10 Comments
User's avatar
David Hope's avatar

I enjoy your thinking.

The frustration you express regarding the "asterisk"—the idea that domestication or specific lineages are simply "exceptions to the rules"—suggests you are brushing up against a genuine problem of language. This is not a matter of philosophical fluff, but rather a structural mismatch between your data and your terminology.

The core of the issue is that you are observing a process, but the system of taxonomy requires a category.

Throughout the piece, you describe evolution as a series of verbs: diverging, mixing, replacing, introgressing. You describe a world of gradients and "taxonomic tangles." Yet, the conclusion of the scientific process is almost always the assignment of a noun: Bos cf. baikalensis or Denisovan.

The "species problem" you mention is often treated as a lack of sufficient data—the idea that if we just had more genomes, the boundaries would become clear.

But your comparison between bovines and hominins suggests the opposite: that the more data we have, the more we realize that the "boundary" itself is a linguistic fiction.

When you ask if Denisovans are "really" a species, you are essentially asking if a noun can accurately describe a fluid, shifting population that exists in a state of constant gene flow. The answer is no—not because the science is incomplete, but because a noun is a static tool, and evolution is a dynamic process.

This is why the "asterisk" feels so unsatisfying. To say that domestication is an "exception to the rules of speciation" is to admit that the "rules" (the nouns) are failing to describe the reality (the process). It frames the biological reality as a "glitch" in the system, rather than framing the system as a clumsy approximation of the reality.

You do not need to turn this into a treatise on the philosophy of language, but there is a powerful opportunity here to pivot the conclusion. Instead of wondering if Yak X is "really" a different species, you could argue that the "species" label is a heuristic—a useful shorthand—that we are now outgrowing.

By pointing out that the "taxonomic tangle" of the bovine is a mirror of the hominin, you are not just comparing two animal groups; you are demonstrating that the tension between "the name" and "the genome" is a universal feature of mammalian evolution. The "problem" is not with the yaks or the Denisovans; it is with the insistence that nature must fit into a dictionary.

I am a low vision communicator who uses assistive technology.

To the memory of the Blessed Margaret Pole

John Hawks's avatar

Really outstanding comment, thanks!

The core of your comment is entirely in line with where I am going with species. Frustration with this disconnect of continuous versus dichotomous variation is not original to me of course, it comes originally from Darwin, who was the first to grapple with how to find words for the continual and (in his view) gradual process of change, when the endpoints had long been considered to be hard and cleanly separated. His examples again and again reinforce that so many forms intergrade continuously that it is reasonable to infer the gradual origin of all species, with discontinuities a result of the extinction of intervening forms.

It is very perceptive of you to emphasize the expression of process using verbs, and species using nouns, as an illustration of the conceptual mismatch. I'll be writing more about this in upcoming weeks, as posts like this one are really only laying groundwork.

David Hope's avatar

We observe that the traditional Darwinian matrix of evolution is anchored in the metaphor of the tree. This topological framework provides a powerful sense of conceptual unity, predicated on a unitary starting point from which all subsequent life diverges. In the tree model, the trajectory of a lineage is linear and unidirectional; once a branch splits, the governing logic of the geometry dictates that these branches never meet again. Evolution, in this view, is a history of divergence—a relentless splitting of the one into the many.

However, the emergence of ancient DNA data from hominins and bovines reveals a profound structural mismatch between this linear geometry, and biological reality. The evidence of introgression, hybridization, and "cytonuclear discordance" suggests that evolution is not merely a process of splitting, but one of reticulation. Reticulate evolution describes a network in which branches that have already diverged merge back together. When lineages converge, the "tree" is broken and replaced by a "web."

The fundamental tension lies in the requirement for a unitary origin. To preserve the conceptual unity of a species, the Darwinian matrix assumes a single, unified point of beginning. This ensures a clean narrative of descent.

But such a model cannot accommodate polyvalence—the idea that a lineage may have multiple, simultaneous starting points. If a species emerges from the hybridization of two divergent populations, its origin is not a point, but a nexus. Its identity is not derived from a single ancestral line, but from the convergence of multiple genetic streams.

In a linear system, any genetic contribution from an outside lineage is treated as a deviation, an anomaly, or a "pollution" of the pure line. This is the source of the "asterisk" in taxonomic classification: the tendency to treat introgression as an exception to the rules of speciation rather than a fundamental mechanism of it. By insisting on a unitary point of origin, the linear model frames the merge as a glitch in the system, rather than recognizing that the system itself is a network.

The transition from a tree to a web represents a shift from a geometry of divergence to a geometry of convergence. While the tree focuses on the split, the web focuses on the mixture. The "species problem"—the difficulty of naming a boundary where one group ends and another begins—is therefore not a problem of missing data, but a problem of topology. It is the result of attempting to map a multivalent, networked history onto a linear, divergent matrix.

