A 'native English speaker' is not what you need! 

This blog is inspired by one of my 'pet hates' that editors (especially non-English speaking editors) so regularly come up with. In my experience, having advice from a 'native English speaker' is no guarantee to getting a well written manuscript. 

Most of the world’s scientists did not grow up speaking English. Yet, rightly or wrongly, English is the language in which science is currently written. So, if English isn’t your mother tongue, should you expect to receive help during peer review?

Should you expect to receive help with your English when you submit a manuscript?

Perhaps I should start from the outset by stating that English is my mother tongue, and that I have spent many hours correcting the language of colleagues for whom it wasn’t. However, most of these hours were spent when I was a postgraduate student or postdoc. I no longer think that correcting English is my role either as an editor, reviewer or as a supervisor.

I did not study English, and would be the first to admit that my grasp of grammar and syntax is by no means perfect. I have read enough correct English to know when something is incorrect, but that doesn’t mean that I know how to correct it. I have spent many long hours trying to decode what others have written, and in some cases this has involved me re-writing entire manuscripts. I still do this as a co-author, although I do remember asking one colleague to please send any further drafts in their native Spanish as it would be easier to translate than it would be to re-write.

The time component is at the crux of my reasoning why, as an editor, reviewer and supervisor, I will not provide an English language service. It is both a time consuming and an unsatisfying experience. We all have our own voice (see blog here), and correcting while maintaining other people’s voices is a painstaking task. There’s an entire profession that specialises in this (think translator). These days there are also services available from publishers to non-English authors to help them correct their English.

In addition, there turns out to be research suggesting why peer review is not the best way to improve writing (Shashok 2008). Peer review works much better at screening technical content than it does at improving the communication of that content. 

Given that I’m not going to correct your English, what remarks would you expect me to make when reviewing a poorly written manuscript?

No matter how you might hope that it isn’t true, a poorly written manuscript will not get a good review. If your reviewer is struggling to understand what you have written, this becomes the overall impression that they will likely pass onto the handling editor. I try to separate my difficulties with English from my review of the science. But this isn’t always possible. Frequently, a poorly written manuscript will mean that I won’t be able to understand why the research was undertaken, what was done or what it means. This is bound to impact the review negatively.

The more I struggle to read, the more negative the review becomes. I see this as inevitable. What shocks me is that some senior researchers consider it to be their right to submit poorly written manuscripts and have reviewers or editors correct them (if you don’t believe me, see here). Worse, I’ve received manuscripts that are co-authored by people I know are native English speakers, but are full of glaring mistakes that appear never to have been checked. For me, this violates the terms that all authors have approved the final manuscript.

So what will I do to help?

Normally, I will highlight poorly written text. If there are just a handful of places, I will report these as minor corrections and there is unlikely to be any comment about English. If there are many (>15) you will probably get your manuscript back with these highlights. If there are more than 20 in the abstract (and I have seen worse), then I will give up even highlighting, but still carry on making an effort to read the manuscript. There have, however, been times when I’ve not been able to finish.

Being a native English speaker, am I above having people complain about my English?

Sadly, no. I do get it wrong, and my English can often be improved. I’m always happy to receive help, and see it as a sign of how I can improve clarity of a manuscript. However, I hope that my manuscript are never so poorly written that a reviewer or editor cannot make sense of them. So it is a matter of degrees.

Do you have to have a ‘native English speaker’ check your work?

No. There are many people who are not native English speakers who write far better than I do. It’s a ‘pet hate’ of mine that reviewers and editors insist that a manuscript must be corrected by a ‘native English speaker’. I’ve seen so many very poorly written essays, theses and manuscript written by native English speakers that I know that having one correct your manuscript is unlikely to be of much help, unless they themselves are good writers.

Being a ‘native English speaker’ doesn’t automatically qualify you to write well, edit well or do any of the things that non-English speaking editors think that it does.

That non-English speaking editors often comment that a ‘native English speaker’ should read my text, simply underlines the problem that many editors themselves are incapable of knowing whether or not something is well written. 

So if not a native English speaker, who should do it?

Anyone you know who writes good English and is willing to help you. One option is the service offered by AuthorAID: http://www.authoraid.info/en/ Using AuthorAID, you can find a long term mentor who will help you with your English. You can read more about this approach here (Freeman & Robbins 2006).

Failing that, I’m afraid that the best route will be to pay for help. 

Congratulations to Zishan Ebrahim, MSc!

