“econ rules”
Having presented in an economics workshop and seeing economists in action, I think the norm is that you can interrupt whenever you want and can ask any question. Fierce argument is encouraged. For a while, I thought it was a cool idea. Who doesn’t want criticism? Aren’t we supposed to mercilessly question our assumptions and let the truth stand out?
After a while, I’ve decided that I don’t like “econ rules.” I have seen too many talks where people never get to their main conclusion. People waste time with questions that are satisfactorily answered in later slides. With experienced authors, interruptions are merely time wasters. They know the limits of the study already, have done reasonable jobs in addressing the issues, and will tell you about them in a minute … if you would just shut your mouth. The norm creates bad incentives for younger scholars as well. Instead of working at smooth delivery and clear presentation, talks can be bloated affairs designed to combat questions.
Don’t misinterpret this. I am not claiming that people should pull their punches in seminars. The quality of our work depends on strong criticism. But can’t criticisms be held until the person actually gets to their conclusion? Isn’t the point to hear a research presentation, not the posturing of audience members? If my time is worth money, don’t these endless questions cost me $40/hour (or more)?
orgtheory.net 
Isn’t the point to hear a research presentation
What a quaint idea.
As you surely know, it’s all a question of what such events are meant to convey to those present. (Hint: no-one is interested in the speaker conveying the content of their paper.)
LikeLike
Kieran
March 25, 2010 at 12:28 am
I agree with Fabio. “Econ rules” are rules made by people with absolutely no interest in substantive or theoretical issues and with a lot of interest in showing off. So talks degenerate into pissing-contests about who can catch the speaker into some sort of procedural or methodological flaw before the poor person can get to that slide in their talk.
Brian Uzzi just gave a talk to a room full of finance types here at the ND business school about a month ago, and by the 30 min. mark I was ready to canonize him as a saint. If it was me up there I would have jumped over and strangled the blowhard who kept coming up with tertiary, meaningless objections regarding methods and measurement every five minutes and seemed to be getting none of the substance of the talk much sooner than that.
LikeLike
Omar
March 25, 2010 at 12:45 am
If the real purpose is to listen to others listen to themselves, then I’d just watch a congressional hearing on C-span. At least I can watch it on my own time.
LikeLike
fabiorojas
March 25, 2010 at 12:51 am
Part deux: If the real purpose is watching dominance displays, I’d rather watch the Animal Planet.
LikeLike
fabiorojas
March 25, 2010 at 1:04 am
While I agree that what you call “interrupt-throughout” seminars can pose problems that need to be managed (e.g., the host should shut down the questions if it looks like the presenter is running out of time; and there should be a general attempt to manage the culture so that the questions are productive), I largely disagree with you guys. Two reasons:
1. The 40 min+q&a model can be very unproductive when at minute 3 the author made a problematic characterization of a literature, at minute 6 the presenter made an error in logic, at minute 11 there were highly questionable issues of data, etc. Well before minute 40, the listener has often lost interest because the paper is built on all kinds of assumptions or moves that are very difficult to digest. And then the q&a session is completely disjointed, moving from theory to method to literature, back to method, etc.
2. I agree that the 40 min +q&a works well when the presenter has put together a good paper (so the above problems don’t crop up). And I would agree that the best presenters are often the most experienced. But there is a great deal of variance, with some of the most senior people often giving awful papers. I just cannot endorse this statement of Fabio’s: “With experienced authors, interruptions are merely time wasters. They know the limits of the study already, have done reasonable jobs in addressing the issues, and will tell you about them in a minute.” Do you guys really believe this? My own experience is that “experienced” authors often get complacent and lose track of the need to present coherent arguments and compelling data (analysis). And I also think that it is very good for students to see that senior and junior people alike do not get treated with kid gloves, but are subject to the same standards.
LikeLiked by 1 person
ezrazuckerman
March 25, 2010 at 2:28 am
There’s nothing worse than when sessions end up like this!
Question timing obviously helps, but (riffing a bit on Omar’s point) isn’t the core issue actually: how do you develop rules, structures and norms that discourage seminars (etc) turning into destructive status auctions/pissing contests?
