26

I am currently reading a paper in my area which is very poorly written in my opinion. It is hard to understand the bigger picture of the paper. Now, I am just a student and my opinions might not be of that value but it is really difficult to understand the paper. It appears that everything in the paper is only implicitly stated. I took help of other better researchers (postdocs, profs) but they also face the difficulty of getting it properly. The authors of the paper are established researchers and the results are really great. But, I really want to tell them how bad the paper is so that when they write their next paper they also think about the readers. I wonder if their advisees also turn out like them. I just want to write a mail to them and tell them how a paper is written but I do not know if they will take it as constructive criticism. I am really mad.

17
  • 19
    What constitutes "acceptable feedback" has a huge cultural aspect to it, so, you might want to state the country this is in. What you can always do is to ask specific clarification questions, instead of phrasing it as feedback. Mar 29, 2023 at 7:46
  • 9
    And what if your opinion about quality of their writing is simply wrong? For instance, in my area, math, it is common to skip on some of the details that a professional mathematician would find easy to fill in, while for a young researcher it would be a very difficult task. This does not make such papers poorly written. Mar 29, 2023 at 14:26
  • 43
    @MoisheKohan I disagree: I think many math papers are poorly written, for exactly this reason. Mar 29, 2023 at 19:51
  • 22
    This famous snippet from Mark Twain is worth considering: "When I was a boy of fourteen, my father was so ignorant I could hardly stand to have the old man around. But when I got to be twenty-one, I was astonished at how much he had learned in seven years." Dunning-Kruger comes for us all, especially as we advance past novice.
    – Tom
    Mar 30, 2023 at 2:55
  • 38
    Suggest waiting until you're less mad before you do anything. Mar 30, 2023 at 6:21

9 Answers 9

16

As a reader

Its worth asking what you stand to gain or lose by emailing the authors or not doing so. A published paper is not going to change, so you have no chance that you will change the paper at this point. It is also not likely that the authors thank you for your completely out of the blue and unsolicited feedback on something outside of the science, so you are not going to impress anyone. There is some (but very small) chance that your feedback means that their writing will improve in subsequent papers, but for that to be the case, your feedback would need to go beyond just "I can't understand your writing", and contain constructive feedback, including specific suggestions for what particular parts of the writers style could be improved, and how they might go about doing that. Take it from someone who has provided feedback on a lot of writing - this is hard and long work. The only other benefit is that you will feel like you have done "something" about the thing that is making you mad, even if that something doesn't actually achieve anything.

On the other hand, you are likely to associate yourself with negative emotions in the minds of the authors - the best you can hope for is that they forget you. They don't need to be deliberately vindictive or retaliatory - this can be entirely subconscious. This can be true irrespective of whether they take your advice on board (or even thank you with the conscious part of their brain). This is just human nature. This could be particularly harmful if they are big names in the field.

This is not to say you should never disagree with a big name, or provide them with feedback at appropriate times. Good scientists will welcome feedback particularly on the science as it will lead to avenues for new investigation. But what is achieve must always be balanced against the downsides.

As a reviewer

As a reviewer of the paper as part of the peer review process things are slightly different. Generally, journals will provide fairly specific guidance on how to feedback on guidance. Usually you are not supposed to feedback on "style", but only those things that genuinely hamper your ability to understand the paper. You might say:

Assessing the work was difficult because I found the way it was written difficult to follow. For example, the meaning of 'Lorem ipsum dolor sit amet, consectetur adipiscing elit.' was unclear.

You should also comment where things are genuinely ambiguous – for example:

The sentence 'Cras suscipit massa ipsum, ac tempus massa viverra et.' could be read as meaning either 'Nullam nec varius odio.' or 'Mauris sed ante posuere, euismod urna id, rhoncus erat' and which is not clear from the context.

If there are many typos in the text, point this out if it makes the text difficult to follow.

