Race and the odd history of human paleontology


  • Jeffrey H. Schwartz

    Corresponding author
    • Departments of Anthropology and History and Philosophy of Science, University of Pittsburgh, Pittsburgh, PA 15260
    Search for more papers by this author
    • A fellow of the World Academy of Art and Science, Dr. Schwartz is professor of physical anthropology and of history and philosophy of science at the University of Pittsburgh and a research associate at the American Museum of Natural History. He recently published the first study of virtually the entire human fossil record (The Human Fossil Record with I. Tattersall) and a revised edition of The Red Ape, which explores the assumptions underlying molecular and morphological approaches to phylogenetic reconstruction. Just published is a revised edition of his textbook on human osteology, Skeleton Keys, is forthcoming.

    • Fax: 412-648-7509


Although the late 17th century witnessed the recognition of fossils as the remains of extinct organisms—because they could be incorporated into the creation story embodied in the Great Chain of Being—acceptance of human antiquity through the indisputable demonstration of the contemporaneity of human bones, stone tools, and accepted fossils was not forthcoming for nearly 2 centuries thereafter. When it did occur, however, ancient humans were not seen as presenting a pattern of diversity similar to that seen in the fossil records of nonhuman organisms. Instead, human evolution then, as now, has typically been interpreted as being unilinear. This belief can be traced to Huxley (1863), who argued that the Feldhofer Grotto Neanderthal skullcap was merely an extension into the past of morphology seen in the Australian Aborigine, whom he took to represent the primitive end of an extreme range of variation he thought characterized Homo sapiens. During the mid-20th century, Mayr and Dobzhansky (mis)used their clout as founders of the evolutionary synthesis to cement in paleoanthropology the idea that human evolutionary history was characterized by nonspeciation. As such, anything that could be interpreted as potentially representing taxic diversity was relegated to the status of individual variation. Lack of understanding of the history of human paleontology, and the biases that constrained its perspective on human evolution, continue to affect the ways in which most paleoanthropologists pigeonhole human fossils. Anat Rec (Part B: New Anat) 289B:225–240, 2006. © 2006 Wiley-Liss, Inc.


To most observers of the discovery and interpretation of human fossils, the field of human paleontology—paleoanthropology—probably seems exotic. And why not? What other discipline can declare as its goal documenting the emergence of modern humans from a trail of increasingly human-like forms? Perhaps no other area of paleontology grabs the imagination so completely, especially because it deals with a basic question all humans share: Where did I come from?

The practice of human fossil hunting also captures our attention, and that of the media, which plays on (perhaps more precisely, preys off) the personalities and aura of romanticism associated with fieldwork. Sunburned but dedicated men and women who throw caution to the wind as they spend field season after grueling field season coping with the parching sun, lack of water and food, dangerous predators and disease, and even attack by bands of nomads. After years of crawling around on hands and knees, picking up every tiny fragment of animal bone, stone, or root, they eventually find “it”—the “it” often being the specimen or specimens that its discoverers then put forth as that year's no longer “missing link” in the human evolutionary saga. And as often as not this now critical ancestor of our own species is recognized as a new species if not also as a new genus. Even if “it” turns out to be only a handful of teeth or a fragmentary jaw, the discovery is usually touted as the basis for the most definitive declarations about the evolutionary relationships of this undoubted new species to previously known human relatives (to which, with humans, I will refer as hominids). The cachet that accompanies the successful discovery of a fossil hominid is more than worth its weight in academic gold.

On a more serious note, however, the interpretation of a new hominid fossil is typically bounded by, and filtered through, a received wisdom that is largely unspoken, and probably not even understood, by the very paleoanthropologists who make these pronouncements. Indeed, an increasing number of paleoanthropologists appear to believe that only they have the right to publish on “their” specimens, and to decide whether or not they should be allocated to new taxa. In addition, especially over the last decade, the frequency with which noted paleoanthropologists have kept their precious fossils off limits to other paleoanthropologists has increased at an alarming rate. Ignoring the abuse of the rules of the International Code of Zoological Nomenclature through the claim that newly named taxa have merely been “announced” and not properly described, such paleontologists often justify denying access to specimens by declaring that since they, the discoverers, had to endure the hardships in the field that led to the recovery of these fossils, they have the right to determine access to them (Gibbons,2002). Unfortunately, this mindset, well known for years as “discoverer's bias,” is only as good as the discoverer's background in comparative morphology and, more importantly, in systematic theory and practice.

With regard to paleoanthropology, it is an odd and also tragic state of affairs that the rigor of systematic practice typically prevailing in other areas of comparative evolutionary morphology is not now, and actually has never been, central to the phylogenetic and taxonomic interpretation of human fossils. Yet in almost reverse correlation with this history, during the past decade there has been a marked increase in the number of newly proposed hominid taxa (genera and/or species, and even the elusive notion of subspecies) and in a self-imposed pressure on the part of the discoverers to situate their finds definitively in a schema of hominid evolution (e.g., White et al.,1994; Leakey et al.,2001; Senut et al.,2001; Brunet,2002; Mallegni et al.,2003). I seriously question the claims about how these specimens (especially the recently published very early specimens) are related to other hominids—if, indeed, they are even hominid (Schwartz,2004a,2004b; Schwartz and Tattersall,2005). Nevertheless, the rash of discoveries, accompanied with the media blitz that coincides with their publication in a prestigious journal, gives the general public and even most physical anthropologists the impression that systematics is alive and well in paleoanthropology. Unfortunately, once a specimen is named, its claimed taxonomic status assumes a life of immortality, which, given the relative inaccessibility of certain specimens, only the discoverers will later be able to alter or reevaluate.

