Image: You know what you’re walking into. © Gary J. Wood via flicrk.com, CC BY-SA 2.0
This is a joint post (argument and rejoinder) from Steve Heard and Simon Leather. You can find it on either of their blogs.
Should a paper title tell you what the paper is about? Yes, but not the way Simon thinks.
Steve opens with – A few weeks ago, Simon Leather blogged about one of his writing pet peeves: “titles of papers that give you no clue as to what the paper is about”. I read this with great interest, for a couple of reasons – first, Simon is consistently thoughtful; and second, I’m terrible at titles and need to learn as much about good ones as I can! Much to my surprise, I found myself disagreeing strongly, and Simon was kind enough to engage with me in this joint post.
I don’t mean that I disagree that a paper’s title should tell you what it’s about. That’s exactly what a good title does! My disagreement is, I think, more interesting. Simon offered some examples of titles he scored as failing his tell-you-what-it’s-about criterion, and some he scored as passing. I found myself scoring those examples exactly the opposite way: the ones that failed for him, succeeded for me; and vice versa.
What gives? Well, most likely, I’m just wrong. Simon has a couple of years more experience than me in science, has published many more papers than I have, and has significantly more editorial experience. But “oh, I guess I’m just wrong” doesn’t make a very interesting blog post; so I’m going to work through my thinking here.
Here are two titles from Simon’s disliked list:*
And here’s one from Simon’s liked list:
They’re on exactly opposite lists for me. Simon dislikes the first one because “it takes until line 9 of the Abstract before you find out it’s about an insect herbivore, [and] until the Introduction to find out which species” (he dislikes the second for the same reason). Simon likes the third because “you know exactly what this paper is all about”. I think this is all wrong (sorry, Simon). Since I’ve been writing about scientific writing as storytelling lately, let me put it this way. Simon would like to know that the paper is “about” an insect herbivore, or “about” the British Swallowtail Butterfly. But to me, that isn’t what it means to say a paper is “about” something – the study species is character, not plot. Would you say that The Old Man and the Sea is “about” Santiago, or that Slaughterhouse-Five is “about” Billy Pilgrim? Well, maybe in casual conversation, but not in a book review you were getting graded on.
I want a paper’s title to tell me about its plot. By “plot”, I mean the questions the authors ask, and the way the experiments (or observations, or models) answer them. That’s what a paper is “about” – the way The Old Man and the Sea is about a man’s struggle with his catch, his failing career, and his mortality (but I should stop before I venture further into literary criticism for which I am poorly qualified). The “unified framework” and “seasonal life-history” titles tell me what questions the papers ask and answer. It’s true that they don’t tell me which characters (species) they answer them with, but that’s not what I’m looking for in my first pass at a title. And the swallowtail title? It tells me nothing other than that the paper has to do with conservation of the swallowtail. It mentions “questions”, but doesn’t say what they are; and it mentions “challenges” and “opportunities”, but these remain similarly shrouded.
A title that announces what species a paper is about doesn’t grab me, unless I already work on the species (or a similar one). Who would pick up the swallowtail paper, except someone already interested in swallowtails or similar butterflies? Is that the only audience the authors want? What if the paper asks questions with implications for the conservation of mammals, or birds, or orchids? Those audiences won’t be engaged. With a title that announces what question a paper is about (and if possible, what the answer is), authors can recruit a broader audience.** And readers can find out what species the question is asked with (and ponder whether the answer applies more broadly) at their leisure.
Should a paper title tell you what the paper is about? Yes, but not the way Steve thinks.
Simon replies – I totally see where Steve is coming from with his point about plots and storylines and his references to Slaughterhouse-Five and the The Old Man and the Sea (although I could of course, somewhat tongue in cheek, riposte with a whole slew of titles such as Nicholas Nickleby, Martin Chuzzlewit, Oliver Twist and David Copperfield to name just a few.***) I think that I come at paper titles from two aspects of my academic profile. First as an applied entomologist, I really do want to know if the paper is about the particular species or related group of species that I am working on – so referring back to Steve’s footnote about Tables of Contents (or even Current Contents)****, both of which I remember – yes, the title needs to be highly specific. Second, this is a debate I have had with conservation biologists working with vertebrate animals.
