partly overlap with the contents of one's current belief set, but cannot just replicate a subset of the latter.” (148)

I have five questions or comments concerning this picture. The first has to do with passages of fictional works which are not nonfictional but somehow less fictional than others. Take for example Goethe’s complaint about certain parts of Manzoni’s novel I promessi sposi:

“I told you lately,” Goethe began, “that the historian had been of great use to the poet in this novel; but now, in the third volume, I find that the historian hurts the poet, for Signor Manzoni throws off at once the poet's mantle, and stands for some time as a naked historian.” (Eckerman: Conversations with Goethe, 435-6 (23.July 1827))

I take it that Goethe sorts certain parts of I promessi sposi as nonfiction. In a sense, he is right to do so, for in chapters 31-32 on the black plague the protagonists are almost never mentioned, all traces of irony vanish, indeed nearly all literary design is missing. There also is a clear sense, in which Goethe is wrong, though: the events described in chapters 31-32 contribute to the storyworld.

If Stock's theory gave us the only relevant concept of fiction, it could not deal with these cases. Although Manzoni intends his readers to believe that the events of chapters 31-32 happened, he also intends his readers to F-imagine them as well as the connected story of Renzo and Lucia, and so Stock’s theory has to sort chapters 31-32 as fictional utterances. We cannot explain the expression that they are less fictional than the rest of the book.

I see three ways of dealing with such cases. We can either adopt a multifactorial theory of fiction (e.g. along the lines Stacie Friend gives). Or we adopt Stock’s theory but recognize other notions of fiction (and indeed that seems to be her idea (157)). Or we could give up the notion of fiction and instead talk directly about the different aspects involved. Why should we prefer the second picture, as Stock does?

Secondly, Stock restricts fictional content to propositional imagining (e.g. 26). She gives two reasons for this, (a) that “authors very often do signal that sensory and/or phenomenal imagining is appropriate for certain fictional passages: most obviously, by giving detailed descriptions of the perceptible properties of people and things. But even here, propositional imagining is clearly appropriate as well,” (26) and (b) “assuming the reader forms, and is intended to form, a relatively determinate image in response to a given descriptive passage, the precise determinate aspects of her image will nonetheless not correspond exactly to anything she is intended to imagine. In response to the passage above, for instance, the precise shade of crimson, the size of the windows, and the precise spatial relations of the furniture, as represented in one's image, are not dictated by the fiction but decided by the reader.” (26 Fn. 19)

Unfortunately, I cannot see why (a) excludes nonpropositional imaginings from fictional content. And (b) just assumes that there cannot be intended individual imaginings. Why not, though? Manzoni, e.g., in some passages explicitly asks his readers to imagine something, but leaves the details of the imaginings to his readers: “Imagine what a blow this was for our good father.” (Manzoni, The Betrothed, 231) “The reader may imagine the feelings of Renzo at these words. His eyes expressed the warmth of his gratitude to him who had uttered them; but they sought in vain for Lucy's.” (434) “One fine evening Agnes heard a carriage drive up to the door of her cottage. It was Lucy and the good widow. We can easily imagine the joy of the meeting.” (443) Such intended individual imaginings are not even restricted to nonpropositional imaginings. The examples therefore suggest that the fictional content is not entirely fixed by the intentions of the author.

Thirdly, in contrast to earlier writings, Stock no longer attempts to answer the question what distinguishes fictional from nonfictional works (e.g. 150). I find this unfortunate, since it is the question at the heart of any theory of fiction. Given that Stock’s theory successfully tells us what distinguishes fictions from non-fictions, it does so at the price of being no longer a theory about differences between certain classes of works.

Fourthly, I don't share the intuition that one cannot F-imagine a subset of one's beliefs, which is an important pillar of Stock’s theory. Because it is mostly unproductive to just insist on one’s own intuitions, let me at least sow some doubt: We can bracket our beliefs by assuming for the sake of the argument that they are wrong, and can temporarily abstain from judgement concerning their truth. It seems that in this manner we can assume subsets of our beliefs to be false. Why then should it be otherwise for imagination? Notice also, that one can reject the intuition while still believing that imagining a subset of one’s beliefs is not enough for fiction. Might it be that the intuition concerning not being able to imagine subsets of one's beliefs gets its force from here? And finally, should such a central pillar of the theory not rest on more than just an intuition?

