His take: "Every year, America embarks on a massive project to carefully recreate the Christmases of Baby Boomers' childhoods."
This doesn't quite explain why almost no Christmas songs seem to have been added to the canon since 1970. (Still true, I believe, more than a decade after I observed it.)
It would be nice to get a large database of Christmas songs (such as a list of cuts on Christmas albums) and plot frequency against "year of composition" for each song. I still haven't found an easy way to do this, nor have I pursued hard ways (Amazon API?). But Mr. Munroe's chart illustrates the effect well with just a few data points.
Edited to add: Of all the Google archives, the Google Groups archive is the nearest to the Slow Zone, and sometimes zone disturbances cause Usenet to drop out of the Beyond for days at a time. I think it wise to include the text of my Christmas canon essay here, so as not to frustrate readers who encounter erratic links.
Path: supernews.google.com!sn-xit-02!supernews.com!nntp-relay.ihug.net!ihug.co.nz!news-hog.berkeley.edu!ucberkeley!newsfeed.stanford.edu!info1.fnal.gov!fsgi02.fnal.gov!higgins From: Bill Higgins-- Beam Jockey <firstname.lastname@example.org> Newsgroups: rec.arts.sf.fandom Subject: Canon of Christmas Songs: Well Gone Dry? Date: Thu, 14 Dec 2000 12:11:28 -0600 Organization: Fermi National Accelerator Laboratory Lines: 103 Message-ID: <email@example.com> NNTP-Posting-Host: fsgi02.fnal.gov Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: info1.fnal.gov 976817491 17333 126.96.36.199 (14 Dec 2000 18:11:31 GMT) X-Complaints-To: firstname.lastname@example.org NNTP-Posting-Date: 14 Dec 2000 18:11:31 GMT Here's something that's bugged me for many years. It's about Christmas carols. There are a lot of Christmas songs, and people are always writing new ones. But the set of "standards," widely accepted, widely performed Christmas carols, hasn't gotten any bigger since the early 1960s. (I'm speaking of the United States. Your nation's culture may vary.) There are really old ones, like "Silent Night" or "The First Noel." There are Tin Pan Alley ones added in the first half of the 20th Century, like "White Christmas," "Rudolph the Red-Nosed Reindeer," or "Santa Claus Is Coming to Town." There are a few that were added to the canon in the Fifties and Sixties, like Mel Torme's "Christmas Song." And then it stops. The last real Christmas "standard" I can think of was "Do You Hear What I Hear," from 1963.  I think I could prove my hypothesis by more than one method. One way would be to build a database of Christmas albums, listing the songs that are performed on each one, and of dates that Christmas songs were written. Make a histogram, showing frequency of appearance of a song on albums vs. year the song was composed. I argue that post-1965 songs would show a very abrupt cutoff in the distribution. Yes, one would expect older songs to be more frequent in general, but I'm talking about a really sharp drop. I don't know where I could find a discography database from which I could compile such a database... but I imagine there might be a suitable one somewhere on the Web. Composition dates would not necessarily be logged in a discography. I could probably get some of the information I need by going through five or six books of Christmas sheet music. At least that would give dates for all the highly popular songs. Others might be available by tediously slogging through the ASCAP/BMI Web site, one song at a time... Another way would be to count songs in songbooks. I tried looking at a listing of *Your Hit Parade* songs, but it turns out that Christmas songs hardly ever make it into the top 10, so data are sparse there. I'd be interested in tools. Suggestions welcomed on sources, Web sites, databases, reference books, etc. that would make this research job more painless. Counterexamples-- you may be able to point out songs, composed 1965-2000, which you claim have achieved "standard" status-- would be mildly interesting, but not nearly so interesting as help with a quantitative way of proving or disproving my hypothesis. I'm not saying "NO songs have been added to the canon since 1965," I'm saying "hardly any." I'll maybe grant you "Grandma Got Run over by a Reindeer" (1979). The next question is: If I am right, why? What happened to stop the widespread acceptance of new Christmas songs? What stopped the formation of consensus on which songs are classics-- as measured by criteria such as "played frequently on the radio," "recorded by many famous artists on their Xmas albums," or "appearing in many printed collections of Christmas songs?" Certainly songwriters haven't stopped writing 'em. There must be hundreds each year in Nashville alone. Perhaps the Christmas standard is a victim of the earthquake in American popular music triggered by the rise of rock'n'roll. If so, the connection isn't clear to me. * * * Footnote : Except for "We Need a Little Christmas," from *Mame* (1966), which is really a song about celebrating Christmas at a totally inappropriate time, like in the summer. But of course people only play it at Christmastime. Seems wrong. Even though it has become one of the standards. Misses the point. Footnote : Another thing: There's a set of songs that aren't really about Christmas, but are lumped in with Christmas songs anyway. Mostly they are about winter itself: "Sleigh Ride," "Walking in a Winter Wonderland," "Let It Snow," and "Frosty the Snowman." I don't think including them in a study would distort the figures much. Footnote : Smart-alecs who point out that Three Dog Night had a #1 hit with Hoyt Axton's "Joy to the World" in 1971 will not be tolerated. -- Bill Higgins | Favorite carol around Higgins house: Fermi National Accelerator Laboratory| Glooooooooooooooria Internet: email@example.com | In excelsis Deo Bitnet: Sic transit gloria mundi | Deo SPAN/Hepnet: [OBSOLETE-ADDRESS] | Daylight come and me wanna go home