30 Mar 2010 9:46, CY

I’m quite surprised that this practice exists. One would expect, as you mentioned, the first author i.e. the PhD / Post-Doc student to write his own article, or at least the first draft of it. However, with non-native English users, this task becomes trickier. Anyhow, wouldn’t the non-native English users have at least passed the IELTS/TOEFL exams and can therefore write sufficiently well already?

If so, the argument that they can’t write in English with proper grammar and punctuation does not hold!

30 Mar 2010 10:18, Mirjam

In my opinion the situation is very clear: learning how to write a scientific article in proper English is an essential part of your training as a PhD/postdoc and the group leader is responsible for giving you the opportunity to practice. Therefore, the first author should at least write a first draft, also of the response to the referees. If their English is poor they should also take some courses to improve their English. If the professor always does everything him/herself because (s)he can do it much faster then (s)he should not pretend to be training young scientists (in this case they probably just see PhDs and postdocs as cheap measuring slaves)

30 Mar 2010 10:57, Jacopo Bertolotti

I can understand the professor writing the paper only in particular cases like when you have a very large amount of people who worked on the project (e.g. for a nuclear physics paper) and no real “first author” can be identified or if you are really in hurry to publish.
Yet both cases should be an exception, not the rule. In 99.9% of cases delaying submission by one month will not hurt you too much so you can give the post-doc, PhD or even to an undergraduate student the task to write the paper.
Beside writing a paper is not just the matter of knowledge about english grammar: writing is the moment when you finally come to summarize your work, when you decide what it is important/interesting and what is just a detail that can be skipped. In my experience a paper usually contains not even half of the work that was done and choosing what to include is as important as being able to write in an intelligible way.
And if the student/post-doc english skill is not good who cares? As long as he/she is able to write something there is always the possibility to correct mistakes and/or to rephrase some paragraph. Anyway every paper is going to be rewritten a huge amount of times so who really cares if one of them is devoted to grammar corrections?

As a concluding remark I would like to ask why, if the professor thinks his PhDs/post-docs are so untrustworthy that they cannot even write a first draft, he got them in the first place.

30 Mar 2010 10:59, Karen

I was a doc student and I gained little writing skills during my doc time. So I think I can say something about that.

My opinion is that, as a doc student, if you have a professor who are willing to teach you writing skills in the right way , well, you are lucky. However, if your professor does not want to “waste ” time, you should manage to learn to write in other ways which you should find and find really soon.

However, I still can not write well, although, I graduated. Bad for me.

30 Mar 2010 13:43, Jean-Luc Lebrun

The native language of many of the researchers and postdocs where I work is not English. Speaking as trainer in scientific writing skills, I really enjoy working in a small country which systematically trains its scientists in academic writing, considering it a core skill. As CY commented, a scientific writing course should not dwell on but go beyond English grammar to address issues specific to writing a scientific paper: writing strategy in a publish or perish world, reviewer viewpoints, crafting of the title and abstract, wrapping the structure around the title, etc… This is something professors and PIs know how to do well, and they would not mind helping the writers so long as rewriting did not take so much of their time! That is why a writing class is essential. It cuts down the number of mistakes leading to additional rewrites! By the way, I am working with a European university on a semi-automatic scientific paper evaluation tool (software) based on 8 years of training on this topic. The preliminary results are VERY PROMISING even though the tool still relies on human assistance and interpretation, being based on Natural Language Processing.

30 Mar 2010 13:46, Ad Lagendijk

I will quote from my book Survival Guide for Scientists

The first author writes the manuscript. Other authors comment and give suggestions. Never should a senior scientist write part of the paper himself. He should keep on discussing and suggesting modifications until he can live with the manuscript. A senior scientist, not being the first author, who writes part of the manuscript himself will seriously depress and indeed insult the first author

-
And that is still my opinion

30 Mar 2010 16:59, Phil

“Ii must be horrible to have to wait for months for a PhD student to finish an article, knowing that you yourself would have done the job in one or two weeks.”

Ehm, that might happen in the opposite way. I wrote my first paper draft in january and had to wait until the end of february to get input from my supervisor. Hope that now after my revised edition everything goes faster. I mean, I work in a group which has some experience concerning ultrafast phenomena (yes, pun intended).

30 Mar 2010 20:41, Klaas Wynne

I agree with just about everything that’s been said so far. I’d like to add though that good writing has little to do with one’s mastery of English (well, up to a point). Good writing means being able to formulate your thoughts and write them down in a coherent manner. A good story can be told well (or badly) in any language. Your ‘Prof’ should be able to iron out the English mistakes. I have written the majority of a paper as a Prof twice but in those circumstances I put my name first as per Ad’s rule above. Otherwise it should be the main responsibility of the student or postdoc, definitely.

30 Mar 2010 22:44, Ad Lagendijk

I have answered this by sending you a private email. But indeed I made the mistake. Sorry.

