I Saw What You Did There, Andy

Now I know I have already spoken about Andrew Schlafly's "Best New Conservative Words" nonsense. Still, Andy keeps on coming back to it, in much the same way you can't help playing with a rotten tooth. As you may, or may not, know Andy has surmised that ever since the 17th century, more and more conservative words have been created, essentially doubling in number every century.

There are, of course, two problems with this:

• The so-called "conservative words" appear to be a random collection of terms pulled out of Andy's ass, and augmented by several parodists. Thus we get such entries as "Radar" and "transistor" being conservative terms.


• Andy applies his own selection bias to the project. Thus if the 1700s had 5 new "conservative" words, the 1800s must have 10. So, Andy stops counting when the words he has decided on match his criteria. He hardly ever goes further back, because finding a new word in 1600 would mean 2 new 1700s words, 4 new 1800s words, 8 new 1900s words and 16 new 2000s words. And even Andy could have trouble coming up with so many words.

Still, despite the unscientific nature of the whole thing (but then we all know science is the Devil's work!) Andy has issued a challenge on Conservapedia's font page, which reads:

"Liberals doubt our observed doubling per century for these words, yet every layer reproves the remarkable growth pattern. Still looking for 6 more. Can any liberals disprove the pattern?"

Oh by the way, the "looking for 6 more" is Andy calling out for help from the parodists best of the public to help complete the progression. I hope somebody adds a whole bunch to the 1600s, so we can watch him flannel away in the 2000s.
However, one so-called liberal took Andy up on his challenge. I'm not going to repeat everything he said, but please read the article here. It's informative, and because it's aimed at Conservapedia is easy to read and has pictures. Suffice to say, once he had finished his analysis, he wrote a letter to Andrew Schlafly, explaining just why he was wrong. I'm sure he won't mind if I repeat it here.

Dear Andy Schlafly,

at the Main page of Conservapedia you write:

That is the new number of Best New Conservative Terms by century as we find the next layer. Liberals doubt our observed doubling per century for these words, yet every layer reproves the remarkable growth pattern. Still looking for 6 more. Can any liberals disprove the pattern?

Of course, there are difficulties to disprove such a pattern. But it can be shown quite easily that it is were unlikely that the selection of words you label as conservative is randomly taken from a population of words which follow your remarkable growth pattern.

As you asked for a liberal to disprove your pattern, you can find the analysis at a place you believe to be liberal:

http://rationalwiki.org/wiki/Conservapedia:Best_New_Conservative_Words

Summary:

• if the creation of conservative words follows a geometric distribution, this should be true not only for centuries, but for smaller classes: more words should be created in the second half of a century than in the first half, etc.

• statistical tests applied to the classification into decades, 20 years, 25 years, and 50 years show that it is very improbable that there is a geometric growth pattern

• this implicates that the pattern which is observed for the centuries is indeed artificial, i.e., it is created by actively (though maybe subconsciously) searching for words to follow the very pattern.

Hope this helps.

Now, that's a fairly reasonable and polite reply. How will Andy react?

Typically, it would appear. Realising that somebody has once again made a fool out of him, Andy hurriedly changes the main page. Please pay close attention to his edit summary - "Trimming." Not "removing erroneous claim", not "oops I made a mistake'. Nope, just "trimming.' Why/ Because quite simply Andrew Schlafly lacks the ma-cheese-mo to admit in public (or in the ZB sysop's space for that matter) that he made a mistake. Suddenly, there's no mention at all of the "doubling every century" nonsense and of course, no admission that a "liberal" proved him wrong.

By the way (h/t to Susan) I see the Harvard educated lawyer uses the phrase "Every layer reproves the remarkable growth pattern." Now I'm assuming that what Andy meant to say was that each successive century has double the conservative words. However, the word "reprove" does not express that.

In fact, turning to Andy's favourite Merriam-Webster, we see the following:

1. to scold or correct usually gently or with kindly intent


2. to express disapproval of

and certainly nothing to do with affirmation. However it does add that "reprove implies an often kindly intent to correct a fault". Which is what we're trying to do.

Coming up next time:

Does Terry Koeckritz smoke hash and was he responsible for hiring Diane Feinstein's staff? People want to know.