Aquilna reflects on the reflections in Callus and da Silva's "Strange Metapaper."
How to answer the invitation by the editors of electronic book review to provide a response of some kind to 'A Strange Metapaper on Computing Natural Language'? Should I write a report or review meant to evaluate the suitability of this 'strange metapaper' for publication? If so, then in which ways would my report be essentially different to those of the three reviewers whose reports already assess what Portela and da Silva call the 'embedded paper', 'If then or else: Who for whom about what in which'? In the version that appears on electronic book review, the authors have added – to the 'embedded paper' – a Metaintroduction and another abstract, but the 'strange metapaper' now also includes, within its realigned textual frames (as they appear on screen), the three reviewer's reports and an editorial abstract by the editors of electronic book review, all of which now appear under the new title of the 'matapaper'. Who then would a review to be written at this point be reviewing? Would it be Portela and da Silva? Or would the review also have to consider the work of the three reviewers and the editors of electronic book review? The review genre, in most institutional contexts, gives the dispensation to the reviewer to write something that positions itself outside that which it reviews, and it also offers the mask of anonymity. Outside, objective, and anonymous, then. Or I could choose to write what the editors call a 'publishable riPOSTe, which is usually about 3-5 paragraphs' long (I am still within the conventional limits for this option, especially since I decided to join what should really have been two or three paragraphs in order to stay within the limit). If I opt for the riPOSTe, would my response be a comment about the 'strange metapaper' or a further addition to it, in the way the reviewers' reviews are incorporated into it?
This long preamble is perhaps a roundabout and somewhat inelegant way of saying that, as Derrida writes in 'The Law of Genre', 'as soon as genre announces itself, one must respect a norm, one must not cross a line of demarcation, one must not risk impurity, anomaly or monstrosity' (LG, 224-5), but also that there is 'lodged within the heart of the law itself, a law of impurity or a principle of contamination' (LG, 225). It is the hyper-awareness of genre and its concurrent contamination that the 'strange metapaper' performs.
By moving the reviews, the new Metaintroduction, the new abstract and the electronic book review introductory paragraph within the frames of the same title, the paper turns the parataxtual into the textual, the supplementary into the constitutive, and the countersignature into a signature. The internal re-mark engineered by Portela and da Silva (through the addition of a 'meta-textual level' and the 'networked text') gives way to a recursive series of remarks that disrupt and devour any attempts to stabilise the text in order to respond to it objectively, from a distance, the way a reviewer is expected to review the text. Indeed, the author of Review 1, who was 'not tending towards recommending acceptance at the workshop' is unwittingly describing what would become of their own intervention when asking, 'do they comment on the writing process while being also a part of it?'
This 'strange metapaper' is primarily a paper about framing. It is about how self-commentary and self-reflexivity display the inherent permeability of genre. It is about the way the various institutional guardians of the law of genre (authors, reviewers, organisers of the INLG Workshop on the Computational Creativity n Natural Language Generation, the NAACLHLT, the electronic book review editors, the conventions around academic discourse, etc.) seek to legislate not only about the text but also its frames and borders, seeking to clarify what is in the text and what is parasitic or supplementary to it.
In this respect, the metapaper is successful in showing how any accounts of textuality (included that generated by computers) that are exclusively formal in nature fail to consider what Portela and da Silva call 'the cybernetic logic of social control'. And in its suggestion that the self-reflexive awareness of the writers 'cannot be automated' and is possibly beyond 'computational creativity', they touch on an essential question: How does one create, not simply a machine for the creative generation of language, but a machine for the generation of critical, possibly evaluative, riPOSTes that are conscious of the institutional framing of the text they generate?