A comment about comments

« previous post | next post »

Earlier today, someone calling himself (?) Baishui submitted this comment on Victor Mair's post Burlesque Matinée at the Max Planck Gesellschaft:

My comments were deleted twice here. Apparently, someone is offended by me saying 'this incident shows how ignorant the West (and its academics) are of the non-Western world'. What pettiness!

I deleted this comment, just as I had deleted the same individual's first attempt, which consisted only of a one-sentence indictment of Western academics, and the second attempt, which added the accusation of censorship.

In an attempt to maintain a reasonable signal-to-noise ratio in the comments section, I'll continue to be skeptical of comments that lack specific and relevant content. To put this skepticism into context, though, you need to understand something about how comments work in a standard WordPress blog.

The first time that a comment is registered from a given email address, it's held up for moderation — it won't appear until and unless one of the blog contributors approves it. After that, any comments from the same email address will appear immediately, without moderation, though contributors can delete them after the fact.

My own policy as a moderator of first comments is to reject those that seem to me to lack content, whether I agree with them or not. Therefore I reject "Me too" comments, unless the context is a poll of opinions; and likewise "This is disgusting", "I disagree", and "Bravo!". Similarly, I tend to reject first comments that simply restate some aspect of the post, or a previous comment, without adding any content. This is partly because such comments are often spam or near-spam, but mostly because, well, they're intellectually empty. Other LL contributors have more — or mostly less — tolerance for vacuity.

I rarely delete comments that have not been held up for moderation, unless they're gratuitously off topic, offensively contentless, or contentlessly offensive. If someone has shown that they have interesting things to say, an occasional "me too" or "oh come on" is fine, especially because I don't want to spend time editing our comments section, since this would cut intolerably into the hour a day or so that I can devote to blogging. Again, other contributors have somewhat different standards, mostly in the direction of preferring not to deal with comments at all.

In the case under discussion, Victor's post explained in detail one instance of ignorance of written Chinese. In my opinion, Baishui's comment added nothing except for an unsupported generalization about "the West". And Victor himself is one of the millions of Westerners, and tens of thousands of Western academics, who are not ignorant in any of the relevant ways, so that the comment seemed especially unhelpful in that context.

If Baishui had (for example) written something like "Western academic journals display a shocking ignorance of China, as shown by X, Y, and Z", I would happily have allowed the comment, even if I disagreed with the analysis or the conclusion. If he'd linked to his (as far I know non-existent) blog posts on the subject, and invited us to read them, I'd have been even happier. If he proposed to start a discussion of whether Western academics are more ignorant of China than Chinese academics are of the West, that would be fine with me as well.

As it is, his contribution struck me as below the threshold. If he'd like to explain his ideas on the subject in a more contentful way, comments are open.


  1. Lane said,

    December 16, 2008 @ 8:30 am

    I for one welcome our comment overlords.

    My experience (at The Economist's American politics blog) is that a comment environment becomes self-perpetuating. We have fewer commenters, probably because we require an initial quickie registration, than most big political blogs. But the fact that early commenters were a) Economist readers and b) committed enough to register meant that a small but excellent coterie of commenters got a good environment started early. Traffic surged last year with the election. And yet we rarely had to delete offensive comments; even on the touchiest topics, things stayed on point and decent.

    So not only do Mark et al. have the right to oversee comments how they see fit, I think they're smart to do so. People tempted to write something dumb will see the thoughtful and content-packed comments before theirs, and realize (correctly) that they're out of place.

  2. Ray Girvan said,

    December 16, 2008 @ 10:08 am

    It's a difficult decision. For sites I run, I'm a strong believer in allowing comments (having no vehicle for open discussion is one of the egregious aspects of sites relating to pseudoscience, political polemic, news media resistant to correction, etc). But conversely, I'm not letting that sucker me logically into a situation of having to accept everything: personal insults, thinly-veiled linkspam, unsupported assertions that I'm wrong, etc. I think moderation is fine as long as it steers between those boundaries.

  3. Olga said,

    December 16, 2008 @ 1:30 pm

    I have to mention the lovely xkcd that was under discussion here a while ago, about the virus that automatically reads back every comment (the discussion was about the one-word-reply "virus"). My name here links to it (I hope; here's the link again, just in case: http://xkcd.com/481/ )

    On a different note, I like the comments policy here. It means that it's actually possible to read and enjoy the comments section. It actually functions as a forum for linguistic discussion — great! And a comment like the one mentioned in the article does not advance the discussion in a significant way.

  4. Virgil Ikari said,

    December 16, 2008 @ 3:53 pm

    The one I think I dislike about moderated comments is that the time delay prevents any actual discussions from occurring between commenters, and instead frames the comments section only as a response to the post itself.

