Six ways to get it wrong, and four steps to get it right
Image credit: CC0 Public Domain.
When you want to get your paper or chapter read and appreciated by a wide audience, adopted for courses, and hopefully cited by great authors in good journals — in short, when you want to ‘sell’ your writing to colleagues — titles can play a key role. It is obvious too that a title is how you ‘brand’ your text, how you attract readers.
Most people find articles, chapters and papers now via Google Scholar or other online sources, for instance, by searching for key or ‘trigger’ words. The search algorithms used by Google and other search engines assign extra importance to words appearing in a title, compared with an abstract, or the body text of a paper. So if your article title includes key words that other academics and researchers in your field are likely to search for, then your text is much more likely to show up high on their search returns. For readers more generally, (such as business or public policy folk, media people and others interested in your field) using some widely used but subject-specific vocabulary in your titles will tend to improve the visibility of your work.
When readers first see a search return for your paper or chapter, it is usually just the title itself (for instance, on a journal or book contents page) or at best in a ‘snippet’ form, showing the title and perhaps a couple of lines of text. If the title looks dull, routine, like a hundred others, or if it seems enigmatic or obscure, then the odds are strong that people will pass it by and never even read the abstract in a journal, or try to find out what the chapter says on Google Books. By contrast, if the title looks interesting and relevant for their interests, potential readers will next click through to download the abstract or look for some accessible window onto your book chapter. If the these materials are also interesting and relevant, and the paper is open access, then potential readers will download it. If the paper or chapter sits behind a paywall, practitioner or general readers generally give up immediately. They either forget your text straightaway or try to retain for their purposes just whatever they gleaned from the abstract or preview. Only academic readers (with big libraries) will bear the time costs of trying to find the paper or chapter on their library systems, so as to download the full text. It’s a final (huge) sweat nowadays for an academic to leave their study and go search for a book chapter in their library’s stacks — so naturally they economize on the effort unless the title and any Web-visible materials strongly suggest relevance and value for their needs.
Even after other researchers have found and read your text, titles remain important. If they liked your piece they may enter it into a bibliographic system or save it as a PDF in a PC archive or on Mendeley or similar systems. Very rarely they might make notes on it. Now the issue is: will they cite your work in their own professional publications, often written months or years later, by which time they have scanned lots of other publication details and their memory of your work is dim and vague? To re-find it they must search their PDF library, or if they haven’t saved it, recall that the paper exists out there in the ether. In both circumstances a great, informative title for an article or chapter maximizes the likelihood that they correctly remember enough to re-discover what they are looking for. If your paper is ‘grey’ literature — such as a working paper, research paper, conference presentation, or a report for an outside body — remembering your name and something about the title will both be crucial. Without these cues your work will sit undisturbed on other scholars’ PDF libraries, or languish unread among hundreds of millions of other documents on the Web.
Yet, over and over again, academics and (perhaps even more) PhD students choose titles for their journal articles, chapters in books and research or working papers that are almost completely uninformative. Clearly many authors believe that
there is some kind of professional obligation on them as academics to be deliberately and carefully obscure, to choose titles that convey as little as possible to potential readers about what their text says; and
they will be penalized or viewed as ‘racey’, reckless, or over-claiming if they do anything like give a clear picture of their argument or findings in the article or chapter title.
How to design a completely uninformative title (irony warning)
Rather than batter my head against a brick wall on this subject yet again, I’ve decided to write this section throughout in ironic mode, as if I was going with the grain of existing practice. So here I advise you on how to get to the ultimate ineffective title for academic work, one that utterly fails to communicate what it is about, let alone ‘sell’, the ideas involved. Hopefully seeing things in this extreme way will illuminate what’s wrong with the over-caution and lack of imagination that afflicts most of us, most of the time. (Yes, I’ve done everything below here myself at some time).
A completely ineffective title should systematically repel and put off potential readers, to ensure that as few as possible are motivated to look beyond the title to the abstract, or the full text. If anyone has by mischance persevered and read the abstract or saved a PDF, the title should deprive them of any memorable cues to help them recall the paper or chapter in context when it comes to citing sources or influences in their own work.
