BCS is a registered charity: No 292786
Electronic Visualisation and the Arts (EVA 2016)
London, UK, 12 - 14 July 2016
Poetry conventionally is the liberated utterance of the feeling subject, “the spontaneous overflow of powerful feelings” (Wordsworth). The computer has taken poetry in several significant directions. With poetry on the page (electronic or otherwise), the computer has emulated human-authored texts. But it has also been used to create more experimental poetries. A further, better known, initiative has been to abet the departure of poetry from the page in various forms of visual, kinetic, hypertext, and hypermedia poetries. The first enterprise this paper engages with, rejects and attempts to exceed. The second it advocates. The third is not its subject.
Whitby (2002) argues that the Turing Test (the ability of the computer to deceive a human interrogator) had historical value but has since become a distraction. The utility of computers therefore lies rather in achieving what humans alone cannot do rather than what they can. Whitby argues further that The Imitation Game tests not for intelligence but for cultural similarity. In this there is a failure of design based upon an attempt to test for intelligence without quantifying what this is. There have also been a variety of Turing Tests for poetry. These continue to comprise attempts to make computer poetry pass for human authored text. Involved in this, is a frequent appeal to conventional poetics. Computer poems may emulate traditional verse forms and the poetic styles of established poets. Comparable to the familiar Turing Test, the ‘Poetry Turing Test’ checks for a cultural match based on accepted poetics, which in turn evokes the eloquent, speaking subject.
This paper proposes a development of poetics by the adoption of algorithmically oriented approaches to create poems based on the potential of the computer for such things as speed, or the handling of large quantities of data, or the precise control of text. Furthermore, this suggests the construal of ‘the poetic’ as an internally unstable and unfixed category, rather than a completed entity. This argues further for the computer as a collaborator in the further development of poetic forms. There have been a few initiatives. One of these is DIASTEXT by Jackson Mac Low. Exceptions prove rules. The combination of the algorithmic and the experimental in poetry is yet to be fully explored. To do so might add to the scope of poetry on the page.