Is there, however, a necessary correlation between taxonomy, which is the art of naming a new taxon, and systematics, which involves trying to delineate groups of specimens on the basis of shared derived features (i.e., synapomorphies), and deciphering a pattern of relationships between these hypothesized groups? I think not (Schwartz,2003). Nevertheless, there is a tendency not only for the media-conscious general public, but also for the professional consumers of taxonomists' pronouncements, to presume that taxonomy and systematics are synonymous. In turn, this leads to an unwarranted conflation of these historically, theoretically, and methodologically distinct endeavors. From my perspective as a practicing systematist and taxonomist, the results of this confusion are particularly easily seen in paleoanthropology.


The general public, and even the interested but not necessarily well-informed academic community, are frequently led to believe that most of the debates in paleoanthropology revolve around whether hominids ever were a taxically diverse group. I do not mean to suggest that the question, “how much taxic diversity is there and has there been?” is not of general concern with regard to metazoans (as opposed to bacteria, for instance, in which cellular lineages typically transform without diversification). Yet, unlike the paleontological and systematic study of other organisms, the study of hominids has developed a uniquely odd tradition, which continues to be perpetuated. Namely, in contrast to the typical paleontological experience of discovering new taxa as new sites are opened or as already-known sites continue to be excavated, it is not uncommon to find paleoanthropologists (e.g., Wolpoff,1996; Curnoe and Thorne,2003; White,2003) arguing against the possibility that hominids could have been as speciose in the past as undoubtedly appears to have been the case for other groups of organisms, whether vertebrate or invertebrate.

Because of this bias against recognizing taxic diversity in the human fossil record, the motivations underlying the presentation of a new hominid genus and/or species are yet further distinguished from the normal practice of taxonomy and systematics. This is because, for many paleoanthropologists, a newly designated taxon rarely embraces a biological entity that arose via branching speciation (i.e., cladogenesis). Were this to be the case, the presentation of this new taxon would be in the form of a testable hypothesis. The typical paleoanthropological notion of a “transitional form” is, however, untestable. In addition, a theory of relatedness integrating this new hominid taxon with other taxa that also emerged via cladogenesis would also be formulated as a testable hypothesis, at a higher level within the monophyletic group or clade (i.e., a group of organisms descended from a common ancestor). However, because paleoanthropologists typically think in terms of nondiversity (i.e., nonbranching speciation), a new genus and/or species is often used as a place marker in what is seen from the beginning as an ongoing process of unilinear (i.e., anagenetic) transformation, which, by its very nature of continual transformation, precludes the recognition of distinct taxa.

Such a perspective—in which unilinearity and taxic nondiversity are heavily emphasized—is reminiscent of a mindset that was popular during the 1970s among a small number of vertebrate paleontologists. As Darwin (1859) had done before them, these paleontologists argued that the only reason one thinks one can delineate species as distinct entities is because of taphonomic effects on, and other inconvenient distortions of, what had once been a seamless record of organismal transformation (e.g., Gingerich,1976). A complete fossil record would reveal a picture of transforming lineages with fewer instances of branching.

Nevertheless, the emphasis on searching for ever-elusive transitional forms was steadily replaced by the theoretically and methodologically more rigorous procedure of phylogenetic reconstruction afforded by cladism. As any recent issue of the journal Systematic Biology and the Journal of Vertebrate Paleontology will demonstrate, the hypothetico-deductive strengths of cladistic analysis have led to its becoming the methodology of choice not only among paleontologists, but among systematists in general. This is in large part because cladism's emphasis on extensive cross-taxic comparison and hypothesis testing precludes requiring a priori that either diversity or nondiversity characterized the evolutionary history of any group of organisms (Schwartz,2003). Curiously, paleoanthropology, while a discipline that must also confront evolutionary history, and that therefore should share the same theoretical and methodological concerns as other systematic enterprises, does not (Tattersall,1986; Tattersall and Schwartz,2000).

Typically in paleoanthropology, taxa are defined not on the basis of uniquenesses or apomorphies that are delineated through a process of broad morphological comparison, but in the context of an a priori hypothesis of how one's specimen(s) is related to previously known hominid fossils. Increasingly, the determination of relatedness is through the temporal sequencing of fossils (e.g., White,2003; White et al.,2006). Thus, a newly described hominid “taxon” is likely to represent a phase in a transforming lineage [e.g., the middle-late Pleistocene specimens from Herto, Ethiopia, that were declared to represent the immediate ancestor of living humans and placed in a new subspecies, H. sapiens indaltu (White et al.,2003)], or the antecedent of more than one descendent [e.g., specimens from the c. 780 kya TD-6 level of Gran Dolina, Atapuerca, that were claimed to represent the ancestor of both H. neanderthalensis and H. sapiens and consequently placed in the new species H. antecessor (Bermudez de Castro et al.,1997)].

White (2003: p. 1997) has criticized other paleoanthropologists for practicing “diversity systematics” which, he believes, derives from a zeal for recognizing taxic diversity in the human fossil record that overrides the assessment of diversity through “the canons of modern biology.” As a result of suggesting that many taxa, such as Australopithecus anamensis (Leakey et al.,1995) and Homo antecessor, were named prematurely and/or on the basis of inadequate specimens (and may even represent previously diagnosed taxa), White was obliged to question the validity of two taxa he had created, i.e., Ardipithecus ramidus (White et al.,1994,1995) and the geologically younger Australopithecus garhi (Asfaw et al.,1999). He also argued that another cause of overestimating taxic diversity was “expanding matrix distortion” (EMD). Because of EMD, specimens become so reshaped that they “fool” paleoanthropologists into thinking they represent new taxa. His primary example from the human fossil record was the skull of Kenyanthropus platyops (Leakey et al.,2001), which is, indeed, composed of a myriad fragments held together by matrix.