If we can re-frame the problem in this manner, then a fair bit of literature exists to help us begin thinking about the sorts of rules, structure or norms that are effective at containing such action. Sutton and Hargadon (ASQ 1996: 706-7) write a bit about this and Emmanuel Lazega often touches on such problems in his studies (e.g. with Pip Pattison in Lin, Cook and Burt, 2001).
LikeLike
sam_macaulay
March 25, 2010 at 2:33 am
E-Z: Unless the paper is a complete mess, I think I’d prefer to see the dismemberment at the end. Too many papers are in the middle of the quality range, and I think I’d really need to see the whole paper before making a judgment. Social science papers are not mathematics, where one logical mis-step undermines the entire story. Instead, social science is a patchwork of imperfect models and data. So I really need to see the whole thing.
BTW, senior scholars do not have a monopoly on good or bad work. I never said that. But my experience is that, on the average, they have thought through typical objections. That’s all I meant. I did not say we treat senior folks with kid gloves, but, at the least, let them finish the sentence!
Finally, let me empirical, I’ve been to dozens of talks over the last few years. And I think there have only been two or three cases where it was a complete disaster from beginning to end that needed immediate intervention. Instead, scholars just get interrupted on insanely minute points that lead to no where. Once, as I was leaving a talk, the speaker asked if she could finally address my question after nearly an hour of interruptions. I bluntly said, “I am no longer interested, I have to leave.” That is how I honestly feel about talks these days. I am not here to listen to grade school contests. I love research, but my enjoyment of talks has plummeted when people employ “econ rules.” If people love talks with no conclusion, just endless bickering, just perhaps I’m just too weird.
LikeLike
fabiorojas
March 25, 2010 at 2:46 am
Fabio,
How do you feel about, for example, the Michigan Econ Soc Workshop, which uses an “econ-style”, but premised on the understanding that everyone has already read the paper? Perhaps it’s a totally different beast, because there is no need to “get to the conclusion”, as everyone has already read it, but I just want to check to see if you are lumping them together.
LikeLike
Dan Hirschman
March 25, 2010 at 3:25 am
Workshops where people know the point and then argue are cool with me. At UCLA, I had a crazy time in a seminar, but it was cool. I gave a short presentation and then the knives came out. But if people are expecting a “talk,” doesn’t it seem logical to like, hear the talk?
LikeLike
fabiorojas
March 25, 2010 at 3:42 am
Dan: I’d also add that I pay attention to the quality of question. The type that Omar mentioned drives me bananas. At least in 40+QA you don’t have to listen to Dr. Whack till the end of the talk.
LikeLike
fabiorojas
March 25, 2010 at 3:48 am
As a presenter, I actually prefer giving a talk under the econ rules. It’s more lively and feels like an academic discussion rather than a lecture. At its best, the format is intended to create a discussion about the research, much like a workshop, which can potentially help the presenter identify holes in the paper’s logic, improve the study design, etc. The feedback you get from others’ gut reactions to something you’ve said, which is what is happening in most econ rules talks, is often right on target. Not always, but often. Even if it misses the overall point, someone’s gut reaction can tell the presenter that he has framed the paper in a confusing way.
At its worst, econ rules can lead to tangents, usually resulting from poorly chosen questions (e.g., why didn’t you ask this other question that I’m really interested in instead?). In most seminars, there is one person with a reputation for asking that derailing question. The presenter just has to be prepared for it, try to get something out of it, and move on. If you have an audience that is engaged with the work and is really trying to help the presenter improve the work, then econ rules can lead to a good experience for everyone. I’d say that the 2 or 3 most productive talks I’ve given, where I learned the most about my work or got new ideas for future research, have all been in talks where questions were flying around from the get-go.
LikeLike
brayden
March 25, 2010 at 4:25 am
@Dan: If everyone really has read the paper, it would be absurd for the speaker to give a 40-minute talk reiterating it (unless the point is to practice for a presentation elsewhere); I assume that’s not how your Econ Soc workshop proceeds. I’d think the speaker should then give a 5-minute summary of the key point and the main issues that they’re most interested in getting feedback on, then let the audience get to whatever they want to get to. (Or if the speaker had things to say that were truly extensions of the paper, that would work too.)