Ebrahim, Z.C. (2017) Amphibian conservation in an urban park: A spatial approach to quantifying threats to Anura on the Cape peninsula. MSc thesis. Stellenbosch University.

Zishan Ebrahim  has been a man of mystery in the lab for just over two years. Zish works as a GIS technician for South African National Parks at their research centre in Tokai, and so only came to the lab one day a week or so. You probably won't have noticed him in many other blog posts has he was usually at work. Nevertheless, Zishan is passionate about frogs, and was dedicated to his quest of how to use GIS in order to help with looking at a fine scale of conservation planning. We enjoyed having Zish in the lab for his thought and his insight into our discussions. 

Zish was one of a lot of scientists that graduated from Stellenbosch University today.

Are researchers writing more, is more better and who should be an author?

The concept of some kind of ‘literature inflation’ in science has interested me for a while. The idiom ‘publish or perish’ suggests that researchers will increase their output in order to obtain positions and promotions. And if a researcher’s productivity is measured by their publication output, shouldn't we all be writing more papers?

Similarly, if we should all be writing more, then wouldn't some people start publishing two (or more) papers, when one would be adequate? This idea of ‘salami slicing’ to inflate outputs would be an understandable strategy if researchers were all trying to increase their output.

A new study by Daniele Fanelli and Vincent Larivière (2017) has a new take on the above questions, by asking whether researchers are actually writing more papers now than they did 100 years ago. They used Web of Science to look for unique authors (more than half a million of them) and determine whether the first year of publication and the total number of publications resulted in an increasing trend.

 

The trend line for biology (bi) is very stable at around 5.5 publications whether you started publishing in 1900 or 2000 (note that earth science es and chemistry ch do both increase dramatically).  

However, they found that the number of collaborators is increasing, and so they adjusted publication rates for co-authorship. Their finding was then that there is no increasing trend in researchers publishing. This then poses another question. Who are these people that are publishing so much more than 5.5 publications, and are they unfeasibly prolific?

If we are all writing the same number of papers, are some authors unfeasibly prolific?

This was the question posed in a study that examined prolific authors in four fields of medicine (Wager et al. 2015). This publication piqued my interest as it turns out that they decided that researchers with more than 25 publications in a year were “unfeasibly prolific” as this would be the equivalent of “>1 publication per 10 working days”. Their angle was to suggest that publication fraud was likely, and that funders should be more circumspect when accepting researchers productivity as a metric. Looking back through the peer review of this article (which is a great aspect of many PeerJ articles), I’m astounded that only one reviewer questioned the premise that it’s unfeasible to author that number of papers in a year.

I have not published >25 papers in a year, but I know people who have and I do not question that (a) it is possible and (b) that they really are the authors. Firstly, the idea that prolific authors constrain their activity to “working days” is naïve. Most will be working throughout a normal weekend, and working in the early morning and late evening. A hallmark of a prolific author would be emails early in the morning and/or late at night. This gives you an indication of their working hours, and how they are struggling to keep up with correspondence on top of writing papers. Having authored >20 papers this year (2017), I can attest to the fact that it’s a lot of work and that it would not be unfeasible to have authored five more.

Authorship of a publication is often the result of several years of work. Thus, publications that I co-authored in 2017 frequently had research conducted in 2014 or earlier. For example, one of the publications, Measey et al (2017) is the product of aSCR work that started in 2009, funded in 2011 with fieldwork in 2012, and required the development of software for analysis by Ben Stevenson in 2015, before it could be completed and submitted. Thus, from my perspective, when I look at authoring a lot of publications it reflects the activity of the initial concept for the work, raising of money, conducting the field work or experiment, analysing the data and then writing it up (with the subsequent submission and peer review time). Thus, publications in 2017 result from a lot of work done for 3 or more years. 

Who should be an author?

There is an increasing number of journals that now give clear instructions on who should author a paper, and these have been formalised by the ICMJE. For some time, I have explained to MeaseyLab members that authors need to participate in at least three of the following five points before they can be considered for inclusion in the author line.

• initial conceptualisation for the work (hypothesis and/or question)
• raising of money (which often involves writing and submitting several research proposals)
• conducting the field work or experiment (the hard slog that many people will recognise)
• analysing the data (often much more difficult than anyone realises)
• writing it up (see lots of postings on this blog about the many requirements of writing)

It’s worth taking some time to think through each of these aspects of a piece of scientific work, especially when considering your own authorship. One point that students (in particular) often fail to recognise is that by the time they start on day 1, the first two points in this list are often completed. It’ll only be much later, when you have to raise your own money to conduct research, that you will appreciate all the work that went on before day 1.