Although perhaps too dismissive of this specimen's preservation of relevant morphology (many facial features are preserved bilaterally), White ultimately rests his case for rejecting claims of hominid taxic diversity on an article by John Noble Wilford in the 25 March 2001 issue of the New York Times. There, Wilford argued that because modern anthropologists “have been steeped in multiculturalism and diversity, and see [newly discovered hominid fossils] in a favorable light,” an appreciation of modern ethnic diversity may have predisposed paleoanthropologists to looking for and subsequently “discovering” diversity in the human fossil record. In other words, as White apparently sees the situation, the embracing of ethnic diversity (“race”) by anthropologists in general (and by cultural anthropologists specifically) led to a strange version of politically correct paleoanthropology, in which fossils that were really only variants of the same species could be recognized as distinct species. In reaction to this purported motive, White strongly urges paleoanthropologists to return to the taxonomic conformity (in reality, taxonomic minimalism) that the founders of the evolutionary synthesis had imposed decades earlier on the human fossil record.

It is not my intention here to discuss systematic theory or how one might deal practically with the question of taxic diversity, whether among hominids or any other group of organisms. I have addressed many of these issues elsewhere (e.g., Schwartz,1999a,2003). Of relevance here is the cultural element of race by which White, via Wilford, accuses paleoanthropologists of being unduly influenced in their attempts to decipher pattern in the human fossil record. Also significant is White's promotion of Mayr (1950)— one of the “fathers” of the evolutionary synthesis—as an example of taxonomic level-headedness who tried to bring order to paleoanthropology, which Mayr characterized as a field of a “bewildering diversity of names.”

Superficially, it might seem reasonable that White should invoke the authority of Mayr in resolving taxonomic issues in paleoanthropology. For, after all, Mayr trained as an ornithologist and systematist and his major monographic contribution to the evolutionary synthesis, Systematics and the Origin of Species (Mayr,1942), did present his view of that subject matter. But in view of his acceptance of taxic diversity among birds and other organisms, it is odd that Mayr argued so adamantly for whittling down the “bewildering diversity” of potential hominid taxa to only three species within a single genus, Homo (Mayr,1950). Was the impetus purely scientific, merely the neutral act of one whose primary interest lay solely in the science of evolutionary biology? Or did Mayr, as others before and after him, fall victim to the mindset that comes from belonging to a species, Homo sapiens, that is so unique (especially in being the only species of its kind alive today) that biological objectivity can easily be submerged by nonbiological considerations?

The only way in which we can begin to understand why the study of human fossils and the interpretation of human evolution is and has been so different from systematics and paleontology in general is by understanding the disparate histories of these disciplines.


The fossilized remains of once-living organisms or the stone tools manufactured by earlier humans were not always recognized for what they are. In 1565, Konrad Gesner, creator of the genus (Gesner,1551–1558), classified objects that came from the ground. He referred stone tools, which were often called “thunderbolts” or “thunderstones” because of their presumed extraterrestrial origin, to “ceraunia,” a term whose Greek and Latin etymological underpinnings refer to thunder. Some eight decades later, Ulisse Aldrovandi (1648) rejected Gesner's ceraunia and argued that like minerals, natural stones and rocks, potter's clay, and fossils, stone tools were formed in the earth.

Aldrovandi's and other 17th-century Great Chain of Being scholars' belief that all earth-borne objects had a similar origin is at odds with Michel Mercati's widely circulated manuscript of 1574 (published only in 1717), in which he argued, by analogy with stone implements that explorers saw “primitive” non-Westerners using, that these unnatural-looking stone objects were actually human-made stone tools (Grayson,1983).

The demise of the practice of lumping together all objects that came from the earth began in 1669, with the publication of Steno's studies on the reality of fossils (Winter,1968). Combining chemistry with the nascent principles of geology and micropaleontology, Steno demonstrated not only that the crystalline structure of earth-derived minerals was totally unlike that of the shells of fossilized invertebrates, but also that the crystalline structures of the shells of fossilized and extant invertebrates were similar. Thus, rock-hard objects with shapes reminiscent of biological entities—such as clamshells, shark's teeth, and mammal bones—were truly the remains of once-living organisms. In turn, this understanding permitted considering life in a broader and deeper historical context—at least nonhuman life, and within the time constraints of the creation myth.

Once the notion that all earth-derived objects had a common origin began to crumble, the reality of stone-tool production by earlier generations of humans could also be embraced, although not without certain consequences. For Mercati's argument, based as it was on analogy with the stone implements of “primitive” societies, led to the “scholarly” construction of a technological hierarchy that proceeded from the lowliest of stone implements to the metallurgic sophistication of western Europe (Schwartz,1999b). In turn, this bias made it possible for natural historians such as Woodward (1728), to argue that, whenever one found stone tools, one was dealing with savage and barbaric humans, whether they were of the uncivilized past and had no choice in available tool-making material, or inhabitants of the present who, for whatever reason, had failed to advance technologically to metallurgy as “true” civilized human societies had done.