In fact, there’s almost nothing I find more annoying than when someone asks me to read a paper and then lectures at me as though I had not read or understood it. But that’s different from a talk where the material is new to you.
Status games aside, I wonder if part of the problem is a disagreement about whether talks are mostly to present new material that the audience will need to hear in full to comment on, or an opportunity for discussion that takes advantage of the audience mix.
Regardless, it seems that the worst compromise is to structure talks as though they’ll be delivered uninterrupted for 40 minutes (so the main points aren’t necessarily clear at the beginning) but have the audience contribute from the beginning. Is that really how econ talks work, though? It’s far from my field, but I’m surprised the talks aren’t just set up to work as seminars rather than lectures.
LikeLike
Elizabeth
March 25, 2010 at 6:10 am
“Regardless, it seems that the worst compromise is to structure talks as though they’ll be delivered uninterrupted for 40 minutes (so the main points aren’t necessarily clear at the beginning) but have the audience contribute from the beginning. Is that really how econ talks work, though? It’s far from my field, but I’m surprised the talks aren’t just set up to work as seminars rather than lectures.”
Agreed. I think I’d be so much happier with the 5 minute intro, followed by open discussion.
LikeLike
fabiorojas
March 25, 2010 at 6:27 am
I agree with everything Brayden said about which format is better for the presenter. As does every presenter, I typically like to think that I am in the position that Fabio described as true for the “experienced presenter”– i.e., “know the limits of the study already, have done reasonable jobs in addressing the issues”, etc. but of course I always overestimate the extent to which this is true, and I’m much more likely to figure out where I went wrong if someone asks me what’s bothering them early on. And as Brayden says, experience with this format gives the presenter a facility with managing the crowd (and the annoying guy in the room who won’t shut up; [a tip: the key to managing this guy is that everyone else in the room hates him too; so they will love it if you politely but firmly get him to shut up). More generally, I think the divergence in this discussion has more to do with relative experience in the two different systems than anything else. Another point is that while there is relatively little variance in what 40+q&a feels like [precisely because the 40 min are completely controlled by the speaker] there is more variation in how ‘interrupt-throughout’ works or does not work. So yes, higher variance and it is up to those running the seminar to manage the norms.
LikeLike
ezrazuckerman
March 25, 2010 at 9:35 am
Fabio: Having trouble with this line: “Social science papers are not mathematics, where one logical misstep undermines the entire story. Instead, social science is a patchwork of imperfect models and data.” I agree that social science often must rely on observational data that are imperfect, but it’s hard for me to see why our theories cannot be as tight as anyone else’s. Models are models after all. They may be externally invalid, but it would be nice if they were internally valid. That being said, I agree that many of the theories that have been very productive for social scientific progress have been logically flawed in one way or another (often because of hidden scope conditions/assumptions). But they would be even more productive if their flaws had been noted and ironed out when they were first advanced, and progress comes from ironing them out now. And this seems just as true for social science as any other branch of science.
LikeLike
ezrazuckerman
March 25, 2010 at 9:46 am
I’m with Ezra on this one.
It’s worth thinking about why we have talks anyway (beyond as an excuse for flying the person out to create social ties to him or her). The “interrupt-throughout” seminar serves multiple purposes:
(1) It’s feedback for the author that can help the author to improve the paper. Even if questions arise that the author has already anticipated, the fact that those questions come up where they do suggests that the presenter probably did not order the ideas in the paper (or at least in the presentation) effectively.
(2) It can allow the audience to move past problematic points of the paper (as Ezra notes). Often times, a paper that on reading seems weak gains credibility if it becomes clear that the problems stem from the language or the writing (as opposed to from a flaw in the logic or the methodology).
(3) It allows brainstorming among the audience. Even bad papers sometimes stimulate a really interesting flow of ideas. Those ideas might help the presenter but they might also become topics that someone in the audience chooses to pursue.
