The Deep-Time Face

Chapter 4

The Deep-Time Face

A face that kept getting smaller

Zoom out past the last few centuries of measuring tape and calipers, and the human face has a much longer story — one written in bone rather than instruments. Across roughly seven million years of hominin evolution, the trend line for the face is strikingly consistent: it shrinks, and it retreats. Early hominins like Australopithecus carried large, forward-projecting jaws and heavy chewing muscles built for grinding tough, fibrous plant material. Later Homo species show a long, gradual reduction — smaller molars, a flatter face tucked further under the braincase, a retreating lower jaw. By the time anatomically modern Homo sapiens appears, the face has become smaller and more vertically oriented relative to the skull than in any earlier hominin. The arc did not run in a straight line across every lineage — Neanderthals, for instance, retained a large midface even while other traits diverged from ours — but the broad direction, smaller jaws and teeth doing less mechanical work, holds across most of the hominin family tree.

Recovering that story from bone has been error-prone in ways worth remembering. In 1924 a fossilized skull surfaced at a South African limeworks and reached the anatomist Raymond Dart, who recognized it as a young Australopithecus — one of the first early-hominin fossils found in Africa, and evidence that upright, small-brained ancestors came before big brains.¹ Dart's "Taung Child" was largely dismissed for two decades, in part because the reigning "missing link" was Piltdown Man, an English specimen unveiled in 1912 that seemed to show the opposite arrangement: a large modern braincase paired with a primitive ape-like jaw. Piltdown was finally exposed in the 1950s as a deliberate forgery — a medieval human cranium fitted with a filed and stained orangutan jaw — and a 2016 forensic study concluded a single hoaxer had likely built it.² The genuine African fossil was doubted for decades while the European fake shaped the field, a reminder that even the fossil record is read through what its readers already expect to find.

Even the celebrated real discovery got tidied in the retelling. Dart's own 1959 memoir has the fossil-laden boxes arriving while he stood dressed as a wedding's best man, and has Dart himself prising a crate open to make the find. A 2025 historiographical review found that the geologist Robert Young had personally carried the critical breccia back from Taung and hand-delivered it to Dart — a role Dart's account steadily wrote out over the following decades.³ The same lens the Piltdown story turns outward, this one turns inward.

Cooking, softer food, and a shrinking chewing apparatus

One durable explanatory thread for that reduction is dietary. As hominins developed more effective food processing — stone-tool butchery and pounding, later the controlled use of fire for cooking — the mechanical demands on the jaw dropped. Cooked food is softer, requires less chewing force and fewer chewing cycles, and yields more calories per bite than raw, fibrous plant matter. A jaw and chewing musculature built to grind tough raw material is expensive to grow and maintain once the diet no longer requires that capacity, and structures that stop being mechanically necessary tend to shrink over evolutionary time. Applied to hominins, this predicts exactly the pattern in the fossil record: as food became easier to process outside the mouth, the jaw, teeth, and chewing muscles gradually reduced. This dietary-mechanical story is not a complete explanation on its own — it describes a pressure on the chewing apparatus generally, not every facial feature — but it is one of the load-bearing threads running through this chapter.

In 2016 researchers tested the mechanical half of that story directly: volunteers chewed raw and cooked goat meat and starchy roots while their jaw-muscle force and chewing cycles were measured. Adding meat to the diet and processing it with simple stone-age-style slicing and pounding — no fire required — cut chewing force and the number of chews per year substantially, evidence that mechanical food processing, well before cooking became routine, could have relaxed enough of the demand on the jaw to make a smaller one possible.⁴

Self-domestication: a face shaped like a tame animal's

A second, more recent thread comes from the biology of domesticated animals. In the 1950s, the Soviet geneticist Dmitri Belyayev began breeding silver foxes for a single trait — tameness toward humans — with no deliberate selection on appearance. Within a few generations, the tame foxes looked different from their wild-type relatives: floppy ears, curled tails, patches of white fur, and, notably, shorter snouts and smaller, more juvenile-shaped faces. Belyayev's foxes became the reference case for "domestication syndrome" — a cluster of physical traits that keeps showing up as a side effect of selection for tameness across domesticated species. In 2014, Adam Wilkins, Richard Wrangham, and Tecumseh Fitch proposed a unifying explanation: many domestication-syndrome traits, including facial shortening, can be traced to mild deficits in neural crest cell migration during embryonic development — the same cell population that builds much of the face and skull.⁵ Some researchers propose that modern humans underwent a comparable process — "self-domestication," selection against aggression and for social tolerance, without an external breeder — and that some of our face's juvenile, gracile characteristics relative to earlier hominins and Neanderthals are a byproduct of that same neural-crest pathway. In 2019, Matteo Zanella and colleagues reported experimental evidence connecting the gene BAZ1B, known from Williams syndrome research, to neural crest cell function and to genetic signatures in ancient and modern human genomes, offering a molecular candidate for how self-domestication could act on facial development.⁶ The hypothesis remains actively debated rather than settled, but it has become one of the more productive frameworks for connecting behavioral evolution to the shape of the modern face.