Steno's work had a truly significant impact on biological science. The reality of fossils meant that the search for “missing links” to fill in the gaps of the Great Chain of Being could be expanded into the past (Blumenbach,1969; Schwartz,1999b). But this did not apply to humans. Humans, being the last to be created in accordance with the creation myth and the Book of Genesis, could not be (appreciably) antediluvian (Grayson,1983). Nevertheless, the availability of hordes of previously unacknowledged extinct life forms opened up new avenues of study. In particular, the nascent fields of vertebrate and invertebrate paleontology allowed the Great Chain of Being-motivated taxonomist to revel in the machinations of a divine creator. As the diversity of extant life had been recognized by earlier taxonomists, so, too, could diversity of past life now be documented (Schwartz,1999b). And because of this, the acts of describing, cataloguing, and organizing fossils in a systematic manner—albeit not from an evolutionary perspective but according to a classificatory representation of the Great Chain of Being (Gregory,1910; Lovejoy,1942)—became the backbone of biology and systematics, reaching an apogee by the early 19th century, especially through the field, comparative, and interpretive work of Cuvier and other European paleontologists (Gregory,1910).

The legacy of such a long history of allocating fossils to new or previously named taxa and of systematically organizing them is still with us today, e.g., in museums of natural history and university departments that foster paleontological research. As anyone visiting these collections knows, fossils, like specimens of extant taxa, are extensively curated, with jaws, teeth, and other skeletal remains allocated to genus and species and arranged in drawers, and drawers arranged within cabinets, according to a larger classificatory scheme. This is not to say that one might not find that the occasional specimen of living or extinct animal was incorrectly identified to taxon or was not recognized as a different genus and/or species (Schwartz,1995,1996). But, when one does use collections to pursue a systematic study of a group of animals, it is usually the case that at least most of the initial identification of genus and species (i.e., alpha taxonomy) has already been quite efficiently done in accordance with a procedure that began centuries ago.


Such regularized taxonomic and systematic practice has not, however, carried over completely to human fossils (Schwartz and Tattersall,2002b,2003,2005). During my nearly decade-long study with I. Tattersall of virtually the entire human fossil record, and in contrast to my experience with collections of nonhuman fossil primates and other mammals, I have found that the level of attention to morphological detail in the curation of human fossils has often been minimal. In some cases, as with middle to late Pleistocene specimens, fossils are often identified by site or provenance (e.g., the Steinheim, Petralona, or Kabwe skulls), by an informal name corresponding to a perceived time period of that hominid's existence (e.g., the Krapina, La-Chapelle-aux-Saints, or Spy Neanderthals), or by a formal species name that is associated with a time period in which that taxon is supposed to have existed (e.g., Homo erectus, to encompass all hominid fossils from c. 1.8- to 0.5-million-year-old deposits of the Old World). The situation with earlier, Plio-Pleistocene, presumed hominid fossils is more tenuous and is further complicated by the fact that there has been no useful definition of what “hominid” means (Schwartz,2001,2007). Thus, it is not uncommon to find boxes of isolated teeth whose catalogue numbers identify them simply as “hominid” or refer them all to a specific hominid taxon, such as “Australopithecus africanus.” Often, however, even cursory inspection reveals that many or even none of these specimens can unequivocally be identified as hominid (Grine and Franzen,1994; Schwartz et al.,1995; Schwartz and Tattersall,2005). How could the approach to human fossils be so dramatically superficial compared to the way in which other extinct taxa are so routinely scrutinized systematically? Interestingly, one can actually point to two specific publications that had this effect on paleoanthropology.

In 1965, Simons and Pilbeam published a major taxonomic revision of the then-known Miocene hominoid (= the group consisting of hominids and apes) fossils in which they not only sank or synonymized a considerable number of species and genera, but also further constrained hominoid diversity by taking the three genera that survived their taxonomic lumping—the Indo-Pakistani Sivapithecus, the East African Proconsul, and the European Dryopithecus—and demoting each to a subgenus of Dryopithecus. As they (also Pilbeam,1970) saw it, Dryopithecus (Sivapithecus) was the ancestor of the living Asian orangutan as well as of the large and extinct Gigantopithecus from the Plio-Pleistocene of Indo-Pakistan and southern China; large and small species of Dryopithecus (Proconsul) gave rise, respectively, to the gorilla and chimpanzee; and Dryopithecus (Dryopithecus) went extinct without issue. But while they collapsed most Miocene taxa into one genus, Simons and Pilbeam also recognized Ramapithecus, which at the time and for at least 15 years thereafter was interpreted as the ancestor of “proper” hominids because of similarities in premolar and molar morphology, thickness of molar enamel, and also, for a while, palatal shape (Simons,1964,1965).

As drastic as Simons and Pilbeam's revision of Miocene hominoids was in reducing the number of accepted taxa, it had been preceded by any even more drastic revision. For Ernst Mayr had already truncated the picture of human evolution, first to three and then to four time-successive species of hominid.

It is clear from Mayr's (1950) initial appraisal of paleoanthropology that he was displeased with what he considered the paleoanthropologists' lack of standards when naming newly discovered human fossils. In large part, his disapproval is still warranted. From the beginning, paleoanthropology was essentially devoid of the procedural rigor that characterized nonhominid paleontology, particularly with regard to adhering to the rules of the International Code of Zoological Nomenclature (Tattersall and Schwartz,2003).

But, aside from the fact that its practitioners were schooled primarily in biomedicine rather than comparative morphology and systematics, there are other reasons why paleoanthropology differs so markedly in practice from its sister disciplines. Much of this difference derives first from the preevolutionary theory history of human paleontology, and then from Huxley's (1863) assessment of the then-known human fossils. But we must also return to Mayr. For, despite his complaints about paleoanthropologists' shoddy practice of systematics, Mayr himself could not have done more to deter scientifically healthy, alternative thinking in paleoanthropology.


In 1863, Thomas Henry Huxley published Man's Place in Nature, which contained three essays: “On the Natural History of the Man-like Apes” (a brief review of historical references to, and conceptions about, the apes); “On the Relations of Man to the Lower Animals” (the first developmental and comparative morphological argument for why humans should be classified not only as primates, but in a subgroup with the apes); and “On Some Fossil Remains of Man” (a discussion of the Feldhofer Grotto Neanderthal skullcap and other real or presumed fossil crania).

