On occasion during talks, presenters will say, "as you know from reading my papers" or during a Q&A someone will say "you should know from my papers".

The above clearly seems inappropriate for most settings outside of a group meeting. However, to what extent should I expect my audience to know what I am talking about vs. completely assuming that my audience know nothing?

  • 13
    As a basic talk-giving etiquette: it is much better to say "I dealt with that problem separately in my paper X" as opposed to "as you know from reading my papers" or "you should know from my papers". Jul 17, 2012 at 13:48
  • 5
    The Law of the Ignorant Audience states: "There is always someone important in the audience who doesn't know a vital piece of information, even if you take into account the Law of the Ignorant Audience.". I first read it in the Latex Beamer user's guide.
    – gerrit
    Jul 17, 2012 at 17:16

5 Answers 5


This really depends on your audience.

  • At a department colloquium, I assume very little background, probably an undergrad degree in math, but not even an undergrad class in my research area. Here the audience will often be smart, but ignorant of the relevant background.
  • At a research seminar with lots of undergrad and/or masters students, I still don't assume much background, maybe an undergrad class in the area, but even then I "remind" the audience of important information they "should" know.
  • At a research seminar with mainly PhD students and active researchers I assume more, but even here I usually don't assume that they're familiar with the problem or the relevant literature or techniques.
  • A conference special session (or minisymposium) is similar to a research seminar with mainly PhD students and active researchers. Here I often expect that much of the audience is familiar with the background and the techniques, but I still usually review them at least briefly. I've never gotten complaints about giving too much background and context.

I never assume the audience has read any of my papers. I typically view my talks (at least in part) as advertisements for my papers. The audience will never want to read my paper if they don't see why it's important (so that's my job during the talk). For those few that already want to read the paper, I hope they leave my talk with an outline of the paper, which will make it easier to read. Any time you're talking to an audience with varied background, it's good to briefly describe the key definitions. If you can do this quickly, perhaps only verbally, you don't bore those who already know, and you give the rest a fighting chance to keep up.

  • 5
    I agree 100% The only time when assuming the audience has read the papers is justifiable is (a) when presenting for a journal club/group meeting or (b) presenting at a workshop where participants are specifically requested to read certain papers. And even in the case of (b) an actual expectation that they've read the papers in question is often naive. Jul 17, 2012 at 13:44

In general, when giving a talk, my only assumption about the audience is that they found the title (or me) compelling enough that they decided they didn't have anything better to do with that hour (or 15 minutes).

Or at a conference, that they can't be bothered to get up and leave before the session ends, and decided to stay.

While I will give references to papers I've written, it's always a "For more information..." rather than a "As you all know..."


I think the causality is probably reversed: a talk at any level is a chance to "advertise" or "promote" ideas (whether "your own" or due to a larger enterprise) far more dynamically than in a paper. Thus, a good talk will motivate people to read your (or others') papers. Papers and talks are not in the same currency.

Further, if you have any reason to believe that most people in the audience have read your paper that you intend to talk about... you'll surely be boring them, and making them sorry they attended, no? Certainly if the talk is just an abbreviated form of a formal paper.


I guess that this would strongly depend on the culture within your scientific domain. In my case (software engineering) making this assumption would be inappropriate. You can suggest to read your paper if, e.g., there are some technical details you are not going to present since they might not be of general interest, but you cannot assume that this already happened.


In general, the broader the audience, the less you're able to assume about the audience's knowledge of your work. For example, at a general meeting for your profession or a department-wide seminar, you'd have to be more careful to provide background and context than at a smaller, more specialized event like a Gordon Research Conference.

However, time limitations will also be a major consideration—and in some cases, can outweigh the specialization issue. If you only have ten or twelve minutes to give your talk, then you don't have time for more than a minute or two of background information, regardless of how complex or novel your topic may be.

  • 16
    You left out the two most important words: If you only have ten or twelve minutes to give your talk, then you don't have time for more than a minute or two of anything except background information...
    – JeffE
    Jul 17, 2012 at 12:02

You must log in to answer this question.

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