Be aware that it is not acceptable anymore to tell the authors to have a native speaker proofread the text. This is because:

  • a) You have no idea if a native speaker has already proofread the text.
  • b) Many non-native speakers are actually better at technical writing in English than many native English speakers.

In both cases it is never acceptable to criticize the authors. Make sure always that your are criticizing the paper, not the authors. The difference might seem meaningless, but it is vitally important.

5
  • 1
    Some of these are matters of personal opinion. I have both commented on matters of style and received reports with helpful feedback on my own style. Similarly, the claim that "it is not acceptable any more to tell the authors to have their text proof read" is a statement of personal opinion, not a statement of fact (just like many other claims of the form "... is not acceptable any more"). If anything, the problem is that the authors might not have access to a competent enough speaker, not the irrelevant fact that some non-native speakers may be better writers than some native speakers. Mar 30, 2023 at 14:55
  • 1
    @AdamPřenosil You misquote. It's not "tell the authors to have their text proof read" it's "tell the authors to have a native speaker proofread" (em. added). -- While it was once very common to make such requests to authors with "foreign sounding" names, it's now generally considered gauche, due to the implications. Its still acceptable for a reviewer to suggest a paper needs further proofreading, what's not acceptable is to suggest/demand that a "native speaker" do it.
    – R.M.
    Mar 30, 2023 at 15:22
  • 1
    @adamprenosil it's not just about considering it gauche, it's explicitly against the guidelines at many journals, and may get your comment removed/rephrased by the editor, or your review being disregarded. Mar 30, 2023 at 18:20
  • 2
    @IanSudbery I've seen guidelines asking, essentially, to be tolerant towards non-native speakers, but I have not seen guidelines stating that referees should not suggest that the text be proof-read by a native speaker. I'm not questioning your statement, but if you recall a journal which has guidelines of this sort, I'd be curious to take a look at them for the sake of my own curiosity. (Btw I agree that doing this isn't a great idea because presumably the authors would have already done so if they could. I just disagree that the two reasons that you gave have much of anything do with this.) Mar 30, 2023 at 22:37
  • 1
    @R.M. Sorry, my bad! I changed the quote to fit the character limit, but then I should have also removed the quotes. In my experience, the phrase "it is now generally considered gauche/outdated/refuted" often performs a function similar to the phrase "clearly/obviously" in mathematics. What follows is often not in fact a generally held opinion. The "implications" of the phrase are debatable. Of course, people looking for "implications" can find them anywhere, but in a normal, charitable reading it simply means "have someone highly competent in the target variety of English take a look at it". Mar 30, 2023 at 22:44
29

Historical context: It's by no means unusual for important papers to be difficult to understand. It's rare that the seminal papers in a subject are the place to go to learn that subject. For example, few people choose Einstein's original papers as their place to go to learn relativity. Few people go to Feynman's original paper on path integrals to learn that. Few people go to Dirac's original paper to learn about the Dirac equation. A lot of people will read these articles for personal reasons or historical reasons. But they are not likely to be the text for those learning those subjects.

There are a few possible contributing reasons for this, some of them these:

  • The first explanation may arise from the process the author used to get there. This is often not the easiest to understand explanation.
  • The journal probably has some harsh limits on word count, diagrams, tables, etc.
  • There has not been a process of refining notation for ease of reading.
  • There may be language issues if the authors are not native speakers of the language they are publishing in.

So, there is quite a lot of room for writing pedagogical versions of research papers. If you have talent for writing clear prose that gets the idea across, maybe there is a future for you in writing textbooks. Or at least review articles that wind up on every grad student's desk for the next thirty years. I recall a Physics Reports issue on Yang Mills theory that nearly every grad student in particle physics had a copy of.

So a poorly written paper that yet has interesting results might be viewed as an opportunity to produce a better written, possibly expanded, version for a journal somewhat more focused on instruction. You might even get a collab with one or more of the authors of the article that confused you.

That is, turn an annoyance into an opportunity. Much like an oyster turns a grain of sand into a pearl.