With regard to the second essay, students of human evolution continue to claim incorrectly that Huxley demonstrated a close evolutionary relationship between humans and the African apes (e.g., Satta et al.,2000). Yet, as Huxley (1896) stated in the preface to the second edition of this volume, and as is clear in the original essay, he did not intend to provide an evolutionary argument for the relatedness of humans and any ape (Schwartz,1987). Rather, as seen in the quote below, Huxley's conclusion was taxonomic in nature, and consistent with his stated goal of validating Linnaeus' classifying humans with other animals:

“The structural differences between Man and the Man-like apes certainly justify our regarding him as constituting a family apart from them; though, inasmuch as he differs less from them than they do from other families of the same order, there can be no justification for placing him in a distinct order” (Huxley,1863: p. 125).

J. Marks (personal communication) has suggested that we should read this passage in light of Darwin's (1859) view of the Linnaean taxonomic hierarchy as reflecting evolutionary processes that yield lines of descent. If we were to do so, however, we would not conclude that humans are closely related to the African apes. Rather, we would interpret Huxley's passage as indicating that humans are related to a group consisting of all large-bodied (“Man-like”) apes. Nevertheless, it became commonplace in the 20th century for paleoanthropologists to cite this essay as the first demonstration of a close evolutionary relationship between human and African apes (Schwartz,2005), and this misconception continues to the present.

In the essay “On Some Fossil Remains of Man,” Huxley focused on two partial adult crania and a somewhat disarticulated juvenile skull, all of which he embraced as being of some antiquity because of Lyell's assessment of the geology and fauna of the sites from which they came. Schmerling had excavated the juvenile and one adult specimen in the 1830s at Engis, near Liège, Belgium (Figs. 1 and 2). The other adult cranium had been unearthed during limestone quarrying at Feldhofer Grotto, near Düsseldorf, Germany, in the Neander Valley (Neander Thal). It is significant that Huxley considered these partial crania to be ancient because the still largely biblically oriented scholarly community believed that humans could not be “antediluvian.” Indeed, only 5 years earlier, when the Feldhofer Grotto Neanderthal was presented publicly, the eminent German human anatomist Hermann Schaaffhausen argued vociferously that this specimen was neither fossilized nor contemporaneous with the bones of extinct animals also uncovered at the site. Proof of the association of human remains with the bones of acknowledged extinct species did not emerge until 1886, when two partial Neanderthal skeletons were discovered at Spy, Belgium (Fraipont and Lohest,1886). Thus, it was only in the late 19th century that the broader scientific community “allowed” human history to extend into the distant past (Grayson,1983; Schwartz,1999b).

Figure 1.

Views of the child partial cranium from Engis (not to scale).

Figure 2.

Views of the adult partial cranium from Engis (not to scale).

Huxley did not dwell on the juvenile from Engis. He did, however, provide an extensive translation of Schmerling's description of the adult, to which he added his own remarks and drawings based on a plaster cast of the specimen. In general, as Huxley translated it, Schmerling was convinced that these as well as human-like bones from Engihoul, a site across the river Meuse from Engis, were fossilized and contemporaneous with the numerous bones of extinct animals also excavated. In particular, Huxley (1863: p. 145) focused on Schmerling's finding that “a broken fragment of an ulna was soldered to a like fragment of a radius by stalagmite, a condition frequently observed among the bones of the [extinct] Cave Bear (Ursus spelaeus), found in the Belgian caverns,” and also that “Professor Schmerling found, incrusted with stalagmite and joined to a stone, the pointed bone implement … and worked flints.”

Although, according to Huxley, Schmerling remarked that even if the face of the Engis adult had been preserved it “would not have been possible to pronounce, with certainty, upon a single specimen” (Huxley,1863: p. 149), he proceeded to state that “the elongated and narrow form of the forehead,” together with “the slight elevation of the frontal, its narrowness, and the form of the orbit, approximate it more nearly to the cranium of an Ethiopian than to that of an European” (Huxley,1863: p. 150). Given the apparent association of the bones of a presumed primitive human (which “approximate” an Ethiopian's) not only with the bones of acknowledged extinct animals, but also with bone and stone tools (hallmarks of “savages”), it is not surprising that Schmerling then declared:

“At whatever conclusion we may arrive as to the origin of the man from whence this fossil skull proceeded, we may express an opinion without exposing ourselves to a fruitless controversy. Each may adopt the hypothesis which seems to him most probable: for my own part, I hold it to be demonstrated that this cranium has belonged to a person of limited intellectual faculties, and we conclude thence that it belonged to a man of a low degree of civilization: a deduction which is borne out by contrasting the capacity of the frontal with that of the occipital region” (Huxley,1863: p. 150).

With regard to the overall morphology of the skull, I agree with Huxley's description: it is relatively long, but not very narrow (especially across the front), with a high-domed frontal, well-arced sagittal profile, and little muscle scarring (Schwartz and Tattersall,2002b). Of particular importance for discussion here is that the only region of the skull for which Huxley (1863: p. 148–149) provides much morphological detail is the supraorbital region:

“The supraciliary prominences or brow-ridges … are well, but not excessively, developed, and are separated by a median depression. Their principal elevation is disposed so obliquely that I judge them to be due to large frontal sinuses.”

