From my friend Greg, whose opinions matter to me greatly:
Can the solution to the hermeneutical fixation *solely* on the meaning of the text be legitimately compensated by an assertion that meaning is *not* found in words [as signs]? It seems to me that it would be much more reasonable given the problem to say that meaning is not *only* found in the semiotic function but also in the performative function of words (to say nothing of the author's subjectivity).
Just to be sure I'm accurately representing Fairclough, Jessop, and Sayer's 2001 article "Critical Realism and Semiosis," let me make clear that what they prescribe is exactly what you suggest: that even-handed attention should be given to both the constantive (semiotic) and performative (extra-semiotic) functions of a text. I am going to differ with them, however, in suggesting that focusing on the constative (or denotative) function of a text reveals that we still believe that texts somehow hold meaning when no one is reading them. I'd like to debunk this notion. I don't believe that texts magically contain meaning within their margins. This misconception is similar to the misconception that people have about batteries, that somehow energy is coursing within the battery casing like a hamster on an exercise wheel.
In both cases (meaning in text and energy in battery), a process has been misidentified as an object. In the case of a battery, what one might mistake for "energy" is actually a process in which two chemicals interact with each other once the positive and negative terminals form a loop. This produces an electric current, which can then be used as energy. In the case of a text, a collection of signs lies there on the pages. Once read by a subject, the reading of each word sets off a kind of Pavlovian reflex in the mind of the reader, conjuring a meme that the reader associates with this word. This meme (or idea-gene) is shaped by the reader's experiences with those words (how the word has been used by others [social component] and how the word has been used successfully by the reader [human agency component]). Each of these memes are filtered through the reader's current mental/emotional milieu, and have a feedback effect, in which they in turn affect the reader's meme->word association and mental/emotional milieu. What I'm describing here is a series of events that form a process (semiosis, or the creation of meaning), not an object.
Critical Realism is helpful here, because it asserts that reality is comprised of objects (texts) that possess emergent powers (the power to catalyze semiosis in the mind of the reader) which interact to form processes (semiosis). But what does thinking in terms of semiosis instead of meaning buy us?
It helps debunk the idea that texts are a magical flagon of meaning to be poured out by English professors or theologians. Texts become more like a cultural artifact whose exhaustive study must also involve sociologists, historians, linguists, etc. I envision the text to be like a patient who interfaces with a wide variety of medical professionals (clerks, nurses, physicians, surgeons, x-ray technicians, anesthesiologists, etc). Except, instead of trying to cure the text of a disease (hermeneutics sees a text as a puzzle that needs solving, a patient that needs curing), the text becomes the cure--as a locus for interdisciplinary studies (involving all of the humanities). The text forces us to read humanity itself.