How to read

Literary analysis should limit itself, more or less, to the methods of of philology, in order to avoid doing damage to the text. When tasting a wine, it can be useful to pick apart its flavors and so on—this directs one’s attention to the richness of the taste, and then the taste can work better on the tongue. But trying to actually capture the experience itself is an impossible talk; inevitably, one will end up saying things like “I like this one,” or “huh, you know, I really like this one” or something only marginally less banal.

Likewise, it can be useful to look at the words and sentences and structures and so on, because by tuning in to these details, they can better be seen, and hence retain more of their richness. However, we must limit ourselves to a purely literal analysis of the text—that is, analysis on the level of words. Attempting to explain what it “really” means, the symbology, will hurt the text. Or in any case, good literature simply cannot be explained in this way. Good literature works like magic: it works; it is not itself a thing in the world, but something that acts from outside the world on it, or on the soul, and this process necessarily happens subconsciously and therefore is necessarily ineffable.

How does explaining a text damage it?

Firstly, whereas the original text could be used in any number of ways, acting on the soul differently depending on the individual reader (and their state of mind and the context in general), when it is reduced and packaged up into an analytic (or “logical”) discourse (or “logos”), it is forced to take only one of its many potential forms. Already there, much of its power is removed. Even worse, however, is the fact that packaging up the text in this way actually causes it to lose its power altogether! It is like replacing the wine itself with a description of what it was like to drink the wine. Sure, perhaps whoever is writing the analysis will not (if they are sufficiently careful) lose what they have gained from the text—but certainly, the experience of merely reading the analysis, cannot possibly substitute for the experience of reading the text. Therefore, any attempt to produce an analysis that substitutes the thing itself will not succeed; and so, when analyzing a work of art, the analysis must be kept strictly supplemental—that is, in the case of a literary work, literal.

At best, the analyst could themselves have written a new work of literature whose subject is the experience of reading or having read the original work. But then, it would no longer really be called “analysis”. So while such an endeavor might in principle be unobjectionable, it would be best for the author to think of themselves not as writing in the genre of literary analysis.

Note on good literature

There are useful written works that are not “good literature” in the sense above. In other words, it is possible to write something that is worth reading, but which is not good literature. Such works themselves usually proceed according to the formal mode of analysis, rather than of fiction, and therefore, they can be safely subjected to analysis without risk of harm. Examples include essays and non-fiction works.

But when a writer chooses to write in the fictional mode, they assert that they are attempting to do something that cannot be done in another mode—otherwise, analytic writing (i.e. non-fiction) would have been the superior choice, since it has numerous other advantages (for example, it does not take as much skill on the part of the reader to understand non-fiction as it takes to understand fiction). So anyone who subjects a work of fiction to analysis—beyond the philological, surface-level analysis of words—repudiates the author’s intention in a way that seeks to debase the work itself. (Of course, many literary analysts and critics do not know that this is what they are doing, and in fact may believe themselves, falsely, to be helping the author and their work. However, it is impossible to help a text do its work on the reader by explaining the text on its behalf.)