The Science of Discworld

The Science of	Discworld
Ebury Press, London 1999		now available in paperback
left to right:	Jack Cohen Terry Pratchett Ian Stewart	on the occasion of Terry Pratchett's honorary D. Litt at Warwick University

Terry Pratchett, Ian Stewart, and Jack Cohen
Ebury Press, London 1999
ISBN 0-09-186515-8, £14.99 (hardback)

One person has had a major influence on my life as a writer and mathematician: Jack Cohen. And Terry Pratchett. Two --- two people have had a major influence on my life as a writer and mathematician: Jack Cohen and Terry Pratchett. And Marty Golubitsky. Three --- three people have had a major influence on my life as a writer and mathematician: Jack Cohen, Terry Pratchett, and Martin Golubitsky. Oh, and Tim Poston. Look, I didn't expect the Spanish Inquisition... Let's try it again.

Most of my work, both mathematics and writing, has been done in collaboration. A lot of my colleagues have tried writing books with other people, and found it a demoralizing and wearying experience. You put the semicolons in, your coauthor edits them out again, you put the semicolons back, your coauthor edits them out again... The book goes round and round in circles and never converges; in the words of Tracy Kidder in Soul of a New Machine, you don't get the product 'out the door'. (Funny how Americans don't understand the word 'of.) I've been fortunate --- no, let me take the credit: I've been cunning enough to choose my coauthors sensibly. My experience of writing collaborations has been one of unalloyed pleasure. I put the semicolons in, my coauthor edits a few of them out again, and I agree it's better that way.

This being an SF event, I won't mention Tim Poston or Marty Golubitsky again, except to say that they are both mathematicians, that my collaboration with Tim gave birth to the epic Pitman Lecture Notes Taylor Expansions and Catastrophes and the widely acclaimed monograph Catastrophe Theory and Its Applications, while my collaboration with Marty produced the classic Singularities and Groups in Bifurcation Theory Volume II (I'm sure you all own a copy --- and in case you wonder, I wasn't involved in writing Volume I) and Fearful Symmetry, which is pop science. In contrast, I will mention Jack and Terry again, because (a) I've just done so, and (b) the organisers of Novacon 29 asked me to write about my collaboration with them, and I might be struggling if I didn't refer to them by name occasionally.

As with so much in life (especially lateral suspension) a great deal that has happened to me would never have happened were it not for a fateful telephone call from Jack Cohen. It was so fateful that I now realise I can't actually place the year when it happened --- so, as a special service to Novacon 29, I shall stop writing and delve into the desk drawer with all the old diaries in a daring act of historical research, and report my findings in a moment...

...OK, well --- the exercise proved quite valuable. For example, I found a whole stack of blank labels for floppy discs which I mislaid about six years ago. I also found eleven pocket diaries, including the 1991 one that went into the washing-machine by mistake. (Let me explain: for many years I kept my diary in the back pocket of my trousers, and sometimes I forgot to check before putting them in the ali baba to be washed. My wife always used to buy me two diaries for Christmas: one to start the year, and the other to replace the first when it got washed by mistake.) There was a lot of useful historical information --- the gas man called on 17 March 1987, for instance, and somebody (or some thing?) called ARCHILLA warranted three entries on 11, 12 and 13 December 1989. I think he was a physicist --- but with a name like that she could be a villainess in Blake's Seven, or a world-girdling entity from the Lensman series.

As well as these two still-amazingly-well-preserved items of Stewart family history, and countless others equally intriguing which I shall reserve for another occasion, I found the first diary entry relating to Jack, on 3 April 1990. It reads, in its entirety:

That 'or so' shows that Jack's punctuality and awareness of time then were no better than they have ever been, before or since, and adds credibility to my belief that this was the occasion of First Contact --- the alternative being that I failed to record that event in the diary, which I find inconceivable. The 'Cohen' adds conviction: forever after, the entry reads just 'Jack'. But my main point here is that no subsequent 'Jack' entry ever includes 'or so'. Obviously I learned fast.