(4) It’s a form of practice. Students attend seminars so that they can practice forming logical arguments and critiques. As faculty, we usually stop taking classes. But that doesn’t mean that we cannot benefit from practice. Interactive talks help to keep the faculty sharp.
It’s not to say that the interactive style is perfect. Presenters unpracticed in running discussions can find it difficult to get through their material. Groups that adopt the format of continuous interruption but not the intent of improving scholarship can devolve into petty one-upmanship. But it seems a reasonable risk when one considers the alternative.
LikeLike
olavsorenson
March 25, 2010 at 9:49 am
Complete agreement with Olav, esp on points 3 and 4. And I’d add that that kind of brainstorming is invaluable for building a community of people who, while still diverse in their interests and research modalities, have a shared sense for what makes for strong and weak research.
LikeLike
ezrazuckerman
March 25, 2010 at 9:59 am
Unsurprisingly, all the B-school people like the econ model. Socialization or selection effect? Oh noes!
A well-run seminar on the econ model can be great. But a well-run traditional talk can be great, too, and provide the same benefits Olav lists — what’s making the difference there is that it’s a well-run event, not that it’s of a particular form. I should say that I’ve often presented in econ-style settings myself (including in front of Ezra), and only had one experience where I really wanted to tell someone to shut up and listen (not in front of Ezra). Then again, I spent four years in college standing up in front of and being heckled by large groups of Irish students, so I have a pretty thick skin.
I do think it’s a little too charitable to characterize things as “there’ll just be this one problematic guy everyone else hates so just deal with him”. In my view the model encourages posturing that’s quite disconnected from any actual high-quality critique or intellectual engagement or whatnot. It’s perhaps an advantage of the econ model that the quality norm can be managed at the seminar- rather than the field-level, as a well-run seminar can’t be hijacked by a badly-prepared speaker.
LikeLike
Kieran
March 25, 2010 at 11:03 am
Hmm… all the bschool people who’ve commented have their degrees from soc departments. So we have a lot of experience with both models. But yet, I’d say that the reason i continue to be in a bschool is not unrelated to a preference for this style of engagement.
LikeLike
ezrazuckerman
March 25, 2010 at 11:22 am
I should hasten to add that I agree with Kieran that a successful seminar is less about the form than it is about how well-run the seminar is. Also, I didn’t mean to imply that managing the a-hole in the room is the only challenge that one faces; but it is not an infrequent challenge. Finally (back to work!), I’d quibble with the idea that the interrupt-throughout model (shocking that I don’t like the moniker ‘econ model’!) “encourages” posturing, but I will agree that it gives such posturing room, such that it needs to be managed; on the other hand, I have found a great deal of an equally annoying form of posturing in the 40+qa model (which I believe Olav was alluding to with his line about the ‘social ties’) in which the hosts spend an unseemly amount of time kissing the speaker’s rear-end (and vice versa).
LikeLike
ezrazuckerman
March 25, 2010 at 11:35 am
Other than job talks, I *never* see the point of no-one’s-read-the-paper 40-minute-lecture sessions among faculty. I can read the paper faster than you can read it out loud. The point of all getting together in person is to exchange questions and ideas (gladiatorial style or otherwise), so I sympathize with the urge to jump right in– but for that to result in a non-stupid discussion, reading the paper in advance seems like a sine qua non.
The very best workshop I’ve ever regularly attended, works in progress at Chicago Law, has the mood of an econ-rules seminar, but with 5 minutes of opening remarks following everyone having read the paper, not a 40-minute power-point to interrupt. Sometimes the speaker doesn’t get through the 5 minutes, but that’s not the same kind of problem Fabio’s talking about.
LikeLike
Jacob T. Levy
March 25, 2010 at 12:19 pm
One point that hasn’t been mentioned is that, in my view at least, criticism isn’t only about identifying mischaracterizations of literature (or historical background, etc.), confusing concept use, logical flaws in argument or model specification, etc. (as troublesome as these may be), but also about assessing the consequences of these issues for the overall claims being made. Sometimes the consequences are major, but sometimes quite minimal. Assuming no paper in advance, it seems that there’s likely to be a lot more precious discussion time used up on trivial considerations if heated debate preempts the presentation of the full argument. (Although disciplined interruption along the way for clarification, with criticism saved for the end, probably does make for better critical discussion later.)
