Presenting a conference paper: deviating from the accepted version?

In our March "how can we help you?" thread, a reader asks:

When presenting a paper at a conference (e.g. APA), how much is it acceptable to deviate from the version of the paper that was reviewed and accepted at the conference?

This is a good and important question to answer. One time when I served as a conference commentator, the presenter basically addressed all of the issues that I had raised in the comments that I sent to them, making my commentary useless. It felt like a really underhanded thing to do, put me in a really difficult position as their commentator, and I've heard anecdotally that similar things have occasionally happened to others. I'm not sure whether and how far it is acceptable to deviate from an accepted version of a paper if you don't have a commentator, but it seems to me plainly unacceptable (and unprofessional) to do what I describe above. What do you all think? 

Comments

[Team deviate]

I almost never want to present exactly the same paper that I submitted to an APA. For one thing, the APA draft is always cut down from a longer piece that does stuff like respond to objections and give added nuance. For another, the lag time between APA submission deadlines and the actual presentation is what, 9 months? It would be crazy if you didn't continue to work on the paper in that time. Of course it will be different. There is no value to the author in presenting a lesser version of their paper just because the APA is slow.
Granted, it can feel annoying to commentators if they are not informed beforehand. If the author is deviating from the submitted draft, they should send the commentator an updated draft as early as possible, and not make changes after comments are received. That's only fair.

[Emma]

I have been to APA talks where the speaker will mention at the start of the talk that they've been working on the paper and it has changed but that they will present the paper as it was accepted. Then in the Q&A inevitably folks will ask how the paper changed and the speaker has a chance to discuss it then. This seems to me like a good way to approach it to not put commentators in a weird spot. I haven't personally been a speaker in this situation, but it always seemed to me that there was a good Q&A discussion and round of feedback, and as an audience member I've found it interesting to hear the earlier version of the paper then the discussion of how it evolved. So I'm not sure I agree with Team deviate that there's no value in presenting the earlier version of the paper, though I see how it would be more valuable to present the updated paper (were it not for the commentator issue).

At conferences without commentators, I've seen people go ahead and just present the most recent version of the paper (sometimes with a new conclusion), and I can't imagine anything wrong with that.

[What APA says]

I will just copy and paste what APA says on this issue: "Professional Conduct Policy:
All authors of refereed papers must present their papers as they were refereed, without substantial
revisions. Speakers, commentators, and critics in all sessions should ensure that their presentations are
sent to other session participants well in advance of the meeting and are not changed substantially
thereafter. Such practices as withholding prepared texts and changing papers to undermine commentators'
objections are egregious breaches of norms of professional conduct."

[MS]

When I'm told who the commentator will be, I get in touch with them asap to ask if they're willing to read a revised version of the paper. This seems to be an acceptable and professional practice, and I've never heard complaints.

Relatedly, I'm curious to hear what others think should be done when faced with the underhanded situation Marcus describes.

As a grad student I was slated to give commentary on a tenured prof's paper. I sent along my commentary a few days before their talk. Then the prof changed their talk to address or avoid all of my comments. I was too annoyed/surprised to come up with something new to say on the spot. So I just prefaced my commentary by noting what the prof had done, and then delivered the now useless commentary.

[Team Deviate #2]

I think the norms for this sort of thing vary depending on whether you have a commenter on your paper. If there's a commenter, then I'd just run it by them and see if they're willing to comment on an updated version of the paper.

But if there's no commenters, I've seen (and I've also personally done this too!) instances where someone gives a talk that they admit is quite different from what was accepted. In some cases I've seen people give entirely *different* papers, and I think as long as you let the audience know, it's fine.

[Anonymous]

I sign on to everything MS says. A very prominent moral psychologist did the same to me when I was a graduate student. For what it is worth, I continue to resent them, and I (obviously) did not resist the urge to complain about them to everyone I knew. To a person, they all agreed that this more prominent philosopher was a jerk. To this day, I understand everything they say in the light of what I saw of how inconsiderate they could be (rightly or wrongly—I'm only human).

Get in touch with your commenter. Don't undermine their comments. It is often hard work writing comments, and commenters are in significant part doing authors a big favor. Don't be a jerk.

[shrug]

One time a speaker omitted the part of the paper that I had an objection to, so I just prefaced my commentary with "The written version of the paper contained the following additional argument ... My comment on it is ...".

I didn't necessarily think the speaker was being underhanded - I think it's good not to directly read the paper, and then depending on your presentation style you might not get through all of it. But anyways, this seems like a fine thing to do when faced with any sort of change. I'm not going to design a second commentary on the fly.

[Evan]

The APA's position is understandable, but I think the other commentators here are on a better track. You shouldn't have to not work on the paper after it's been accepted, but at the same time you can't put your commentator in a bad position. Just talk with the commentator about any changes with plenty of time in advance.

[Karl]

Just echoing what others have said: present the best version of the paper you can possibly present, UNLESS it will throw the commentator under the the bus. If you get an objection that you MUST change (lest you look ridiculous), tell the commentator. Give her generous credit in the talk (and final written version) and give her enough time to come up with new comments. Do your best to not make more work for a commentator who is getting little CV credit for this and just wants to look like they are doing a competent job.

[Caligula's Goat]

I would add one additional moral variable here that should add at least one reason NOT to modify the paper:

Power relations mattter. A junior presenter with a senior commentator definitely puts themselves at some risk if their commentator holds a grudge against this behavior (which, as has already been pointed out, is 100% against APA paper presenting policy).

Things are even worse if the presenter is senior and the commmenter junior (or, worse, a graduate student). I think many graduate students will find this an offer they can't refuse (especially those, like me, who are first generation and so lack the sort of confidence that comes with those who have deep institutional knowledge). Consent in that case would be strained.

Why not just present the paper you submitted? If you've already done additional work then you'll have excellent responses to your Q&A and you won't be risking treating your commentator as a mere means to your own professional ends.

[TT]

I'm throwing the speaker under the bus, on the spot, if they ever give a version of the paper that preempts the comments I sent them. I'm not a graduate student, so that helps, but I just don't care about being polite in scenarios like that.

Overall, fine to change a paper, just not from the one the comments were based on.

[sprayed]

I have witnessed real douche-bags at the APA. One young presenter once said they just got the comments from the commentator a week before. The commentator then LIED and said they just got the paper the week before that - this is one of the biggest, fastest and successful liars in the profession.