Beside the above entry is the cryptic remark 'Exit 1', in what is probably different ink, and almost certainly bears no relation to Jack's visit. I haven't the foggiest idea what it refers to. Exit 1 of what?

I have always told the story of our first meeting as Jack turning up at my office door unannounced, saying: "I'm Jack Cohen, you're Ian Stewart, it's time we met." He has always sworn that he rang me up a few days before to arrange to drop by. Having on many occasions witnessed his formidable ability to wield the telephone at the drop of a hat, I am convinced that his recollection is the correct one. At any rate, we both agree we went to a pub --- the Coventry Cross in Kenilworth --- and discovered that despite him being a biologist and me being a mathematician, we discovered we had a lot in common --- among them chaos, SF, and an interest in philosophical issues in science.

Out of that first meeting eventually (1994) came The Collapse of Chaos. My diaries record numerous meetings --- in 1990 they are 18 May, 30 May, 21 June, 28 June; then on 29 June I gave a talk at Birmingham University on Chaos and Biology which I know Jack must have arranged; 13 July, 3 August, 3 October (most of the gap was when I went to Kyoto); something weird on one or more of the 29, 30, 31 October ('OK = JACK 1' it says and each of those dates has an entry 'OK'); and then --- hope I'm not boring you, but now we get to the point of all this --- and then, 'NOVACON a.m. Jack' on 9, 10, 11 November. Right. Jack persuaded me to come to Novacon: clearly it was Novacon 20, and my collection of Novacon badges bears this out. In short: the evidence of the diary agrees with what I now remember of our initial interaction. I have many times told people that it took us four years to write The Collapse of Chaos (having long ago forgotten whether that was true) and this seems to be correct --- as long as you include the period between submitting the manuscript and getting the beast into print.

Well, that was fun, boys and girls, and I have now pinned down the year of our first meeting, which I've been wanting to do for ages, so thanks to Novacon for triggering an uncharacteristic burst of action.

At Novacon 20, Jack introduced me to Terry Pratchett, and the three of us had lunch in the Birmingham airport hotel where the con was being held. Little did I know that... but I get ahead of myself.

Implicit in those pub lunch discussions was the idea that Jack and I might end up writing a book together, and within a few months we'd pretty much decided to do so. There didn't seem to be any rush, and we developed a method of collaboration that we both found congenial. One week, I would go over to Jack's house for most of the day, and we would discuss various issues in science (and out of it). We would repair to a pub for lunch, and continue the discussion. The next week, Jack would come over to Coventry, and we'd do much the same thing. This alternation of venues went on for about two years, and might still be going on to this day, with The Collapse of Chaos still existing only in embryonic pre-verbal form, but then the fifth person out of the four I mentioned at the start made a major impact on my life. This was the redoubtable John Brockman, a literary agent and New York jew who had decided to corner the market in popular science authors. (He succeeded, and scares publishers silly to this day.) John was visiting London and he invited me down to discuss the possibility of signing up. This panicked my then publisher (no names, but an Antarctic bird springs to mind) who phoned me out of the blue to offer an advance on whatever my next book might be, the sum being 25 times as big as any previous advance I'd ever had. (Bear in mind that before that my typical advance was £0, but the last couple had been in the four-figure bracket, and I don't mean £10.00.) The offer was roughly 2.5 to 3 times my annual salary at that time. (Bear in mind that I was working in a university, not the City, too.)

The publisher's pre-emptive strike was seriously off-target, though, because once people started talking serious money I knew I was out of my depth and immediately concluded that I needed an agent. By this point Jack and I already had quite a bit of Collapse down on paper, and it became clear that Jack would need an agent too, and the same one as me. Needing good advice, we (no, Jack --- of course) phoned Terry, and said that the publisher had offered £x, where x is large, but the agent had held out hopes of £y, where y was considerably larger than x. Terry said that in his experience it was wise to accept large amounts of money, and even wiser to accept considerably larger ones, and pointed out that if the publisher was offering us £x before we had an agent, then they would still offer us £x after we had an agent, so all we were risking was the agent's percentage. So we signed up with the agent, and duly received something like £(x+y)/2, which the mathematically minded among you will recognise as being better than £x but not as good as £y.

