This may generate some excitement. The authors have come up with a way to quantify the amount of selection needed to create a molecule.
I will comment further after I’ve read the supplementary section.
1 Like
This appears to be a comment on the above, but I don’t have access.
How purposeless physics underlies purposeful life
Evolution by natural selection peerlessly describes how life’s complexity develops — but can it be explained in terms of physics? A new approach suggests it can.
1 Like
I have access to the 2nd one (maybe because the link recognizes that this computer is sometimes on university property)?
Anyway, In perusing the articles, I find myself rather split down the middle. On one side I can see well enough that this stuff is well beyond my pay grade, and so I should be very hesitant to pass judgement on it. On the other side, I am seeing a LOT of hand-wavy modeling done by people who show no signs that they know much about biology, so they are “winging it”. That seems perhaps uncharitable, and maybe I don’t understand it. So I would be interested in learning what other people think.
2 Likes
A recent development on this is that there is a bit of a dust-up on Twitter (X) that the paper is pure B.S., which I find heartening since it felt that way to me. Also it was funded by the Templeton Foundation, which raises some suspicions.
1 Like
And here I was getting ready to say something nice to say about it.
After reading the supplement I don’t think it is BS, but I’m not convinced it is useful either, at least not yet. Then my weekend got busy, so I haven’t followed up the references.
The basic framework for the Assembly Index (AI) looks OK to me. The title seems like a grand claim, but if we set biology aside, I think the “explains and quantifies selection” part is OK. If we understand “and evolution” to mean adding more parts as the paper describes, again without biology, that might be OK too.
Can this be usefully applied to biology? I still don’t know.
There is a citation to a paper an algorithmic information theory which I am familiar with (Wallace 1999). I think AI and Minimum Message Length (MML*) are related in how they reduce the data to essential parts.
*MML is an important statistic if you want to make statistical inferences using Algorithmic Information. Nobody does that because it doesn’t work very well, or so I’ve read. I don’t think this bodes well for applications of AI either, but that remains to be seen.
Note: I’m pretty sure Dembski (2005) cited this same paper by Wallace. Dembski didn’t actually use MML, but it would be a wonderful irony if MML or something like it went on to be a key to explaining biological evolution. 
Another citation is to is an article on Constructor theory (Deutsch 2015), which is a topic I’ve been trying to follow. Constructor theory requires enumerating all the states a system can change (or evolve) into, and the AI seems like a step toward doing that.
I will need to look closer at the connection between MML and AI, and the other references. Stay tuned.
2 Likes
So why the fuss? Part of the friction comes from terminology. When Walker and Cronin talk about selection and evolution, they mean something that overlaps but does not coincide with Darwinian evolution by natural selection. That could have been made more explicit – far from ‘explaining’ selection, their paper seems barely even to discuss it in a way evolutionary biologists recognise.
I personally know at least one fantastic scientist who was (and still is) JTF funded, and for that and more basic reasons, I reject these kinds of aspersions. I dislike JTF’s aims and influence but feel strongly that this should not prejudice us toward scholarship that they fund. And besides, one of the best summaries and critiques of the paper, which is straightforwardly critical, comes from a JTF-funded scholar:
1 Like
Let’s face it: the abstract was impenetrable and probably doubly so for biologists because it uses some standard vocabulary in a non-standard way. Nature and Science papers are supposed to be decipherable by a science-literate audience in any field. This was at least a major editorial failure. Whether the paper itself makes sense is unknown to me, and that’s part of the failure.
2 Likes
There is also interesting discussion in the comments section at Nature, particularly the comment by “professor_dave”, who seems to have a good grasp of how assembly theory might be applied to detect signs of extraterrestrial life.
Agreed - It’s not a biology paper at all. The abstract is a train wreck.
4 Likes
I rather like the comments by Kasper Kepp at the end of the paper. He takes it apart rather nicely.
2 Likes
Lee Cronin weighs-in on Assembly Theory and how he intends to use it.
Assembly Theory as a theory about assemblages of molecules may be correct and useful. But the stuff in the first part of the paper about how physics is somehow in conflict with life is both wildly dramatic and ridiculously wrong. There actually is a whole field called “biochemistry” – and it’s almost 200 years old. And last I heard there was no congtradiction between physics and chemistry.
7 Likes
This topic was automatically closed 16 days after the last reply. New replies are no longer allowed.
Reopening this topic to note recent developments.
It’s been a while, but I had been curious to see if William Dembski would weigh in on the subject, and he finally has.
1 Like
I quickly read the ENV article, and I mostly agree with Dembski. He’s a little late to be getting on the Criticize Assembly Theory (AT) wagon, but that’s OK.
Deeper into the article, the more I find disagreement with Dembski, and these all have to do with the Assembly Index, which is supposed to be a measure of object complexity.
-
Dembski objects to the reversible or symmetric property of the basic elements which more complex objects are assembled from. I disagree; this is just allowing for symmetrical sequences to be counted as a single element. Worst case you need to add reversed elements to the set unique elements, and problem solved.
-
Dembski objects to the LEGOS example, claiming that “flexible” connection cannot be simply described. I think the LEGOS example was fine, and he is making a big deal out of not much.
-
He also objects to nested hierarchies, but I don’t see why. Hierarchies are built-in to the recursive nature of the assembly index.
-
Dembski may be right about application to assembling proteins, noting that binding energy makes some connection more difficult than others. I’ve already seen discussion of including “binding energies” in the Assembly Index, and I think that hurdle can be cleared.
Note that clearing these objections still doesn’t mean the Assembly Index is going to be useful. I can’t really speak in favor of Assembly Theory, I’m just pointing out where I think Dembski’s criticisms are unfounded. MOST of his criticisms are valid, IMO.
I find it ironic that many of Dembski’s criticisms of AT applies equally well to his own work on Complex Specified Information. The Assembly Index makes use of Minimum Message Length (MML) which is closely related so Kolmogorov or Algorithmic Information. The methods used to approximate Kolmogorov Information are closely related to calculation of MML.
Note that MML plays a similar role in Algorithmic Information Theory as Maximum Likelihood Estimation plays in Shannon Information and Statistical theory. By the accounts I’ve read MML “doesn’t work very well”. Therefore my expectation is that Assembly Theory and the Assembly Index won’t work very well either. Still, I think Cronin and his collaborators are making a serious effort, and I remain curious to further developments.
Irony unintentional?
2 Likes
This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.