Wednesday, March 23, 2011

The Unspecificness of Specified Complexity

Just a couple of days ago a poster on Topix tried to use William Dembski's Explanatory Filter as something real and useful. Now if you know anything about it, you probably know my feeling. Anyway, here is my reply. But that's not the point of this post. I find the timing funny because of what I found today. Please understand that the central core of his 'explanatory filter' is the idea of 'complex specified information', or CSI. The problem I, and others, have with CSI is no one seems to be able to calculate it.

So where would one go to get the low-down on how to calculate CSI? Why I would think Dembski's own blog might have a clue. So while wandering around I came across "On The Calculation Of CSI". So what does it tell me? That absolutely no one has a clue how to calculate the value of CSI. Someone called 'MathGrrl' wrote the post, but you also have to look at the comments, about 60 of them when I caught the article. It's the biggest collection of just plain junk I have ever seen. Have you ever seen people trying to nail jello to a wall? Well this blog post is the literary equivalent.

My advice it that before you can nail jello, you have to freeze it. That's what CSI needs -- it needs to be frozen, that is to be specified [pun intended]. Dembski needs to sh** or get off the pot and provide a clear, understandable, and repeatable explanatory filter that lays out CSI is a useful and usable fashion. His work needs to be complete, well supported, and understandable. There will be critics, and those need to be dealt with, not dismissed in his usual off-handed way. Mathgrrl's post is more an embarrassment than anything. If Dembski's filter and CSI concept is more than a philosophical argument, then this post should have never happened. But it did, which tells a very different story than anything Dembski has published in his marketing pieces . . a very different and telling story.


  1. They're past 200 comments now, and still no closer to answering what should be an FAQ. It reminds me of nothing so much as a group of astrologers arguing over how precisely to calculate someone's horoscope.

    It's also amusing to see how the arguments have evolved, if I may use that word. From "evolution can't explain it!" to "you're doing it wrong/it doesn't apply in this case/try this other metric" to, around post #200, "why should we do all the hard work of defining CSI? Math is hard! You do it!"

  2. And every now and then someone wonders why I enjoy this so much. thanks for that great analogy!

  3. PZ Myers caught this story too (Shades of Ontogenetic Depth! and the post over on Uncommon Descent is over 300 posts and, as PZ said, they are spinning their wheels. Gotta love it!

  4. Thanks, Ted - I'll be linking to this in a post of my own: we have an on-line ID 'resource' store starting up here in NZ & the only thing on their shelves at the moment is a 'resource' on things like...
    ... specified complexity!

  5. Really, an ID Resource Store? What a concept. it sounds positively Monty-Pythonesque! I'll keep an eye out for your post.

  6. Done. They're Monty Pythonesque, yes, but unfortunately hard-pressed teachers may not realise that (and yes, some may incline that way anyway...). A couple of years ago we got The Privileged Planet & before that a bunch of videos along the same line as this latest effort. And back in 2003 (from memory) there was a sponsored workshop on intelligent design - for teachers & parents - that actually featured Dembski as a speaker. It was even advertised in the Education Gazette, which is the official Ministry of Education document - I was sooooo cheesed off about that!