Tuesday, December 25, 2012

The Pierre Menard code

Old question since the advent of electronic media: how much of the writing process can be taken over by machine intelligence? And how much credit can be given to an "author" in this case? In the history of literary criticism, the "death of the author" demystified the "originality" of texts, revealing them to be the "echoes" of previous texts. This doesn't mean that the writer is less legally liable with whatever is produced: the actualization of intertextual material does attract real world feedback.

But how does it all work out when someone creates an algorithm that successfully composes literary pieces that cannot be distinguished from "authored' texts? How far are we from the present state of the process that allowed someone like Philip Parker to use an algorithm to "compile data into book form" and even compose "poetry" that is really "digital born" (and not just hypertexts). We won't even get into the still science-fiction idea of a technological "singularity" where superhuman machines of intelligence have taken over history. Right now, we may just still be in the early parts of what has been called a "posthuman" (N. K. Hayles) rubric in writing and literature. Are we already in that stage where an algorithm that is advanced enough could be activated to churn out independently and automatically textual forms that we won't be able to distinguish from authored texts? And is this question still relevant today? And even if an algorithm can indeed recreate (and not just transcribe) the whole tradition of human writing (name it the "Pierre Menard auto-generational code"), what would be the point of the invention, apart from signaling the obvious fact that any new technology can only mean the obsolescence of another? Which ones, we ask, will we see surviving in the end?

"The next area of formulaic writing to which Parker wants to adapt his algorithm is romance novels, which are widely (perhaps unfairly) denigrated as "cookie-cutter" literature. Parker believes their simplicity and limited plot structure suggest romances as the best target for an early attack on fiction writing....

- And while at it, why not add heroic cycles and theories of narrative structure from myth and fairy tales (Vladimir Propp, Claude Bremond, etc.) to the whole algorithmic recipe?

"Regardless of his level of success, human authors are likely to face progressively more competition from algorithmic authors over the next decade or so. At this point it seems likely that the place of the best human writers is probably safe, but for how long? Time will tell." (http://www.gizmag.com/writing-algorithm/25539/)

*See also the Writing Machine Collective site, http://www.writingmachine-collective.net/about.html.

No comments:

Post a Comment