|
RIPR is a (paper) newsletter that looks at local, state and federal policy issues that affect life here in the Ocean State. Each issue focuses on particular policy areas of interest. Future issues will examine controversial aspects of environmental policy, health care, state tax reform, and education spending. The intention is to look at action rather than talk. RIPR also issues a weekly column about public policy, carried by ten of Rhode Island's finer newspapers. See here for an archive of recent columns. If you'd like to help, please contribute an item, suggest an issue topic, or buy a subscription. If you can, buy two or three (subscribe here). Search this siteAvailable Back Issues:
Subscription information:
Contact:For those of you who can read english and understand it, the following is an email address you are welcome to use. If you are a web bot, we hope you can't understand it, and that's the point of writing it this way. editor at whatcheer dot net Archive:
Links:Links pageRSSRIPR is primarily a print publication (yikes! how 20th century!), and the work it represents is supported by its subscribers. Feel free to use this link to an RSS feed for the blog, but the real meat is in the newsletter, so come back and subscribe when you have a chance. Responsibility: Tom Sgouros
|
Mon, 29 Nov 2004
What to make of an article in the National Review about amazing advances in genetics and what that means for cherished liberal ideas? The article, predictably, was all about how advances are really going to uncover important differences between races, and deterministic facts about intelligence, and therefore it has to be conducted in secrecy, lest the researchers be branded bigots. The writer told about a friend of his, a "datanaut" who was up to his neck in complex math to analyze the genome.
Well, there is a lot to say about computational genomics, but not really what the NR author had in mind. The message of modern genetics is both encouraging and discouraging, but not in the way he had it.
The encouraging part is that while the "datanaut" in that NR article knows a lot about the math of genetics, math is hardly everything. A developmental biologist (whose name I wish I could remember, but maybe someone will help me) made this point famously many years ago by taking a chick embryo, pureeing it, and centrifuging it. He pointed out that after pureeing, it still contained as much genetic information as it had before, and that the centrifuging added order, but that no one thought it would still turn into a chick. A living egg is a complicated thing, and no one has yet figured out how to reduce its complexity to math, which is why there is no field of "Computational Development," with computer scientists pretending to "slum" as development biologists. So the encouraging part is that there is still a great deal of mystery in life, and most of the claims you read about in the Science Times or in the National Review are at least partly crap.
Here's an example: you hear pundits moaning about the possibility of cloning babies. Our Congressman Langevin says this is why he voted against stem-cell research: over the fear of reproductive cloning. Well, I guess we've cloned a sheep, so why not a baby? What you never read about is that Dolly the sheep was the first successful result after almost 300 failures. That's 300 stillborn, deformed, or died-as-infant sheep to produce just one now-that-we-think-about-it-not-very-healthy adult, to translate "failure" into something more concrete. Presumably this record will improve, probably dramatically. But will it improve enough for even the most morally obtuse person to want to try it? Maybe, but it hasn't yet, so far as I know.
The discouraging part, on the other hand, is that nothing seems to stop people in this kind of science from overstating the power of their results. Scientific hubris is an old tradition, and the successes of datanauts at finding things like the gene for Huntington's disease has only emboldened them. I don't doubt for a minute that they may have found "genes" that have some correlation with intelligence. (Such a claim was made in the NR article.) But so what? Good diet correlates with intelligence, too. I doubt very much that there's any kind of profound understanding why some stretch of DNA correlates that way. Maybe they've found a section of DNA that regulates something else that has to do with efficient digestion. Who knows? I don't. And I stongly suspect they don't, either. They can't even agree on what constitutes a gene.
Humility is a virtue, but it's not one shared by most computer scientists, who think they have a monopoly on understanding issues involving computation. Genetics and psychology both seem to touch on related subjects to some degree, but does that give computer programmers the keys to the kingdom?
The NR author writes that he is snowed by the sophisticated math of his datanaut. What he doesn't realize is that the math is sophisticated because the understanding isn't. They're using the math to look for correlations between distant parts of the genome, and looking for all different kinds of ways to massage the genetic data so that it is readable. But it's all based on the assumption that all the information is in there, which anyone who spends five minutes thinking about the subject can tell isn't true. (Think of that poor chick again.)
But, like the NR author fears, word will get out, and pundits who don't know squat about the science will be persuaded by the simple and straightforward and mostly wrong story told by datanauts. They'll find some gene that correlates with intelligence and then find that Asians have more of it, and there will be a to-do and cry in our newspapers and on our television sets about how it's inevitable that China overtake the US in everything. And then in 20 years we'll learn that it's really a bit of DNA that makes something to do with your lungs (which are also important to intelligence), and there are two other genes that do the same work, and are just as good. But in the meantime, we'll have made all kinds of stupid policy decisions based on that first interpretation.
So that's the discouraging part. What makes me fearful is not that we'll eventually understand too much of what makes our selves and our races. It's that on the way there, we'll fool ourselves into thinking that we know more than we do and we will do some very very stupid things to ourselves.
Ads and the like:
