Why I love a good argument (and what makes an argument good)

We don’t have a formal comment policy on this blog. I don’t think they’re very useful, and I don’t think we’ve ever needed one. But the amazing, huge comment thread on Brian’s “statistical machismo” post convinced me that I ought to elaborate a bit on what we’re hoping for in our comment threads. The comment thread on Brian’s post got rather heated in a couple of places, not least because of some sharply-worded comments from yours truly. I did end up blocking one comment, and one or two readers expressed discomfort with how “personal” the thread got. I actually thought it was a great thread that never got personal except in one or two isolated places. I love a good argument in the comments; let me explain why.

Attention conservation warning: this post turned out much longer than I originally intended. Sorry! As I was writing, I kept thinking of more and more points I wanted to make. So if you’re not super-keen to hear every single opinion of mine on blog commenting, you may want to just skim, or go read another post.

I love it when lots of folks comment and a conversation gets going. That’s what blogs are for. It’s much faster and easier for people to exchange views on a blog rather than via things like comment-reply exchanges in a journal. And I love it when people are passionate and excited in their comments. That’s part of what makes blog posts and comments fun to read and attracts the critical mass of people you need to have a discussion (since the vast majority of readers never comment). Why would anyone want to read a science blog on which the posts and comments were written in the dry, formal style of journal articles? That a discussion is conducted in strong, passionate language does not imply that it is just a pointless shouting match, like rock fans vs. rap fans (this is something I think folks who don’t read blogs often fail to appreciate).

Passionate, excited comments also are a reliable signal that the commenters care a lot about the topic and find it important and worth discussing. Which is a good thing. Indeed, if your commenters aren’t passionate and excited about a topic, you won’t end up with a dry, boringly-written comment thread. You’ll end up with no thread at all, because no one will bother commenting.

But if those passionate, excited commenters disagree (with one another and/or with the post author), there’s a risk that things will get heated. Here are my thoughts on that:

• Racism, sexism, insults, etc. are out. Obviously! I’ve never run into anything like that on this blog and I’d be stunned if I ever did.
• You can’t prevent heated debate. People will mostly write how they want to write. And if they really care about what they’re writing about, they’re going to use strongly-worded language sometimes. My experience reading and writing blogs is that you mostly can’t prevent that by telling people how to express themselves.
• Opinionated is good. Or at least, not necessarily bad. Just because someone has strongly held opinions does not mean that they don’t appreciate or understand opposing opinions, or that they don’t respect those who hold opposing opinions. And sharp, opinionated writing is fun to read, and prods the reader to pay attention. Conversely, having only weakly-held or tentative opinions, or no opinions at all, isn’t necessarily good. It may just mean you don’t know much about the issue, or are confused, or are just trying to avoid conflict at all costs, or are unwilling to take sides on an issue on which there is no sensible middle ground, or etc.
• You can’t please everyone. Some people are much more easily upset or offended than others. If you block every comment that could possibly upset anyone, you’ll end up blocking pretty many comments, and so won’t have many people commenting. In response, lots of readers and commenters will leave and never return. I don’t like that trade-off.
• Strong language is fine as long as it’s directed towards ideas, not people. It’s not ok to call someone an idiot, that’s strong language directed at a person. But it’s perfectly ok to, say, tell them that whatever they just said is a totally fallacious zombie idea that deserved to die years ago, for reasons X, Y, and Z.* This sort of thing may not be the most effective rhetorical technique, but it’s not personal and not offensive. I recognize that this isn’t a universal point of view. People who are passionate about ideas often feel attached to their own ideas, and so may take it personally when their ideas are criticized. But I think people can, and ought to, set that feeling aside. You and your ideas are two different things, and no matter how wrong anyone says your ideas are, you shouldn’t take it to imply anything about you. Science is hard, and so everyone gets some things wrong. Like, really wrong. Even in their area of expertise. Einstein was wrong about the randomness of quantum mechanics (saying “God does not play dice.”)–which didn’t make him any less of a genius. I’m not saying you need to “grow a thick skin”; I’m saying you need to recognize the difference between your skin (which is part of you), and your ideas (which are not).
• Taking criticism of your ideas personally makes you especially vulnerable to cognitive biases. And cognitive biases prevent you from participating constructively in debates. If you feel like defending your ideas is the same as defending yourself, you’ll be prepared to find any excuse to cling to what you believe, to ignore contrary evidence and arguments, and to dismiss the beliefs of others. It’s pretty much impossible for people to free themselves of cognitive biases–but as scientists I think we’re obliged to do the best we can.
• It’s not polite or respectful to pull your punches. In my experience, it’s a bad idea to try to be polite by not coming out and saying exactly what you mean. Far from preventing a debate from becoming heated or personal, it just introduces confusion into the debate by making unclear precisely what you’re really saying. It’s also likely to come off as patronizing, as a refusal to engage with your opponents and take them seriously. This is what happened in a heated debate I was involved in on another blog once. I got so frustrated that I stepped over the line, for which I make no excuse. But that doesn’t change the fact that I wouldn’t have gotten frustrated in the first place if my opponent had stopped trying to be polite and focused on the issue at hand (in this care, whether a statistical statement in a Science paper was correct or not). So don’t try to be oblique, or drop hints, or be understated. None of that works on blogs the way it does in face-to-face conversations.
• Just because someone uses strong language doesn’t mean they’re “arguing to win” or “showing off” or trying to “shout their opponents down”. I push hard in arguments not because I insist on “winning”, but because I want to get things right. I’m more than happy to change my mind, because that means I’ve really learned something. But if you want me to change my mind, you’re going to have to give me good evidence and good arguments. I think the same is true of pretty much everyone I know. And no, when I push hard in arguments I’m not showing off, as evidenced by the fact that I push hard even in private conversations with close friends. I’ve never met someone who argues just to show off. And as for the notion that using strong language amounts to “shouting someone down”, um, no. Blogs are written, not spoken. Nothing anyone says in posts or comments physically prevents other people from making comments, or reading anyone else’s comments. The blogging equivalent of shouting someone down would be blocking comments I disagree with rather than replying to them. I don’t do that, and again, neither does anyone I know.
• There are lots of ways to help prevent strong language about ideas from derailing a productive discussion. Even on a blog, where you lack lots of tools available in face-to-face conversation, like facial expressions and gestures. Here are some ways to ensure that a heated debate remains productive:

