I am having trouble asking questions in seminars, conferences, and public talks.

As a graduate math student I often fail to keep up with speaker and more mature members of the audience at events like seminars and conferences. It is very frustrating to lose track of the talk simply because I am not familiar with one key theorem/definition referenced by the speaker.

On the one hand, many people encourage me to raise my hand whenever something is unclear, motivating it by "no such thing as a stupid question" quote. On the other, very often I observe people being annoyed by a student asking an "obvious/elementary" question and wasting everyone else's time.

I have seen several related posts on this website, most of which are asked on behalf of the speaker. However, my question comes for the opposite side of the barricades. I want to know how to find balance between not annoying the speaker and the audience too much on the one hand, and catching up with the talk on the other hand.

More formally, my questions are:

  1. Are there any (semi-, non-) official recommendations on professional etiquette for mathematicians?

  2. I understand that every situation is unique and highly subjective. However, I would be glad if someone gave me advice on when to ask, how to ask, and what to ask at research talks.

  3. How can I tell if my question is "dumb" (i.e. the answer is well-known or searchable), or if it addresses legitimate ambiguity?

    • In particular, how can I quickly determine whether a definition/theorem/lemma mentioned by speaker is a part of common knowledge?

All relevant links or examples are appreciated.

  • 140
    Having reviewed over 2000 first posts on this site, I must say this is one of the best questions I ever reviewed. I can imagine many graduate students would have the same problem. One sugegestion, start your question by saying "I am a graduate student." so that everyone in the room will understand.
    – Nobody
    Commented Aug 17, 2015 at 6:37
  • 123
    I do wish people would stop saying "there's no such thing as a stupid question". It's simply untrue. In any given context, there are stupid questions. And given almost any specific answerable question, there exist contexts where it is a stupid question.
    – 410 gone
    Commented Aug 17, 2015 at 8:05
  • 13
    It would be great if you could find a seminar buddy. You sit next to each other, you write a brief question on your pad of paper, tilt it toward your buddy, and your buddy scribbles a quick answer, or gives you a nonverbal signal -- shoulder shrug, shaky of head, thumbs up. Commented Aug 17, 2015 at 13:30
  • 22
    Remember that sometimes there are other people in the audience who wanted to ask the very same question, but were afraid to do so. But in other cases the question might indeed be so basic, that asking it would show that you don't have a chance of understanding the talk anyway. If you only need to ask a single question during a talk, and if that question can be answered in less than 10 words, and if that answer makes the difference between you understanding the rest of the talk and you learning nothing from the talk, then you should not feel stupid for having asked it.
    – kasperd
    Commented Aug 17, 2015 at 16:24
  • 45
    @EnergyNumbers Yes, in theory there are such things as ``dumb questions''. In practice, I have seen many talks where the speaker lost all of the audience, and no one dared asking a question, but I have not yet seen a single talk that was derailed because of someone asking too many dumb questions. It seems that the real dumb questions aren't a problem yet.
    – Boris Bukh
    Commented Aug 17, 2015 at 16:31

14 Answers 14


There is no such thing as a dumb question is a good adage for the classroom, where our mission is to teach students, and we have a number of weeks to accomplish the learning objectives. We use this maxim to encourage students to ask questions rather than fall behind.

However, there is such a thing as an annoying question can be an equally true corollary, particularly in a conference setting where someone is trying to cram months worth of research into a 45-minute talk in front of presumed experts in the field. In such cases, it might be preferable to not derail the speaker's presentation.

How can I tell if my question is "dumb" (i.e. the answer is well-known or searchable), or if it addresses legitimate ambiguity?

Really, there's no way to tell for sure during the talk. However, if you venture to ask your question, you can preface it with something like:

"I'm sorry if I'm asking something obvious here, but..."

So far, my answer doesn't really differ from some of the other advice you've gotten in other answers. However, I want to address the professional etiquette part of your question. While you are asking your question, and in the immediate wake of getting the question out of your mouth, pay careful attention to the body language of the audience at large. If you see several heads nodding affirmatively, that might be a good indication that you were brave enough to ask something that was nagging in the minds of everyone else, and the speaker has made some erroneous assumptions about what was presumed to be fundamental or obvious. However, if you notice some sideways glances accompanied by grimaces or eye-rolls, then maybe you've touched on something that would be better left until the end of the session, or until the next break, in which case you can quickly add:

"If you'd rather discuss that with me off-line, that's okay."