The idea has a living, non-human test case: bonobos. Compared with their closest relatives, common chimpanzees, bonobos carry a smaller, more juvenile-shaped braincase alongside markedly reduced aggression — and because no human ever bred them, primatologists have proposed them as a wild natural experiment for whether selection against aggression alone, with no breeder at all, can reshape a primate's skull.⁷

The chin puzzle

No feature of the deep-time face has resisted a settled explanation longer than the chin. A prominent, forward-projecting bony chin is unique to anatomically modern Homo sapiens; no other hominin, including Neanderthals, robust-jawed Homo erectus, or earlier australopithecines, has one. Reviewing the evidence in a widely cited 2016 survey, James Pampush and David Daegling laid out the difficulty plainly: the chin is one of the most-studied, least-agreed-upon features in human evolution, with competing explanations that have circled the same bone for decades without a consensus victor.⁸ Three broad hypothesis families currently compete, and the honest state of the field is that all three remain live.

The masticatory hypothesis holds that the chin is a structural buttress that helps the lower jaw resist bending and twisting stresses generated during chewing, once the jaw itself had become smaller and more gracile. Daegling's own biomechanical modeling has tested this idea directly, and the results have been mixed at best: several analyses found that a chin does little to measurably improve the mandible's resistance to chewing-related stress, part of what has kept the hypothesis from becoming the accepted answer despite its intuitive appeal.

The developmental hypothesis reframes the chin as a byproduct of how the growing face is assembled, rather than a feature selected for a mechanical function. Ontogenetic work by Michael Coquerelle and colleagues traced how the tongue, suprahyoid muscles, and surrounding soft tissue interact with the growing jaw and face during infancy, and found that chin prominence tracks these developmental spatial relationships rather than chewing-related stress.⁹ A related, earlier line of thinking associated with Donald Enlow's growth-field models of the craniofacial skeleton likewise treats the chin as an emergent outcome of differential growth between the jaw and the shortening face, rather than a trait shaped directly by selection.

The spandrel hypothesis — named for the architectural byproducts Stephen Jay Gould and Richard Lewontin used as their central example of a non-adaptive evolutionary outcome — proposes that the chin is not "for" anything, but a side effect of changes happening nearby. Nathan Holton and colleagues' 2015 analysis of chin growth found that allometric and biomechanical scaling patterns did not support a clean adaptive story, consistent with the chin being a passive consequence of surrounding facial and dental reduction.¹⁰ A 2026 study comparing craniomandibular form across apes and humans reinforces this reading: Noreen von Cramon-Taubadel and colleagues found the chin evolved too fast to be neutral drift, yet most of its individual traits showed no direct selection — a pattern they read as the chin arising largely as a byproduct of selection acting on nearby traits, reduced anterior tooth size and craniofacial changes associated with bipedalism, rather than as a direct target of selection.¹¹

None of the three families has won. Each has published support and published counter-evidence; each explains part of the pattern and struggles with another part of it. This is not a gap waiting to be filled by the next clever study — it is the field's honest current position, and this chapter will not manufacture a winner it does not have. The chin remains one of the clearest examples in human evolution of a feature that is easy to describe and unusually hard to explain.

Two wildcards worth knowing

Two smaller stories round out the deep-time face. The "Habsburg jaw" — the pronounced mandibular prognathism visible across generations of Habsburg royal portraiture — got an actual clinical test in 2019, when a team scored dysmorphic features from dozens of historical portraits and found the degree of prognathism rose in step with each individual's calculated inbreeding coefficient: a jaw trait amplified within a single family line by sustained close-relative marriage, distinct from the population-level, million-year story told above.¹² And the cleft or dimpled chin — the visible midline groove some people have, not the bony chin prominence discussed above — is partly heritable too, a reminder that even a small cosmetic facial detail traces to genetic variation, though the specific markers marketed by consumer DNA-testing services are not well established in the peer-reviewed literature. Both sit layered on top of the deeper chin puzzle rather than resolving it.

Further reading