Tag Archives: academic life

The Law of Good Enough (or why your thesis will never be finished)

I spent quite a bit of time recently meeting our section’s post-graduate students for tutorials. In some cases this is to welcome new arrivals, or to catch up on progress from those who have been away on lengthy field seasons. The ones I most enjoy seeing are those  who are busy writing up — because they’re the ones I’m most able to help.

It can be difficult to persuade a postgrad staring down their thesis deadline that 15 minutes in my office is time well spent, which I fully understand. Usually they are stressed, feeling the pressure and unable to focus on anything other than the thesis. Much of this derives from a sentiment I hear echoed again and again in various forms: “I just want to do the best job I can”.

No. Stop. This is not the way to approach a thesis. You need your thesis to be good enough.

This shift in attitude is hard to accomplish when your whole academic career has been geared towards achieving the highest mark possible, or at the end of four years when you want to have something on your shelf to be proud of, that you can look at and think “I wrote that” and feel a warm glow inside. Allowing yourself to fall into this vanity trap is pathological, and the root cause of a lot of unnecessary stress on the part of post-graduates.

Your thesis is the means to an end, which is graduation. When the day comes, you will walk across a stage for 20 seconds, shake someone’s hand, collect a piece of paper and get a photo taken in a silly gown. It doesn’t matter if you’ve written the most perlucid, inspiring and impressive thesis of all time. No-one will clap any louder, or any longer. No-one will ever judge you on the quality of that thesis, good or bad. All that matters is that it was good enough.

In one of the labs I worked in we had a thesis that did the rounds of the post-graduates who were writing up. You might think that they were sharing a particularly wonderful thesis so as to learn best practice and be inspired by the achievements of others. I’m sure all the supervisors would have preferred that. But no, the thesis everyone wanted to see was singularly atrocious. No-one reading it could fail to spot glaring errors, hideous formatting and some of the worst figures ever committed to print. That’s exactly why everyone was so keen to read it — if this person passed then surely there was hope for others!

I’m not going to reveal whose thesis it was, because that doesn’t matter. They have gone on to a successful academic career where they are respected in their field with an international profile. Does anyone care that they submitted a shoddy thesis? Of course not. It was good enough. On the other hand, the best thesis I ever read remains that by Mike Shanahan, who preceded me by a couple of years and even worked at the same desk. Nothing could be more demoralising than to witness a standard of writing to which I had no hope of aspiring (at the time). Perhaps he still looks with satisfaction upon that thesis. He might do so again if he reads this. My bet is that it hasn’t crossed his mind in a decade or more. Did it benefit his career? Maybe, but probably not that much.

There is an argument that a better thesis will lead to an easier viva, and that’s perhaps the case, but my suspicion is that the correlation is not strong. How a viva goes depends on the personality of the examiners, their particular bugbears, the wind direction and the alignment of the stars. You can no more predict the questions than you can anticipate how many corrections you’re likely to get. The time to be a perfectionist, or at least to aim for the highest standards you can, is when you’re preparing a manuscript for publication. Then you know it’s going to be pored over in great detail. A publication is your contribution to the legacy of science, a work that will be forever associated with you. The thesis? That’s a bookend.

The best advice I ever received while writing up was from another old hand in the group who told me that a thesis is never finished. Eventually you just relinquish it to the examiners. Bear this in mind if you’re tempted to read and reread chapters, add more references, or tinker endlessly with the figures. There’s always something else you could do. Just get it done, make sure it’s good enough, then move on to the rest of your career.


Edit: @ZarahPattison made an interesting point on Twitter about thesis by publication. Although this is arguably the best possible way to prepare a thesis, it’s not for everyone, and many universities don’t even allow it. I wouldn’t like to give any student the idea that it was an expectation, not least because I didn’t manage it myself. It’s certainly true that a well-written chapter is easier to turn into a manuscript, but that’s missing the point. If you want to write a manuscript, write a manuscript. If you have a manuscript then turning it into a chapter is easy. If you need to finish a thesis then get the chapters done and worry about the manuscripts later.

