Assembly Theory pops up again

This morning I was surprised to see Assembly Theory popping up all over in my social media. Did somebody find evidence for it? Did the authors clarify what their babble meant? No, nothing so interesting: another evolutionary biologist took a hard look at the original paper, and tried to figure out what Cronin was talking about.

In October, a paper titled “Assembly theory explains and quantifies selection and evolution” appeared in the top science journal Nature. The authors – a team led by Lee Cronin at the University of Glasgow and Sara Walker at Arizona State University – claim their theory is an “interface between physics and biology” which explains how complex biological forms can evolve.

The paper provoked strong responses. On the one hand were headlines like “Bold New ‘Theory of Everything’ Could Unite Physics And Evolution”.

On the other were reactions from scientists. One evolutionary biologist tweeted “after multiple reads I still have absolutely no idea what [this paper] is doing”. Another said “I read the paper and I feel more confused […] I think reading that paper has made me forget my own name.”

As a biologist who studies evolution, I felt I had to read the paper myself. Was assembly theory really the radical new paradigm its authors suggested? Or was it the “abject wankwaffle” its critics decried?

He highlights some of the weird stuff in the paper, like this observation that leapt out at me when I read it.

In the words of one Nature commenter: “Why so many creationist tropes in the first few sentences?”

Yeah, that was odd: either the author was so totally unaware of how creationists make really bad arguments, so he made one himself (the generous interpretation), or it was intentional, and he’s underhandedly trying to sneak creationist nonsense in the literature (the uncharitable interpretation). Either way it was a warning sign.

So the critic works through the paper, trying to answer the question, is it “abject wankwaffle” or not? The answer is phrased politely.

However, as a sweeping new paradigm aiming to unify evolution and physics, assembly theory appears – to me and many others – to be addressing a problem that does not exist.

“abject wankwaffle” it is!


  1. Matt G says

    Physics and biology are already unified without Assembly Wankwaffle, but thanks for the effort!

  2. says

    The authors … claim their theory is an “interface between physics and biology”…

    It sounds like they’re talking about living things as if they’re assembled machines, thus baking an unspoken assumption of “design” into their “theory” early on.

    So yeah, I’m going with “abject wankwaffle” too — but are they really worth so many syllables?

  3. Matt G says

    Raging Bee@2- I’m trying to remember the Third Way, and their attempt to inject design into evolution. I somehow downloaded the first chapter or two of their “textbook” for free before I realized it was nonsense. I think it was 11 years ago when I got my first iPad and it popped up on the Apple ebook store.

  4. sophiab says

    I just…what? WHAT???

    There is no problem between biology or evolution in particular and physics. What’s meant to be the problem?

    There was at least one time in history when there was some bafflement between biology (and geology) and physics in history. At some point in the 1800s continental drift and evolution understanding existed, but “how the fuck does the sun work for over 1 billion years” was really confusing.

    1900s answered that. Physicist, high-school biology but as far as I know there is no problem in any way or form between evolution and physics. Or is dark energy cosmic jellyfish pushing everything away?

  5. says

    Scientists have grappled with reconciling biological evolution with the immutable laws of the Universe defined by physics. These laws underpin life’s origin, evolution and the development of human culture and technology, yet they do not predict the emergence of these phenomena. Evolutionary theory explains why some things exist and others do not through the lens of selection

    They could have ended it right there.

  6. Reginald Selkirk says

    I think reading that paper has made me forget my own name.

    And who wrote that? We don’t know.

  7. says

    Put that wankwaffle in your toaster and smoke it!
    Sounds like they are trying to show how making sausage explains how a diesel engine works.

  8. kome says

    I know peer review is imperfect, but damn Nature, you really screwed the pooch on this one. Whoever approved this for publication should be publicly shamed.

  9. garnetstar says

    You are so right, kome @10. What is Nature coming to? Going for eye-catching titles, instead of reviewing what comes after the title?

