One of Murphy’s Laws of Technology facetiously avows, “Under the most rigorously controlled conditions of pressure, temperature, volume, humidity, and other variables the organism will do as it darn well pleases.” Can such a law be tested with an ammeter or a differential equation?

“Physics envy,” a term sometimes used to disparage the “soft sciences,” expresses the desire of biologists to share the scientific prestige of physics which has been highly esteemed since Newton for its mathematical precision. Launch a cannonball, calculate the force on a dam, or fly a spacecraft, and if the initial conditions and variables can be measured accurately, physicists can usually predict the outcomes meticulously. The laws of physics, expressible with differential equations and tensor calculus, can even predict new laws. But can evolutionary biologists predict the size and time of emergence of a small predator in an ecosystem overpopulated by mice? Hypothesizing “methinks it is like a weasel” is uncomfortably vague.

Organisms Obey Physics, But…

To be sure, the bodies of organisms will obey the laws of physics. When launching a human cannonball, physicists can predict where to place the net. The power output of an electric eel can be measured with voltmeters. The luminosity of fireflies submits to photometry. Terms such as entropy and enthalpy can also be used to model processes of photosynthesis in plants.¹

That’s not the kind of physics that evolutionary biologists envy. They want to predict what evolution will do, or post-dict what it has done. Sometimes they try too hard to imitate physicists in their stories by borrowing their terms. One recent paper,² for instance, attempted to describe “The emergence of eukaryotes as an evolutionary algorithmic phase transition.” They’ve compared a saltational evolutionary change, in effect, to water going from liquid to steam.

During the first phase of this process, corresponding to prokaryotes, protein length follows gene growth. At the onset of the eukaryotic cell, however, mean protein length stabilizes around 500 amino acids. [Emphasis added.]

Supposedly this is when the water starts exhibiting convection before boiling.

While genes continued growing at the same rate as before, this growth primarily involved noncoding sequences that complemented proteins in regulating gene activity. Our analysis indicates that this shift at the origin of the eukaryotic cell was due to an algorithmic phase transition equivalent to that of certain search algorithms triggered by the constraints in finding increasingly larger proteins.

In physics, phase transitions are reversible. Steam can condense into liquid water again. The authors of this paper want to present a major step up the ladder of evolutionary progress.

Perhaps their “phase transition” language can be forgiven for its pictorial value. They never explain, though, how the prokaryotes broke through this phase to emerge with new organelles, a nucleus, and new molecular machines like the spliceosome. With the magic word “emergence,” they simply allege that it happened.

The emergence of the eukaryotic cell, most likely arising from the symbiosis between two previously unrelated organisms — an archaeon host cell and a bacterium that would become mitochondria — brought a new cellular structure with membrane-bound nucleus and organelles. Without this evolutionary event, the posterior evolution of multicellular organisms represented by animals, land plants, and the majority of fungi would not have been possible.

Feel the tension (another physics term measurable with instruments). The authors struggle to make their model look like a physics paper. They even derive the tension with equations.

Our blend of theoretical and empirical approaches will ultimately uncover how the tension between a conserved process of gene growth and the constraints on the increasingly longer proteins resolved in a phase transition signaling the emergence of the eukaryotic cell.

Later they are even more explicit that their model belongs in the physics category, even like astrophysics. But in drawing this comparison, they point out an embarrassing gap.

Our portrait of a phase transition is in agreement with the lack of intermediate forms behind the emergence of eukaryotes — what has been termed a black hole at the heart of biology. Previous work has also highlighted the shift between prokaryotes and eukaryotes on the basis of energetic constraints, or metabolic allocation. Our results add an algorithmic dimension to this view emphasizing the role of constraints. It reconciles the contingency of evolution — exemplified by the random exploration of the search space — and the universality of physics.

Evolution, they hope, has reached the pinnacle of physics. Time to self-congratulate!

Importantly, our framework has an unparalleled predictive power, as shown by its ability to predict not only the specific laws governing the growth of genes and proteins across the entire evolutionary history, but also the precise moment in time at which eukaryotes emerged and the critical mean gene length at which this occurred.

Unsurprisingly, the press release from Gutenberg University celebrates the achievement of its homeboys. “The study recently published in PNAS not only answers essential questions, but is interdisciplinary, combining computational biology, evolutionary biology, and physics.” Great. Now use the equation to calculate the evolutionary emergence of the weasel in deep time with 6-sigma precision.

No Weasel Words Allowed

At The Conversation, Kathleen Garland and Alistair Evans from Monash University also appeal to physics for their evolutionary theory. “A secret mathematical rule has shaped the beaks of birds and other dinosaurs for 200 million years,” their article proclaims — which is a bit odd, for it is not the tradition of physicists to keep their mathematical rules secret. 

We can feel their physics envy in the essay:

Finding universal rules in biology is rare and difficult — there seem to be few instances where physical laws are so pervasive across all organisms.

But when we do find a rule, it’s a powerful way to explain the patterns we see in nature. Our team previously discovered a new rule of biology that explains the shape and growth of many pointed structures, including teeth, horns, hooves, shells and, of course, beaks.