5
  • 5
    Yes, some papers are hard to read and understand because they are about difficult topics or have a lot of assumed knowledge. I was going to suggest: try to write a version of the paper that makes it understandable to students at your level. When you are happy with that it might be appropriate to share the "translation" with the author though I would have someone in the field who I know read it first.
    – Elin
    Mar 29, 2023 at 15:22
  • 3
    If it is a recent article, you could try reaching out to the authors for clarification. Not all authors answer these queries, but some do. Mar 29, 2023 at 16:21
  • 3
    To add to this -- sometimes the first paper written on a subject is breaking new ground on terminology or other ways of thinking about that subject. Over time a community will develop better ways to think about or describe work. But, it's hard to be the first breaking ground, and thus papers doing so are often much harder to read.
    – Nathan S.
    Mar 30, 2023 at 5:09
  • 5
    "For example, few people choose Einstein's original papers as their place to go to learn relativity." It might be worth to give it a try, though: I read the first part of "Zur Elektrodynamik bewegter Körper" in the final year of high school, and remember it as really well written and easy to follow.
    – Heinzi
    Mar 30, 2023 at 5:57
  • There's a difference between the first paper ever on a subject, and a paper that is just poorly written, yet contains good science. This question is about the latter.
    – masher
    Apr 2, 2023 at 2:57
21

From How to Win Friends and Influence People:

  • The easiest way to make enemies is to criticize them.
  • Criticizing people puts them on the defensive. They will dig in their heels and rationalize their actions.
    • Just consider, how often do you see people respond to criticism with "Yeah, I was wrong. I'll do better next time"? Then compare vs. the number of times people respond with "I did nothing wrong, but you did X, Y, and Z".
    • Abraham Lincoln is one of history's greatest leaders. In 1863, during the American Civil War, the South launched an invasion of the North that was repelled at the Battle of Gettysburg. Because the Southern general (Robert E. Lee) had appeared hitherto invincible, this was a major victory for the North, but the Northern general (George G. Meade) failed to pursue and destroy Lee's army. Lincoln was upset, and wrote a critical letter to Meade that you can read here. The most remarkable thing about this letter however isn't how restrained it is, but rather that Lincoln never sent it. He knew he wasn't there, he didn't see the battle and death firsthand, or know what it's like to trudge across soggy ground under threat of enemy fire, and he knew all these things didn't apply to Meade. Therefore he shouldn't criticize.
  • If you absolutely must criticize, praise first. Wax lyrical about how the results are really great, and then say you wished the paper were better written so you could've understood it quicker.
6
  • I'm not sure criticizing a manuscript as part of the peer-review process is like sending a letter to a general still bloody from battle, but the sentiment of your post is quite correct: if you cannot be constructive in your criticism, better let it go. Mar 29, 2023 at 14:15
  • 11
    @ZeroTheHero I don't think the question was about criticism in peer review, it seems to be about unsolicited criticism post-publication.
    – Bryan Krause
    Mar 29, 2023 at 17:36
  • 1
    I actually don't like the military language that is often used on both sides of peer review, it seems unnecessary to me.
    – Tom
    Mar 29, 2023 at 19:04
  • 2
    >"If you absolutely must criticize, praise first. Wax lyrical about how the results are really great, and then say you wished the paper were better written so you could've understood it quicker." There's a name for that: a "compliment sandwich." Compliment, criticism, compliment.
    – cgb5436
    Mar 30, 2023 at 22:16
  • 2
    Which should really be called a "criticism sandwich"... after all, we don't call bread-cheese-bread a "bread sandwich". Mar 31, 2023 at 18:58
6

Papers are not written for students but for experts in the field. It is not uncommon for a bachelor thesis (depending on subject of course) to work out all the details in a paper and show that the student managed to work through it in a couple of month. This working out things, helps you to slowly become an expert in your chosen field. So long story short, my answer is: absolutely no.

What as a student of course you can do is point out errors (if there are) in the publication.