In summary, be aware of your environment:

  • What is the purpose of the talk?
  • Who is the intended audience?
  • What are the speaker's time constraints?

Moreover, be cognizant of non-verbal audience reaction to your initial question, and use that as a barometer before venturing to ask follow-on questions. In my experience, people are rather forgiving the first time a presentation is interrupted by an elementary question, but they begin feeling exasperated when that one question transforms into a hijacking of the presentation as a whole.

  • 35
    I really like the "read the audience" part of this answer. Nodding heads or exasperated looks are easy enough to distinguish.
    – Floris
    Commented Aug 21, 2015 at 19:25
  • 1
    +1 "This maybe something obvious that we could discuss after, but why is it that..." is a usual goto for me when attending graduate seminars even though I am an undergraduate student. Its most helpful when the speaker knows you and your background knowledge.
    – user84539
    Commented Jul 15, 2018 at 3:23
  • Wouldn't perhaps the better method not be to screen for content of the questions, but rather for how much time of the presenter is taken up by asking them? I.e. if you're asking too many questions, to the point that it's being bogged down, then that's too many. But if you only ever ask one or two questions, it shouldn't matter how "stupid" it is or isn't (And moreover, if those one/two questions take the whole thing to answer and the lecturer actually chooses to go that far, chances are that means they are NOT "stupid" questions!). Commented Feb 20, 2019 at 9:31
  • Moreover, the term "stupid question" does not admit a well-defined meaning, and thus one should/can only (it is only possible to) learn from experience and don't be afraid to make mistakes and yes, do "bad" things - that's what being human is about. Just ASK the damned things, but limit the number. Commented Feb 20, 2019 at 9:32
  • 2
    In theory, trying to read the audience while asking a question is great advice. Personally I find that very hard to do though--depending on the number of people in the room as well as their seniority, my heart will be racing as soon as I raise a hand and then a lot of focus will be required basically to not stammer..
    – smcs
    Commented Feb 5, 2021 at 15:43

Keep asking the dumb questions! It is better to look like a fool, than to be one.

You worry that many speakers are annoyed at the `elementary' questions. Some speakers do it because they are stressed about public talking, and any question upsets them. For some, communication of mathematics is not the aim of the talk; they give it because it is a condition of travelling to the venue, or simply because everyone else gives the talks. Instead of feeling joy at an opportunity to clear the confusion, they might get annoyed at having to do the extra work of explaining some of the background. The annoyance has no long-term effect --- nobody holds grudges for asking dumb questions. To give up actually understanding math for such a petty reason is just not worth it.

There is only one situation in which you should refrain from asking a question. That is when you are representative neither of the actual audience nor of the intended audience. So, if you are graduate student at your department's colloquium, it is OK to ask anything. If you are a graduate student at a seminar in your field, it is OK to ask anything. If you are at a seminar in another field, and there are several other students in your field in the room, again it is OK to ask anything. Only if you a lone outsider at a seminar or a conference that is not in your field, there is a reason not to ask questions.

  • 42
    This is good advice. I will add that "when you are not the audience" works the other way too. For the first year or two of my tenure track faculty job, I attended the Graduate Student Seminar. There were a lot of talks in different areas of mathematics, which I enjoyed, and I thought "Well, if these talks are meant to be understood by students, then I should really be able to understand them. So a few clarifying questions have to be okay..." It took me way too long to realize that my questions often pulled the talk away from its intended audience and actually had an intimidating effect. Commented Aug 17, 2015 at 16:08
  • 4
    Well, I can think of other cases where asking (more) questions is not a good idea. For instance, if a talk is really hard to follow and/or ridden with mistakes, and any answers you get only exacerbate the confusion, it may be a better idea to just let it go instead of prolonging the agony. But yeah, that's good advice IMO.
    – tomasz
    Commented Aug 19, 2015 at 13:48
  • 6
    There has been a rare occasion where someone who wasn't paying attention will ask a question that was just covered. I find that incredibly annoying, both as the speaker and as a listener. Please don't ask unless you're actually paying attention. Commented Aug 19, 2015 at 16:53
  • 6
    @DoubleDouble: I somewhat agree, but sometimes when you are paying attention, just lose focus for a second (because a colleague asks you something, or because you focused on a different part of the presentation), you still miss something, sometimes even realising that something was just covered and you missed it. In this case, I don't think you should refrain from asking just because of that.
    – tomasz
    Commented Aug 22, 2015 at 3:53
  • 1
    "You worry that many speakers are annoyed at the `elementary' questions." - it's indeed rather the audience than the speaker that I'd worry about. As a speaker, I am usually relieved when some of the questions I get are actually easy to answer for me. It gives me an opportunity to explain something that may have been mentioned before in different words. And quite possibly, I'm too mentally busy to think about whether it was a really sharp question, anyway. In a way, I actually prefer somewhat basic questions about my topic to some other types of questions I regularly get - questions about ... Commented Nov 17, 2016 at 14:40

