Why I write the Introduction last

The White Rabbit put on his spectacles. “Where shall I begin, please your Majesty?” he asked. “Begin at the beginning,” the King said gravely, “and go on till you come to the end: then stop.”

– Lewis Carroll, Alice’s Adventures in Wonderland

Don’t listen to the King.

At least, not when writing a scientific paper. There are, to be sure, good times to start at the beginning and continue to the end: telling your child a bedtime story, making flan, singing a song from a treacly 1960s musical.* But writing isn’t one of those times. At least, I don’t write my papers that way; it doesn’t work for me.

That’s actually a slight exaggeration. Once, a long time ago, I wrote a paper by starting with the first line of the Introduction and continuing until I came to the last line of the Discussion. It was this paper, and unfortunately, it worked. I say “unfortunately”, because that was the first paper I ever wrote on my own, and it came out smoothly and easily in a single weekend from start to finish – which set me up in exactly the wrong way for the next paper, and the one after that. Every single other time, I’ve struggled; but I’ve struggled less since I realized that the King’s advice to Alice is terrible advice.

Here’s what I do now: I write the Introduction last. The reason is simple: the job of the Introduction is (obviously) to introduce the paper. But until I’ve written the Discussion, I don’t know what it is I’m introducing! Every piece of the Introduction, in a well-written paper, points to an important question that’s answered, or at least fleshed out, in the Discussion. So, I write the Discussion, and then craft an Introduction (not “the” Introduction) to point straight at it.

So, I don’t begin at the beginning. But I don’t begin with the Discussion, either. Here, for what it’s worth, is my usual sequence.

  • I start with the title page and the Acknowledgements. These are trivially easy**, but they suck my brain into thinking I’m making progress.
  • Then I take that momentum into the Results, which are a little harder, but not bad, as usually I’ve been playing with the data and I’m excited about it.
  • Next comes the Methods section – although that’s in large part a matter of mashing together bits I wrote while actually doing the work. (There’s no better time to write up methods, since while you’re actually doing them, you haven’t forgotten them yet.)
  • Now it’s time for the Discussion. Fairly often, writing the Discussion will suggest another analysis that’s worth doing, so I’ll circle back to Results and Methods as necessary.
  • With the Discussion put together, I can write the Introduction.
  • Finally, remaining front and back matter: an Abstract, a formatted Literature Cited, any Supplemental Materials, and so on. OK, so I guess I don’t quite “write the Introduction last”, but these bits feel more like typing than writing to me. (The Abstract is important! But with a manuscript completed, it’s pretty straightforward to put together.)

Iit took me a while – too long – to figure this out. Once I had, though, it made writing enormously easier, and in hindsight it was a blindingly obvious idea. I’ve been surprised, therefore, to discover that a remarkable fraction of my colleagues still push their mentees to begin at the beginning. This issue rears its head in my scientific writing course. I have my students working on drafts of thesis chapters (mostly), and I ask them to hand in a section at a time. Ideally, I’d have them hand in a Discussion quite early, and an Introduction only late in the semester. But this runs head-on into the students’ supervisors, who want to look at drafts of the same material, and (mostly) insist on starting with an Introduction. Some of my colleagues have an odd attitude to my writing course: they want my students to take it, but not to actually learn what I have to teach them. Sigh.

What about you? Introduction first, or Introduction last? How does your writing sequence differ from mine?

© Stephen Heard  May 17, 2022

Image: IMRaD format; own work, CC BY 4.0

This post is partly, and loosely, based on material from my guidebook for scientific writers, The Scientist’s Guide to Writing. But you probably guessed that.

*^I’m sorry, but it was in my head, and misery loves company.

**^Especially as I’ll almost always change the title later, when someone who is actually good at titles tells me how. For a recent example, see this post, which only got its title because a reader kindly pointed out the pitch-right-down-the-centre-of-the-plate I hadn’t even noticed…



3 thoughts on “Why I write the Introduction last

  1. Marco Mello

    Nice post, Stephen! It’s always amazing to see the many different writing strategies used by scientists. Surely there is no single way to do things, and I have changed strategies throughout my career. Currently, the first piece I write is the first paragraph of the discussion. That’s where I synthesize the main finding of the study. With that finding clearly stated, I have a story to tell. Then I proceed to writing the cast of characters: i.e., the main characters involved in the main plot of my story, such as the working hypothesis, the theoretical and operational variables, the study model, the study site, the analyses, etc. The next step is to develop the story arch and distribute the characters along it. Having organized the plot, I write the rest of the discussion. The next section is the results, which I fine-tune with the discussion: which results are needed to support the central argument? Which will become figures or tables in the main text? Which will go to the supplement? Then it’s time to move to the methods, so I can explain where those results came from. Finally, I write the introduction and, then, the abstract.

    Liked by 1 person

  2. John Pastor

    I think I read somewhere, sometime, that Paul Dirac (who was a fantastic writer) gave the same answer as Lewis Carroll’s King gave when asked the same question. I don’t know whether this is true or whether Dirac was just riffing on Alice in Wonderland (which he very well could have been). Either way, if you have never read anything by Dirac, pick up a copy of his Principles of Quantum Mechanics or Theory of General Relativity and read the first several paragraphs. You don’t have to be a physicist to appreciate how cleanly he gets to the point.


  3. Robin Heinen

    I guess for me it’s Methods, intro, results, discussion, intro revised, abstract. I guess we all have to find strategies that work and match our personalities and workflows. The most important thing is getting words on paper anyway. Revising the intro always makes me very conscious of story and flow, which I find important matter in my writing process. It works for me 😊.

    Liked by 1 person


Comment on this post:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.