How to respond to referees’ comments

The first time I submitted a manuscript, it came back months later with a lengthy list of recommended changes and an equivocal response from the editor which implied that he was reluctant to hear from me again but might deign to respond if I proved myself worthy. I was devastated. It had been an immense amount of work and effort to prepare, and by now I’d moved on to other things. I glumly sloped into my supervisor’s office and was taken aback by his enthusiasm. Apparently this was what passed as good news in science *.

Since then I’ve been through the manuscript submission mill many times and always prepare my students in advance for the likely tone of what they will receive. It doesn’t get any easier. I still can’t read comments as soon as they arrive. Normally I’ll read what the editor says, skim the rest, then go for a short walk around the lake to calm down. Sometimes it takes several laps.

Eventually, however, you need to brace yourself and get down to the revisions. Clear your diary, close the door, unplug the phone (and the internet) and make sure there are no distractions. Don’t leave until it’s done. Unpleasant jobs are always the easiest ones to procrastinate from, and revising a manuscript comes pretty low on my list of favourite ways to spend an afternoon.

Assuming you have an invited resubmission (rather than an outright rejection **), here is a quick guide to how to respond to manuscript reviews that I wrote for my PhD students:

  1. Write back to the editor immediately, thanking them and the referees for their time and helpful comments. Even if you’re not grateful and they weren’t helpful. Even if they rejected the manuscript ***. Being nice works wonders in the long term because they will see your work again. They have also taken their own limited time, usually unpaid, to look at what you’ve submitted.
  2. Compose a response letter, starting in much the same way. List and address every single comment made by the editor and referees sequentially and in full. Keep in the positive ones too, it makes you feel better.
  3. Make it as easy as possible for the editor to tell that you’ve made the changes requested. This means that instead of saying ‘This has been done’, or ‘A paragraph on this has been added to the discussion’, say ‘This is a very helpful comment. We have therefore inserted a new paragraph in lines 283–292 which explains how…’ etc. Editors are busy and don’t like to have to work harder to check whether you’ve followed instructions.
  4. Tread carefully if you disagree with any comment. If it makes no material difference then make the change, even if it’s only a matter of preference. Only contest if you are convinced that the referee is wrong and you can back it up. Even so, apologise for not making the manuscript clear enough and specify where you have added clarifications or extra evidence in the text. If you’ve failed to convince them first time around then it implies that you need to change something.
  5. Try not to use track changes, comments, bold type or other formatting to note changes to the manuscript itself. In my experience (usually when requested to by editors…) this leads to errors in the final copy. Refer to line numbers instead.
  6. Take extra time on the figures. Clear, high-quality figures give your paper a greater chance of being read, cited and used by others. If the figures look amateur then no-one will bother reading the text. Use this opportunity to redraw and tweak them using proper tools (e.g. inkscape, sK1, ImageMagick, gimp). Don’t rely on Microsoft Office products to create publication-quality images.
  7. Never play referees off against each other. If they disagree on a point then compromise and ask the editor for guidance. Also note that if only one referee picks up on something, this does not imply that all the others are on your side. They may simply not have noticed.

Finally, in almost all cases reviewers are doing it because they genuinely care about maintaining standards in the scientific literature and improving the quality of work that gets published. There are some cases when a reviewer might block something too close to their own work, which contradicts them, or out of some personal vendetta against you or your collaborators. This is exceptionally rare though, and can seldom be demonstrated. Even if you suspect it, you’re most likely wrong, and should never say so in your response. No-one is out to thwart you.

Good luck, and remember, we all go through this. If it starts to get you down then go and vent to a colleague. Everyone has stories to share.


* A friend at a university in a developing country once related that the modal number of papers among his faculty colleagues was zero. Exploring the causes of this, it transpired that in many cases they had once submitted something to an international journal and been so offended by the audacity of the response that they had vowed to never subject themselves to such humiliation again. This was true of even senior professors.

** I would recommend doing all this even if you’ve been rejected. Partly because you have a high risk of coming across the same referees again at a different journal, but mainly because it forces you to confront the criticisms of your work.

*** Don’t contest a rejection unless one of two things apply. Either there has been a gross mistake made by one of the referees, and you can unequivocably demonstrate this. Or you’re submitting to one of the big journals (Nature, Science) when putting up a fight can make a difference. Apparently. It’s never worked for me.