LikeLike
Rob Jansen
March 25, 2010 at 1:54 pm
I don’t agree with Omar at all. “Econ rules” were not made by anyone. They are the natural evolution of what is thought to be the best approach to present research in the area. Nobody sat and said: let us create this rules. The just do not work in that way.
Contrary to sociologists, economists are more concerned about finding the truth than in revealing it through an argumentative process. The former requires technical and theoretical debate from the listeners, the latter a passive attitude from the audience. If you are not up to the job, don’t take it.
LikeLike
somecondude
March 25, 2010 at 5:33 pm
It seems to me that somecondude’s model of the truth-finding process is severely underidentified. His confident bluster is of reasonable quality, however, and so he should do quite well.
LikeLike
Kieran
March 25, 2010 at 5:39 pm
A few responses:
@ Ezra:
“Having trouble with this line: “Social science papers are not mathematics, where one logical misstep undermines the entire story. Instead, social science is a patchwork of imperfect models and data.”
I agree that social science often must rely on observational data that are imperfect, but it’s hard for me to see why our theories cannot be as tight as anyone else’s. Models are models after all.”
I’m sorry, but social science (including econ) is simply not as “tight” as mathematics. Math is pure logic based on a handful of assumptions. That’s qualitatively different than having a verbal model, which is translated into a mathematical model, which is then translated, imperfectly, into a set of empirical hypotheses, which are then tested with data which is imperfectly collected and imperfectly measures what we care about. And that data is crunched, imperfectly, into a regression or computer simulation.
The way it works into seminars is that a pure math workshop must identify every single imperfection. If any is found in a mathematical proof, it must be immediately corrected. If it remains uncorrected, there is simply no point in continuing the talk. In contrast, social science can still continue with imperfections and much can be accomplished by looking at the whole package. In other words, math is absolute, social science is Bayesian.
@ Jacob and others: I feel that the ultimate issue is management. Sure, it’s *possible* that econ rules encourage great interaction. *Some* people manage it well. But my average experience is underwhelming. I’ll be blunt: “econ rules” encourage a handful of senior faculty to posture and bicker and hog up time.
LikeLike
fabiorojas
March 25, 2010 at 5:43 pm
Perhaps predicable, since I’m at a b-school as well (and have signed the oath), but I like both sides of “econ rules” — as an audience member I like being able to ask questions/address points/challenge assumptions (while the matters are still pertinent and fresh, rather than 40+ min later), to engage in dialogue and debate, etc. From the presenter side, I enjoy getting immediate feedback and working through the logic via questions, defending the approach and assumptions, etc. For me, “econ rules”-type interaction is a big part of what scholarly interaction is about (of course, depending on purposes, needs, etc).
Now, “econ rules” can of course deteriorate into utter chaos (there’s sort of meta-rules for how econ rules play out in given contexts, depending on interactions and who is present), which can be (at times) outside the control of the presenter — e.g., someone in the audience is out to challenge the very foundations of the discipline and you just can’t shut them up. But, on the whole, we’re all big kids, I think.
LikeLike
tf
March 25, 2010 at 6:20 pm
Incidentally, I think the implicit commenting rules and culture here at orgtheory.net is similar to “econ rules.” (Well, that’s probably not the best comparison and I’m not sure there are meaningful alternatives — though I think different blogs essentially have different commenting cultures and orgtheory.net’s is definitely more “econ rules”-like.)
LikeLike
tf
March 25, 2010 at 6:33 pm
TF: Orgtheory is not “Econ rules” – nobody interrupts me while I write the post!
Seriously, “econ rules” doesn’t mean “critical” or “tough.” You can be tough and critical in many ways. “Econ rules” means you can jump in and wrestle as the talk progresses, instead of letting people finish the point.