Then we suddenly had a contract, which committed us to finishing Collapse by the agreed deadline, so a certain sense of urgency accompanied subsequent pub lunches. We evolved a way of working together which basically we still use, which goes like this. First, we discuss the general area of the book, agree some themes, and Jack produces an outline --- say half a page per chapter. We tinker with that until we're both happy, and then Jack expands the agreed version into about four pages per chapter. This version ranges from fully written discussions of macaques or mycorrhiza to plaintive pleas akin to: "Ian, write something here about chaos in pendulums". Then I take over, and using Jack's version as an informal basis, I turn it into about 20 pages per chapter. (One of the rules here is that I can ignore anything Jack has said if I want to.) By that point most of the material is in close to final form --- checked for spelling and grammar, reasonably organised --- but it may include "Jack, we need a section here about the role of dung in evolution". All such gaps are filled, and finally we have the first close-to-final beta-test version.

Now we both read the manuscript, and scrawl all over it. "What's this bit doing HERE?" "Hang on, we've already done dinosaur sex two chapters ago." "A haggis will only be stable at a unique height up the hill if the sides of the hill are concave, not convex." "I don't like this example of genetic assimilation: can't we do it with Darwin's finches and some snails?" And so on. Then we get together (an entire day, or two, works best) and work through the two annotated piles of paper, transferring anything that we agree about on to a third, previously virgin copy of the manuscript. We shout a lot at this stage. Then I go away and implement all the changes. Then we take another run at it. Eventually we reach a point where both of us are happy: we deem the book finished, and send it in to the publisher.

At that point their editor does a number on it, anything from pointing out obscurities to suggesting we choose a new main theme and rearrange six chapters into five in totally different places in the book, and the same sort of process happens again.

That's how Jack and I wrote Collapse, and it's pretty much how we wrote Figments of Reality. With one exception. When we'd finished Collapse, the agent wanted another book from us, and we sent a proposal for an early version of Figments. Unfortunately there were several 'mind' books in the pipeline from other people, such as Daniel Dennett, so the idea wasn't terribly popular with publishers. So I went off and wrote Nature's Numbers and various other things, until one day Jack told me he'd sold Figments to Cambridge University Press. Moreover, he'd reworked the outline into a much improved form. So for that book, the initial outline was much more a one-person effort.

We like to have fun with our writing, and we like to push out a few boundaries. Leaving aside the whole of The Science of Discworld, the Zarathustrans in Collapse and Figments are the most extreme examples of this tendency --- weird aliens, obsessed with the number 8 (Pratchett influence here) whose role is to allow us to indulge in wild speculation without having to say so explicitly. We realise that by doing this we may put off some potential readers, but what the heck. When the reviewer of Figments in Nature (Henry Gee) started babbling about Z-shirts, we knew we'd done the right thing. And in a sense the Zarathustrans previewed the approach that led to The Science of Discworld. Which spent 13 consecutive weeks in the top ten of the Sunday Times nonfiction bestseller list, so there.

Throughout this time, Terry Pratchett played a substantial role behind the scenes. We felt that in many ways he was the person best qualified to advise us on alternative views of modern science, because Discworld was itself an alternative view on just about everything, and Terry was Wise on Matters Philosphickal. He read both Collapse and Figments in manuscript, and had a major effect on both. Jack and I visited Terry for a day's discussion about three or four times a year, asking him for thoughts about our next book, and offering our own suggestions for his. Out of this informal interaction grew a sense that the three of us were on much the same wavelength. So when The Physics of Star Trek opened publishers' eyes to a new kind of media tie-in popular science, it was only a matter of time before we came up with the idea of a pop science book hung around Discworld. The title The Science of Discworld seemed unavoidable.

