For the past six months I've been working with K.M.Payne on a large
(indeed monstrous)
hypertext poem. The process has inspired the following thoughts. I'd be
glad of any response to or criticism of them.

For those interested, the poem itself is at Snakeskin webzine, or
directly at:
http://homepages.nildram.co.uk/~simmers/maze/index.htm

So here goes:

Large claims have been made for hypertext. To leapfrog around a body 
of writing by clicking on links is certainly a different experience from 
reading a piece sequentially from start to finish. Yet it's notable that
the 
most popular hypertext documents so far have been encyclopaedias, 
where the reader can follow a train of thought or whimsy from one 
summarised and superficial document to another. Encarta and its 
brothers have breadth, no argument - but they are not notable for their 
depth.

Yet writers from the Oulipians onward have been fascinated by the 
literary potential of hypertext. One senses the possibility of a whole
new 
relationship with the reader - the reader as  explorer; the reader as 
game-player; the reader as  co-author. One senses the possibility of a 
whole new type of poem. The poem as infinitude of choices; the poem 
liberated of the tyranny of sequence; the poem as a matrix of 
possibilities.

So far it's easier to feel the potential of the medium than to point to 
wholly successful achievements. Texts that I have seen tend to have the 
limitations of one or other of two genres - the Fighting Fantasy or the 
imagist gallimaufry.

Prose hypertext has actually found a market in printed books, in the 
Fighting Fantasy series for children. These books (like the computer 
games to which they bear a close resemblance) show the possibility of 
interactive fiction, while cruelly showing its limitations. The reader 
pursues her fate avidly, because the text is event-driven, one damn
thing 
after another. Puzzles may be momentarily intriguing, incidental details 
may be delightful, or striking or amusing, but story is all. One hardly 
notices the dearth of deeper drama. Characters slot into three simple 
roles - protagonist, helper or enemy. Subtlety or development might 
theoretically be possible, but the genre seems to be working against it.

Verse hypertext has tended to go an opposite way, away from games-
playing and puzzles, towards imagism. Most hyper-verse that I have 
come across lets you jump from one short piece to another by clicking 
on  keywords. A universe of images lies there before the explorer, yet 
without narrative thrust the process can become tedious. One clicks to 
the next screen because it is there, not because one has an urgent 
desire to see what comes next. My own experience is that after a while i 
stop clicking, say "That was quite nice," and feel no regret that I
haven't 
reached the end of the text, if end there is.

The poets, I suggest, have something to learn from the fighting 
fantasists. A page of hypertext verse should not be an end in itself,
but 
should be clearly part of a greater whole. But the whole is a different 
kind of unity from that which poets are used to.

The hypertext poet needs new ways of thinking about his work. No 
longer is he like a composer, with a right to expect his audience to 
remain attentive from the opening bars of a sonata to the last.  Rather
he 
is like an architect. Entering his building will be different for each
visitor. 
Few will systematically explore the entire edifice from servants
quarters 
to wine cellar. One may enter the main hallway, and then see nothing 
but the main reception rooms and the conservatory. Another may sneak 
through a side door, passing the kitchens and the billiard-room, and 
head through the secret passage to the master bedroom.  their 
experiences will hardly coincide - yet both follow routes made possible 
by the architect.  The visitor feels free and unconstrained, yet the 
architect has constructed her experience, determining the nature of 
entrances and exits, and designing juxtapositions of rooms and 
passageways that are appropriate, or impressive, or surprising.   A 
house should be more than just a place where one can wander 
aimlessly, though; it should also be a place where one can walk with 
purpose.

What kinds of purpose suit a hypertext poem? Let us try to answer this 
by thinking of poems of the past that have some affinity with hypertext.

Some people have suggested that Eliot's Wasteland is  already almost a 
hypertext poem, with its leaps of subject, tone and register. It
exploits 
the poetic value of links - from present to past, sublime to grotesque, 
text to notes. Yet in other respects it is not hypertextual. Its
succession 
of voices come in a strictly controlled and artistically necessary
order. It 
moves from beginning to determined end.

Pound's Cantos might be a better example of a poem that verges on 
hypertext. Here surely is a poem that nobody has ever sat down and 
read through sequentially from start to finish. It leaps from thought to 
thought with all the quirkiness of Old Ezra's busy mind.  Yet the
difficulty 
many readers have with the Cantos shows a danger of hypertext. How 
does such a work avoid the impression of formlessness - of local glories 
but a distinct possibility that readers will lose their sense of
direction and 
purpose as elliptical leap follows leap follows leap?

A poem that I have sometimes had in mind whilst creating my literary 
monstrosity with Ken Payne is Browning's Childe Roland. Its hero 
crosses the dreadful landscape, and whichever way he goes finds 
nothing but despair. Had Browning had the technology, mightn't his ludic 
mind have enjoyed letting the reader choose her own optimistic 
alternatives, whilst making sure that every road led to the Dark Tower?

Ken and I have followed the Fighting Fantasists in having a "you" as the 
protagonist (in fact we have made that "you" and its self-concept the 
focus of the poem). We have used narrative devices to swing the reader 
from page to page. On the other hand, we  have set ourselves the 
constraint of a strict verse form (ottava rima) that we hope will give
the 
poem a unity as we switch from genre to genre and mood to mood. We 
have tried to give the reader the feeling of complete freedom of choice 
(though a re-reading may show the limitations of that freedom). 

Ours are certainly not the only choices; they are probably rather 
conservative ones. What we have tried to do though, is to produce a 
poem as environment, in which every visitor can find his or her own 
reading, from a huge possibility of permutated choices. Some journeys 
through the poem will be richer than others; readings will vary hugely
in 
length. Some will seem like quick romps through the action; others will 
be leisurely strolls through peripheral material.  It has occasionally 
bothered us to think that some of our favourite lines and images are 
hidden away in distant rooms that the average visitor is unlikely to
find. 

What else could we have done?

Hypertext does not have to be narrative. If it is, then the genre need
not 
be Gothic, and the central character need not be "you". Our choice of a 
set verse form is very much the product of our own prejudices and 
preferences. There is room for all sorts of experiment - mixing prose 
with verse; setting narrative interludes against meditative ones;
finding 
new ways of letting the reader discover the motifs that can be developed 
within the work. Maybe  hypertextual techniques could mingle with  
those of computer-generated text. 

Poetry begins in wordplay, and at the moment hypertext is something of 
a computer game. But there are techniques and possibilities here worth 
exploring, until one day a large intelligence arrives who will use this 
techniques in such a way that they become his own, and say things that 
could be said in no other medium.

____________________________
George Simmers

SNAKESKIN webzine is at
http://homepages.nildram.co.uk/~simmers


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%