LikeLike
fabiorojas
March 25, 2010 at 6:41 pm
somecondude,
the problem with the “just evolved that way” idea is that soc rules are an unstable equilibrium that has to be enforced rather than allowed to emerge. as soon as one or two people start interrupting the speaker, it becomes clear that there will be no time for Q&A at the end, which means that people who would have preferred to wait for the end will have to ask now. so even if a strong majority of the audience prefers soc rules, it can quickly tip to econ rules.
ezra and olav,
as a speaker, i usually prefer to get feedback at the end because then i’m in “notetaking” mode rather than “presenting” mode and i find it hard to switch back and forth from the active to passive mode and back again every two minutes. a big part of this is that when i’m still giving the talk, i’m trying to get back to the presentation ASAP.
fwiw, a couple years ago i saw one speaker give the same talk twice, once at a soc department and the other time at a b-school (using econ rules). at the b-school the talk didn’t flow well, the speaker kept getting flustered, and the audience came away somewhat skeptical. i don’t think the issue is that the econ rules were better at exposing weaknesses in the study, but rather (and i say this as someone very familiar with the research at issue) it simply threw up innumerable mostly spurious distractions that made it hard to follow the presenter’s argument.
LikeLike
gabrielrossman
March 25, 2010 at 6:52 pm
a few years ago Mankiw had a conversation with an HMS professor who said that economists have many good qualities but asked why:
The economists are the most aggressive people in the room. They have little patience for introductions, motivation, or “being nice”. They want to spend the first 10 minutes trying to figure out the -entire- talk. If they’re not happy, they tend to disengage. I will note that they’re like this with each other also. Why are things this way in economics? There must be pluses and minuses to this way of interacting.
LikeLike
gabrielrossman
March 25, 2010 at 7:00 pm
The economists are the most aggressive people in the room. They have little patience for introductions, motivation, or “being nice”. They want to spend the first 10 minutes trying to figure out the -entire- talk. If they’re not happy, they tend to disengage.
See, I really think this is self-serving tripe. Economists in my experience are in practice highly attentive to status, and can be as deferential as the next supplicant if the moment requires. But the status dynamic is expressed by way of this “no bullshit / just give me the argument / cut to the chase / who’s the smartest guy in the room / I don’t suffer fools gladly” schtick that is itself a distinctive kind of BS. You see it in other fields, too. The interactional style rewards a certain kind of quickness and flashiness in conversation that’s not that strongly correlated with talent, and it tends to reinforce a collective commitment to punishing deviation from key methodological or theoretical premises with instant negative sanctioning. Even worse, in many cases it encourages people to develop various tics (head-grabbing, face-pulling, incredulous stares, intensive slouching, etc) meant to signal one’s high degree of personal genius.
LikeLike
Kieran
March 25, 2010 at 7:09 pm
“Orgtheory is not “Econ rules” – nobody interrupts me while I write the post!”
Not yet, but you’ve just come up with a great idea for a new interactive blog tool!
LikeLike
brayden
March 25, 2010 at 7:11 pm
Three points:
1. Re Gabriel’s comment and related earlier points about how talks tend to get bogged down under “interrupt-throughout” rules:
(a) I agree that there is a risk to this which must be managed; and (b) my experience is that this is invariably the sign of a bad presentation/paper in which the author/presenter has failed to make clear what is important or unimportant– and this is usually because the author/presenter does not actually know what kind of contribution s/he is trying to make or cannot commit him/herself to a particular point. A presenter/author who does know what is at stake generally does not have difficulty saying– no, you guys don’t get it; that point is relatively unimportant given what I’m trying to do here, and here’s why. Now, this can sometimes be difficult in an interdisciplinary setting where the presenter and audience differ on what makes for a significant contribution, but that communication divide obtains regardless of the format of the seminar. And when a speaker does not know why his/her paper is important (or has the wrong idea about it), the paper stinks regardless of format.