    I hate not being able to talk about it with others, which is one of the main reasons I post in comment sections. Websites that have moderation enabled for their comments, requiring approval before anything is posted, rarely attract my long term attention.

    Additionally, I don't understand why the Comments are closed for some posts–they seem closed here far, far more often than on other sites. If you want t ocut down on linkspam, add a captcha of some sort. Closing the comments section only hurts readers who were eager to add their views to the topic and discuss it with others.

  5. Paul Wilkins said,

    December 16, 2008 @ 3:54 pm

    That commenter reminds me of someone I once knew.

    Way back when I was wanting to be a writer and brimming with proufound ideas and no way to express them, one of the the most prominent characteristics of my writing was the reliance on what can be best described as specious self-evident truths.

    I eventually came to learn that expressing prejudices and relying on unfounded, so called common knowledge was not academic exploration of ideas. Epistemologically speaking, I think it was not until I was in graduate school until that little gold nugget sunk in.

    Periodically, I revisit some of the pieces I wrote back then. So much for those brilliant early manifestos…now they're not much more than a fart in a mitten.

  6. Stephen Jones said,

    December 16, 2008 @ 3:57 pm

    The problem certain of the illuminati who withdrew comments had was with the normal flow of conversation. A blog is like a conversation. There is the lead article to set things off but things develop and of course people end up responding to the last post, not the original article. That is what happens in any intelligent conversation; to expect posters only to comment on the original post is naive. And counterproductive. Why should I be interested in what unknown posters have to say about an article by Zwicky, or Pullen, or Liberman or Poser. A forum (and a blog that allows comments is in effect a forum where only the original subject is limited) is of interest because of the dynamics of the conversation, and unfortunately that does lead on occasion to the thread meandering and ending up as a tangled ball on the floor.

    There seem to me to be two rules that can fairly strictly be followed. Firstly idiotic generalizations like the one referred to by Mark should be nuked. Secondly there should be a general attempt to keep the topic linguistically based. I, and most others, come to this forum, because the majority of the posts will deal with linguistic topics; if I wanted politics or cookery I would go elsewhere. Thread hijacking should be allowed as long as it is on general linguistic topics. Off topic asides, like gentle explorations in the sidewaters, should be allowed, but if the canoe looks like going up another creek, then it should be cut off from the flotilla.

  7. Mark Liberman said,

    December 16, 2008 @ 4:21 pm

    I agree with Stephen Jones that "topic drift" can be a fine thing, if the conversation is an interesting one. And the analogy with the flow of a multi-person conversation, though obvious, seems to me to be helpful in thinking about the etiquette of commenting.

    It's not easy to reduce this to simple rules — participants in a conversation are trying to optimize an ill-defined non-linear function of many variables.

    But interesting "explorations in the sidewaters" are welcome, as far as I'm concerned.

  8. Catanea said,

    December 16, 2008 @ 4:51 pm

    And conversational drift, maybe, is a linguistic concept in itself?

  9. peter said,

    December 16, 2008 @ 5:05 pm

    I guess part of the problem here is that any pedantic quibble on a matter of linguistics is certainly within-scope of the blog and of interest to other readers, but pedantic quibbles on non-linguistic matters may or may not be. For example, Mark has just written:

    "participants in a conversation are trying to optimize an ill-defined non-linear function of many variables.

    Of course, the overwhelming majority of participants in conversations are doing no such thing, although they may be acting in a manner which could be readily modeled as such. I think it very important, as scientists, to be careful to distinguish metaphorical talk from descriptive talk, for reasons well outside the scope of this post.

  10. The other Mark P said,

    December 16, 2008 @ 5:06 pm

    The attitude of the deleted poster was pretty typical of a "trolling" attempt. He wasn't interested in why his first post was cut, but immediately engaged in a counter-offensive. So we had a second post that called the entire "West" ignorant, and had a crack at the moderation too.

    On the whole outright ignorance is easy to ignore, but trolling has to be dealt with very firmly. If you don't show people spoiling for a fight the door, pretty soon you get a nasty fight.

  11. Lance said,

    December 16, 2008 @ 7:47 pm

    It's worth noting that

    Mark: But interesting "explorations in the sidewaters" are welcome, as far as I'm concerned.

    runs somewhat counter to

    Arnold: Somewhat subtler are comments that pick up on some phenomenon that is mentioned in passing in the original posting and run with that…. What goes wrong? Our comments policy was designed to keep focus on a particular topic, but Language Log is treated like any blog that allows comments, unmoderated mailing list, or unmoderated newsgroup: everything is open space. So discussion wanders all over the place

    "Somewhat" because (a) Arnold implicates, rather than outright stating, that he doesn't welcome sidewater exploration, and (b) there's no inherent contradiction between Mark not minding something and Arnold minding it.

