To find truth: Evaluate one claim at a time until a pattern emerges

Post of the Month: March 2012

by

Subject:    | Is TO.org actively maintained?
Date:       | 31 Mar 2012
Message-ID: | W4SdnRANBak5yerSRVn_vwA@giganews.com

>> Your original post to this group in Feb. proved to be quite an entrance
>> and was quite interesting. Did that exchange change any of your
>> thinking on genetics?

> It has lead me (in conjunction with other studies) to a certain feeling
> of, I don't know, intellectual despair if you will.
> There is simply so much data, so many arguments, so many studies, so
> many criticisms, so many factors that the layman (who normally lacks the
> technial credentials to even properly evaluate the data) simply feels
> helpless in trying to form an educated opinion. For every point there
> are a hundred counterpoints, and each counterpoint raises a hundred
> replies on its own. So while it is deeply fascinating to me, it
> frustrates me even more.

First caveat: We're now in a world where "educated" means having access to the library of a good research university (and knowing how to use it). If I have access to a copy of "Linkage disequilibrium in the human genome" and you don't, it doesn't matter if you're smarter than I am. You have to rely on me to tell you what's in that paper. The digitization of decades of the primary literature has amplified this disparity. If you want to evaluate claims and counterclaims for yourself, the price of admission is at least part-time university enrollment.

Second caveat: Learning to read the primary literature is a skill that's easiest to pick up in a grad seminar class. You can figure it out on your own, but you'll have a lot more frustration and you'll make a lot of avoidable mistakes. Each community will have an understanding of what can be assumed and what needs to be explicit and you'll have to read several dozen papers before you can start evaluating the quality of the work with any competence. If you still want to dive in, then read the abstract first, then the conclusion, then the introduction, then the discussion. You'll know you're getting good when you start skimming the bibliography early on to see if there's any papers or authors you recognize.

So here's the trick: pick one claim and master it.

Here's how this works in practice.

When you posted your query about Carter's genetics paper I stopped reading when I got to his first citation. He made a claim that "there is abundant evidence that the entire human race came from two people just a few thousand years ago" and I decided to focus my efforts on that particular claim.

That in turn led to Nelson's paper in the journal of creation, and again I focused on one claim there and followed up to the paper he cited (Dorit). Then I read Dorit and read several of the papers that came later that cited Dorit and improved on the work. (I also had to consult an evolution textbook and a few wikipedia articles to figure out what a few of the more technical terms meant.)

Having done all of that reading do I understand linkage disequilibrium? Hell, no. But I have a general idea of what's involved, I have a decent sense that this technique is well-regarded in the community, and most important I can see the error bounds decreasing over time. And based on this I'm confident that Nelson's claim about linkage disequilibrium was wrong.

I then turned to Nelson's next claim, followed up the citation to Reich, and figured out straightaway that Reich didn't say what Nelson said it did. (I don't think Nelson was clever enough to lie about this; he probably just didn't read the paper carefully.)

So let's take a step back. I've invested a few hours of highly technical reading to figure out that the first two claims of Nelson are wrong and that Carter shouldn't have relied on Nelson for support. Nelson makes a lot more claims in his paper that I didn't look at and I haven't even gotten to the meat of Carter's work yet. I could easily spend two months doing this kind of analysis on Carter's paper and that's just one paper out of the thousands of creationists publication that are out there.

But all is not lost.

After doing a couple of dozen deep dives like I've illustrated above you'll begin to realize that, if a creationist makes a scientific claim in support of creationism, the claim is either wrong or trivial. After another couple dozen deep dives you'll start to see patterns in the errors. And at some point you'll be comfortable reaching the (tentative) conclusion that if the first fifty claims you investigated were wrong or trivial then then you can start making increasingly confident predictions about creationist claims in general.

Let's take another step back.

You mentioned points and counterpoints. That's very much a debating approach. If you're talking to folks who don't have access to the peer- reviewed literature then that's probably the only model for a conversation you have. You can enumerate your beliefs, they can enumerate their beliefs, and there's really no way for one person to convince the other.

The approach I've outlined doesn't have that problem:

1. Nelson says Dorit's linkage disequilibrium work supports a population bottleneck consistent with a worldwide flood. Dorit was doing exploratory work and his error bars were so wide that it supports a population bottleneck last Tuesday. Subsequent work (that was available to Nelson) improved on this technique and excludes a flood-related bottleneck. Nelson got it wrong, full stop.
   
   
2. Nelson says Reich's population bottleneck supports a worldwide flood. I say Reich specifically mentioned that no bottleneck was found in the Nigerian population he studied, and I'm happy to send you a copy of Reich with the relevant passage highlighted. Nelson got it wrong, full stop.

And that, in microcosm, is the evolution/creation "debate".

To sum up: don't waste your time with points and counterpoints. Find a claim that interests you and run it to ground. Then repeat. There are lots of folks here who would be eager to help. If you need a paper that's paywalled, drop us a line and it will magically appear in your inbox. If you like, write up your results and post them here. If you've picked a claim and have no idea where to start, drop me a line or post here. It gets a lot easier with practice.