2. On the ancillary point raised by Fabio about dirty models and dirty hands (cf. http://www.jstor.org/stable/657726), two cavils: (a) You seem to be confusing “is” and “should”. Sure, a lot of social theory is poorly constructed. But I just don’t see what is it about our endeavor that excuses that. It’s hard for me to see how a paper would not benefit (at least the audience, if not the presenter) from identifying a logical flaw in an argument; (b) I also suspect that your model of how math seminars work is idealized. I have in mind a story from a friend of mind who is a prominent MIT computer scientist, who told me how they rush to get papers for their main conference; stuff is published in the conference proceedings; it has an influence on the literature [conference proceedings are more important than journals]; and then years later, it may or may not come out that they couldn’t get the proofs to work. More generally, my experience has been that we idealize natural-science governance mechanisms at our peril. Those in social science tend to be at least as good (e.g., we do double-blind; they do not), though that may be damnation with faint praise.
3. Complete agreement with Kieran about posturing in econ seminars. But as I have said, we sociologists have our own brand of posturing. And as I hope the comments from “interrupt-throughout” backers should have articulated, that model need not devolve into such posturing. There is variance, and the risks need to be managed.
LikeLike
ezrazuckerman
March 25, 2010 at 7:21 pm
But as I have said, we sociologists have our own brand of posturing
Yeah. Unfortunately it takes the form of passive-aggression. Like regular aggression, only not even enjoyable for the instigator!
LikeLike
Kieran
March 25, 2010 at 7:25 pm
LOL on the passive-aggression. Though passive-aggression is so popular, you’d like to think someone is enjoying it. (BTW, I had in mind the ass-kissing; ah– so many forms of posturing, so little time…)
LikeLike
ezrazuckerman
March 25, 2010 at 7:30 pm
For what it’s worth, math talks don’t follow econ rules at all, since people present overviews of questions rather than proofs in progress.
LikeLike
yes
March 25, 2010 at 8:05 pm
But unlike economics, math is notoriously unconcerned with rigor and truth.
LikeLike
Kieran
March 25, 2010 at 8:07 pm
Well, while you can’t intercept and speak over posts (yet anyways, I agree with Brayden: someone ought to work on that technology), I’d still say that we’ve got a pretty strong “econ rules”-type culture around here, and yes, I’d equate the tenor of your post/”econ rules” with tough and direct and immediate questioning, not beating around the bush etc when commenting even on blog posts. There can, of course, be a certain hijacking that goes on in the comments section as well, but, it’s a democracy (you can even be anonymous, if you’d like) and anyone can drop in a comment to blast or support or add to whatever points are being made. And, it’s usually all in the right spirit, I think. I remember talking to a potential guest blogger once who expressed very strong reservations about blogging at orgtheory due to an all-too direct commenting culture. Despite being blasted several times myself (with many more times to come, I’m sure), I like it, a lot. So, just like (ok, sort of like) in a seminar, it means that you can’t just say whatever you like (well, ferret etc posts aside), as readers/audience members will provide a “check” on your points/assumptions.
(But, ok, admittedly my comparison between blog posts and commenting culture and seminar talk rules is not the best and breaks down in lots of ways.)
LikeLike
tf
March 25, 2010 at 10:07 pm
A few comments to various folks:
1. You’d think that I recommended refraining from pointed criticism. I didn’t say that at all. Rather, my argument is about management of talks. I think criticism is better when you’ve allowed people to make their basic points, rather than endlessly interrupt them. My sense is that “econ rules” are specific to econ. Other fields, often technical ones like math, use different talk rules. This makes me suspect that interruptions are culturally specific to economics, not a required practice for producing quality work.
2. To Ezra: I never said anything about computer science, which is fundamentally different than math. CS is engeineering, math, stats, logic, and stats all rolled into one. It’s also a super fast field, which incentivizes “quick and dirty” talks and publications. Math is purely about deductive proof. Period. Pick up a copy of the Bulletin of the AMS and you will see it is 99% proofs. As “yes,” notes the talk rules are different in math. When I was a math student, “overview” talks were almost never interrupted. Workshop talks were designed to vet proofs, which means literally checking every step of a proof. But the tenor tends to be different. You don’t see the inane practice of econ talks. Unless the person is a complete novice or they say something crazy, you let them at least sketch the proof before arguing over details. People in math are crazy, but they’ll at least wait to see what the argument is. The no-holds intense criticism of works in progress is usually reserved for small, private sessions.
