I feel somewhat less than human right now. It's my fault: That's what you get for living a life of excess. But before you condemn me for indulging my PhD grant in the choicest beverages and 'all-nighter' parties, let me explain. It's an excess of thesis composition, not beer, that's precipitated my current state of stupor.

I defy anyone to argue that there's an easy way to write a thesis. OK, so you can be organised and efficient about it, but as far as I'm concerned, a PhD isn't the top qualification without good reason. And right now it feels like the 'good reason' is the very process of constructing the book.

After 3 years of hard work, and with the prospect of a salary only a couple of months away, this PhD candidate is feeling a little radical toward his thesis. So, I've decided to break all the rules. I'm tackling the whole thing in one go, a sort of shotgun approach to thesis writing. Whilst perhaps unorthodox, I'm finding this method extremely satisfying. The basic idea is to touch everything--every result, every paper--once and only once. You pick it up and you slot it right in where it needs to go. Although this doesn't work all the time, it does let you clear those piles of paper relatively quickly.

Start with your Figures and Tables. Taking them each in turn, describe exactly what you've got. You can use these descriptions as your figure and table legends, and a bit later on as the basis for your results sections. Use a similar approach to tackle the literature. Distil out what you've picked up from each paper--a short sentence is often all you need--and slot it into place, or places. It seems that a little of this spadework goes a long way. Just lump relevant bits of text together and sort them out when you eventually get back to attack that section. It's crude but effective.

Lest you find yourself procrastinating over how to subdivide each lump into smaller and smaller sections, please just get writing. You'll like yourself much more when you've started on this journey. Kick-off by tackling something familiar. If you've written a paper, so much the better. Paste chunks of it straight in where you need it. I soon forgot any ideals of starting to write it all from scratch. Anything I've previously written--even if unread by anyone except me--is now precious. I just chop it up, sort out the chaff, and use it, even if I only manage to save a single sentence. Sometimes you'll find a sentence written in the unhurried days of your first year that's a pure gem of insight.

This approach demands that you have eight or 10 files open simultaneously on your computer screen, so it helps if you have a fast computer with auto-save switched on. If you don't, you'll soon forget which files you haven't recently saved. With words and pictures flying into so many different files, I make a point of remembering to backup my work. The thought of losing even one day's work made up of so many unconnected bits is just too scary.

Even though I'm only halfway through writing my thesis, I already know the cardinal sin for all budding thesis writers, including those attempting the shotgun approach. It's too much, too soon. Detail, that is. It's so very easy to get sidetracked by trivia. Figures, especially, have a way of sucking you in. Before you know what's happened you've effectively wasted 4 or 5 hours of good writing time faffing around trying to get one figure nicely lined up and labelled. Now, the counterargument goes 'you have to do it sometime, so surely it's all to the good'. I disagree for two reasons. One, why seek to attain perfect figures that you or your boss might later amend or scrub altogether? Two, the more you kid yourself that what you're doing is useful work, the harder it'll be to get writing properly. OK, so you may experience a premature sense of triumph once you've finally converted your carefully selected results into Figures and Tables, but they are just the beginning, and the easiest bit. Soon comes the dawning realisation that the real meat of your thesis is not your results themselves, it's what you say about them.

And when it comes to writing, the biggest potential pitfall of the shotgun writing method is never finishing anything. Don't think the shotgun approach means writing quickly. That's a recipe for shoddy work. You might spend an hour or more on one quite short section, but the point is to finish what you start, even if it's only a single paragraph. This way you don't have to waste time rereading what you only half-wrote 2 weeks ago. Naturally, to make this work you have to have faith in what you've already written.

This approach does, however, have the advantage of enabling you to keep up your interest in the whole endeavour. If you don't spend ages stuck on one section, you won't lose inspiration. Just finish off the paragraph and move on to a totally different area.

If you are methodical, you'll eventually reach a point when you have gathered together an incomplete but critical mass of text in each section. When you recognise this point, it instantly becomes time to kill off your thesis chapter by chapter. Warn your boss well in advance of the impending deluge of first drafts.

This shotgun strategy might not suit everyone. Some will find the long wait until the first finished chapter just too freaky. But if you are up for it, you might just find that the long, dark rite of passage called your thesis write up isn't quite the endurance test you first thought.