4
  • 2
    It's a problem for papers to be written only for "experts in the field". They should be written so as to allow a somewhat wider audience, not just, say, the 5-10 people who follow that line of research. Of course there space constraints etc., but effort must be made to keep things as clear as possible, avoid logical jumps which outsiders may easily miss, etc.
    – einpoklum
    Apr 1, 2023 at 8:30
  • @einpoklum I do not think it is a problem and every now and then there is a 100 page introduction article to the topic. You cant add that to every paper.
    – lalala
    Apr 1, 2023 at 16:14
  • 1
    I believe you're presenting a false dichotomy. The idea isn't to include part of that 100 page introduction article; the point is to make the core of your writing more accessible.
    – einpoklum
    Apr 1, 2023 at 19:12
  • @lalala: what has a 100 page introduction article to do with this? You can make the writing of your paper more accessible while still not writing a general long detailed introduction.
    – user111388
    Apr 1, 2023 at 22:59
2

As a student, you're unlikely to be the intended audience of the paper. They may in fact be thinking of their readers; it's just not you. From the context, I'm assuming that the paper is a technical, academic one written by professors of some sort. The authors' goal was probably to announce results in their field to their peers, maybe stake out some territory, improve their odds of tenure or funding, and so on. For publication in a research journal, the authors are presumably writing for an audience of researchers in the same field who already have a substantial background in the subject and are already familiar enough with the area to automatically connect the dots themselves. It could also just be a bad paper; I don't know you or the authors or even the paper involved.

Regardless, I don't think anything would be served by "tell[ing] them how a paper is written." To be honest, I would respond to a unsolicited, angry message like that by summarily throwing it in the garbage, regardless of whether it was from a student or an established professor or a Nobel prize winner. You probably would too, and it's reasonable to do so. It's not even harsh but honest criticism; it's just being angry that someone didn't write the paper that you wanted them to write.

If you like the ideas of the paper but think their presentation is too opaque, you could contact the authors and ask for clarification directly or suggest they write a lighter survey paper. You could even write it yourself. If you do in fact know how a paper should be written, polishing up someone else's badly-phrased good idea for a different audience can be a valuable thing; that's kind of what textbooks are.

3
  • If stated really bx a Nobel price winner (or some famous person in the field), I would really think about it. After all, if not even a Nobel price winner can understand it? (However, not if the person is famous for unnessessary critism).
    – user111388
    Apr 1, 2023 at 23:03
  • For constructive, or even just specific, criticism? Sure. For verbal abuse from someone who considers me an incompetent inferior and just wants to rant at me? Nah, that's going in the garbage. (Or, if it's actually from a Nobel laureate, it's probably going up on the wall as a novelty and conversation starter, if nothing else.)
    – anomaly
    Apr 2, 2023 at 3:17
  • 1
    Indeed, I would do the same - but the OP seems not to be an verbal abuser and does not seem to consider them an incompetent inferior. If really a Nobel laureate (working in my paper's field) wrote just like the OP would like to do "Hey, your paper is too hard to follow for me", I would really think about how I can improve my writing style (and consult people close to me about that).
    – user111388
    Apr 2, 2023 at 7:12
1

Just write to the authors with all the questions you have about their paper: "Do I understand correctly that...?", "What exactly do you mean by...?", "One page X are you implicitly referring to Y?", etc. Do not hesitate to ask follow-up questions until you are sure you understand. Having to answer all these questions will punish the authors for their bad writing!