My list is different from that of the ICMJE, but not mutually exclusive. They appear to have placed all of my first 4 categories into 1 and then added “final approval” and “accountability” as extra requirements. While I agree with the ICMJE list, I wouldn’t add “final approval” and “accountability” to my list above as these are journal requirements that all authors must meet for any publication. By the time authors arrive at this point, their name has already been included on the submitted manuscript throughout peer review.

Is writing a lot of papers a good strategy? 

This is a question of long standing, and one that you may find yourself asking at some point early on in your career. I'd suggest that the answer will be more about the sort of person that you are, over any strategy that you might consciously decide. If you tend toward perfectionism, this will likely result in fewer papers that (I hope) you'd consider to be of high quality. If on the other hand your desire were to finish projects and move on, you'd be more likely to tend toward more papers. 

Given that the 'best' personality type lies somewhere in the middle, you can decide for yourself whether you identify with one side more than the other. But which is the better strategy? Vincent Larivière & Rodrigo Costas (2016) tried to answer this question by considering how many papers unique authors wrote and seeing how this relates to their share of authoring a paper in the top 1% of cited papers. Their result showed clearly that for researchers in the life sciences (bottom graph below), writing a lot of papers was a good strategy if you started back in the 1980s. However, for those starting after 2009, the trend was reversed with those authors writing more papers less likely to have a smash hit paper (in the top 1% of cited papers). Maybe the time scale was too short to know. After all, if you started publishing in 2009 and had >20 papers by 2013 (the inflection point of the curve in top graph below) then you have been incredibly prolific (only 19 authors from what I can see). 

One aspect not considered Larivière & Costas  is that becoming known as a researcher who finishes work (resulting in a publication) is likely to make you more attractive to collaborators. Thus, publishing work is likely to get you invited to participate in more work. Obviously, quality plays a part in invitations to collaborative work too. Thus pulling the argument back to the centre ground. 

If you find yourself becoming pre-occupied about which is the best strategy for you, I'd suggest that you get back to finishing what you were writing before you got distracted!

If more is being published, will Impact Factors increase?

Yes. A number of years ago (October 2013), I asked David Green (Global Journals Publishing Director for Taylor & Francis Group) whether there was an inflation rate for Impact Factor of journals. He responded that it ran at around 5%, but then didn’t come up with a source for this information when I followed up.

Of course, it’s not just that more is being published, but the number of citations within every paper is increasing over time. Who and how should you cite? That's the subject of another blog.

When is it possible to retain your own voice when writing?

Should one use “I” or “we” when writing a scientific paper? Although there was a tradition not to use first-person pronouns when writing scientific articles, this has fallen by the wayside in recent years as the use of I or we makes writing simpler for writers and readers. Use “I” if you are the sole author (or in your thesis) and “we” for two or more authors.

You will regularly see “I” or “we” being used in the last paragraph of the introduction and the first paragraph of the discussion. It’s also quite common to use this in the methods section. The reason why these sections have “I” or “we” is quite obvious, as it allows you to place your own voice on actions and decisions. After all, it is your aim and hypothesis for your study, so you should own it. Using first-person pronouns also allow you to be more concise, and there are some great examples here where you can see what happens when you don’t use them.

The first use of “I” or “we” at the end of the introduction is powerful. Up to that point, the reader has read four or five paragraphs of logical arguments outlining the background information that makes up the reasoning for your study. Using the first-person pronoun now makes the reader sit up as they have you communicating directly with them what you aimed to do. But that power is lost with repetition. So if you were to continuously use it throughout the methodology, the reader may get tired of reading a string of statements stating that you did this, then you did that, and so on. The power of using “I” or “we” is to use it sparingly to boost clarity.

Does writing “I” or “we” help people to understand who you are?

Actually, all of your writing will do this. As long as you are using your own words (and not plagiarising), your writing is likely to be unique to you and hence recognisable as you. It never ceases to amaze me that it is possible to convey the same information in so many different ways. Your way will likely change with time and experience, but remain yours throughout your writing life. If you keep writing, people will be able to recognise your distinct style.

Meeting of astrophysics and frogs

It was great to meet up with physicists Ninan Sajeeth Philip (St. Thomas College) and Ashish Mahabal (CalTech). Although our 'big data' is not nearly as big as theirs, it'll be great to explore how our areas of research overlap. 

Looking forward to getting your great minds working on our little frog problems!