Language Log

Please read, and take to heart, our "Comments policy".

If you have something worthwhile to contribute, say it courteously and cordially.  There is no need to be gratuitously snarky.  That does not contribute to the smooth, productive flow of discussion.

Be respectful.

Do not be repetitive.

Do not harp on some private hobby-horse of your own.

Be aware that your reputation on Language Log will follow you elsewhere.  Realize that your behavior on other internet fora is known to the denizens of Language Log.  The World Wide Web is mercilessly unforgetful.

In previous posts on this subject (see "Readings" below), we have listed a number of traits of the typical troll.  There are a few more items that have not been explicitly covered, so I will mention them here.

First, though, a prefatory remark about the defining nature of a troll and what his / her modus operandi consists of.  Namely, the primary purpose of a troll is to disrupt the smooth, collaborative functioning of a discussion group that is dedicated to the discovery of ideas and free, fruitful, civil exchange of opinions.  Trolls want to inflame others so as to bring a screeching halt to amicable, productive dialogue and discourse.  Sometimes trolls will come perilously close to derailing an interesting discussion, causing a furor of denunciation and recrimination, but then, if the group is fortunate and things calm down, they will end up having a stimulating, enlightening conversation after all.

A conspicuous characteristic of the typical troll is that either they do not read the comments policy of the forums where they deposit their invective or they read the guidelines but choose not to adhere to them.  Our comments policy may be accessed by clicking on the link at the top right of the Language Log (LLog) homepage.  However, since many commenters consistently break these rules, I think it is fitting to list them here for all LLog readers to see:

As you are aware, our fans in China and elsewhere around the world would like to translate "Language Log" into their own languages.  The problem is that there are different words for "language" and "log" in the many languages that they wish to cover.

For example, the Romance languages distinguish between the faculty of language—the human capacity to communicate, using spoken or written signs—from specific oral or written natural languages (French, Mandarin, etc.). One chooses between one word or the other depending on the subject under discussion. In English, the same word can be used for both phenomena.

Just received this from our fan club in Shenzhen, China.

Andrew Garrett is Professor of Linguistics and Nadine M. Tang and Bruce L. Smith Professor of Cross-Cultural Social Sciences at the University of California, Berkeley, and also Director of the Survey of California and Other Indian Languages there. He wrote to me after he saw my post about who has the oldest linguistics department in the USA to give some interesting comments about his department's early history, the relations between linguistics and anthropology, and the vexed question of which is the oldest department of linguistics in the USA. Here's the gist of his email, as a guest post.

Guest post by Andrew Garrett

The first Berkeley Linguistics department was set up in 1901, in fact a few months before even the Anthropology department here. An introduction to linguistics course that is still taught was first taught in Fall 1901, by Benjamin Ide Wheeler, the president of the university and an Indo-Europeanist who had received his Heidelberg PhD as a student of the neogrammarians. "Wheeler's Law" of Greek accentuation is named after him. (Joseph Aoun is another linguist university president, at Northeastern University, but I don't know how many others there have been.)

A year ago, I noted that Language Log was getting about 4000 spam comments per day ("A million (spam) comments", 9/2/2011). Recently, this number has been substantially lower — just 771 in the past 24 hours, for example.

But during the past year, the spam tide rose before it fell. We logged our millionth spam comment at some point early in the morning of September 1, 2011; today, a year later, our spam filters have caught 3,277,574, or an average of (3277574-1000000)/365 = 6240 per day over the intervening time.

Whether the spam filter's daily harvest is 500 or 10,000, it's too many for me to check the whole list for false positives. So if your comment doesn't appear, and it wasn't offensive and/or devoid of relevant content, there's a good chance that you somehow got caught in the spam filter.

LL readers will not be surprised to learn that Daniel Casasanto and Kyle Jasmin disagree with my evaluation of their work on the "QWERTY effect". Yesterday afternoon, they added a comment to that effect on the original post. Since relatively few of the people who read that post are likely to see their comment, I'm reproducing it below. I'll respond at some later point.

Stock markets are down around the world, but the flow of spam comments continues to increase. Three weeks (21 days) ago, our spam comments counter stood at 1,008,782. As of this morning, it's at 1,140,742, for an increment of 131,960 in 21 days. This is 6,284 per day, or 2,293,591 per year.

I've long since given up scanning the spam trap for real comments that the automatic algorithms have caught by mistake, though I'm sure that there are still a few of these. (Note that even if the false alarm rate is very low, say a tenth of a percent, there would still be a half a dozen innocent victims a day. I believe that in fact the rate is a bit lower than that, but it's not zero.)

Worth reading: Josh Marshal, "Should you be you?", TPM; and the comments following it.

Note to readers: Thanks to a link from Andrew Sullivan, our server is maxed out at around 2250 visitors/hour, and things are a little slow. If you come back in an hour or two, response times for browsing or commenting should be better, as we return to our more normal mid-day average of around 1,000 visitors/hour:

(Status as of 3:05 p.m.)

Under the name "Arnold Zwicky" I have returned to the top of the list of Language Log authors, having spent some time in the guise "Zwicky Arnold" at the very bottom of the list. Let there be wild celebrations! Boundless e-Champagne and i-Bûche de Juillet for everyone!

Since our Christmas-themed posts are a bit thin so far this year, here are a few topical blasts from Language Log's past:

2008: "Seven fishes"; "Happy Christmas";
2007: "One Christmas too long"; "Christmas and 'politically correct(ed)ness'", "'Tis the season", "The unkindness of strangers", "Victims and etymology", "Lexical repulsion", "Insert flap 'A' and throw away";
2006: "Merry … umm … Christmas, Will!", "Like, a Christmas gift card", "Happy tensing and coal in sex";
2005: "Christmas trees and holiday trees";
2004: "Talking animals: miracle or curse?", "A boxing day election — or not?"
2003 "'Twas the night before Christmas", "Same-sex Mrs. Santa: 'The semantics are confusing'"