The really useless title must be as similar as possible to a thousand others, or so obscure that its meaning completely evades readers. It could also miscue or mis-direct readers, for instance, appearing as if it is about a completely different topic, or undertaken in a completely different discipline. Including a high quotient of words that no one else is ever likely to use (or search for) can be especially helpful for a useless title. The top five most popular versions are:
A ‘cute’ title using ‘ordinary language’ words with a clear meaning, but taken radically out of context. The essence of a cute title is that the author should know what it means, and as few other people as possible. This is great for academic snobbery — it says to potential readers: ‘I introduce my work in such esoteric ways, because I am so much cleverer than you’. It also ensures that anyone interested in the topic covered would be very unlikely to input these words into a search engine. For instance, an article about not teaching thinking skills in high school education could be entitled: ‘Burning down the pagoda in order to roast the pork’. (This actually quotes an apt analogy from Edward de Bono: but someone who’d not read the source already would never, ever think that these words relate to the topic of high school curricula). However, a cute, understandable title may be a bit memorable for the few searchers who ever find it, if it is quirky or distinctive like this.
A ‘cute’ title that is completely obscure. This is a variant of (1) where even the language the author includes in the title is incomprehensible. My favorite example is a 2004 report by an eminent group of professors at the British Academy, about the role of the humanities and social sciences in promoting economic growth and social development. They chose as a title: ‘That compleat complement of riches’. This is a vague-sounding quotation from the eighteenth century philosopher David Hume, which could be about anything, and with the added advantage of using an archaic English spelling that no one has used for 250 years. The report duly became very little known.
An ultra-vague, vacuous, completely conventional, or wholly formal title, preferably one that could mean almost anything. To be fully obscure here it is vital to pick vocabulary that is as general or unspecific as possible and is capable of multiple possible meanings. It is especially effective to be ambiguous about what field of interest is covered, or what discipline the paper is in. For example: ‘Power and society’ could be about many things in sociology or political science; equally it could be about generating electricity and associated technology. In the same vein, ‘Accounting for ministers’ could be about politicians running government departments in parliamentary countries; or alternatively, a manual for vicars or priests doing their income tax returns.
An empty box title. This is by far the most popular academic approach. Its advantage is that it can look as if the author is being pretty specific, while actually telling readers nothing about what findings have been made, or what line of argument is being followed. For example: ‘Regional development in eastern Uganda, 1975-95′ gives you a location, a date range and a topic. But the key message is still: ‘I have done some work in this box (topic area), and I have some findings. But I’m not going to give you any clues at all about what they are’. Most book contents pages incidentally are nested box titles, all equally opaque as to what argument is being made in a chapter.
The look-alike, empty box title, is a variant of 4 above, where the paper title has lower memorability by closely resembling hundreds of others, and is devoid of any distinguishing or memorable features of its own. For instance: ‘John Stuart Mill on Education’ tells us what author and sub-field you are covering, but that’s it. Is the discipline you are working in philosophy, or history, or education? Combining box titles with formal/vacuous wording also keeps the potential scope really broad. So: ‘Key features of capitalism’, leaves us with a blank sheet to guess about what you have done, in which discipline.
The interrogative title, which must always end with a question mark. Again vagueness is an asset in seeking obscurity. For example: ‘Can democracies compete?’ is suitably non-specific. Compete with whom or what? And in what sphere? At other times an interrogative title may regrettably give away some clues to what you are actually discussing, or glimpses of the slant you might have taken on it. But you are at least completely disguising your answer. For example: ‘Was Jane Austen ever in love?’ Well, was she, or wasn’t she? Many academics write articles and even blogposts with interogative titles in the mistaken belief that they are ‘teasing’ readers, to motivate them to read further. This actually cuts little ice, because jaded expert readers have seen the trick so many times before. As I think Microsoft used to say in their advertising several years ago, the key problem with interrogative titles is that: ‘Questions are everywhere, but answers are few’. Lots of us can frame perfectly decent questions. But far, far fewer of us can generate the interesting, valuable or novel answers that researchers and practitioners are looking for.