I am, as my Twitter handle indicates, an entomologist, and at the risk of being seen as narrowly partisan and parochial, means that I, and all other invertebrate zoologists, work on, until evidence is presented otherwise, the animals most relevant to ecology in general J. A paper on the movement ecology of zebras, for example, is unlikely to give me any insight into the migratory behaviour of aphids (of which there are more species than there are mammals), whereas an insect migration paper might give a mammal ecologist something to think about (incidentally I just realised that this helps Steve’s argument, in that an unwitting mammalogist might read an opaquely titled paper about insects). As a PhD student, when I first got interested in life history traits, I noticed that many vertebrate zoologists were publishing papers addressing concepts that were already well known to entomologists (e.g. Tinkle, 1969*****), but not referring to those studies; so much so that I made rather a point of referring to vertebrate papers in my thesis whenever possible J
And in the spirit of Monty Python’s Spanish Inquisition sketch, third, (yes I know I said two things initially) is the point I made in my blog post about ‘scientific fashion’ and what we now call ‘click bait headlines’ (my example of one of my own titles in that post underlines this very neatly). On the other hand, as Steve and other commentators have pointed out, there is a cost to both download and citation rates when titles of papers are very specific and lengthy (Letchford et al., 2015), which is surely why high impact and more general journals encourage the titles I abhor, and Steve favours. A new pet hate of mine, and something favoured by high impact general ecology journals, are titles with question marks: it is obvious that the answer is always going to be yes!
A thought (oops, now a fourth point – the Spanish Inquisition strikes again) that occurred to me as I was writing this and beginning to feel that I was succumbing to Steve’s cogent and compelling arguments, has to do with science communication. We are being encouraged (some would say forced) to become ever more open access so that in theory the whole world can read our outpourings (although I suspect that most proponents of Open Access are more concerned with their ability to instantly access data, than for the general public to access the ever increasing number of academic papers). If this is the case, then surely, rather than use titles that are said to increase scientific citation rates, we should perhaps be using very explicit titles that will enable the general public to know what to expect?
To wrap up: Steve admits to being terrible at titles, and to Simon being a more experienced author and editor than he is. And yet Simon admits that Steve’s arguments had him (ever so briefly) questioning his own. So we’d like to turn this over to you. Where do you stand on titles, character, and plot? Please tell us in the Replies.
© Stephen Heard and Simon Leather August 27, 2019
*^I decided that I wouldn’t actually read any of the papers. I wanted to react to titles as I would if I encountered them in a Table of Contents (anybody remember those?) or in a Google Scholar alert.
**^The obvious compromise is a title that reveals both of those things. I like that sort of title, although the cost is they can get long, and there’s empirical data suggesting that they reduce citation rates.
***^Steve can’t help himself, and footnotes Simon’s half of the post (chutzpah!) to point out that saying that David Copperfield is a novel about David Copperfield is true, but not particular enlightening. He doubles down on his argument, therefore, while wondering what the Dickens was up with that particular novelist’s penchant for character-based titles.
****^I felt that as this is a joint effort with Steve, parenthetical interjections were essential 🙂
*****^Incidentally, the title of that paper fits Steve’s point under his second – that the ideal paper title reveals both character and plot, although this one does it even better: “Grazing as a conservation management approach leads to a reduction in spider species richness and abundance in acidophilous steppic grasslands on andesite bedrock”.
Letchford, A., Moat, H.S. & Preis, T. (2015) The advantage of short paper titles. Royal Society Open Science, 2, 150266.
Tinkle, D.W. (1969) The concept of reproductive effort and its relation to the evolution of life histories of lizards. American Naturalist, 103, 501-516.