1. The best way is to back up your claims with evidence, logical arguments, and links to sources. That keeps the focus on your substantive points and encourages (indeed, forces) others to engage with your substantive points, even if they’re couched in strong language.** Indeed, if you know you will be criticized if you don’t back up your claims, you’re more likely to back them up, so allowing vigorous debates can actually make those debates more productive. Backing up your claims also shows respect for those with whom you disagree–you’re not just asking them to take your word for anything, the way you would with a small child. They’ll show you the same respect in return. One reason the “statistical machismo thread” was great was that commenters mostly backed up their claims. Another reason it was great was that Brian went to the trouble of checking a lot of those claims. He actually took the time to go back and re-read in detail a bunch of papers commenters cited, in some cases discovering that the commenters (and sometimes most ecologists!) had actually mis-read those papers. That to me is a really productive, focused debate, in which people truly are showing mutual respect by taking each other’s views seriously. Backing up your claims also reduces misunderstandings. And no, backing up your claims need not mean spending scads of time writing your comment. Nobody expects each comment to be a mini journal article. Linking to one or two sources that illustrate or support whatever you’re saying is usually fine. (Grouchy aside: a corollary of this principle is that proof by authority is horrible. This is a personal pet peeve of mine. The fame or expertise or number of publications or height of the person who claims X is totally and utterly irrelevant to evaluating the truth of X. Same for the number of people who believe X. And same for the journal in which X was published. Famous people can be wrong. Lots of people can be wrong. Claims published in Science can be wrong. If you think it’s inherently disrespectful or arrogant to criticize the views of eminent people, or of lots of people, or of authors of Science papers, you’re going to have a rough time in the comments section on this blog.)
2. Explicitly say that your criticisms are directed at someone’s ideas, not at them personally. You can say things like “With all due respect, you’re wrong and here’s why…”.
3. Go out of your way to thank people for taking the time to comment, even if you seriously disagree with them.
4. Write your comment and then sleep on it to see if it still looks ok to you in the morning.
5. If someone says something that sounds really wrong, or confused, or self-contradictory, or for which they’ve offered no evidence, ask them to clarify or elaborate before laying into them. Who knows, that person might be an inexperienced student, or someone who rarely reads blogs and so doesn’t understand the “norms”, or not a native English speaker, or etc. As I said, I think that often the most productive debates are ones in which the debaters push each other really hard. But there absolutely are circumstances in which pushing someone hard isn’t productive or useful.
6. If you’re not very sure of your ground, say so. Say “I could be wrong about this, but I think that…” or “In my limited experience, I’ve found that…” or “I’m not an expert, but…”, etc. No one’s going to bite your head off in reply if you do that. Plus, writing in this way forces you to be self-conscious about your own reasons for saying something, which leads to better debate. If you explicitly say that, e.g., you’re only speaking from personal experience, that will make you less likely to over-generalize. After all, other people’s experiences could be very different.
7. Avoid strong language entirely, and don’t reply in kind when other people use it. If you care about the topic, you may find this hard to do. And you need to be careful not to pull your punches, for the reasons described above.
8. Making jokes can help keep things light or lower the temperature.
9. Also note points on which you agree with someone, rather than talking only about points on which you disagree.
10. If you think you’ve been personally insulted, don’t reply in kind. Instead, just explain why you felt that the comment in question stepped over the line, and that you don’t appreciate it. You can also email me to complain (jefox@ucalgary.ca).
11. Before you comment for the first time on a blog, “lurk”. Read some of the posts and comment threads without commenting yourself. Get a sense of what’s considered acceptable, and what’s expected. Different blogs have different audiences, and so different “norms”. This is a really important piece of advice for people who mostly don’t read blogs.
12. Various other techniques I’ve forgotten or don’t know about.

