why this sky will be presenting their show Portents as part of The Space’s Spring Season. The play began as the result of devised workshops and in this guest post, writer Nat Norland shares his tips for deconstructing and reworking the original text to create something new.
I can’t really talk about how I started writing Portents, because I didn’t start writing it.
The first draft was written by a friend of mine off the back of a number of devised workshops I directed, playing with ideas related to conspiracy theories and self taught artists. I made a slightly clumsy attempt at staging it, and then I put it in a box and forgot about it for quite a long time.
And then I took it out the box, and decided I wanted to hack it to pieces.
Not completely to pieces, maybe. In situations like this, you have to be quite clear with yourself about exactly what you want to keep, and what you want to make your own. In my case, I wanted to hang on to a lot of the content and tone of the piece, while tearing out as much of the characters and structure as I could. Of course, new themes and ideas started to creep in – something that had been largely about conspiracy theories began to be increasingly preoccupied with language and religious experience – but the main thrust of the rewriting was about unmaking the structure, collapsing it from the inside out.
So, with all that said, here’s a few ways of tearing a play’s insides out and putting them back in stranger:
1 – Killing the authorship
Before I start talking about slicing things out, lets talk for a minute about splicing things in. One of the really freeing things about doing a heavy rework of an existing text, rather than starting from scratch, is that it releases you from the idea of a unitary authorial voice. Since you’re immediately dealing with lines you’ve written coexisting with lines you haven’t, authorship starts to feel like another structure that can be played with and bent out of shape. To this end, I started experimenting with including direct quotations from different sources, spliced unceremoniously into the flow of scenes. It felt in some ways similar to music production – if you’re remixing something, why not try throwing in other samples for added texture?
It quickly became apparent that different sources lent very different flavours. Song lyrics were fun – not as finely wrought as poetry, but more stilted than prose – making them stick out immediately in an interestingly jarring way. I also played with transcripts from television interviews, and, owing to the aforementioned religious themes, snatches from the Bible.
I think this is a technique where trial and error is really valuable. Pulling from a lot of different places, and seeing what was pleasantly and unpleasantly incongruous. At the very least, it makes for an enjoyable process.
2 – Gutting the dialogue.
Portents, Portents Mk I, pre-disembowelment Portents, had a number of scenes which were just characters talking to each other. In the same place, about the same things, at the same time. This crops up in plays occasionally, I’ve noticed. There was a flavour to them that I really liked, but I wasn’t that interested in people talking to each other in the same place at the same time any more. Somehow that framework had to go.
There’s a number of ways you can go about this, but I think setting some sort of formal constraint that runs counter to one or more of the scene’s internal rules is often a really good place to start. That kind of rule bending can force your writing into some really interesting and unexpected places, without the burden of knowing exactly where you’re going beforehand. And dialogue scenes have a whole lot of internal rules to tear into – characters talk to each other, roughly alternating lines; they’re generally assumed to be in the same neighbourhood of spacetime; they respond to at least some of the information presented to them by their partner in conversing.
In my case, I tried keeping one character’s lines exactly intact, while stripping out and replacing the other’s. The new lines had to either seem to be from an entirely different time and place as the originals, or to completely ignore the other participant in the conversation. This threw quite a disconcerting sense of distance into the scene, but didn’t completely paper over the original flavour.
3 – Shuffling the grammar
Of course, you can’t throw in a sense of distance between two voices when there’s only one voice talking. Somehow, that distance has to come between the voice and the audience. Finding a rule to twist to that end was less obvious, particularly as the speakers were generally addressing the audience directly
My solution here was based on parsing speed: how quickly an audience can understand what’s being said to them, syntactically and semantically. The slower and less completely an audience can process what’s being said, the more conceptual distance they feel between themselves and the speaker. Like something interfered with the text before it reached them.
I suppose a more extreme way of doing this might have been to have the speaker talk in another language, but too much of the information would have been lost on a predominantly english speaking audience. Also, I’m rather too unilingual to have been able to write it. So instead, I attacked the grammar of the speeches. You if you can rewrite a sentence using malapropism, repetition and start stuttering to make it harder considerably harder to process without as a reader or listener, without taking making it impossible. It can be done to a lesser degree than that, or to a much greater one, till the text feels like half ordered words with their meaning hidden half out of sight.
There are other ways you could experiment this as well, perhaps tacking semantically rather than syntactically, loading sentences with double meanings and incongruous ideas in order to slow parsing speed.
4 – Letting go of determinism
One of the common features of all these structural destabilisations is pulling things out of context, breaking the rules of where words and lines have to be situated. In the final section of Portents, I tried to bring all these different distortions of structure together, and deprioritise context even further, by leaving the order in which lines were spoken up to chance.
I wrote down pages individual, isolated lines – some of them lines of dialogue culled from the original text, some of them quotes from other sources, some of them lines grammatically scrambled using an algorithm called a Markov chain*, and then allowed the performers to read from them in an order of their choosing. Because there could be no predicting what was said moment by moment, all meaning had to emerge slowly, from the lines taken together, en masse.
Although this sounds like giving up an awful lot of control over the text, there are a lot of different ways you can try to manipulate what meaning emerges. The most obvious one is deciding what lines to throw into the mix – different ingredients result in different effects. You can also try adding back in subtler overarching structures. I split the lines the performers could choose from into five different sections, and then had them start selecting from section one, and slowly move all the way down to section five, at their own pace (an idea mostly cribbed from Terry Riley’s in C).
As with everything before, the randomness lives within rules, and the rules can be changed and rewritten. I think, more than anything, you just have to experiment with it. Tearing the structure out of a text can be a really liberating thing. You just have to let yourself enjoy it.
*Markov chain text generators take a piece of source text and calculate the probability of one letter coming after a short string of others across the source. They then use these probabilities to generate a new text. They tend to produce text that retains the tone of the source, but often makes no logical sense at sentence level. In any case, they’re a lot of fun to play with. This is the one I used when writing Portents.
Portents runs at The Space from 26 February – 2 March 2019. Book tickets here