Terry immediately recognised the main obstacle: there is no science in Discworld. It runs on magic and narrative imperative. The eighth son of the eighth son has to become a wizard, even if she is a girl and the midwife has made a mistake (Equal Rites). Yes, there's a lot of science-related stuff: Greebo in Maskerade; there are three states for a cat in a box --- alive, dead, and absolutely bloody furious. But there's no actual science. One morning, sitting in the Mathematics Institute common room, Jack and I had a flash of inspiration, which we quickly passed on: "Terry --- if there isn't any science in Discworld, you'll just have to put some there!" And so a new genre was born: the fact/fantasy fusion. Terry wrote a 30,000 word novelette in which the wizards of Unseen University unwittingly set in train the Roundworld Project, a magical containment zone inside which magic does not work. Our own dear universe unfolds before the wizards' eyes, and being wizards, they interfere. Jack and I added what in effect were Very Big Footnotes on the science behind all the wizardly shenanigans.

That was a three-author collaboration, and I can tell you it's a bloody complicated thing to do. Terry couldn't write his bit until he knew what we intended to put in our bit; and we couldn't write our bit until we knew what he intended to put in his bit. To resolve the impasse, he asked Jack and I to give him a shopping list of scientific events and topics that we'd feel happy commenting on. Then he wrote about 15,000 words of the story, and we took a look and started adding the Very Big Footnotes. Once Terry had seen how that was working out, he carried on with the Discworld story. Then Jack and I spent about four months writing 80,000 words of science chapters. Along the way, the three of us met occasionally to work out where we'd got and plan ahead, and the book evolved as we went along. Terry made suggestions about the science parts, we made suggestions for the Discworld parts, and we put the jigsaw together. Those of you who've seen the book (if not, shame on you, go buy ten copies now!) will know that the chapters alternate between Discworld and Roundworld (though we got Terry to write the final Roundworld chapter summing up the message of the book, because he seemed to know what the message ought to be, and Jack and I didn't). One of the great things that Terry brings to a collaboration is perspective. From the beginning he had his eyes open for the big issues that would be coming up, and how they would affect the development of the story. Jack brings original ideas and a lot of lateral thinking. Finally, Terry administered expert corsetry and a sprinkling of fairy-dust, going over the entire manuscript and making minor but crucial adjustments.

We finished writing SoD, as we referred to it on e-mail, on 4 January 1999. (There may well be a sequel, though. Already we know what we want to put in it --- and this time, we know that the format and the collaboration work.) One side-effect was that Warwick University awarded Terry an honorary Doctor of Letters --- so he awarded Jack and I honorary degrees at Unseen University, an event that made it to the hallowed pages of Nature. And the less hallowed pages of the Independent and the Times Higher (Education Supplement). Since then, Jack and I have been working on (and finished, except for some corsetry and a sprinkling of fairy-dust) an SF novel, Wheelers. It is about a boy who has an intuitive feel for the minds of animals, and incredibly ancient aliens on Jupiter who throw comets around. One comes this way... and the rest of the story is about trying to stop it. Oh, and Jack embedded it all in a deep philosophical theme.

We wrote that in a rather different way. It sort of began during a long car trip to Plymouth, by the end of which we had a plot and some characters. (It had been sort of hanging about in potentia for a while.) I still have some pages of hotel notepaper with these scrawled on them. It developed largely through conversations, and virtually all of the actual writing was done by me, section by section, with no four-page versions from Jack. A lot of the ideas, though, are his, or evolved jointly. Moreover, it was Jack who persuaded his SF agent Ashley Grayson to look at an outline and some sample chapters, and before we knew what had happened, Ashley sold the book to Time Warner. Then it took six months to agree the contract. Ho hum.

So what have I learned, gained, or lost from these collaborations? I've learned a lot --- a lot of offbeat science from Jack, a lot about writing from Terry. I've learned that the trick is to tell a story. I've learned that the biggest obstacle to collaborative writing --- apart from choosing the wrong collaborator, which I never do --- is today's computer technology. (Terry and I transferred files for SoD by e-mail. However, I work on a Mac in WriteNow and he uses a PC with Word Perfect. This led to a curious sequence of file translations, with Claris Works as an intermediary at my end and god knows what at his end. My main advice to would-be collaborators is: start by all buying the same computer systems.) I've gained two friends --- indeed, many others, and a lot of new acquaintances. I've gained a new career or three. And I've lost any inhibitions I might have had about breaking out of the traditional academic world and having some fun.