This simple mathematical rule captures how the width of a pointed structure, like a beak, expands from the tip to the base. We call this rule the “power cascade”.

After this discovery, we were very interested in how the power cascade might explain the shape of bird and other dinosaur beaks.

Few would complain about rule-governed processes in embryonic development, where physical pressures between cells and flows of material follow laws of fluid mechanics under genetic control. They’re not talking about those. They’re talking about Darwinian evolution, as they stated in the preceding sentence: 

By studying beaks in light of this mathematical rule, we can understand how the faces of birds and other dinosaurs evolved over 200 million years. We can also find out why, in rare instances, these rules can be broken.

Their photos show the hooked beak of an eagle, the long bill of a spoonbill, the short snout of an ostrich, and the curved bill of a godwit. “All these bird beaks follow the power cascade rule of growth, despite being used for very different purposes,” they boast. Their physics-like model explains all — except when it doesn’t. “While rare, a few birds we studied were rule-breakers.” The Eurasian spoonbill (pictured at the top) evolves as it darn well pleases. “Perhaps its unique feeding style led to it breaking this common rule,” they sigh, as they return to Just-So Storyland.

Playing Games

A look through another paper in PNAS is sure to satisfy physicists with its equations and matrices.³ Evolutionary Game Theory (EGT) with its apparent mathematical precision comes close to mollifying physics envy. Unfortunately, these authors come to bury Caesar, not to praise him.

We argue that it is crucial for the field of evolutionary game theory to emphatically acknowledge that the replicator dynamics is more limiting than previously acknowledged (or fully appreciated) in terms of the biological systems that its predictions can apply to. Here, we have discussed a fundamental limitation of the replicator dynamics, the implications of which had not been reconsidered or explicitly acknowledged as a substantial caveat, even as the field moved to address other limitations, e.g., that the replicator dynamics ignores stochastic effects, that it muddles growth rates and interactions, or that it apparently ignores more realistic genetics and demography. This fundamental assumption — that the game is the only source of fitness differences — is so likely to be broadly violated across natural systems, that the potential for paradoxical findings and unproductive debates is very high.

Stochastic effects ensure that the organism will do as it darn well pleases. 

A week later in Current Biology,⁴ John Harte discussed another biological field where failure is always an option. He summarized an ecological model that demonstrated only partial success:

A theory-derived ecological equation of state relating biodiversity, productivity, abundance and biomass in ecosystems has been tested with satellite-derived proxy forestry data. Predicted failure of the relationship in disturbed ecosystems is partially supported but further ground-based analysis is needed.

Biology Is Not Physics

Biology, with all of its subdisciplines, is an ancient science worthy of esteem for the wealth of insights it brings. But it’s not physics. Biology is largely observational and taxonomic, and its explanations tend to be statistical (i.e., messy). Like Garland and Evans admitted, rare are the instances where identifiable causes yield predictable effects in law-like regularity. 

Irregularity is due to the nature of the beast. Animal algorithms predict behaviors en masse but there will always be exceptions. Evolutionary biologists invest great hope in the exceptions, the mutations, the innovations, that they believe “nature” will “select” to drive a population to new fitness peaks (but see this). Much as they would like to wrap the magic word “emergence” in equations, the nature of biology balks at being reduced to physics. I asked Paul Nelson about this. He replied:

I don’t think the problem in biology, however, is physics envy, as much as it is reality denial, or, more charitably, trying to use the wrong analytical / explanatory tools for the phenomena at hand. Imagine Caravaggio trying to paint this eye … with house paint rollers, or spray paint. Ain’t happening. “Physics envy” is really the belief that “equations will suffice,” when the data — a single protein, for instance — simply cannot be expressed as any mathematical relation. Amino acid sequences (and their nucleotide templates) are highly incompressible. If you want a histone or a kinase, you need the complete sequence, and math will not contain enough detail to specify that sequence…. If a scientist denies reality by trying to use the wrong tools for the data at hand, he will produce nonsense.

The incompressibility in biology is a feature, not a drawback. It’s what makes our world beautiful with wildflowers, towering redwoods, majestic whales, and individual human faces. When it comes to explaining origins, evolutionary biologists seek to alleviate physics envy in vain. They’ll find more relief, I believe, in teleology. On that, see the new book by David Klinghoffer about the thought of ID biologist Richard Sternberg, Plato’s Revenge: The New Science of the Immaterial Genome.

Image source: Discovery Institute Press.

Notes

1. Hall et al., “Entropy is an important design principle in the photosystem II supercomplex,” PNAS 19 March 2025.
2. Muro et al., “The emergence of eukaryotes as an evolutionary algorithmic phase transition,” PNAS, 27 March 2025.
3. Tarnita and Traulsen, “Reconciling ecology and evolutionary game theory or “When not to think cooperation”, PNAS, 31 March 2025.
4. John Harte, “Ecology: Why failure is success for an ecological theory,” Current Biology, 7 April 2025.