Directed Panspermia Doesn't Belong In Schools (and Probably Never Will)

Opening to the Ideas section of the Boston Globe this past Sunday, I was immediately faced with a grotesque site.  Nothing to do with the appalling terrorism of fifteen years ago, but instead the Globe putting front-and-center in the section a truly awful idea, which the subhead trumpeted "Creationism can have a basis in science - if aliens are involved".
As expected, the article by Blue Marble Space Institute of Science scientist Jacob Haqq-Misra trotted out a typical exposition of some gaps in our understanding of the origin of life, a few big names who have discussed the subject, and a claim that life on Earth did not originate here but was seeded by aliens, aka directed panspermia, is an actual theory worthy of teaching in public schools.  

The first name out is Carl Sagan, who was one individual to propose that an intelligent species along the lines of our own could send life to distant planets.  Francis Crick wondered about the universality of the genetic code, given alternatives. Astronomers Fred Hoyle and Chandra Wickramasinghe published a book on the subject in the early 1980s.   Some unnamed scientists have argued that regularities in the genetic code point to design. And the regular finding of meteorites originating from Mars provide plausibility.

The bit about Martian meteorites is about the only piece of solid evidence, and of course that neither shows that life did arrive here from afar or that it can.   Attempts to find uber-patterns in the canonical genetic code seem to crop up with regularity -- and have absolutely zero impact with equal regularity.  I perused Hoyle & Wickramasinghe's book when I was at Harvard, and it made the same depressing logical flubs as many religious creationists (who indeed, sometimes cite the book).  For example, they spend time discussing the extreme conservation of sequence in one of the electron transport chain cytochromes, claiming the probability of this protein arising by chance is infinitesimal.  However, given that we have no idea how many other proteins could form a functioning electron transport chain in its place, perhaps by changing all the other proteins, means we can't calculate the probability.  

Ultimately, the problem with any such theory is that it provides absolutely no testability and no predictive power.  Take a look in a really large art museum; designed (by humans) artifacts there can take almost any form, from intricate to a pure white square.  Similarly, all possible living systems, any protein, any pathway, any cell, is equally plausible as a designed object -- which means bringing in a designer brings in nothing.  There's also the problem of complexity: the widespread commonality of the core genetic code (or the fact that all life have the same chirality to its DNA and proteins) could simply represent the triumph of one system over other possible systems.  Everything the "same" because all the life forms are related by common descent.  

The most galling claim in the piece is that this theory is something that could and should be taught in public schools.  First, given the level of evidence (aka none), it lowers the bar for what can be taught, encouraging the religious creationists in their never-ending quest to sneak religion into public science classes.  Second, public curricula really are a zero-sum game: there's only so much time for instruction, so much money for materials and a limited amount of opportunity to train teachers.  Time and money spent incorporating a non-theory into curricula steals from valuable topics of real relevance.  You can't have students evaluating evidence that simply doesn't exist.

This sort of sloppy thinking is depressingly prevalent in astrobiology, or at least those who exhibit it seem to have both penchant and success in getting publicity.  Finding extrasolar planets is amazing, hypothesizing that they might be similar to Earth is reasonable -- but too often the discoverers go on to announce they've all-but-found life (or even more effusively certain).  Or the notorious Drake Equation , which purports to estimate the probability of advanced civilizations -- but does so with a number of free parameters which have no empirical basis for estimation. Just today some news item claimed that a calculation showed we have a 50% chance of living in The Matrix, an elaborate computer simulation by a higher power -- but  we could never tell for sure according to the article.  Where did that probability come from?  Parameters with no basis - and the hypothesis is touted as untestable.  The Globe piece mentions the alleged Mars meteorite fossils and does at least say that they are discredited, but somehow those still have relevance.

I'm not systematically against astrobiology, just wild speculations claiming to be science.  I attended a talk last year on the challenge of detecting signs of any life on distant planets using telescopes and spectroscopy, and that was solid stuff.  It's a classic astronomy problem: how can you observe a distant world you can never visit, and how do you distinguish real signal from noise and biological processes from possible non-biological ones.  

Or the question of microbial life on Mars: a perfectly plausible (but probably quite challenging, given the necessary background material) educational approach would be to explore how we might distinguish native microbes on Mars from bugs that rode along with our robotic explorers. 

Jettison the specious claims to extraterrestrial designers, and the genetic code is a fascinating world.  While all the codes are kin to each other, the variations on the theme are multitudinous and still expanding.  Just recently a case was described of a genetic code with no dedicated stop codons.  There's also all the interesting bits of getting more from the code - overlapping open reading frames, RNA editing, ribosomal slippage, context-dependent translation.  Why bring in untestable hand-waving when the field is so rich in real observations?

I could also see students excited by active research that attempts to reduce the complexity of current organisms by finding which genes are dispensable and which are not, and then the real issues of how sets of genes might be dispensable as a set but losing any one of them would not be survivable.  Or how the possible sequences of ur-proteins are inferred from modern proteins, and the caveats involved.  This is real science, with problems that aren't papered over with magical creatures and with actual experiments and data.

For a newspaper in a town often identified with science and technology, the Globe can most charitably said to be uneven in its treatment.  Sometimes there are very good articles, but there are also ones which are truly awful.  Alas, this Sunday the tally of the atrocious ticked up one more.