The "Rulin waishi"
The previous chapter argued that certain authors incorporated functions once the province of the extratextual commentator into the text itself through the creation of narrator-commentators and pointed to a number of concrete and observable features in those narrators to back up this hypothesis. The task of this chapter is far harder, because there is no objective evidence for its central thesis that certain writers, under the influence of fiction commentary, purposefully left interpretive space for readers to fill in through the construction of their own commentary based on unobtrusive clues in the text. I think we can not only describe certain categories of holes in these texts with a fair degree of precision, but also show that they were designed to be filled with mortar prepared under the influence of fiction commentary.
What needs to be proved first is that there were holes in these texts that needed to be filled in by the reader, and that they were the kinds of holes a reader trained by reading fiction commentaries would be prepared to fill. We will also need to prove that the individual works were influenced by fiction commentary. It will also help to show that the works in question were easily understood and accepted by readers acquainted with traditional fiction commentary but caused problems for readers who were not.
This chapter will present the Rulin waishi as an example of fiction written with latent commentary because it seems to me to be the most convincing case. My decision not to discuss other works in this chapter does not mean that the Rulin waishi is unique. It is just that I lack the ingenuity and space at this time to present a convincing argument vis-à-vis other examples (in Chapter 14, however, we will see how latent commentary appears in the Honglou meng as one of several responses to extratextual commentary).
The Rulin waishi is a good example for a number of reasons. First, it represents perhaps the most extreme development of the suppression of the tra