Before asking a question, ask yourself the following:

If I get a nice detailed and understandable answer to this question, will I be able to understand a significant part of the rest of the talk?

If the answer is "no", then you should probably not ask the question even if there really is some ambiguity that could be cleared up, because chances are that asking the question will not benefit anyone (i.e. that those who are able to understand the talk would also be able to answer the question themselves).

If the answer is "yes", then that is a good start. In that case you should probably ask the question unless it is something you really ought to know (if you don't know that it is something you should know, then that is usually good enough). Of course you don't want to become known as "that annoying grad student who keeps asking trivial questions", but you also don't want to miss out on learning something from the talk because you are missing some small detail, and usually the above will limit the number of questions you ask sufficiently that people should not be annoyed.

  • 32
    It seems hard to predict in advance whether you'll be able to "understand a significant part of the rest of the talk"--if it's anything like the talks that I've had trouble following, you're not always lucky enough to know whether a reference to some concept you haven't heard of is a sign that the rest of the talk depends on that concept.
    – Milo P
    Commented Aug 17, 2015 at 15:48
  • 5
    @MiloPrice Yes it can be hard to predict. But a metric which you can only apply retroactively is still better than no metric at all. Additionally usually there is some sort of understanding between speaker and attendees about what the prerequisites for understanding the talk are.
    – kasperd
    Commented Aug 17, 2015 at 16:30
  • 3
    @MiloPrice I think the best you can hope for along those lines is to use inductive reasoning to make a reasonable guess as to whether that definition is a key to "[understanding] a significant part of the rest of the talk." It won't always be right, but you'll learn more about yourself by trying... and that's worth a lot, in and of itself!
    – Cort Ammon
    Commented Aug 17, 2015 at 19:27
  • 3
    @MiloPrice Indeed, it is not a perfect method, and if in doubt, I would usually err on the side of asking the question. Commented Aug 18, 2015 at 7:05
  • @MiloPrice Well, there are some clear cut cases where the answer is unequivocally no. I still ask those questions.
    – Kimball
    Commented Aug 18, 2015 at 13:54

I have recently finished a PhD in Particle Physics. Over my time as a student, particularly early on in my studies I frequently encountered this problem.

I would start by pointing out that, generally, you won't be the only student in the room and there will almost certainly be others thinking of the same 'stupid question' but not asking. Many times these questions arise due to the incompetence of the speaker and not the audience. Secondly, it is an academic environment, questions should be encouraged. Those academics who don't encourage bright young students are the problem, not you.

Saying that, the timing of the question could be considered. From my experience I found that in a seminar it was best to ask questions from the audience which were relevant to the topic, perhaps not so much for your personal understanding but to encourage discussion. More often than not there will be ample opportunity to ask the speaker personally in a coffee break or at dinner, maybe take these opportunities to understand the details you are missing. Often this approach will allow you to spend more time gaining an understanding and benefit you.

  • Thank you! The last paragraph sounds extremely reasonable to me.
    – Vlad
    Commented Sep 28, 2015 at 11:48

First, it is important to identify whether you are the intended audience for the talk or not. If you're not the intended audience (you're going to a seminar well outside your subfield, you're a second-year graduate student at a conference that's mostly not graduate students), then you should be careful not to annoy the audience. However, if you are the intended audience, then you're doing everyone a favor by asking more questions.

Second, if you're generally good at reading social cues, then you should trust your instinct and feel free to ask questions until the speaker says something to shut you down ("we can talk about this afterwards", "that's standard material", etc.) or you notice the audience being annoyed. If you're not as good at reading social cues, then be a bit more careful and try to get honest feedback from a friend about whether your questions were out of line (e.g. "I'm not so great at reading social cues, so I was wondering whether you could honestly tell me whether some of my questions went too far.").


I'm not in your field, so I'm not sure if this'll be helpful or not.

When I was a grad student, I convinced myself that if there was a talk I didn't understand, then it wasn't my fault, but the speaker's, and that there would be other people in the room with the same questions and confusion as me. It might have been overconfident of me and I might have been wrong, but I nevertheless acted according to that belief.

So I was always "that guy" in the audience who asked at least one question after every talk I attended, and many of my questions were just asking for clarifications of things I didn't understand, and if the answer didn't help, I wasn't shy to say "I still don't understand".

And sure, I often wondered what all those big shot professors in the audience were thinking when they listened to me asking all these questions. I still don't know. But after a session at an annual meeting of our society, in which I had again asked a lot of questions, a Berkeley grad student came up to me and thanked me for daring to ask all my questions, which she said helped her a lot in understanding what the presenters were trying to tell us. And she was a pretty smart student to begin with.

It's possible that you don't understand things you should understand, but you should have a reasonably good estimate by now whether you are a normal, brilliant, or terrible grad student. If you're in the two former groups (which I'm sure you are), I can pretty much guarantee you that there will be others in the audience (professors and grad students alike) who will appreciate your questions.

As to what to ask: If you don't understand the presenter's talk, then there must've been a certain point where you got confused. Ask a question about that very thing that made you lose track. It may or may not help you in understanding the rest of the presentation, but ask anyway, because others will have been confused by the same thing.

And for the record, I'm a professor now, but I don't understand more of talks than I did as a grad student ...


For the case you mention of being unsure of a definition or a named theorem, your phone is your friend. Look it up. It should be possible to gather whether the concept is well-known, for example if it is mentioned in a course, a survey, or on Wikipedia. If it is, and the mention is in the research seminar's area, that suggests the question would not be welcome, as the audience is reasonably expected to be familiar with it.

  • 7
    This sounds to me like a good advice in principle but not in practice. If you start concentrating on Wikipedia you are likely to loose track of the talk? It does answer the question asked by the OP though!
    – chris
    Commented Aug 19, 2015 at 9:02

Unfortunately most attendees at most pure maths seminars do not understand most of the talk. A good talk aims to make the first third comprehensible to phd students, second third to faculty, last third to experts. Most talks do not achieve this.

You are generally expected not to ask questions unless they are to the point which is very hard for a phd student.

I try just to get a few ideas across in mine recognizing that no one will follow details. However, I sometimes get sarcastic comments that my talks are too "philosophical." I also sometimes get great compliments.

  • While a given person is not usually expected to ask questions, they are not typically discouraged.
    – Kimball
    Commented Aug 19, 2015 at 2:18
  • my experience was that asking basic questions about the material was discouraged.
    – Mark Joshi
    Commented Aug 19, 2015 at 3:41
  • 6
    Asking basic questions about things introduced in the talk is fine in most seminars I've been to. E.g., speaker defines moufang loops, gives an infinite example and starts to move on. Q: Can you give an example of a finite moufang loop which is not a group?
    – Kimball
    Commented Aug 19, 2015 at 6:15
  • yes but if he says a moufang group is a Lie group and you ask what a Lie group is, you'll irritate the senior academics present who expect you sit at the back and shut up.
    – Mark Joshi
    Commented Aug 20, 2015 at 5:22
  • 2
    I always heard "1/3 understood by everyone, 1/3 understood by specialists, 1/3 understood by no one."
    – James King
    Commented Aug 20, 2015 at 16:34

The balance should be struck in terms of usefulness. Let us assume that the question is genuinly dumb, it might still be useful, first to you if the answer will really allow you to make a significant breakthrough in your own understanding of the field. It might also be useful to other persons like you in the audience, who likewise lack some information to follow the talk properly; it might also be useful to the speaker, who has failed to realize a fraction of is audience is not aware of such and such premises (s)he relies on. In other words, any question is going to cost at least one minute of the speaker's time, but the gain might outweigh the penalty even if the question is dumb.

Typically, even a dumb question may enlighten someone in the audience because it gives them the opportunity to realize that the matter discussed opens itself to some level of misunderstanding when perceived from a different perspective than that intended by the speaker. In the end, exposing such different views is nowadays one of the main reason to attend talks in person.