The illustration of the supraorbital region correctly portrays the “principal” elevation as moderately swollen bilaterally, with right and left portions oriented somewhat superolaterally (Schwartz and Tattersall1996,2000,2002b,2003). Although Huxley did not describe other features of the supraorbital region, they are faithfully represented in the drawing. Specifically, the bone lateral to each swelling is flatter and smoothly continuous with the surrounding frontal bone. This particular configuration, which is apparently unique to Homo sapiens, is identified in the paleoanthropological literature as a “bipartite brow” (Stringer et al.,1984; Schwartz and Tattersall,2002a). Given Huxley's prowess as a comparative anatomist, I doubt that these few morphological details escaped his notice inasmuch as the supraorbital region assumes some importance in his subsequent discussion.


After dealing with the partial adult cranium from Engis, Huxley turned to the Feldhofer Grotto bones (Fig. 3). He quoted at length on the morphology of the specimens from a letter written by Carl Fuhlrott, the school teacher to whom miners had given the first-discovered specimen, the skullcap, and who later salvaged various postcranial bones from the site. The excerpt immediately below, from Fuhlrott, and others that follow are relevant for comparison with the adult Engis cranium:

Figure 3.

Views of the Feldhofer Grotto Neanderthal calotte (not to scale).

“The cranium is of unusual size, and of a long-elliptical form. A most remarkable peculiarity is at once obvious in the extraordinary development of the frontal sinuses, owing to which the superciliary ridges, which coalesce completely in the middle, are rendered so prominent, that the frontal bone exhibits a considerable hollow or depression above or rather behind them, whilst a deep depression is also formed in the situation of the root of the nose. The forehead is narrow and low, though the middle and hinder portions of the cranial arch are well developed…. [T]he superior occipital ridges … are greatly developed, and almost conjoined so as to form a horizontal eminence” (Huxley,1863: p. 151–152).

Huxley (1863: p. 152 et seq) then cites large sections from the anatomist George Busk's translation of Schaaffhausen's report, which was primarily a refutation of Fuhlrott's claim that the Feldhofer Grotto human bones were fossilized and contemporaneous with those of extinct nonhuman animals. After comparing the Feldhofer Grotto skullcap with both recent and what were then considered ancient human crania, Schaaffhausen concluded:

“But the human bones and cranium from the [Feldhofer Grotto] Neanderthal exceed all the rest in those peculiarities of conformation which lead to the conclusion of their belonging to a barbarous and savage race. Whether the cavern in which they were found, unaccompanied with any trace of human art, were [sic] the place of their interment, or whether, like the bones of extinct animals elsewhere, they had been washed into it, they may still be regarded as the most ancient memorial of the early inhabitants of Europe.”

Huxley (1863: p. 161–162) follows the above quote with the comment: “Mr. Busk … has enabled us to form a very vivid conception of the degraded character of the Neanderthal skull, by placing side by side with its outline, that of the skull of a Chimpanzee, drawn to the same absolute size.” And, indeed, his discussion of the Neanderthal skullcap is essentially a justification of this impression of it.

Huxley tells his audience that he came to study a plaster cast of the Neanderthal calotte for a second time, but then in detail, in order to provide Lyell with a diagram that would illustrate its oddities in comparison with other human skulls, which he represented by the adult cranium from Engis. For standardization, Huxley chose two cranial landmarks through which he drew a line: i.e., glabella, which is the most anterior point on the midline of the skull above the nasal region, and inion, which is the most posterior point in the midline of the skull. He chose inion, which he situated at the juncture of the occipital protuberance and the superior semicircular curved line, for two reasons: it corresponds with the intracranial transverse sinuses and the line of attachment of the tentorium, which tethers the brain to the inside of the braincase and also approximates the lower limits of the occipital lobes of the cerebrum (the “thinking” part of the brain).

Huxley aligned outlines of the Neanderthal and Engis partial crania along the glabella-inion axis. To his surprise, he found that “the difference was so vast and the flattening of the Neanderthal skull so prodigious … that I at first imagined I must have fallen into some error” (Huxley,1863: p. 162). He asked: “Was it possible for a human being to have the brain thus flattened and depressed; or, on the other hand, had the muscular ridges [left by the nuchal muscles] shifted their positions?” (Huxley,1863: p. 162). He then cited his correspondence with Fuhlrott, who had corroborated the configuration of the occipital region of the Neanderthal skullcap and had also sent photographs that confirmed Huxley's (1863: p. 164) earlier finding (“the posterior lobe of the brain of the Neanderthal man must have been as much flattened as I suspected it to be”). Huxley (1863: p. 165) concluded this section of the essay on the Neanderthal skullcap with this comment:

“[The Neanderthal skull cap] is particularly interesting, as it suggests that, notwithstanding the flattened condition of the occiput, the posterior cerebral lobes must have projected considerably beyond the cerebellum, and as it constitutes one among several points of similarity between the Neanderthal cranium and certain Australian skulls.”

The choice of “certain Australian skulls” for comment is not without historical significance. For the Great Chain of Being-based hierarchical arrangement of life, which also ranked humans from those closest to the “brutes” up to the most civilized, was not disavowed by 19th-century evolutionists. Among Great Chain of Being taxonomists and certain 19th-century evolutionists, contenders for the lowest of the humans included Hottentots, “bushmen,” and Australian Aborigines, the latter of which was Huxley's choice as the most primitive living human (Schwartz,1987).