    All the same, it means that it can be hard for readers to judge how (un)welcome a tangential (but still linguistics-related) comment will be. (The comments-off nature of some posts suggests a feeling of "utterly unwelcome" to me, though again that's by inference, not assertion.)

  12. Arnold Zwicky said,

    December 16, 2008 @ 10:33 pm

    Lance: ""Somewhat" because (a) Arnold implicates, rather than outright stating, that he doesn't welcome sidewater exploration, and (b) there's no inherent contradiction between Mark not minding something and Arnold minding it.

    All the same, it means that it can be hard for readers to judge how (un)welcome a tangential (but still linguistics-related) comment will be. (The comments-off nature of some posts suggests a feeling of "utterly unwelcome" to me, though again that's by inference, not assertion.)"

    This picks up on an idea in comments in an earlier posting, about how comments from a certain "core" number of Language Loggers constitute a de facto "official" policy of the blog.

    This idea freezes my heart, and goes so far against my idea of what I am doing on this blog (in which blogging here was supposed to be fun and was also supposed to be an expression of individual opinion) that I will now withdraw (as I have proposed doing twice in recent months, though I was dissuaded by the other bloggers) from the blog, though I intend to leave pointers to my writing elsewhere. I DO NOT SPEAK FOR LANGUAGE LOG.

    This is especially galling since in a few weeks I will be accepting (with Mark Liberman) an award from the Linguistic Society of America on behalf of Language Log. But I'm doing this entirely as a stand-in for Geoff Pullum, who's not able to come to the ceremony. If I could think of a graceful way to get out of the event at this point, I would take it.

  13. MJ said,

    December 16, 2008 @ 11:39 pm


    I understand that purpose-clause constructions, like explanations in general, are context-sensitive and interest-relative. As such, you may have other reasons for withdrawing from Language Log, and I don't at all mean to pry.

    But you should not take a few commenters' opinion that there is an "official" policy, position, or even Censurious Totalitarian Regime as the "official" policy or position of commenters to this blog. Some of us want structure, and read it into your smallest whims. Others demand anarchy and fight for it. But it's here as everywhere, and I don't see why here can't be as fun as anywhere.

  14. blahedo said,

    December 17, 2008 @ 1:34 am

    Oh, that would be so sad. I do hope you reconsider—though I too have been occasionally frustrated by closed comments on some posts, I've been glad the posts were there!

  15. dr pepper said,

    December 17, 2008 @ 3:22 am

    How about thinking of it as a type of social contract?

  16. Skullturf Q. Beavispants said,

    December 17, 2008 @ 10:45 am

    I enjoy everyone's posts and so I'm sad to hear Arnold talk about leaving Language Log.

  17. Stephen Jones said,

    December 17, 2008 @ 2:06 pm

    Nobody has ever suggested Arnold speaks for Language Log. And why on earth can't Arnold have one policy for his postings and Mark another? Since when has Arnold been in favour of a 'foolish consistency'?

  18. S Hawkins said,

    December 17, 2008 @ 3:20 pm

    The fact that some commentators have taken the existence of a closed posting as a personal affront is itself intriguing. There seems to be an undercurrent of belief that one has a right to comment, that to not permit comments violates the very nature of the Internet. But doesn't the much ballyhooed freedom of the Internet also extend to the freedom of a poster to not allow comments? The desire for a standardized right to comment on any posting in any location seems to strike at the very ideas of diversity of approach or individual autonomy.

    Comments can be wonderful. But to focus so much attention on them runs the danger of suggesting that the original postings would not be worthwhile without the comments, which seems absurd.

  19. peter said,

    December 17, 2008 @ 4:37 pm

    S. Hawkins wrote: "The fact that some commentators have taken the existence of a closed posting as a personal affront is itself intriguing."

    To my mind, the key issue here is that the comments policy on a blog or on an individual post signals whether the poster is open to persuasion or not. A post-with-comments-open signals that the poster seeks a dialogue and is making (possibly insincerely) an effort to solicit and consider dissenting viewpoints. A post-with-comments-closed signals that the poster cares nought for any dissenting opinion, not even seeking to solicit any, and (thus) is not even going to pretend to be persuaded from the viewpoint expressed in the post. It is hard, and ultimately pointless, to engage in dialogue with people who signal that they don't intend to change their opinion, no matter what is said to them. This is why comment-closed-posting is taken as a "personal affront" by readers.