Fifthly, if we assume with Stock that fictional content is given by author’s intentions alone, we are not done, since author’s can have conflicting intentions. In the Holmes story The Adventure of the Speckled Band the solution presented by Holmes relies on a Russell's viper being able to crawl up a bell-rope, which actually, Russell's vipers can’t. So Conan Doyle intended readers to imagine that Russell's vipers can climb ropes, but also that the usual facts about nature hold in the Holmes stories: they are not supernatural stories. Therefore, Conan Doyle intends readers to imagine that Russell's vipers have the abilities they actually have. In this case, one intention obviously trumps the other. However, the theory should be able to explain why. One should like to get a more complete picture here.

Notice that it is not enough to deny that Doyle intended his readers to imagine the usual facts about nature, for there are examples in which such intentions are important, as when people rightly complain that Scott lets the full moon rise in the north-west in The Heart of Midlothian or that Hugo makes errors about Paris in Notre Dame de Paris.

Response to Tobias Klauk from Kathleen Stock

In his response to Chapter 5 of my book, Tobias Klauk offers several worries about the theory of fiction I outline there.  I’ll respond to the first, second and fourth ones. Klauk’s third point – that I don’t explain what makes a fictional work, a work – I concede. Klauk’s final objection – that I don’t offer an adequate account of cases where an author’s intentions apparently conflict – does need to be addressed, but requires more space than I have.

Klauk’s first worry concerns fictional texts that contain long passages of history, such as Alessandro Manzoni’s I Promessi Sposi. He worries that my theory of fiction is unable to accommodate such cases, since they are “less fictional” than other passages, and yet my account can’t mark this.

I don’t know Manzoni’s work, but Klauk’s description stresses that, in the relevant passages, “the events described.. contribute to the storyworld”. So it sounds like the passages are intended by Manzoni to be imagined, though full of historical information. On my view, one of the things that would confirm this is, roughly, that the propositions they contain - though the reader may well believe them and be intended by the author to do so - are also intended to be conjoined in thought with further propositions, where those further propositions - propositions about the invented characters in the wider work, and so forth - aren’t currently believed by the reader, or intended to be. Single passages within fiction can serve multiple functions, after all.

Now, it may be true that for many, Klauk included, passages like this seem “less fictional” than those that are purely invented. I’m sanguine about this, however, since as I make clear in the Introduction to my book, I’m not intent on providing a theory of fiction that could satisfy all intuitions (none could!). Of course I want my theory to accommodate many of the things which we already say about fictions, for fear of missing the target altogether. But to try for more than this is impractical.

For similar reasons, I’m not especially troubled by Klauk’s intuition, as articulated in his fourth worry, that we might imagine only that which we also simultaneously believe, something I deny. I’m already aware of those who share this intuition, but, again as emphasized in my Introduction, my aim is not just to introspect/ intuit and pronounce accordingly, nor to get into a stand off with those who intuit differently. Rather, I wish to build up coherent theories of fictional content, fiction, and imagination simultaneously, each of which complements and so supports the others, and with which we can then do good explanatory work. (So it isn’t true that this “central pillar” of my theory rests only on intuition, as Klauk implies. In fact, it is grounded in observation of what is standard for imaginative engagement with actual novels and stories).

I turn finally, and anachronistically, to Klauk’s second worry. This concerns passages in fictions that seem to leave it up to the reader what exactly to imagine. Again, Manzoni provides a good source of examples: as in “Imagine what a blow this was for our good father”. Don’t such examples suggest “the fictional content is not entirely fixed by the intentions of the author”? I don’t think so, at least by my own lights. Effectively, I think such passages are to be read principally in terms of what they imply: which is, in the sentence just quoted, that this was a blow for the father of the narrator. Additionally, such passages function to emphasize a legitimate feature of the reader’s practice, generally: namely, she is permitted, and in this case actively encouraged, to imaginatively ‘add’ determinate aspects to a fictional scenario, that have not been specified for it by the author. If a heroine has beautiful eyes, readers may try to imagine determinately how they are beautiful; if a man is wearing a green coat, they may try to imagine what shade of green. But even so, on my view these particular determinates still don’t get to count as part of the plot/ what is fictionally true in a given case. After all, the author hasn’t intended readers to imagine that any particular determinate is present in the scenario, but only that some determinate is, without specifying or even hinting. If a rival view insists that unintended specifics ‘added’ by the reader to a fictional scenario are, in fact, part of the associated work’s content, then we’re clearly dealing with a very different theory to mine; and the two must then be compared for their general explanatory power.