Evaluation, translation and baseball

To go back to that Valve argument, again, I want to expand another aspect of what I was arguing for in that discussion, on the subject of evaluation. This is a topic that has come up a fair amount recently; Gawain and I, for instance, have floated it a couple of times over at his site. Therefore, I strongly invite comments to this post from Gawain, Richard, A Little Thought, and the rest of you. Here is the context of the debate:

Me: "Why should the fact that a book was Intelligently Designed (I use this to designate the product of a mind, rather than specifically of a good mind—that would be begging the question) affect your analysis of it?"

John Emerson, replying with a rhetorical question: "Why would our awareness that certain events in [baseball] count as "runs" and are attributed to "teams", and that one of the "teams" will win and the other lose—why should that affect our interpretation of the physical events on the field? Isn’t it ethnocentric and subjective to apply the normative framework of "baseball" to these events?"

Just to spell this out, John is drawing an analogy between baseball and literature. Baseball consists of concepts and rules (or, in Euclidean terms, definitions and axioms), which are agreed upon by convention. We could interpret a baseball game according to different concepts and rules, but we wouldn't have an accurate understanding of what was going on. Just as baseball is incomprehensible without an a priori knowledge of the concepts and rules, so literature is incomprehensible without an a priori knowledge of X. The content of X is never made entirely clear; but it seems that the intention of the author (analogous to the intentions of the batters and pitchers, which are in turn based on the concepts and rules of baseball) is part of it. For John, (part of) studying literature is assessing how well a given author achieved what he was trying to do, how well he realised his intention—just as part of studying baseball is assessing how well the players realised their intentions, strategies and so forth. In his own words, "our understanding of baseball events is enhanced by our understanding [of] what baseball players are trying to do. Similiarly [sic] for literature." This assessment is what we are calling 'evaluation', and it has a central place in literary studies.


This is a respectable argument, but I think that it is completely wrong; here's why. All participatory activities, I submit, have two aspects: the conventional and the creative. The former is a manipulation of 'concepts and rules', and is wholly formulable; the latter is a free, individualistic activity, and wholly non-formulable. Baseball is an excellent example of an activity which is entirely (or almost entirely) conventional. We judge the ability of a given player or team by how well they manipulate the concepts and rules of baseball: how well they perform according to fixed, objective standards, for instance scoring home-runs, which can be completely defined according to those concepts and rules.

Everyday language is an excellent example of a mixed activity. It is largely conventional, practised according to concepts (words) and rules (grammar) agreed upon a priori by those speaking. However, as Chomsky is now famous for noting, one of the fundamental characteristics of human language is that it is creative, able to generate an indefinite variety from a finite lexicon and grammar. Moreover, the creation of new conventions, primarily of words but also of grammar, can be seen constantly in fertile vernaculars and among specialists such as scientific communities. People both need to and like to coin new phrases, and to use language in new ways. Thus everyday language use is a mixed activity, at once conventional and creative, and not reducible to either.

Literature, on the other hand, is an activity which is predominantly creative. There are genre conventions, but these are not fixed, and no serious critic would evaluate a literary work by how closely it adhered to them; they are, in fact, more like expectations. Beyond such expectations, there are no 'rules and concepts' in literature, no 'fixed, objective standards' against which we can measure literary efforts. The best proposition for such a standard so far is 'the author's intention'. I refer the reader here to the standard demolition of the interest in literary intentions: Wimsatt and Beardsley's much-cited 1946 'The Intentional Fallacy'. Their argument is twofold: firstly, the intention is unavailable—we can never know what an author was trying to achieve, and even if he purports to tell us, as with the self-commentaries of Dante and the Dantophile Eliot, we can never be sure how trustworthy he is, and in any case his explanation is just more language of uncertain intention!—and secondly, and more importantly, there is no objective reason why we should base our evaluation on the author's intention, were we to know it.

In baseball, we base our evaluation of a player's performance on standards governed by the rules of the game because it gives that player's actions some objective content. But there is no objective content to be gleaned from basing our evaluation of, say, Ulysses, on Joyce's authorial intentions. It is true that we will get more references if we know about Joyce's life, but not true that our understanding of Ulysses will be altered by knowing Joyce's intentions. And if you say to me, 'I like Ulysses more because I know what Joyce was trying to achieve, and he was successful', I can reply, 'Well, why should I?' And you could not answer this. Ulysses is not made any less intelligible without such a criterion.


Compare translation, which I think makes an interesting analogy. Translations, it is generally agreed, can be either faithful or beautiful. If I am studying Plato and want to understand the whole of his thought, then I will want to read a faithful translation of his works. But if not—why shouldn't I read the most beautiful or brilliant translation, irrespective of its fidelity? In terms of pure aesthetics, shouldn't we evaluate the best finished product over the most accurate? But all literature is a translation: of the content of an author's head. Without Greek, Plato's original text is unavailable to me; but we are all without Greek when it comes to the author's intention, and ultimately, that intention can only be all Greek to us. Let us focus on the product.


I conclude by saying this. Literature is literary insofar as it emphasises the creative over the conventional. We evaluate works of literature subjectively, and on a case-by-case basis, because there are no rules and concepts, no fixed standards, against which we can read a work, except arbitrarily. In its purest form there really are no conventions at all—a work is the product of an individual spirit and nothing else. Given that there are no fixed standards, we cannot discuss how well the work has met those standards. We can express our reaction to the work, our evaluation of it, but (as philosophers put it) we can make no claims on another's evaluation of the work, in the same way that we can with baseball.

Therefore, such evaluation has no place in literary studies.