Yesterday I started drawing some diagrams of novels and short stories with complicated fictional structures. This diagram is typical of the structures I’m looking at, in which there are a number of nested frames accounting for the relative distance of the real world from information presented in the text.

This one is based on An Examination of the work of Herbert Quain by J. L. Borges. It is a work in which some writer celebrates the work of an experimental crime writer, who – we deduce from real-world knowledge – exists only within the conceit of the work itself. (The writer of the Examination is unnamed so it is unclear whether the work is directly a text written by Borges as a fictional trick, or to consider the work as the creation of a fictional writer writing a text that is true in his fictional world.)

The God of the Labyrinth is one of the books summarily described from memory  by the writer of the text. It is a detective story in which:

Once the enigma is cleared up, there is a long and retrospective passage which contains the following phrase:

‘Everyone thought that the encounter of the two chess players was accidental.’

Everyone in the story – with the detective as their spokesman – has it wrong, and it is the narrator who retrospectively provides the reader with a clue that allows the truth of the mystery to be solved outside of (and upon?) the fiction. The reader takes the place of the detective and inserts their own conclusion into the story. What troubles me, now that I’ve tried to draw all this out in frames, is the question of which interpretation is more fictional.

In my diagram I’m not sure whether frame (1), the innermost fiction, should be the detective’s or the reader’s interpretation. The detective’s solution is ‘wrong’ and the reader’s solution is ‘right’. The detective’s solution is a misreading of events in the book, creating an additional layer of fictional cohesion on top of the already fictional series of events in the story. The layer of cohesion that the reader spreads over the fictional events is a faithful extrapolation of those events, and so I could have subsumed it into frame (3) along with the rest of the plot information. But the reader’s solution is more deeply concealed than either the general plot or the detective’s solution.

Indeed the reader’s solution is only available from above the fiction, with the special insight of the narrator’s clue about chess players. The clue alone is insufficient to solve the riddle: we read that “the unquiet reader re-reads the pertinent chapters and discovers another solution, the right one.” The reader is invited to re-read, and in doing so to open out the fiction into a set of personally involving events that can be stepped around and moved about, and it is the reader’s own movement that closes the story. Perhaps my diagram needs to be redrawn with the reader’s solution as far out as frame (5), on the same plane as the putative existence of the very reader.

And then there’s the option of the reader being insufficiently disquieted by the narrator’s clue to read back. This reader will still understand that the detective’s interpretation is inadequate and will cross it out as a possible solution, and will furthermore assume that the correct solution is in the book somewhere to be found, but will abandon the story in a headless state.

We must further remember that these possible readers are all fictional, and cannot creep out beyond frame (5). One thing this does is to keep any decisions, movements or conclusions of these readers as strict potentialities, which are all are possible at once. Had Borges (or anyone else) really written The God of the Labyrinth, the real-world reader would have been obliged to enact one of these possibilities making it actual rather than possible, and causing the remaining possibilities to dry up.