    I think much of the debate expressed in this topic at LL arises because some of us, narrowly, see the function of communication only in the direct transmission of information content, and do not realize (or ignore) its many other functions. The phatic function of some comments which otherwise provide little information-content tranmission (eg, commenters saying only "I agree") is an instance of another function of communications. If the author-posters at LL seek to sustain the community of writers and readers about matters linguistic which you have by now created (which, of course, you may not wish to do), then I think it behooves you to consider ALL the functions of communications in these interactions, and not merely that of the direct transfer of new information content.

  20. Stephen Jones said,

    December 17, 2008 @ 6:26 pm

    I can see excellent reasons for not having comments open. Somebody must oversee the comments, and Zwicky, or anybody else, is quite within his right to decide his time is more usefully spent writing more articles than organizing the comments.

    And as Zwicky has pointed out if you disagree with him, send him an email.

  21. Ray Girvan said,

    December 17, 2008 @ 7:19 pm

    To my mind, the key issue here is that the comments policy on a blog or on an individual post signals whether the poster is open to persuasion or not.

    I don't read it that way at all. As I said in an earlier comment, I do fairly despise sites/media that are notoriously impervious to feedback, for clearly self-serving purposes. But LL posters aren't remotely like that. On several occasions they've amended comments-off posts when I've e-mailed a comment that I was sure added to the picture enough not to waste their time. Since they're actually reading (and if necessary responding to) e-mails as a more selective channel of communication, I don't see any problem with comments being off on grounds of, for example, how much time the poster has to spare, and how high signal-noise they want.

  22. dr pepper said,

    December 18, 2008 @ 1:30 am

    For me, an article with open comments means that the author believes that what they wrote might inspire a fruitful discussion.

  23. Lance said,

    December 19, 2008 @ 1:27 am

    …whoa. I go away for a day or two on vacation and come back to discover this.

    Though it's late enough now that no one may see it, let me clarify what I said: in no way did I mean, either by assertion or implication, that I thought that Arnold Zwicky was speaking ex officio. In fact, I more or less intended to remove that implication entirely with "point (b)", i.e. that Mark and Arnold were welcome to believe different things without contradiction. I meant only what I said: that insofar as Mark and Arnold feel differently about the way comments might drift, it's sometimes hard to judge whether a comment on any particular post will be considered appropriate or not.

    I think I can see how it happened that people drew the wrong conclusion about Arnold's relation to the policy. Since the comments policy is unsigned by any particular LanguageLogger, I wouldn't fault people for (mistakenly) inferring that there was general agreement among bloggers here about its implementation. And since Arnold was the only one to discuss the comments policy (Mark was the first to post about comments, but without mentioning the policy), I wouldn't fault people for (mistakenly) inferring that Arnold's views in particular were representative of the general LL opinion.

    Regardless of how it happened, though, I'm personally saddened that the misunderstanding has caused him to decide to withdraw from the Log, and even more saddened to think I may have been, if not the cause, at least the catalyst. I'd like to stress that it was never my intention to criticize either Mark or Arnold, or to complain about their work on LL in general or their approach to comments in particular—only to express my own confusion about how to tell which comments were appropriate when. If I inadvertently expressed anything offensive, I apologize for doing so.

  24. mollymooly said,

    December 19, 2008 @ 3:36 pm

    I've just subscribed to Arnold's new blog's RSS; and I was thinking that, if he sometimes gives the benefit of his expertise in the comments section of LL posts, and vice versa, it would be pretty much like before.

    But already his third post contains manual backlinks to a bunch of LL posts of a similar theme. This Madness Must End! Come back Arnold, we love you!

    I am crazy enough to think a wee bit of tweaking of the Comments Policy might suffice: along the lines of "different contributors approach this differently; deal with it".

  25. KYL said,

    December 19, 2008 @ 3:47 pm

    The commenting policy makes perfect sense, as long as it's applied reasonably neutrally.

    If, for instance, a comment that simply stated "This incident shows how the Chinese treat all Westerners as barbarians and are ignorant." were let through, then I think there is a problem. But if Mark applies his rule in a viewpoint-neutral way, then I applaud the effort.

    Of course he doesn't have to follow any standards at all if he doesn't want to, but I would imagine that in the spirit of the language log, he would want a neutral policy.

  26. สัตว์แปลก said,

    November 21, 2013 @ 3:48 am

    My comments are randomly failing… I think the ones where I include my personal web page in the link field are more likely to vanish into the ether than ones where I don't, but some of those disappear too.

  27. ทัวร์ลาว said,

    November 22, 2013 @ 12:06 am

    Victor's post explained in detail one instance of ignorance of written Chinese.

RSS feed for comments on this post