Wednesday, November 15, 2006

Eye of Newt and blur of science

Double double toil and trouble.
Fire burn and cauldron bubble.


One of the problems facing scientists and science promoters is that the human brain has not evolved to naturally gravitate towards clever experiments or robust theorems. As a result, journalists and educators often resort to using a variety of non-scientific techniques to catch the public’s interest or to explain the significance of a particular discovery. The problem with this approach is that it blurs the borders between science and the non-scientific tools used for education. It is difficult to complain about liturgists, fraudsters, or entertainers intruding into the arena of science when their traditional techniques are used for science promotion.

I suspect that entire journals could be filled with theories, explanations, or apologies for various aspects of and solutions to this problem, but being an illiterate lab techo, I haven’t read any of them. So I will ignorantly suggest a simple way of judging the usefulness of any particular unscientific selling technique: the cost-benefit analysis. If a particular method has limited explanatory power, and has a high chance of causing confusion or misunderstanding, then it should be avoided. One such high cost, low benefit literary device is nomenclatural overload, and its poor cousin, technobabble.

Modern science is complicated. But despite that complication, the particular experiments and observations that are undertaken are generally done for specific, definable reasons. Simply dumping the names or generalizations about a field onto the page without explaining them merely creates the illusion that science is a collection of arcane trivia, and not the amalgamation of knowledge based on prediction and observation. The practice of name dropping and information overloading has the effect of reducing science to alchemy, where arcane ingredients are combined and channeled without any overarching principles. This is obviously a risky approach:

Adder’s fork, and blind worm’s sting,
Lizard’s leg, and howlet’s wing-
For a charm of pow’rful trouble,
Like a hell-broth boil and bubble.


The solution is fairly obvious- if you’re writing an article, explain why things are important before you discuss what the result of measuring them was. An example of how this should, but doesn’t happen can be found in this week’s New York Times science section, in the article, "Ancient Crash, Epic Wave."

The offending excerpt:
“When a chondritic meteor, the most common kind, vaporizes upon impact in the ocean, those three metals [iron, nickel and chrome] are formed in the same relative proportions as seen in the microfossils, Dr. Abbott said.”

The reason behind these analyses is fairly simple, so it is a shame that this otherwise excellent article didn’t bother explaining it. For any non-geologists who have wandered into this site by accident, here is the story:

When the planets and asteroids originally formed early in the solar system’s history, the larger ones heated up and melted, and the immiscible liquid metal and molten rock differentiated to form a metallic core and a silicate mantle. Small asteroids did not generate enough heat to melt, and remained undifferentiated. These are the chondrites, and they contain a mix of metal and silicate that condensed from the primordial solar nebula.

Because the Earth is differentiated, most of the elements that dissolve into metal instead of magma are contained in the core, and thus are relatively rare at the surface. These elements, called “siderophiles” (iron lovers) are thus more common in undifferentiated bodies than they are in the Earth’s crust. So when a chondrite hits the earth, the ejecta is enriched in siderophiles such as Fe, Cr, and Ni. Iridium is also a siderophile, so the search for nickel and chromium is based on the same principle as the search for the Ir anomaly at the K/T boundary, which marked the end of the Mesozoic (the age of dinosaurs).

Obviously, these elements are not “formed” in the impact. They are just particles from the impactor, which can be identified because the impactor and the silicate Earth have very different siderophile concentrations.

Of course, getting scared off by terminology and avoiding it is just as problematic as simply namedropping undefined terms. But in science, things often have names for a reason, so explaining what that reason is can go a long way towards illuminating the subject at hand.

4 comments:

  1. Proliferation of undefined scientific terminology in a newspaper or magazine article is a fairly reliable indicator that it's a recycled press release - and the person doing the recycling has no idea what it means.

    Nice of you to explain it to them!

    ReplyDelete
  2. John Fleck says -

    There's a problem with your criticism of Sandy Blakeslee's story and your proposed solution.

    Any particular scientific bit is the tip of an enormous pyramid. It seems obvious to you that she should have gone on to explain the formation of the chondrites, and your explanation is both delightful and helpful. But you've taken her 36 words and tacked on another 178. And within those 178 are a number of fascinating threads worth explaining with more helpful detail, for precisely the same reason you think the idea in the original paragraph deserves elaboration. So any solution you offer would be subject to exactly the same criticism you've made of Blakeslee. And you've singled out chondrite formation, but there are quite a few other points in the story where you could level exactly the same criticism. You seem to have picked the one relevant to your area of scientific understanding. I happen to be fascinated by geomorphology, and would have loved more detail there, for example.

    It's hard to say what the right stopping point is, but at any time one picks that stopping point (which is generally defined by the space constraints of the medium), the same criticism you make could be applied.

    In fact, I happen to think Blakeslee picked the right stopping point in this case. A story like hers needs to try very hard to pick just one thing and be about that. It would be nice to know the details of how chondrites form, and how that underpins the scientists' understanding. But it's not essential to the article's central point, which is about Earth's impact history.

    ReplyDelete
  3. Dear John,
    I admit that this article was more of a camel-breaking straw than a flagrant violation, but here is a more succinct argument:

    If you aren't going to explain what chondrites are, then why use the word? Why not just say "meteorites?"

    ReplyDelete
  4. John Fleck says -

    If it were my article, I certainly would not have used the word "chondrite".

    ReplyDelete