    This guy is just unbelievable: he’s a synthetic chemist, he calls himself interested in “complex” chemisty. All chemistry is complex, in his meaning of the word! (WTF does he think matter is?) All chemistry involves atoms and molecules instantly falling together in “designed” self-assemblies that sure do follow the laws of physics. And he calls his clusters “complex”: clusters of similar sorts have been done for decades now.

    The rest of us call “chemical evolution” “reactions”. That the structures of proteins and such in living systems get more and more efficient and fine-tuned over time is just called “evolution”.

    I might force myself to watch his TED talk: he thinks he can produce life that is not based on carbon. I want to see if he’s got anything to propose that wasn’t explored and discarded decades ago. Also, the whole idea has long been deemed highly unlikely, as no other element has the properties of carbon. That’s why it works, get it?

    He needs to start teaching general chemistry again, where all the thermodyamics, kinetics, and quantum mechanics laws of physics get hammered home to the students, and he’d have to re-learn it.

  10. says

    Doc Wankwaffle is my new porn star name!

    And from that name alone, the stupid porn-movie plots just write themselves…or so I’m told…

    The authors…claim their theory is an “interface between physics and biology”…

    Um…we already have an interface between physics and biology — it’s called CHEMISTRY. And I don’t remember any chemists showing how evolution and physics need to be “unified”…

  11. says

    …he thinks he can produce life that is not based on carbon.

    Some people have said that life based on silicon might be doable, because silicon can bond with four other atoms at once like carbon; and I remember some vague rambling about how it might be possible on Titan, because Titan has lots of silicon, and it’s colder so silicon bonds, which are weaker than carbon bonds, would last longer.

  12. wzrd1 says

    So, another fine paper that was left in the wankwaffle iron for too long.
    Perhaps peer review should be provided via cattle prod…

  13. Dr. Pablito says

    Yeah, this nonsense got published in American Scientist a couple months back, and while it’s no Nature, it’s still a pretty good magazine for the Science/Engineering literate. I read it in AmSci and I was just completely befuddled as to how they decided to publish it. It made no goddam sense. It was all this flabby writing about nothing with circular reasoning about entropy and information and complexity and it went nowhere and said nothing. It’s not science. Doesn’t predict anything, doesn’t advance understanding on any current problems. It’s not clear what problem they’re trying to address. There’s no physics content in it at all.

  14. nomdeplume says

    Reminds me of the YEC nonsense about the second law of Thermodynamics. And of how DNA is “information” like a computer code. And of the media tendency to treat every new announcement, or preliminary tentative hypotheses, as having “rewritten Darwin” or “caused a rethink of evolution”. There is no problem with evolution. Certainly no problem of physics and evolution (or chemistry and evolution or geology and evolution). If these clowns aren’t creationists then they are trying to get publicity, and either way shame on Nature for facilitating and legitimising them.

  15. Dr. Pablito says

    Ah! Here it is online, though subscription only.
    But this came out a couple of months back, which explains why I saw it before this latest publication in Nature. The title of the article is “Time is an Object,” and I had thought, “Oh, great, somebody new thinking about the deep physics problem of what is time, exactly?” and I was anticipating an “It from Bit” approach (which is its own type of nonsense, but nevermind), but they never actually got around to discussing really anything about time, it was all this complexity theory nonsense and I couldn’t understand why they wanted that title on the article.

  16. StevoR says

    Assembly not so much Assembly “Theory” as “Assembly doesn’t even really qualify as decent speculation” instead.

  17. imback says

    I had to google it to remind myself that “abject wankwaffle” was coined by Adam Rutherford last month in response to the assembly theory paper.

  18. bcw bcw says

    …so, did this wankwaffle paper get written before or after ChatGPT came out? ChatGPt’s a much more likely origin story for the paper than if real people wrote it. A pastiche of partially real concepts with highly touted populist internet misconceptions – that’s ChatGPT to the core.

  19. chrislawson says


    Yep, the very fact the authors phrased this as a unification of biology and physics was a huge red flag that should have alerted the Nature editorial team. Especially as it’s not even really physics they’re doing. Their method doesn’t even look at the binding energy of any of the bonds involved and is really just combinatorial probability theory, badly applied. Seriously, where is the physics in counting the number of bonds in a molecule?

  20. chrislawson says

    And I didn’t say this the first time around, but you can tell their entire approach in intellectually dodgy from this quote alone:

    These results show that for small molecules (mass < ~250 Daltons) the MA is strongly constrained by their mass. This is understandable because small molecules have limited compositional diversity and few structural asymmetries. The MA of molecules with a mass greater than ~250 Daltons appear to be less determined by molecular weight, indicating that they can display vastly more compositional and structural heterogeneity. This is significant because it gives us insight into how to develop an experimental measure of MA based on tandem mass spectrometry by focusing on fragmenting molecules that have a mass greater than 250 Daltons.

    They explain that their method fails for larger molecules — a Dalton is 1/12 the mass of an unbounded carbon atom, so 250 Daltons means 20 or fewer carbon atoms (possibly much fewer depending on the number of H, O, N and other atoms). This means their system fails for almost all of biochemistry. And their brilliant solution? Break up the large molecules and measure the fragments. To put this into perspective, the common biological molecule albumin is about 66.5 kD, that is, 266 times the upper threshold of their analytic method. And yet they think by breaking it into tiny fragments they can still measure its complexity despite excluding all of the complexity of how those fragments can be arranged.

  21. wzrd1 says

    bcw bcw @ 21, well, Artificial Idiocy vs Actual Idiocy do have parallel reflections, nearly perfect enough to permit lasing.
    Well, save that the intervening media is de-energizing, incapable of population inversion, only population perversion.
    And well, perversions, I prefer those pure and distilled, entirely consensual and extraordinary in feats performed.
    Typically requiring ICU care shortly afterward.
    Yeah, my eyes are brown for a reason… ;)

  22. wzrd1 says

    chrislawson @ 23, nonsense and poppycock! Testosterone is around 288 Daltons of mass, so Estrogen is literally the same, as is well, most cholesterol molecules.
    Point those out, with full numbers it’s a tad too late for my to dig up and watch their brains melt down when men and women are equal in those idiotic metrics.
    Attack the mindset and origin of their position, using actual data to fully emasculate them.
    Yes, my choice of words was quite intentional.
    Words are weapons and weapons should be precisely delivered to target, for striking the small finger isn’t of great effect on a charging beast, whereas striking a vital area immediately recognized as a major blow is.
    If charged by a dangerous beast, one engages CNS, if one can neutralize it instantly, lacking that, circulatory central centers, if all else fails, peritoneal regions, largely in hope of overloading the CNS via damaging the largest innervation center in the mammalian body. The latter, a Hail Mary, but with humans, always a good resource to attack even metaphorically.
    And a major vertebrata weakness anyway.
    If a tiger attacked me from the front, get my legs between its and its chin pinned to its chest, it’s helpless. I can do a hell of a lot of damage while it tries to figure out only its front claws, originally working to old prey are now its only weapons.
    From behind, well, I’d likely be toast. I’d need some warning in advance, such as a mirror.
    Hind claws go for evisceration by evolutionary design, separating them gives an advantage, weakening any possible attack and leverage gives me an advantage briefly. The same is true with jaw muscles in a fairly common design, potentially lose function in an upper limb, while neutralizing attack and weakening the entirety of bite force and respiration.
    Effective strategy, I did it once, out of desperation in a bad situation.

    One has a mind, that’s the only real weapon one needs, if properly focused on one’s target and not distracted by sparkles.

  23. birgerjohansson says

    Wzrd1 @ 25
    I just realised; the hind claws of a tiger is just about the only thing the thin metal plates* in Russian body armor might defend against!
    *The good body armor is routinely stolen and sold by officers.

  24. StevoR says

    Is it ironic that this paper on so-called Assembly theory seems to be disassembling, er, disembling?

  25. chrislawson says

    @26– maybe, but it doesn’t feel like Sokal to me. My bet is on undeclared theistic evolutionism.

  26. Louis says

    When I read the paper, the overwhelming sense I had was familiarity. I read S Wolfram’s “A New Kind of Science” years ago, and despite the book’s contents being way out of my “comfort zone”/professional experience, I didn’t see what all the fuss was about. Kind of a “the parts that were good weren’t new, and the parts that were new weren’t good” sense.

    Now, I will be happy to be corrected re: Wolfram’s work by someone more knowledgeable than I am, and it is donkey’s years since I read the book, so forgive me! However, that was my strong takeaway about the internal logic of the ideas contained within.

    The same was the case with the Assembly Theory paper.

    I know it won’t get much sympathy here, but I know Prof Cronin a little from The Olde Days when I was a chemist (as opposed to an ex-ish chemist), and he’s a very good chemist. I’m not sure why/how he’s gotten out over his skis on this one, but my impression is that he has. I don’t know if this is a “pre-Nobel”-Syndrome thing/”Physicist’s Disease” issue, i.e. “Hey! I’m pretty damned good at this science lark, I must be good at all of it!”. I’m not saying he’s pre-Nobel btw, just that he hasn’t won one! I know being a good scientist doesn’t prevent crankiness, and this could be another data point in the broader Salem hypothesis, but it all struck me as…odd.

    The paper itself was not good (in terms of how written/structured/contextualised re: the field it was intended to address), and the ideas contained within struck me as confused, perhaps because of a lack of field familiarity by Cronin. I can’t be sure. I am a little worried that the responses to criticism (much more substantial and detailed than what I have written here…not tough! I’ve described my feelings!) have also been…”sub-optimal” and a bit “creationist-y” in mechanism.

    I do agree that in trying to “Push Back The Boundaries Of Science(TM)” one must “Think The Unthinkable(TM) and “Overturn The Paradigms(TM)”, all of this is true, although a surprisingly minor part of how modern science advances, except in the popular beliefs about the workings of science. But I am always reminded of Sagan:

    “But the fact that some geniuses were laughed at does not imply that all who are laughed at are geniuses. They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown.”
    ― Carl Sagan, Broca’s Brain: Reflections on the Romance of Science


  27. abb3w says

    Now, I will be happy to be corrected re: Wolfram’s work by someone more knowledgeable than I am, and it is donkey’s years since I read the book, so forgive me! However, that was my strong takeaway about the internal logic of the ideas contained within.

    I’d say you’re not far off on Wolfram; contrariwise, I found he did give a unified presentation regarding some far-flung corners of mathematics that don’t get a lot of attention in the conventional K-12 and college undergraduate core curricula, did provide some pretty pictures to attract the attention of more visual learners, and did actually come across two or three new mathematical findings during the development of the book. Nohow, my impression was that the work became less sound when he attempted to transition from abstract mathematics to particular physics.

    My impression is that the authors of the Assembly Theory paper appear to be trying to re-invent Computational Complexity theory – far more ineptly. While Wolfram presents some introductions for some of the notions of that framework which might in parts (pretty pictures) be gentle enough for bright K-8 students, standard textbooks such as Linz and Sipser give a more rigorous notion; and (unlike Wolfram) the Assembly Theory paper doesn’t seem to have made any connections from what they are attempting to the broader work done in mathematics over the last century.

    In both cases, someone seems to be attempting to re-invent the wheel. Wolfram’s attempt seems to have at least developed some pretty spandrels along the way; the Assembly Theory paper seeming has not.

  28. Louis says

    @Abb3w, #31,

    Thanks for that. :-)

    You remind me that the particle physics bit was the bit I knew more about than the mathematics bit…I need to reread the book!

    The Assembly Theory thing strikes me (as hinted at above) as “Senior Scientists Have Idea And Get Published (when they shouldn’t have been because this wasn’t rigorous enough)”. I wish I had a bit more time to go through it in appropriate detail, and since today seems to be quote day:

    The Assembly Theory paper seems to be in this class of “book”:

    “Not a book to be lightly thrown aside. Should be thrown with great force.”

    William Eli Miller, attrib to Frank Dolan, also attrib Dorothy Parker…


  29. says

    Or we could just consider that at the real interface between physics and biology — chemistry — we have silly little things like “stereoisomers”/”chirality,” “intermediate states and persistent intermediate products,” “incomplete consumption of reactants,” “side-chain reactions,” “intramolecular isotopic decay”… that is, the very messy real world that reflects that an easy case like exposing free methane to ultraviolet light does not result in all of the methane being turned into methyl radicals† and free hydrogen atoms (which, since this is a theoretical exercise completely divorced from reality, don’t then immediately combine to form a significant proportion of stable hydrogen molecules) in biologically-interesting conditions, at biologically-interesting times and reaction rates and energies.

    Go ahead, guys. Try and figure out the actual proportions of reactants involved in the Krebs cycle, and ponder how that relates to doing any of those reactions in a pure test-tube environment (bonus to make this easier: you get to do it in a pure, unadulterated solution of your choice, and we’ll let you ignore both equilibrium-state dynamics and activation energy with some handwaving misuse of unchanging-state math).

    † Radicals. What is this, the sixties all over again? Or did you forget that radicals don’t behave like electrochemically stable compounds… but are definitely part of the state of nature, if only part?

  30. Louis says

    Jaws, Jaws, Jaws…

    Go on. Tell ’em about surfaces and their interactions with solutions. Go on.

    Oh oh oh! I know! Tell ’em about solution-phase structure of biomolecules and hydration!



    P.S. I’m not making fun of YOU, I am remembering how best to really upset (we) chemists!

  31. says

    Chirality may be down to acetate (and glycine) having a polarity (structural and charge based) that enabled attachment to a substrate/adaptor, and substrates only coming from one direction. Like hydrothermal fluid flow. Or donation through a cofactor like biotin.
    Hence lots of molecules only existing in one stereoisomer.
    Attachment through one end and receiving from one direction. At the level of biosynthesis of 3-4 carbon amino acids you start seeing functional group swapping between similar sized molecules. Like cystine donating it’s sulfur for methionine through cystathionine.

    Or maybe the electromagnetic right hand rule is involved (for the direction of magnetic field relative to charge motion, I think).

  32. Louis says

    Don’t make me tap the sign…

    [Sign reads: Origins of homochirality are likely a happy accident due to reactions similar, but not necessarily identical, to the Soai reaction, which, in essence, amplifies the first chiral product it produces]



  33. wzrd1 says

    Jaws @ 33, I recall a failed replication attempt of the 1952 organic molecules from inorganic precursor experiment and testing as to why a replicated experiment was suddenly failing. It turned out that the latest replication attempt was utilizing tempered soda glass, rather than Pyrex and when crushed Pyrex was added into the bath, replication was successful. This lead on to other experiments, where more complex protein components were produced, again, with boron as a catalyst.
    Yeah, things can get a bit complex and overlooked, until later work is done and goes unexpectedly sideways and further investigated.

    Why, it’s almost as if nature itself is complicated! ;)

    As for chirality, some chiral compounds may react in a similar, but more rapid or slower rate than its racemate, with what works better in that specific niche being preferentially selected for. Just as deuterium reacts similar to hydrogen in complex reactions, but a bit “sluggish” (for the lack of a better word) and tritium, radiation aside, can easily reach poisonous levels for life, despite all three being hydrogen isotopes. After all, within places like mitochondria, reaction timing is also quite important.