3. To Ezra II: This is not a “dirty hands” issue at all. My claim is that every single social science paper ever has imperfections, even though I will admit some papers are stronger than others. I am very much Bayesian in how I evaluate social science. If you buy my point, and you don’t have to, then it makes sense to listen to the whole argument. For example, all survey research has limits. Period. But I am not going to harrass a speaker for using survey data. I’d prefer to listen to the argument and see if the final results are affected by the limits of survey data, or if he can glean some insight from survey data. That’s because social science papers have a different logical structure than mathematics. Social science is a series of plausible steps, math is a sequence of completely logical steps. Social science, as a few commenters have noted, can also be judged in terms of impact – we’ll accept imperfect data for good, but hard, questions. These considerations indicate that trying to grasp the whole picture is an element of evaluation that isn’t quite as urgent as in purely analytical fields, like math or philosophy. For that reason, I believe in harsh criticism, but usually after the whole argument is revealed.
LikeLike
fabiorojas
March 26, 2010 at 1:36 am
I want to qualify my previous statement. I am not categorically against econ rules. My problem with the whole thing is the “econ” not the “rules.” I’ve given talks under so-called “econ rules” and actually enjoyed them a lot (the last one was at MIT econ soc seminar; hi Ezra!). In an epistemic culture in which everything is about “identification” and nothing is about substance and theory then econ rules degenerate into something highly dysfunctional. So I’m coming out as absolutely bigoted on this: econ rules are OK, as long as it is people that I trust to be asking good, pointed questions in the audience and not people trying to turn the whole notion of organized skepticism into a caricature of itself.
LikeLike
Omar
March 26, 2010 at 2:26 am
Fabio:
1. I confess ignorance re how things work in mathematics. Happy to be enlightened. (And that distinction from CS is reasonable, though not sure folks in CS would be so comfortable with it. My friend was basically discussing it as dirty laundry, not as a reasonable trade-off they made in the name of [fast] progress).
2. The issue was models, not hands. That is, I was objecting to the idea that we should tolerate errors of logic in the course of a presentation or paper. (So not sure I or anyone else implied that it might be legitimate to object to somebody’s paper bc they were using survey data). Anyway, I’m happy to concede that there are some papers for which it is worthwhile to hear the whole argument before delving into it. It happens to be my belief however that this is not true for the majority of papers. But the data available to me to prove my case are quite dirty…
A final note is that while I think fine to accept Omar’s distinction between econ rules and actual econ talks, I have to say that the econ talks I have attended have been quite civil. On the other hand, there was that seminar I went to where a hotshot young economist introduced her model by saying that ‘there has been a lot of work lately in which economists use ideas from psychology. But I have always thought that the discipline that we are closest to is sociology. That’s why it is too bad it is such a weak field.” Whereupon she put up a model that was a total mess and a feeding frenzy ensued. I got up and walked out. Sometimes those econ rules ain’t so bad… :-)
LikeLike
ezrazuckerman
March 26, 2010 at 2:42 am
Omar, While I appreciate the distinction between interruptions during a talk and having an audience of economists, I’m not comfortable with dismissing concerns about identification as lacking substance.
When economists raise issues about “identification”, they are essentially questioning whether the results might be spurious (i.e. that the results might reflect some process other than the one being claimed by the author/presenter). Unless someone’s presenting a purely theoretical argument, the validity of the empirical analysis is certainly a substantive issue.
Ezra: I think that we saw the same talk…
LikeLike
olavsorenson
March 26, 2010 at 10:57 am
I prefer the econ rules both as presenter and listener. However, I wish there would be some self-imposed restrictions by people asking questions. Some questions I personally find very irritating:
1. Questions relating to empirics when theory is discussed.
2. Questions that are likely to be answered within three minutes anyway.
3. Repeating the same question in a disguised way three-four times during the seminar.
4. Minor questions when there is 15 minutes left of the seminar and the presenter hasn’t shown the key results yet.
In my experience, the above is not a massive problem but it does occur too often.
LikeLike
matslingblad
March 28, 2010 at 2:43 pm