But Huxley did not immediately offer an interpretation of the points of similarity between the Neanderthal skullcap and Aboriginal skulls. Rather, he spent considerable time first emphasizing, as Blumenbach and others had done before him, the enormous range of variability that a global comparison of human crania supposedly revealed. He then dismissed the ways in which other scholars had tried to equate human variation with racial distinctiveness, which, he believed, failed to represent the fundamental aspects that determine cranial shape because they were not “founded upon the establishment of a relatively fixed base line, to which the measurements, in all cases, must be referred” (Huxley,1863: p. 171). By arguing that skull growth in mammals followed a general pattern, he concluded “that the base of the skull may be demonstrated developmentally to be its relatively fixed part, the roof and sides being relatively moveable … [which] … is exemplified by the study of the modifications which the skull undergoes in ascending from the lower animals up to man” (Huxley,1863: p. 171). Essentially, Huxley found that from “lower” to “higher” crania, the basicranial axis became shorter relative to skull length, the angle of the face to the neurocranium became more acute, as the face became increasingly flexed downward relative to the braincase, and the braincase became increasingly more vaulted or arched, with concomitant heightening and backward extension of the cerebrum.

It is one thing to generalize from a series of mammalian skulls that one lines up in a hypothetical series from the lowest (e.g., Huxley's choice, the beaver) to the highest (“man”). It is another, however, to continue this hierarchical projection within the species Homo sapiens. Huxley (1863: p. 175) raises the question and answers it:

“Now comes the important question, can we discern, between the lowest and the highest forms of the human cranium, anything answering, in however slight a degree, to this revolution of the side and roof bones of the skull upon the basicranial axis observed upon so great a scale in the mammalian series? Numerous observations lead me to believe that we must answer this question in the affirmative.”

Huxley's “demonstration” is based in part on comparing in one image the sagittal outlines of the skull of an Australian Aborigine and that of a Tartar and, in another, the sagittal outlines of the skull of a “Negro” with one from a cemetery in Constantinople. The Aborigine and the “Negro” crania are markedly prognathic, with their anteriorly projecting lower faces and upper jaws, whereas the Tartar and especially the other “modern-looking” skull are short-faced or orthognathic. Huxley likens these differences to those between “lower animals” and “Man,” but admits that the prognathic individuals are less ape-like because their frontal cerebral lobes are situated a bit more anteriorly than in the orthognathic individuals. He also states that a round skull (the skull from Constantinople) may exhibit greater posterior projection than a long one (the “Negro”). The accompanying drawing, however, represents the “Negro” skull as the round one and the other skull, which actually does have a prominent posterior extension, as the long one. Perhaps the artist confused the outlines of these two skulls and the error escaped Huxley's notice.

Nevertheless, Huxley continues in this vein. He argues that because gradations of human skull shape appear to parallel the global gradients of longitude and latitude, this obviates colleagues' claims that any particular skull belonged to a “low” versus a “high” type. In spite of this assertion, he then dwells at length on the skull of the Australian Aborigine (Fig. 4):

Figure 4.

Views of the skull of an Australian Aborigine (not to scale; American Museum of Natural History VL/1579). In spite of the robust supraorbital region, this specimen nonetheless possesses the bipartite brow characteristic of Homo sapiens.

“The Australian skull is remarkable for its narrowness and for the thickness of its walls, especially in the region of the supraciliary ridge, which is frequently, though not by any means invariably, solid throughout, the frontal sinuses remaining undeveloped. The nasal depression [a feature he earlier noted in the Feldhofer Grotto Neanderthal skullcap], again, is extremely sudden, so that the brows overhang and give the countenance a particularly lower, threatening expression. The occipital region of the skull, also, not unfrequently becomes less prominent; so that it not only fails to project beyond a line drawn perpendicular to the hinder extremity of the glabello-occipital line, but even, in some cases, begins to shelve away from it, forwards, almost immediately. In consequence of this circumstance the parts of the occipital bone which lie above and below the tuberosity make a much more acute angle with one another than is usual, whereby the hinder part of the base of the skull appears obliquely truncated. Many Australian skulls have a considerable height, quite equal to that of the average of any other race, but there are others in which the cranial roof becomes remarkably depressed” (Huxley,1863: p. 179).

Also on this page is a profile line drawing of an Aborigine's skull and mandible with the outline of the anteroposteriorly much longer Feldhofer Grotto skullcap superimposed on it (Fig. 5). The cranial roof of the latter is much lower than in the extant human. The brow (described and illustrated earlier by Huxley as continuous along the orbits and confluent across glabella, and not bipartite) is more massive, superoinferiorly thickened, and somewhat anteriorly protrusive. And the occipital region (also previously described and illustrated by Huxley as quite different from other humans, including the Engis adult cranium) is distinctive, even in profile. Yet in spite of the number and magnitude of differences that Huxley as a critical comparative nonhuman vertebrate anatomist would have embraced as worthy of distinction at the species if not genus level, he proclaims: “A small additional amount of flattening and lengthening, with a corresponding increase of the supraciliary ridges, would convert the Australian brain case into a form identical with that of the aberrant fossil” (Huxley,1863: p. 179–180). A few pages later, after commenting on how the Neanderthal cranium is rather pithecoid or ape-like in some ways and yet, in others, compares well with “primitive” variants of human crania, Huxley (1863: p. 183) turns his hypothetical continuum of morphological transformation the other way around and states:

Figure 5.

Profile of Feldhofer Grotto skull cap superimposed on drawing skull of Australian Aborigine (modified from Huxley,1983).

“[I]ndeed, though truly the most pithecoid of known human skulls, the Neanderthal cranium is by no means so isolated as it appears to be at first, but forms, in reality, the extreme term of a series leading gradually from it to the highest and best developed of human crania.”

