If the impresario P T Barnum was right to say that there’s no such thing as bad publicity, then the paper just published in Nature by Sara Walker of Arizona State University, Lee Cronin of the University of Glasgow, and their coworkers has been a runaway success.1

The paper claims that an idea called assembly theory (AT) ‘explains selection and evolution’. This has drawn a clamour of responses from scientists on social media – many of them offended, some baffled – and prompted unusually vigorous debate in the online ‘comments’ section of the Nature site. Evolutionary biologists in particular have expressed outrage – denouncing the paper as nonsense, and even a Trojan horse for creationism.

Assembly theory might point to new directions in understanding molecular complexity

It’s not hard to see why. From the first sentence of the abstract, the paper seems to imply that the authors have cracked a foundational problem for biology: ‘Scientists have grappled with reconciling biological evolution with the immutable laws of the Universe defined by physics.’ No we haven’t, biologists respond – we have never found the slightest contradiction between them, and to suggest otherwise opens the door to intelligent design.

Some of the responses are also vituperative, often directed at Cronin himself, who has rarely held back from expressing bold and controversial views. Others see this as evidence that Nature is no longer a serious journal. That’s all unfortunate, both because AT might contain some fertile ideas, and because there are some useful lessons here for cross-disciplinary science.

It’s evolution, maybe

The potential virtue of AT is that it offers a quantitative framework for understanding how complexity arises – which is not, in general, by chance but as a hierarchical process of assembly from less complex building blocks. At first glance, your genome sequence looks like a random string of bases, but the same sequence appears with barely any deviation in each of your trillions of cells. This very high ’copy number’ is possible only because the sequence was selected, in this case by being templated on existing DNA and carefully proofread by enzymes. What’s more, that selection is deeply historically embedded: as molecular phylogeny shows, your DNA is a record of your evolutionary past.

No Theory of Everything can predict the existence of bananas

On the one hand, that’s all Genetics 101. On the other hand, articulating this view in AT might point to new directions in understanding molecular complexity. One of the problems for origin-of-life theories is that there seems to be a gap between the raw ingredients of the prebiotic Earth and the complexity of even the most primitive molecular systems capable of Darwinian evolution. Walker and Cronin believe that some kind of selectivity – perhaps due to chemical kinetics and catalysis, or the thermodynamics of self-organisation – must have helped to restrict the combinatorial explosion of prebiotic molecules and bridge that gap. By the same token, they have argued that there is a molecular complexity threshold, above which only life-like processes can generate high copy numbers.2 If so, the experimentally measurable “assembly index” of molecules could act as a biosignature for astrobiological searches that is agnostic about the chemical basis of alien life. In short, AT might help us decide, when we encounter apparent complexity, how surprised we should be by it.

Selective thinking

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.

And partly the paper does appear to contain a fair amount of hubris. However, I believe that provocative first sentence was not meant to suggest Darwinian evolution ever seemed to violate physics, but rather that physics itself struggles to accommodate the kind of historical contingency and path-dependence that is the hallmark of biology. At every step of the way, Darwinian evolution depends on what precedes it. You simply can’t calculate outcomes a priori: no Theory of Everything can predict the existence of bananas. Could physical laws predict Darwinian evolution itself, though? Could they predict life? Biologists mostly don’t care; for them, life is just what we have. But it shouldn’t count against AT that it raises the question.

The uproar is also an expression of a longstanding tension. There is an inglorious history of other scientists (physicists in particular) barging into biology only to contribute very little because they haven’t done their homework. On the other hand, sometimes these intrusions have made all the difference: ex-physicists Francis Crick, Max Delbrück and Seymour Benzer loomed large in molecular biology from the mid-20th century. Conflicts arise not so much because the intruders claim to have all the answers, but because they often insist on asking questions that don’t mean anything to biologists; as the late science historian and philosopher Evelyn Fox Keller put it, it is a matter of what counts as an explanation of phenomena. 3 While it’s possible for outsiders to ask important but overlooked questions, more often the truth is that they make the greatest scientific contributions when they are able to formulate their ideas in a way that is meaningful to the daily practice of those in the field. Perhaps that is now the challenge for advocates of assembly theory.