4
  • 8
    Unless you personally know the author(s) very well, or you are reviewing the paper for publication, I strongly doubt listing every single question you might have (especially when new to the field) will get you anywhere. Also you don't go around punishing authors of published papers for their (bad) writing by sending them angry emails, that's rude. If one or two questions remain after OP has looked at it with others, a polite email to the authors is, of course, fine. Whether they'll respond, who knows. I would.
    – user53923
    Mar 29, 2023 at 14:18
  • 2
    After two or three of these, there's a good chance you'll just get a brush-off. "I'm sorry, I'm busy and am not sure when I'll be able to take time to look at this." "I suggest studying references [X,Y,Z] for background, which should help make things more clear" (where X,Y,Z are notoriously long and dense textbooks or monographs). Or just no further replies. Not really much of a "punishment" for them. Mar 29, 2023 at 14:44
  • I think that if you have done reasonable due diligence (talk to other people about the papers, read background) then emails the authors with questions is absolutely fine! That's why there is a "corresponding author"!
    – jerlich
    Mar 30, 2023 at 14:29
  • 1
    You should take a good week (or two) to write that email, to ask really good questions. By that time, you probably don't need to send it any more, because you have worked out by yourself what the authors were unclear about. Perfect preparation to write your own paper later.
    – Karl
    Mar 31, 2023 at 20:56
1

You said you sought help from postdocs and professors, but they also faced the difficulty of understanding that paper. This makes me think that the paper is indeed poorly written.

Had your senior colleagues not stumbled on that paper, my sympathies would still be with you. In my opinion (with which some will disagree), a good paper must be written in a language understandable to a graduate student, ideally to an advanced undergrad.

I know an outstanding scholar who is having problems with writing clearly. This is not his fault but his problem of quite a psychological nature: it is difficult for him to model inside his mind how difficult his writings may be for a beginner or for a colleague from an adjacent area. On the other hand, I also knew people for whom writing solely for the anointed was a method of self-affirmation -- which, again, has a lot to do with human psychology.

For whatever reason an author is unable to write clearly, it will be useless to preach to them. Don't waste your time. Also, you may anger the author -- which will do you no good, especially at the start of your career.

Instead, you may send them a polite e-mail, praising their publication, and humbly asking to help you better grasp their so great and important result. After this, the author will probably respond and help you go through his so great and outstanding and ingenious work. Praise is more efficient than reproach.

1

I generally agree with most of the answers. If you write some people and say "Your paper is terribly written", nothing good will come of that.

But, if you write them and say, "This result is amazing! I struggled a bit with this bit... if I paraphrase that with , is that correct?".

Then i think it might be appreciated, and maybe even borrowed.

In big labs, students/postdocs can be the main authors of papers and the group leader may barely comment on the paper before publication. Depending on the language This might result in "unpolished" writing.

0

It's always tempting to immediately critique mercilessly (for me mostly to boost my ego), but first stop and consider why you're reading the paper in the first place. Clearly you're interested in the topic and content they're writing about and in understanding and replicating their results.

Kindness is King. If this is why you're interested in the paper then think, could there be potential opportunities for collaboration? Could an author on this paper be a future author on a paper you're hoping to publish? If so, then you don't want to ruin relationships now. It's always good to build interest and respect for another person's work whatever stage. You'll find you can make a lot more collaborations and faster progress this way, and a lot more people are willing to collaborate (however big their name) than you might think.

If you're hoping to discuss more with the author, I've found it particularly effective to send an email expressing thanks for their work (Say "Thank you for your paper..." as an email header or something similar). Express honestly what you liked, and/or any intentions for possible citation by your own upcoming work (if that's your true intent). I've seen these kinds of emails are very often replied to, and authors are generally fairly open about responding to concerns for a reply or maybe two.

If you're just upset for personal reasons (i.e., maybe the paper is from a well-known university that you were not accepted to, and you feel you could have written something better) then you're going to find you may be very miserable in a winner-takes-all academic environment. Hopefully this isn't the case.

Why might well-known authors allow this? Other than what @BobaFit said, some professors are far too busy to comb through student's work, or choose to let them publish their level of work in order for them to grow on their own. Sometimes it's used as a strategy (unfortunately) to claim some space so as not to give away too many ideas or keep competing labs away from their knowledge. This allows them to publish more papers in the space without worrying as much about competition. Yes, there are fields where publishing is that competitive and moves that fast.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .