Hecklers And Twecklers At Technical Talks
Handling hecklers at technical talks
John F. Kennedy was not a theorist of course, but was the 
Today I want to talk about talks, technical talks, but especially hecklers at technical talks.
We have all been at talks where someone interrupts the speaker, and asks a tough question. The definition of a heckler:
is someone who tries to embarrass you with gibes and questions and objections.
I have to admit that I have been both the heckler and the hecklee over the years—is “hecklee” a word? I sometimes have been the heckler without meaning to be, sometimes I simply have asked a question that turned out to be embarrassing. Sometimes I have been on the receiving end of a heckle.
Kennedy was brilliant in how he handled tough questions and heckles. I recall at one press conference a certain reporter was trying hard to get the President to call on him—waving his hand wildly. The President kept ignoring him and calling on others. Finally Kennedy did call on him, and the reporter asked a long, complex, and nasty question. Kennedy looked at him during the question, waited for it to end, and after a short pause pointed to another reporter and said:
Next.
What a great way to handle a tough question. Perhaps only the President can get away with this stratagem.
Heckles
Here are some examples of heckles, organized by type. In some cases I have protected the identity of the people, since my intention is not to be mean. I just think it is important for all of us to know that we should be prepared for tough comments.
That is a crock, and I want to completely disassociate myself from that.
That is pretty unusual heckle, and it came from a co-author.
I believe that this is all known.
The senior person proceeded to give a precise citation to an article including the year and the journal where he claimed the previous work was. The job candidate was shocked, and tried to explain why his results were new.
The next day the candidate confronted the senior faculty after checking the citation, which had nothing to do with his work. The senior faculty said: “I fell asleep. I thought if I challenged you in this way, you would have to repeat the details that I had missed while I was asleep.” The candidate was speechless and ran to see his friend DeMillo for advice. Rich’s advice I think was optimal: basically forget it and get a job somewhere else. The candidate did just that.
Freeman Dyson explains that the hall was packed on Thursday, September 2, 1954. It was abuzz with excitement—what did the great von Neumann have to say about the future? Unfortunately von Neumann had forgotten about the talk, and he had almost missed being there. So he pulled out an old talk on rings of operators, an old talk from the 1930’s. Nothing new.
As Dyson explains the audience soon became restless. Finally a voice from the hall said for all to hear:
Aufgewärmte Suppe.
Which means in German: warmed-over soup. The talk was just a replay of an old one, not the peek into the future they all had hoped to hear. The heckler was rude to von Neumann, who had no answer and ended the talk immediately. I think in a sense it was a tribute to him: they expected the great mathematician to help give them a view of the future. They were upset at the lost opportunity.
Yeah, Yeah.
As an old chess-wise friend used to tell me: sac, sac, mate. By the way in some logics, such as Intuitionistic Logic, 
Let’s thank our speaker and now let’s go have tea.
Game, set, and match. The talk was over. Apparently the visiting professor was an expert in point-set topology, but not an expert on the Poincaré Conjecture.
During the talk Perlis got interrupted by an audience member, who stood up, and started to read a list of questions. They were all of the form: Why did you do X in Formula Algol? Each question hit right on an issue with the language—it was not a good mating between Algol and formulas. After listening to several questions Alan interrupted and said, “do you want me to answer?” But the questioner just keep reading more of his list. Finally Perlis could take it no longer and repeated “do you want me to answer?” The questioner reluctantly said “okay.” Perlis then said:
Poor design.
He then proceeded to finish his talk.
Open Problems
Please share your best stories. Please share your best heckles. Please be polite at technical talks.
I have one for your “Heckled in another language” category. This story is second-hand, but I believe it to be true.
The distinguished model-theorist Bruno Poizat is well-known for wishing to deliver his talks in French, wherever they take place. (This much is certainly true!)
I have heard that after one such talk, a disgruntled audience-member who spoke no French raised his hand and asked a question in Hebrew (which of course Poizat could not understand), as a means of communicating his dissatisfaction with the talk.
John Austin: Grammar Nazi. Double negative “should be avoided”? Why, will God kill a kitten otherwise? Someone should tell John that linguistics went descriptive quite a while ago! Guess this is one more reason why “we don’t need no education, we don’t need no thought control” 😉
Good-natured disagreements are fun and everyone enjoys them (speaker, questioner, and audience alike). Bad-tempered disagreements evolve swiftly into intransigence, then epithets, then censorship … behaviors that carry us inexorably toward Oded Goldreich’s policy of silence. Sadly, it seems (to me) that Oded’s policy is becoming widespread and that too many of humanity’s best voices are falling silent because of it.
With respect to purely technical questions, forums like TCS StackExchange and MathOverflow provide an effective and well-respected remedy. For broader issues, there’s no substitute for weblogs that set a generous-hearted example … and (IMHO) everyone appreciates—students especially—that Gödel’s Lost Letter and P=NP is one of the best.
Just as kindly humor is scarce relative to abusive humor (and consequently kindly humor is more respected), kindly academic heckles are scarce relative to abusive heckles (and kindly heckles are more respected). In this regard, a traditional heckle in academic medicine is for a resident to begin a case presentations at clinical conference with the set phrase:
“students especially”
And amateur wanna-be mathematicians, like myself 🙂
For all the intelligence and knowledge available in this world, it seem remarkably difficult to find forums for real discussion. For me, I know that it’s only by testing what I am thinking that I’ll be able to grow beyond it. I don’t want to blindly believe that I’m right, but rather learn some of the truth. A good explanation of why I’m wrong is often the most valuable bits of knowledge that I can receive (and appreciate).
Paul.
The speaker, Oxford philosopher of language John Austin, was making the point that in almost all languages a double negative should be avoided. Sometimes a double negative can be positive and sometimes a weak negative.
The version I’ve heard wasn’t about whether a double negative should be avoided, but just how to interpret it. In some cases a double negative is positive, while in others it is a particularly intense negative (i.e., the negatives get added rather than multiplied): “You ain’t seen nothing yet” or “ne … pas” in French. The talk was about this puzzling difference, and the speaker was surveying which interpretation was standard in different languages or dialects. However, he made the mistake of saying there was never any issue about positives: in all languages, two positives are even more positive, because the multiplicative and additive rules agrees. This when Morgenbesser chimed in with “yeah, yeah”.
I have heard the story ending with “yeah, right”
“ne… pas” in French is not a double negative, it’s the normal negation structure. For example, “vous ne devez pas manger” translate as “you must not eat”. A double negative could be “vous ne devez pas ne pas manger” which reads as “you must not not eat” meaning “you must eat” of course.
What always confuses people is the fact that a positive answer to a negative question should be “no”: “tu ne viendras pas ? (you won’t come?)” -> “non”.
Let us also not forget that in French, merci or “thanks” in answer to a question really means “no, thanks.”
My favorite is when a heckler asks a very critical question, which is given a sound response, and you hear the heckler in the background going ‘oooooOOOOOooohhh, now I see’
Hmmm … when the audience does it, it’s called “heckling” … and when the speaker/author does it, it’s called “hectoring” … so, what’s the common etymology?
According to the Oxford English Dictionary:
There are numerous celebrated examples of authorial hectoring in mathematics and computer science: GOTO statement considered harmful (Edsger W. Dijkstra),The mythical man-month (Fred Brooks),You and your research (Richard Hamming) … not to mention Doron Zeilberger’s dozens of essays … no doubt everyone has their own personal favorites.
What distinguishes these essays is that the authors intended no harm … quite the opposite, in fact.
“Heckled on a job talk:” I heard such things several times. It seems like a number of senior researchers, kind of sure of their omniscience, like to ask such a question to applicants in domains they are not expert in: “Everything you’re talking about is known since decades, isn’t it?” Sometimes, it is like they think Complexity Theory (or replace by the field you like) stopped at Cook’s paper in 1971, just because they have no real interest in the field.
I just precise that this is not an angry comment, as I never had this problem!
I do not think giving up on the job was good advice, but I don’t know what would be better. It would be a real shame for him to go work for Goldman Sachs because of a misinformed heckler at his faculty job talk.
A fairly common type of heckling during talks is the line (and variations): “Tell us why we should care about any of this.”… as if to say, “What you have said is not interesting to me, and is in fact not interesting.”
A much more polite version of this, though ultimately a somewhat different question, is, “Could you perhaps tell us why this is a natural thing to consider?” When I hear a question like that (rarely am I on the receiving end of it, thankfully) I often try to think what the true motivation of it is: is the questioner sincerely wanting an explanation that is more natural to his or her mode of thinking, or is it a more-or-less rhetorical question designed to make the point “what you have said here is artificial (and not interesting to me).”
In one of my early talks on Tornado codes for data distribution, a famous senior coding theorist interrupted multiple times to say, essentially, that Reed-Solomon codes and striping were clearly a better approach that had been known for years, and when I tried to explain the limitations of that approach and the (natural) settings where our approach would be better, he (roughly, I don’t remember the exact wording) told me I needed to go back and do my homework on Reed-Solomon codes. I somehow stumbled through the rest of the talk after that, but it was pretty difficult.
History, I think, vindicates us. We met a reasonable amount of resistance from coding theorists more generally when we introduced Tornado codes, undoubtedly because we were outsiders. But that was the most public, and painful, experience.
Now that I’m much older, I’m sure that would bother me much less, or not at all.
I think the best story is Pascal vs. C. C++ was the heckler 😉
Glad you made the post. It was fun to read. As for Opinion 76, I think Scott replied very well to the heckle, on behalf of complexity theorists.
Something elementary I’ve seen people stumble over is answering a question like “why would the following trivial solution not work?”. It is a question you will get asked all the time, and you should definitely have an answer. Actually, the best talks I’ve seen anticipate this, go through a few (not just one) trivial attempts at solutions and explain why they don’t work and go on to show how to make things work from there.
You forgot the most painful example: heckled by your own PhD supervisor. I’ve seen this happen a couple of times and the tension was truly palpable…
And the old chestnut, for pure mathematicians: what is this used for? – usually asked as the first question at the end of a talk and causing several minutes of awkward silence.
One of the most flustering questions I saw at a talk (though not a proper heckle) was “What is n?”
The speaker (Maxim Kontsevich) scoured the board, trying to figure out where he might have ever so briefly needed a variable, used n of course, and forgotten all about it.
Minutes later it transpired that the questioner had misread a mark; Maxim hadn’t used or written n anywhere.
Allen,
Good story. Usually “n” is clear, but have sometimes used the same letter twice.
The job-talk example (“wild goose chase” after facts that don’t exist) is particularly painful to read. Hopefully this will become easier to disprove as we have more capable information technology with us all the time — I’m finding some of my conversations get more efficient as, on points-of-fact, I can whip out my smart phone and just look up the necessary facts.
My single biggest professional regret is, when being told by a dean during a job-talk that “all he heard was bullshit”, I didn’t just end the meeting and walk out immediately, a la “I said good day!”
I was once heckled for using the term PAC in a talk on learning theory. The heckler opposed this term on principle and became very agitated — even when I offered to call this learning model by any name he wanted.
It’s every speaker’s nightmare … a heckler stands up in the middle of your talk and shouts:
Whoops … very sorry for the “blockquote” typo … please don’t hesitate to fix it. In any case, the story is true, and it is perhaps the single most famous “heckle” in all of physics.
John,
Great story about Feynman. And a great answer. It is hard to argue with success.
thanks very much
I will add Sir Eddington’s argument/nonbelief of Dr. Chandrasekhar over Chandrasekhar’s limit/blackholes during his talk in 1935. Unfortunately even the big name scientists who believed in Dr. Chandrasekhar’s calculations, they could not come to defend him because of Sir Eddington’s stature. You can read more on http://en.wikipedia.org/wiki/Chandrasekhar_limit
I was once heckled as a graduate student by a senior chemist at Berkeley for including the word “God” in the title of my talk (the title was in reference to a paper with a large number of authors who were bigwigs.) The title set him off because he was a dedicated atheist (so he started off by chewing me out), and then he spent 15 minutes harping on the first slide, whose content really merited only brief mention. It was great!
In my field of quantum computing one often has to deal with hecklers who think the entire field is crap and will say so directly to you, or imply as much in their questions, especially in physics colloquiums (computer scientists often feel the same way, but I’ve always had far fewer heckles in talks in front of computer scientists.) I actually really like these comments because it is a rare day that the heckler has enough knowledge to do a good job heckling (for instance they usually don’t even know the basics of quantum error correction, which puts them at a large disadvantage when arguing.) In fact, I think I had a good attempted heckler when I gave the physics colloquium at Georgia Tech 🙂
Dave, has it ever happened (to you) that a heckler subsequently apologized?
I have personally witnessed such an apology … on a grandiose scale! The then-director of our Seattle/Harborview level 1 trauma center, surgeon Marc Swiontkowski, invited the Russian orthopaedist Gavril Abramovich Ilizarov to lecture on his revolutionary new limb-lengthening methods. Prior to Prof. Ilizarov’s lecture, Dr. Swiontkowski stood up and said words to this effect:
Needless to say, Swiontkowski’s apology created a sensation … and Ilizarov’s lecture that followed was terrific! This was academia at its very best.
Another apology that comes to mind is Ed Wilson’s public apology to James Watson … in Wilson’s autobiography Naturalist we read:
These various quotes, by the way, derive largely from my database of STEM roadmaps, which historically have been lightening-rods for academic vituperation and apology.
It is greatly to Feynman’s credit that he made a similar acknowledgement to Hanbury Brown … I wonder how common heckler apologies are?
I have three examples: On me (nicely) hackling a speaker, on me being (nicely) hackled by a host, and a famous one on a speaker hackling the audience.
1) This was in 1985 and I sat in Richard Stanley’s class on summetric functions at MIT. Richard developed a complicated formula which involved all sort of expressions including
. At some point Richard made some complicated manipulation and divided by 
. “You cannot do this” I said, and pointed to the blackboard “you cannot divide by 
. Richard went over the computation and did not detect anythig wrong, but I insisted that dividing by 
is not a legal operation.
2) In summer 2006 (or 2005), one of the happy Seattle MSR summers, I gave a talk at Microsoft Research whose title was “Are there interesting?/ realistic? /universal? noise models that cause quantum computers to fail?” Jennifer Chayes introduced me and when she got to the title she said: “Are there interesting?/ realistic? /universal?” then she paused and continued “ ridiculous ? noise models that cause quantum computer to fail.” This was funny and I was quite happy with it because not only that coming up with all sort of strange and ridiculous adversarial models is in the tradition of TCS, but also I truly believed that ridiculously looking models is the right place to start and this is an expected long story.
3. Larry Shepp gave a talk at Columbia University and presented a new statistical model. Transparency 3 of his lecture ended with the claim: “This model is as profound and as interesting as any model in statistical physics…” This claim was regarded as an insult, caused some discomfort in the audience that soon turned into a heated debate. The lecture was interrupted for 10 minutes and the situation nearly escalated to a fist fight. When things calmed down and the lecture resumed, Shepp put the next transparency, which began with the phrase: ” …. well, perhaps not quite.”
When I was a student, our Data Base Theory professor told us the following hilarious story (I do apologize if it has almost nothing to do with hecklers, but I’d like to share it anyway).
During a conference, several invited speakers talked one after the other as usual, illustrating their articles. Each time a speaker finished his talk, the audience applauded clapping the hands, the Chairman stood up clapping his hands too and he introduced the next speaker. Now, for some reason, it happened that the Chairman fell asleep during a talk. The speaker who was talking in that moment was using several slides printed on those transparent plastic sheets. While he was manipulating his slides, unfortunately all such slides fell on the ground. So the speaker stopped talking and started to recollect all the sheets from the ground. There were long seconds of silence.
Abruptly, the Chairman woke up, stood up and vigorously clapped his hands, thanking the current speaker. The audience remained speechless. The Chairman realized that the talk was not finished yet, he silently sat down, visibly embarassed, and the talk continued.
Any example where what initially appeared to be heckling, later on turned out to be a beautiful and insightful question/observations ? In short – any instances where the speaker/listeners/heckler/science was greatly benefited by heckling ?
On a lighter note I wonder if there is a case where the speaker was heckled by his/her own grad students ?? If anyone knows such a case please share the fate of the grad students as well 😉
I recall a Christmas skit at Fermilab in the mid-1970s, during which an on-stage foreign graduate student joked of his advisor “He makes me check the tension in my shoelaces 90 times per day!” The audience of high-energy experimentalists laughed, understanding the joking reference to an obsession with photomultiplier high-voltage power supply calibration.
At the end of that academic quarter, the student’s fellowship was terminated, and he was forcibly deported to a home nation that was torn by a violent cultural revolution.
Several other sad cases come to mind … it would serve little purpose to post details.
Students should be aware that rather commonly academic advisors *appear* humorless, for the common-sense reason, that they are in fact humorless. If there is any doubt whatsoever in this regard, it is prudent for students planning a jocular public exchange to obtain explicit permission in advance … and even practice it.
This cuts both ways … it commonly happens that sober-minded students have their feelings badly hurt by glib-tongued advisors. If there is any doubt, ask the advisor “Do you mean that seriously?”
Thanks Dick for an interesting read. BTW, I think most of your anecdotes have appeared in an earlier blog post by Hank Whitt (http://tinyurl.com/3s5yr).
I found Zeilberger’s opinion very interesting, and I think (ignoring some minor inaccuracies) he makes some good points. It is a blatant wrong (no offense meant to Avi or anyone else who has asserted this) to associate the P vs NP question with solving the Riemann hypothesis (a constant sized problem), and I find (my fellow former postdoc of Avi) Scott’s defense of this point unconvincing. Scott suggests that we should talk about the problem of whether or not the hypothesis has a short proof (of length < n), but this is a completely different problem (whose efficient asymptotic resolution has nothing to do with whether or not RH is true or not), and is not relevant to solving the Riemann Hypothesis.
It also touches on a point that I have started to find annoying in complexity theory. We (in complexity theory) have built an industry around the assumption that P is different from NP. The fraction of papers that appear in the typical STOC/FOCS whose main results would become trivial if P=NP is very high, and the number of people working on such problems is much much larger than the number of people working on research that at least aims to move us closer to resolving P vs NP. That, to me, is a pretty sad state of affairs. This is only perpetuated by hyping up both the purported hardness of resolving P vs NP, and the forgone conclusion that P is not equal to NP.
The reason Zoron's response reminded me of this is that he points out another practice (the use asymptotics) that most of us in complexity theory rarely discuss, and I would guess rarely think about.
Anyway, I ramble. Good thing I don't have a blog.
I have always enjoyed:
http://en.wikipedia.org/wiki/David_Korn_(computer_scientist)#Korn_shell_and_Microsoft
“I sometimes have been the heckler without meaning to be”
Then, by the given definition, you are no heckler.
Tony Hey tells the story of what might be considered some famous heckling: http://tedxcaltech.com/speakers/tony-hey
For 13 years I was National Chairman of the Motorcycle Action Group UK, a bikers’ lobby group. At our annual conference I answered ‘Questions to the National Chairman’ and every year some perpetual malcontent would rise to ask long-winded multi-pointed – and usually nasty – questions aimed at somehow embarrassing me and showing how clever he or she was, invariably failing on both counts. On the most enjoyable occasion a ‘questioner’ went on at inordinate length in a whining voice, reading from a prepared but confused statement disguised as a series of multi-pointed questions in no particular order. The audience were sighing heavily and appeared to be losing the will to live. After about three minutes he lost his place in the sheaf of papers from which he was reading so seizing the moment I simply replied ‘NO’ and ‘Next Question’, at which point the entire audience roared with laughter. I am sure I could equally have answered ‘YES’ and they would have been equally as grateful.
I once attended a talk, directed to a mathematical audience but by a political scientist, on voting methods. The speaker, having taken Arrow’s theorem to heart, had decided that some of the criteria had to be thrown out, and showed how, miraculously, the remaining criteria on which he settled also perfectly supported the method on which he had settled. I forget what the method was, but it wasn’t approval voting.
Well, that’s fine, but the political aspect of the talk had been played up (it was near the 2004 elections), and it was well attended by non-mathematicians, and even non-academics. At the end of the talk, one of the audience members, clearly bewildered by all the mathematics, stood up and said “I’ve spent the last 20 years lobbying the state legislature to adopt approval voting, but you showed an example where approval voting can fail very badly. Are you saying that I’ve wasted my life?” The speaker responded simply “Yes.”
Hurrah for outreach!