I've been asking myself (to no avail) this question for many years: What is writing? As most rhet/comp teachers and scholars will say, it's both a verb and a noun, a somewhat coherent set of processes and a resulting text. This is precisely where our problems started.
          I want to suggest that we begin thinking about writing from a slightly (but crucially) different framework: WRITING AS THE RECURSIVE, SHARED, (AND SOMETIMES ABSCONDED WITH) COORDINATION OR BUILDING OF SPACES AND FIELDS. In other words, writers are not individuals (or even groups) who produce texts, but participants within spaces who are recursively, continually, restructuring those (and other) spaces.
          This seems like a relatively minor (and quibbling) shift to make. After all, haven't we been talking about collaborative writing, about social construction, about postmodernism, about building MOOs and other virtual spaces, for years and years? Yes, we have. But we continue to value, as a discipline, the production of individual, discrete texts: essays, papers, email messages, presentations, Web sites. The more interesting and less traditionally textual examples are still too often constructed as special instances, exceptions, or interesting thought experiments. In part, this can be tied back to our still commonly held notion that the production of text begins with an individual creating an artifact (scrawling a letter in crayon, pressing individual keys [then send] on a keyboard, etc.), then moves outward to more complex and social forms. This tendential force – the notion that writing originates within the writer, then is sent out – powerfully affects everything else we think and do, discourages us (and everyone we work with) to consider writing as separate from social activity, as separate from thought. Perhaps more importantly, it continues to support the debilitating notion that writing should represent some external and objective truth, that writing is somehow an imperfect but necessary lens through which other (more important and real) things can be glimpsed.
          Here are some quick and incomplete implications of beginning with the notion that writing is the recursive, shared coordination of spaces and fields.

1. Writing begins with the assumption of something like a second law of communicative thermodynamics: symbols are never really created or destroyed, it only changes form, relation, or becomes duplicated. That is, writing doesn't spring full formed from the brow of isolated creative geniuses, but is always the ongoing manipulation of intertextual bits.
2. Writing has no meaning outside of those fields and spaces (because it literally *is* the fields and spaces). This, again, is something we've argued for years: that meaning is contextual. But in this new formation, writing is part and parcel of the context. (To think of it another way, writing is the medium in which reality is constructed and moves.)
3. Gone: The Shannon and Weaver model in which communication moves neatly from sender to passive receiver. Instead, writing requires the participation of the sender and receiver in order to even come into being.
4. MOOs (and, in fact, all architecture and "realworld" construction) are texts.
5. We do away with the idea that students must paraphrase quotes in order to "make the material their own" (an idea that makes sense only if you value individual creativity over use). Instead, research (and any communication) could be understood as symbolic-analytic work: the collection, filtering, connecting and disconnecting of information.
6. We do away with the notion of "voice" as the distinctive, inherent fingerprint of individual identity. Instead, we acknowledge that people reconstruct themselves and situations within and as situations dynamically. Some constructions might be more commonly used than others, but all are equally valid within their own contexts (and equally valid to be changed by appropriation, duplication, and modification by anyone to other contexts).
7. Citation is used only as a way to connect spaces/texts rhizomatically, not (as it currently often is) to denote ownership and false origination of ideas.
8. We begin to understand communication as shared space rather than either (a) discrete objects within space or (b) individually owned spaces that are rented by other individuals. (These are the two misconceptions on which currently mangled and out of control IP law is based.)

There are more. And I don't believe all of them are necessarily true; they're objects for thinking with.