How many words in a chapter – convention and purpose


The how-many-words-in-a-chapter question plagues book and thesis writers alike.

When supervisors are asked about chapter word length, many of us hedge. It depends, we unhelpfully say, on how you have organised your results. The number of chapters and therefore words follow the way you construct your argument. You might have three cases and a cross case analysis, or three or four big themes each of which contain results and discussion, or you might have a quant and a qual chapter and then one that brings them together. Or something else entirely depending on your research and your analysis. So think about your argument in relation to the overall word count.

While the it’s-about-the-argument answer is perhaps unsatisfactory, it does point to the importance of not thinking about word count in isolation. The way you organise a text is first and foremost about what you want to say, rather than how you break it up. But how you break it up is an important question.

One way to begin to think about the chapter word count is to consider textual balance. Patrick Dunleavy helpfully points to the problems of

  • frontloading – that’s when you spend a lot of words getting to your own research. So there’s a very long literature chapter, what other people have said. Followed by far too much about methodology and methods.


  • backloading – that’s when you talk mainly about your own research and don’t say enough about the basis on which you got there. So there’s pretty skimpy literature work and not enough said about your research for it to be trustworthy.

Dunleavy’s load metaphor reminds us of the problems of frontloading and backloading a vehicle – they are likely to tip over if they are unbalanced. Avoiding a text that doesn’t stand up to scrutiny means checking for balance between setting up your work and reporting your actual analysis. And watching out in particular for the word count allocated to literatures and design.

So given balance, one possible answer to the word length question for thesis writers is pragmatic – long division. You simply work out the number of chapters you need (particularly how many for your results and discussions) and divide your total word allowance by the number of chapters. Indeed, this is what a lot of thesis writers do.  And they end up with chapters of somewhere between 12-17k words, depending on the number of Results chapters they have. And they do then have a pretty standard thesis.

However, you may want to do something other than this. Then you are in the land of unspoken conventions. Unspoken conventions are often what supervisors draw on when they say that a chapter seems a bit long or a bit short. Yes it may be about the content. But there may well also be an internalised supervisory measuring stick about how many words constitutes an acceptable chapter. It’s as well not to ignore this tacit knowledge as your examiner may well have the same views.

But I do see theses where the results and discussion are divided into shorter-than-usual chapters (5-7 k words). This chapter length shift often differentiates the results from the introductory sections – not only signalling here is my research but also my research needs to be organised explicitly and differently.  I’ve thought about this and have decided on a less common word count which fits my analysis and argument.  A  big caveat here – as usual, because the short chapters mildly break unspoken conventions it’s as well to be confident that examiners will “get” what you are doing and perhaps you need to explain it.

However, the just-divide it up approach doesn’t always work so well for books. Not many people write about book chapter lengths, what they should be and how they vary between disciplines and between genres. So there’s not much out there for me or you to go on. It’s not hard to find online material saying that chapters in novels vary from one page to an entire book, and that the average novel chapter word count is either 2-4k or 3-5k depending where you look. But there isn’t much about academic chapters. At least not that I could put my hands on – I didn’t do a systematic search.

I did find references to chapter length in Pamela Haag’s book Revise. Haag is a professional editor and has worked a lot on helping academic writers turn their research into books. Haag suggests that scholarly books follow conventions including that

Chapter and subsection lengths are roughly consistent. A hundred page chapter that follows a twenty five page one, which I’ve seen, breaks the conventions of the scholarly book. 


Each element within the scholarly book conforms to its own conventions. A preface for example is unlike the introduction: it’s shorter and sets an evocative, distinctive and less detailed tone, through a vivid moment, story or characters. (p. 84)

Haag also mentions 80 pages as long, I suspect that this is about 20 k words double spaced. She refers to fifteen chapters as the short and thematically concise and my guess is that she is probably referring to chapters that are about 4-5 k words. (My personal rough guide to monograph chapters in my discipline is somewhere around the  8-10k mark although I have written books with shorter chapters. And there is some blog/column advice on-line that says the same.)

Haag notes that although scholars often submit manuscripts to presses with long chapters, publishers almost always want shorter ones. She describes these as more thematically concise, thoughtful and shorter (p 85).

Editor Haag directs people not to think solely about book and disciplinary conventions but to consider them in concert with what they want their writing to do. Playing with word length she says can help achieve better focus and momentum – she says It’s hard to be analytically nimble if your chapter units are too long – as with an ocean liner, it’s laborious to change direction when you have large lumbering chapters. Smaller chapters and shorter sections may offer the flexibility needed to shift deftly among subjects and topics. (p 86)

So for Haag, form always follows function. If you’ve got a manuscript whose major contribution if its new archival findings, easily narrated, then you might be fine with somewhat longer chapters. If there are several moving parts, then  consider shorter chapter lengths that can refine key parts of your argument. (p 86)

Haag says that there are author considerations too. Some people feel more comfortable writing shorter thematically focused chapters, while others relish the lengthy. Choose a chapter length that suits you. Nevertheless, Haag says – and contrary to those theses where chapter lengths are varied – don’t forget convention. She reminds, Write chapters roughly equivalent in length. Consistency is the hallmark of disciplined editorial and revision work. (p 87) In other words, once you have decided what your book chapter lengths are, stick with them.

So on the question of how long should a chapter be, the answer actually is it depends. It depends on how you, the writer, want to construct your argument and how you want it to read, bearing in mind what conventions that you decide to meet or break.

All of which brings me to the question of the reader. Next post, word length and the reader.