Ultimately, the struggle to define species like the Denisovans or the Yak X clade is a symptom of this geometric conflict. The insistence on conceptual unity through a single ancestral point creates a friction that the current language of taxonomy cannot resolve.

Until the framework moves beyond the linear constraints of the tree and accepts a polyvalent model of origin, biological reality will continue to be viewed as a series of exceptions to a rule, rather than a manifestation of a more complex, networked truth.

I am a low vision communicator who uses assistive technology.

In memory of the Blessed Margaret Pole

Rothari's avatar

Wondering how many potential domestication or semi domestication efforts started and stopped, or even succeeded before failing eventually and getting wiped from the record.

John Hawks's avatar

This is a fantastic question. It seems likely to me that *taming* by raising orphaned cubs was invented many times and probably within the behavioral experience of Neanderthals and contemporary people. The domestication of the first dogs was maybe the first of these to stick over many generations, but maybe not.

Rothari's avatar

I've also been wondering...what if before stone we started off with *wood*. Very decayable wood-stab a fish with it, pick a fruit off it, carve out soft flesh with it and the logic of working and transporting heavier stones simply didn't occur too early on for a very long time. Until it eventually did, and past that, each age was considered sufficient for what it did. For untold years before the lomekwian then oldowan era. If I'm not mistaken, Lucretius does assume human tech started with wood and unworked stone first.

It seems like innovation as a notion in and of itself is just completely not obvious to us until we have enough of a history of it to treat it, label it as its own thing. The same for a number of other concepts-abstraction not as a capacity which must have pre-existed but as a practice seems to have come quite late in our species. To put it rudimentarily: you kind of have to have done it or stumbled into it before there's an "it" to know. Things like literacy just don't come to humans naturally like speech does.

Regarding cubs, with the way even very non-domesticated species like wolves, cheetahs interact with humans *today* it seems highly likely we have a long history of humans and other species being somewhat species curious, and perhaps for a lot of the time even cooperative in some ways with each other. I almost wonder if there is a bit of hidden history here-the more cooperative of their ancestors being less likely to attack humans, and thus less likely to be attacked back by our ancestors and then killed in the process. Perhaps hundreds of thousands of years of repeat mini instances here were necessary before domestication of cats and dogs was even possible.

However, our latest effort at floral and faunal domestication, apparently stemming from the levant/euphrates/anatolian region clearly was unparalleled as far as we can tell in all of history in terms of its sticking power. the right passed-on-habits encountered the right time and place with enough people with the right social technologies for it to allow for it to spread and stick around until now. At the very least they've allowed us to hit eras like the iron age which probably were not approached beforehand.

Rothari's avatar

"In addition to the weapons, they brought attention to small wooden tools used for domestic tasks like smoothing hides or punching holes."

wow just realized we have no clue how long animal fur and skin has been used, but it was quite likely early even if lice got to us a lot later. i wonder how many events we can trace by detecting human parasite evolution.

Zach Elfers's avatar

Really interesting findings, thanks for sharing.

I'm interested in the origins of domestication for both plants and animals. Plant domestication is dare-I-say more "straightforward" in that seeds from congeneric species may be carried by humans along migration and trade pathways, across geographical boundaries usually insurmountable to said species left to their own devices. This doesn't need to be an intentional process on the part of humans, but can be wholly incidental. Thus we see the origins of wheat domestication in the advent of hybrid swarms of Triticums from grains carried over one mountain pass to another, and the origin of citrus fruits and apples arises from the same phenomenon wherein fruits from far flung places arrive in the same midden piles.

As we explore the genomic history of basically "every" domesticated species, we find the same pattern: the domesticated form arises after hybridization events. Not just between congeneric species, but often three. Occasionally even across genuses.

There is of course a strong case to be made for "pre-domestication cultivation" too in these instances. Before the domesticated forms arose, wild crops ancestors were encouraged with various levels of management. Thus the hybridization events are not purely epiphenomenal, but enabled through human ecosystem engineering and habitat modification.

The case for animal domestication raises additional challenges because it pre-supposes a level of closeness with wild prey animals verging on pastoralism.

So I ask myself, how far back have humans been influencing the transhumance of wild bovines? Were young animals captured, tamed, and woven into the rhythms of daily human movements, or did human ecosystem engineering encourage wild bovines further afield from their usual habitat until they came into contact with "the next herd over"? Use of cultural fire, for instance, can influence the growth of grasses and ruderal vegetation, which in turn influences the movements of the wild herds. So even in the case of animal domestication, it didn't have to begin as a process steered by human intentionality. It could have been a slow and steady melding until eventually human populations and wild prey found themselves in lock step rhythm, and pastoralism of hybrid bovine swarms come into being.