Then there is always the possibility that the question is not dumb :-)

So the bring-home message is "Bite your tongue, and ask yourself, how useful will my question be?".


Usually, the in-depth questions come from people who are familiar with the topic. So, if your intent is to play an active part, you can read up on the work of the speakers who will give the most interesting talks. I would not worry too much about the speaker getting annoyed by questions. From the speaker's perspective, what is more annoying are all those famous professors who the speaker wanted to make an impression on, who are sleeping.


There are a couple of factors I use:

  • Have you asked a few questions already?

    And if so were they well received?

    If you have, and they were not, you get away with a lot less, and rightly so. Most of the etiquette is not hogging the speaker. You can avoid this by either asking questions that help everybody or by not using too much time. Either way you are safe.

  • Did you give it long enough to know you are roughly at the right level?

    If you followed the majority of the talk thus far, and that was a non-trivial fraction, then it's reasonable to assume the bit you're stuck on, others might be too. Even if this is not the case, other's will recognise that you have restrained yourself at least a little.

If you are 'clear' on both points, I think it would be hard to object to your questions.


The literary technique known as Lampshade Hanging may be useful here: specifically call attention to it yourself, acknowledge it, and then just move on.

"So in the interests of making sure everyone's on the same page here, let's ask the obvious question first: are we talking about African or European swallows?"

  • This looks intriguing but I don't completely understand what you're suggesting in the seminar context. Could you expand on this a bit, please? Commented Aug 18, 2015 at 15:35
  • 1
    @aparente001: By pointing out yourself that you're asking (what could be seen as) an obvious question, you beat any potential critics to the punch. It basically helps to deflect criticism preemptively, because now if someone says "that's a really obvious question!" it makes them look bad because you just said that. Commented Aug 18, 2015 at 15:47
  • 11
    I see. Thank you. Well, I will suggest a slight modification: "In the interests of making sure I'm on the same page here, are we talking about African or European swallows?" (Reason for modification -- the OP will probably want to avoid seeming to be claiming to speak for others.) Commented Aug 18, 2015 at 15:52

One simple heuristic I use is considering how long it will take to answer the question. For example, if I don't understand the notational conventions or assumptions in a talk and they seem important, I might ask -- even if it's an obvious question, it can be answered in a sentence and won't waste too much of everyone's time. If it's a deeper or more substantive question, then I might save it for the end of the talk or take it offline.


A good way to ask a dumb question is as follows:

Sorry, this is probably a dumb question, but [ask question here]

for example

Sorry, this is probably a dumb question, but isn't a differentiable function continuous?

This signals to the audience that you are just having a momentary lapse.

Another tactic amongst all the people who wish to appear intelligent is to use the word "surely", eg.

Surely that theorem is not true if you don't have any regularity?

  • 31
    Please do not strive to appear intelligent. Freely admit your ignorance and stupidity --- nothing is more intimidating to a student than a bunch of professors who want to appear intelligent. Do not use other pretense tricks either --- if you have a memory lapse, admit it, but if you don't, then don't fake it!
    – Boris Bukh
    Commented Aug 17, 2015 at 19:40
  • @BorisBukh I am not saying that one should do it. I am just saying that this is what happens in real life, and my post tells you how to do it if you wish to emulate it.
    – Upin
    Commented Aug 17, 2015 at 20:09
  • 2
    Your answer starts with "A good way ... is". I felt that I had to make a strong warning since the answer did not come with one.
    – Boris Bukh
    Commented Aug 17, 2015 at 20:22
  • 7
    Actually, I think the "Sorry, this is probably a dumb question, but ..." part is an ok way to do it, if you're feeling too insecure to ask your 'dumb' question without such a preamble. In most cases, the speaker will answer that it's not at all dumb, thus encouraging more people who feel like they have a dumb question to ask it. BUT I won't upvote this answer because the other suggestion "Surely, ..." is just horribly impolite towards the speaker, because it assumes that the speaker got something wrong instead of leaving the option open to explain why he did it that way. Commented Aug 18, 2015 at 6:47
  • 8
    Yes, the theorem is true even without regularity, and don't call me Shirley.
    – JeffE
    Commented Aug 19, 2015 at 4:39

You must log in to answer this question.

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