By the way, I admit I’m pretty bad about using these techniques, except for the first one and sometimes the second and third. I’m terrible about the fifth. I don’t often cross any lines (though I cross them more often than I should, and more often than anyone else I know). But I’m sure I’m a less effective and productive debater than I would be if I made better use of these techniques.

The above points guide my approach to comment moderation on this blog. 99.9% of comments we get are obviously fine. But on the rare occasions when moderation is necessary, this is how I approach it:

• As long as a conversation is productive and moving forward, as a moderator I generally don’t step in. As long as people are writing comments like “Ok, now I understand what you meant…”, “Here’s another issue…”, “Let me try phrasing my point a different way…”, etc., the conversation is going fine, even if the language is sharp. It’s when people start saying the exact same thing but in stronger language that as a moderator I start to get concerned. That’s a sign the conversation isn’t productive anymore and is at risk of devolving into a personal argument. It’s like shouting to try to get someone who doesn’t speak your language to understand you.
• If someone does write a comment with what I think are inappropriate personal remarks, or that is just totally unproductive, I contact the commenter privately (assuming they provided an email address), politely explain my concerns about the comment, and invite them to rephrase and resubmit it. I’ve literally only ever had to do this two or three times out of more than 2000 comments I’ve dealt with.
• I don’t believe that any comment blocking whatsoever “violates free speech rights” or “the open spirit of the internet”. Yes, I once heard a (quite prominent) science blogger express these quite silly views.
• While I do think commenters ought to back up what they say, I don’t police this during comment moderation. I just comment in reply, asking for evidence and arguments.
• When I step over the line myself, I apologize, publicly.

All of the above reflects my personal preferences and experiences. It is my blog, after all. 🙂 There’s no approach to comments that would please everyone, so I just do what seems reasonable to me.*** I’m quite tolerant of strong language in debate, and I enjoy a good argument. I’m gratified that there are lots of people who share my point of view on this sufficiently to want to read and comment on Dynamic Ecology.

I’m sure some of you (possibly including my co-bloggers!) will disagree with at least some of all this. So let the freewheeling, passionate, productive, fast-moving discussion begin! Because that’s what Dynamic Ecology is here for.

*That very strange example just popped into my head just now. I have no idea why. 😉

**As an aside, it’s worth noting that on science blog comment threads there’s a very strong culture of backing up your claims. This blog is far from unique in that respect. People who don’t read blogs often don’t realize this, I think.

***If you think the “statistical machismo” thread was heated, you should see just about any comment thread on just about any prominent economics blog. Which does not prevent productive economic debate, indeed just the opposite. My reading of various economics blogs has really shaped my views on what blogs can and should be. More on that in a future post.