31 Mar 2010 1:32, David Stern

I don’t understand this. As a PhD student you need to submit a dissertation. Shouldn’t you have a large responsibility for writing it, even if the papers represent joint work? Or is it common in the natural sciences to submit papers that you did experiments for but didn’t write as a dissertation?

31 Mar 2010 7:07, Mirjam

@Klaas: ideally the first author is the one who did most of the experimental/theoretical/simulation work *and* is the one who wrote the paper. If a student/postdoc did all of this scientific work but did not manage to write the paper, then my gut feeling is that (s)he still needs to be the first author. At least, when I look at author lists, I always assume that the first author is the one who did most of the actual scientific work (maybe that is a mistake). Obviously, the last author should have been the main supervisor and as such is responsible for the final product (and can thus be expected to whip papers into a publishable shape).

31 Mar 2010 8:27, Klaas Wynne

@Mirjam: You are absolutely right. The two cases I mentioned, I did the theory and that was the bulk of the paper. By the way, the rule that “the first author did most of the work” only holds in “small” physical sciences such as chemistry, optical physics, etc. It doesn’t even hold in all physics: not in particle physics, not even in some solid state physics areas.

31 Mar 2010 10:33, David Stern

In Economics the standard is alphabetical ordering irrespective of contribution. I don’t stick to that though and mainly list names according to size of contribution, largest first.

31 Mar 2010 11:01, Kaan Ozturk

I think it actually IS the job of advisors to teach their students how to write, and how to use proper English. My advisor was exemplary in that sense. I’m grateful that he patiently corrected many, many mistakes I made (I’m a nonnative speaker).

Being a PhD advisor is about teaching how to do research. This includes teaching how to communicate, too. Teaching is a slow process, and the focus should not be on efficiency. If professors can write a paper faster, they can also do the research faster. Why do they bother with taking students then?

I think this professor is not concerned with nurturing and raising young scientists. Apparently he sees his students and postdocs as hired labor only.

1 Apr 2010 9:00, ‌Bahar

I’m not sure whether my second comment will be published or not, but I still think the professor of the above mentioned post kills self confidence of his group members by not only letting them to write themselves but also discouraging them and pointing out their weaknesses about their English skills.

25 May 2010 18:44, Otto Muskens

For group leaders there is clearly a tension between delivering high-impact research at the highest speed possible and training a starting PhD student by guiding him/her go through the process step by step. The group leader with the highest impact is not always the best for his students.

Even worse, I have heard stories about groups where PhD students are not allowed to publish their thesis work at all if it they do not reach sufficiently high impact to the standards of the group leader, as this would damage the reputation of the group.

12 Aug 2010 2:07, Wolfgang

I wrote my first article during PhD and spent quite a time (and lots of thoughts) writing it, only to learn afterwards that my supervisor would take it and rewrite each and everything new, although he assured me that this was not for quality reasons but only because this was his personal style, doing this in each and every occasion. This was very frustrating! The other thing was, that even when you prepared a fully formatted manuscript using the journals template it may took several years (!) until things get published, because my supervisor was unsure about changing things back and forth. This week some paragraph was quite ok, next week it had to be rewritten eventually, another week left it was not that good, but acceptable! Thus, my conclusion would be…there should be only one guy to write the paper and to submit it as corresponding author and to do all the tasks related to it…either this would be the supervisor, with no work left up to any one else…or the first author, with absolutely no influence by the supervisor. Moreover the supervisor should not even be mentioned as author, but rather in the acknowledgement paragraph, if he did not write the paper or did not do any experiment himself.

29 Nov 2011 21:20, sahane

joining this discussion quite late, but this article disturbed me to the extent that i had to comment i found it incredibly unprofessional on the part of your prof to write articles himself only. As an academic, my assumption has always been that senior academics have a duty to raise the next academic generation by giving them a chance to shine. It is selfish to use postdocs merely as labour without letting them to develop. Postdocs need to write, and publish on their own, not only to reach scientific excellence through practice but also so they can get jobs! In the unis I have worked at, postdocs need to show initiative by proposing, carrying out, and reporting original contributions all on their own. And PIs will only contribute as guides. If not, there is no obligation for your PI to add his / her name on the paper. Presumably this person came to where he is by not being co-authors on papers but had ideas of his own? why does not he allow you to do that? I would suggest, although I dont know if you are in the same situation, to publish on your own or with other members of your group. Scientific publishing does not have a rule that your prof should be on the paper. If you have such institutional rules, leave that place.Believe me, there are much better places.

29 Jan 2012 14:37, Otto Muskens

In my experience it is always the student producing the first draft, for various educational reasons but probably equally for the fact that the senior PI does not have time to start from scratch nor the knowledge about the details of the experiment. After that there will usually at least 4-5 iterations where the senior corrects or rewrites part of the manuscript. This process also depends on the self-criticism of the PhD student and his/her capacity of writing good papers. For myself, I never give away a manuscript to a co-author before having at least done 4 full iterations of the manuscript myself.