In this Scientific American article www.scientificamerican.com/article.cfm Michael Shermer states that the odds that "intelligent, technically advanced" aliens would look anything like a bipedal humanoid are close to zero
He claims to argue this in a YouTube video, produced by the "Richard Dawkins Foundation for Reason and Science" [sic].
However, as you can see, he does not reason or argue this, he merely asserts it. His main argument - his only one, really - is that
Of all the millions of species on earth, only one has become a bipedal humanoid.
To which the natural answer is, "Yes, it's the intelligent, technically advanced one."
Of all the millions of species on earth, only one has become intelligent and technically advanced.
Now, which is more likely:
- That two extremely remote chances came up independently for the same species [rare-squared]; or
- That the one might have something to do with the other?
What are the odds that aliens would look like bipedal humanoids, given that they are intelligent and technically advanced.
Paraphrasing Shermer's reasoning,
Of all the technically advanced species on earth, every single one is a bipedal humanoid.Or, alternatively:
Of all the bipedal humanoids on earth, every single one has become technically advanced.
Of course, by "every single one" we mean that have a single one. And while it is always risky to extrapolate from any sample, extrapolating from a sample of one is especially chancy.
Shermer is making the assumption that all possibilities are equally likely. [Perhaps he is also making the assumption that "chance" can be a cause of something.] What if there is some physico-chemical reason why only certain kinds of beings can be intelligent and technologically advanced?
What if there are certain templates or tendencies along which living things develop? For example, in so-called complexity theory [aka "chaos" theory] one can create a set of recursive equations whose outputs will always converge to [say] a picture of a fern leaf as its equilibrium state; and will do so regardless what the original input data were! Even if random inputs are fed into it, the fern leaf always comes out. [Was this Michael Barnesly?? Sp.?] Might genetic algorithms not be of this sort?[*] Consider that in the whole history of the earth, no new phyla have ever emerged since the Cambrian. The fundamental body plans seem to be givens. Ditto, the number of pre-protein folds seems limited to about a thousand [iirc] by the physics and chemistry. Perhaps a "spectral analysis" could show that only certain structures are physically possible!
[*] This may be how such a huge amount of info is packed into the genome. The blueprints aren't in there at all. Only the instructions for creating the blueprints. You don't have to spell out the structure of the fern leaf, only the recursive functions that create fern leaves.
Oh well, when talking about intelligent aliens, it would be nice to know if we are talking about the Null Set or not. There is a remarkable truth in mathematics: all theorems about the null set are true. This is because, in delightfully anti-Popperian fashion, you can never find a counterexample in the set; and if there is no member of the set for which the proposition is false, then the proposition must be taken as true.