Four steps to getting a better title
It’s not hard to improve. The first step is to look seriously, critically and comparatively at a range of possible alternatives. Make a resolution not to be too vague, general, or convention-bound in choosing what words to use. Try and think things through from a reader’s point of view: How will this wording be interpreted by someone scanning on Google Scholar? What will attract them to click through to the abstract?
And what about this title would make a potential citer of my paper find it easily in their PDF library or Mendeley files, or recall it to mind months or years after they first read it? Always makes crystal clear too (from your choice of concepts and vocabulary) what academic discipline you are operating in. I recommend generating a minimum of 10 possible titles and printing them out on a sheet of paper for careful consideration. Compare these alternatives with each other and see if recombining words from different titles might work better. Type your possible titles as search terms into Google Scholar or subject-specific databases and see what existing work comes up. Is this the right company you want to keep?
The second step is to look at whether your title words are picked up in the abstract of the the article or chapter, and in the internal sub-headings. It’s a good sign if the title, abstract and sub-headings all use consistent, linking, meshing or nesting concepts and vocabulary. It’s a very bad sign if the title words and concepts don’t recur at all in the abstract and sub-heads, especially if these other elements use different, rival or non-synonymous concepts or wording from the title.
A third step is to consider using a full narrative title, one that makes completely clear what your argument, conclusions or findings are. Narrative titles take practice to write well. And they rarely work at the level of whole-book or whole-report titles. But they are often very effective for articles and chapters. One of my current best cited journal articles (written with colleagues) is ‘New Public Management is Dead — Long Live Digital Era Governance’. Here the title sums up the whole argument of the paper, and triggers two specific topics (‘New Public Management’ or NPM, and ‘Digital Era Governance’ or DEG). Since NPM has a huge literature whereas DEG was a brand new concept that we’d just invented, it was very helpful to link them together strongly in the title, and to subtly try to put DEG on a par with NPM. The provocative ordinary language terms here (‘dead’ and ‘long live’) are memorable. And their association with the passing of a crown from one monarch to the next helps make clear our highly controversial argument that DEG has displaced NPM as the dominant form of public management in advanced democracies. The title’s advantages don’t stop there either. By summarizing the argument so completely the title lends itself to mini-quotation and citation even by the many conventional public management folk who strongly disagree with it. It is also perfect for people to cite who haven’t even read the paper (from the rest of what they say). So I’ve lost count of the number of times that other authors have said something like: ‘Some commentators have argued unconvincingly that NPM is ‘dead’ (Dunleavy et al, 2006)’. Well, we can’t all agree, and in the meantime a cite is a cite.
Now perhaps some readers will already feel outside their comfort zone. But do give a full narrative heading a try before you reject it. This approach does nott have to be as deliberately provocative as my example. The essence of a narrative heading is that it tries to tell the full story of your paper or chapter. It seeks to summarize the substance or core value-added of your argument, to capture ‘your takeaway’ (as a management consultant might say) — that is, the one key point that you want to stay in readers’ minds a week after they have read your paper and forgotten most of its details. Notice too that a narrative title does not have to be claiming a lot: if yours is a modest paper, then fit the wording closely to the paper.
Even if you reject a full narrative heading, if academic susceptibilities or disciplinary conservatism mean that you cannot quite bring yourself to be so explicit, there is still a fourth step to try . This compromise solution is to at least provide some narrative cues in your title, some helpful hints or signs for readers about the conclusions you have reached or the line of argument you are making. If you have an empty box or an interogative title already, then ask, how can I make this more informative? So: ‘For Mill, should giving women the vote precede or come after implementing ungendered education?’ does not quite tell us your answer. It hints at a potential difficulty, but it does not yet tell us how you think that Mill addressed it.
To put these ideas in a wider context, you might find it helpful to read parts of my book: Patrick Dunleavy, ‘Authoring a PhD’ (Palgrave, 2003). See also useful material on the LSE’s Impact blog and on Twitter @Write4Research.