Clearly this does not represent the kind of morphological transformation Huxley had earlier argued was the norm when he used the basicranial line as the standard on which to compare mammalian skulls. In addition, there is nothing in Huxley's hypothetical sequence of altered cranial shape that would lead to different details of cranial morphology, such as bipartite supraciliary arches (all recent and near-recent human crania, including the adult Engis skull) versus continuous and smoothly rounded supraorbital tori that merge at glabella (Neanderthal skullcap). Or to a sulcus posterior to these supraorbital tori (Neanderthal) rather than a frontal that rises laterally directly from the supraorbital margin (bipartite-browed human crania, including the adult Engis skull). Or to a short but very wide occiput with a relatively straight occipital “torus” and a pitted suprainiac depression above (Neanderthal) versus a tall and triangular occiput with an inferiorly oriented external occipital protuberance (human crania, including the adult Engis skull) (Schwartz and Tattersall,1996,2000,2002b). These are the kinds of details, rather than the general nature of craniofacial hafting, that Huxley, following 2 centuries during which taxonomists and systematists had worked toward perfecting their discipline, should have embraced as important in delineating taxa. In his classifications of mammals, he did. In the case of humans, he did not.

Why did Huxley, one of England's premier comparative anatomists, abandon generally accepted and for the time fairly rigorous systematic procedure when dealing with the Neanderthal specimen? His emphasis on the gradual transformation of what he saw “as the most pithecoid of human crania yet discovered” (Huxley,1863: p. 181) into a more recognizably Homo sapiens-like skull (albeit that of a “primitive” Australian Aborigine) becomes even more perplexing in light of the fact that by inclination he was a saltationist (Huxley,1860,1863,1876). And for saltationists, differences between organisms result from changes in development that lead to the abrupt appearance of morphological novelty and, thus, to the abrupt appearance of species (Mivart,1871). As such, even though Huxley added that the great brain size of the Neanderthal would have reflected an organization not representative of the ape-like features imprinted in the skullcap, one would have expected him to see in the distinctive, non-Homo sapiens-like features of its brow and occiput the hallmarks of a species in its own right. But, to reiterate, he did not.

Insight into Huxley's abandonment of saltationism in favor of gradual morphological transformation—in which the Feldhofer Grotto Neanderthal is appended as the most primitive variant of a continuum of change leading up to the most advanced cranial form—comes from his wondering whether these bones could “be regarded as the remains of a human being intermediate between Men and Apes” (Huxley,1863: p. 181–182). Had these fossilized bones been the remains of any other mammal, doubtless Huxley would not have hesitated to interpret them as representing a species “intermediate” between more “primitive” and more “advanced” species. But in spite of admitting that forms intermediate between humans and the ultimate “pithecoid” ancestor had existed, he is unwilling to fill this evolutionary void with the Feldhofer Grotto Neanderthal. He concludes his essay:

“I may say, that the fossil remains of Man hitherto discovered do not seem to me to take us appreciably nearer to that lower pithecoid form, by the modification of which he has, probably become what he is. And considering what is now known of the most ancient Races of men; seeing that they fashioned flint axes and flint knives and bone-skewers, of much the same pattern as those fabricated by the lowest savages at the present day, and that we have every reason to believe the habits and modes of living of such people to have remained the same from the time of the Mammoth and the tichorhine Rhinoceros till now, I do not know that this result is other than might be expected.

“Where, then, must we look for primaeval Man? Was the oldest Homo sapiens pliocene or miocene, or yet more ancient? In still older strata do the fossilized bones of an Ape more anthropoid, or a Man more pithecoid, than any yet known await the researches of some unborn paleontologist?

“Time will show. But, in the meanwhile, if any form of the doctrine of progressive development is correct, we must extend by long epochs the most liberal estimate that has yet been made of the antiquity of Man” (Huxley,1863: p. 183–184).

In spite of this challenging and even for the time heretical speculation about humans having extinct relatives of considerable antiquity, Huxley subsumes the Feldhofer Grotto Neanderthal within the species Homo sapiens as an extension into the past of a continuum of a racial hierarchy—portrayed in morphology and in intelligence as inferred from associated levels of technological sophistication—that was widely perceived to exist among modern “races” of humans (Darwin,1871). In downplaying, if not just ignoring, the details of morphological difference between the Neanderthal specimen and Homo sapiens, Huxley conflated within-species differences due to individual variation (i.e., differences in the degree of expression of a particular feature, as in robustly versus weakly developed brows) with between-species differences that reflect taxic diversity (i.e., differences of kind or configuration of a feature, as in bipartite versus continuous brows).

With the clarity that hindsight provides, it is obvious that Huxley's emphasis on degrees of difference in overall size and shape of features (e.g., pronounced versus unpronounced brow regions, distended versus flatter occipital regions, or more rounded versus more elongated skulls) laid the groundwork for what became commonplace in paleoanthropological thinking: that common themes in individual variation (e.g., robustness versus gracility) are as systematically relevant, if not more so, than the morphological details of the regions being scrutinized (e.g., bipartite versus continuous brow, whether either is robust or gracile) (e.g., Hrdlička,1927; Broom,1951; Smith and Ranyard,1980; Bräuer,1981; Thorne and Wolpoff,1982; Wolpoff,1989; Rightmire,1990; Rosas and Bermudez de Castro,1998; Asfaw et al.,2002). For, as more specimens of Neanderthals and other hominids became known during the late 19th century and on into the 20th, Huxley's suggested morphological continuum, although based on a racial hierarchy that was seemingly expunged from the domain of paleoanthropology, influenced how these and subsequently discovered fossils were incorporated into what was seen as an almost limitless realm of human variability (e.g., Hrdlička,1927; McCown and Keith,1939; Dobzhansky,1944; Weidenreich,1947; Mayr,1950; Howell,1952,1960; Thorne and Wolpoff,1982; Wolpoff,1989). In turn, this perception of unbounded human variability was transformed into a self-fulfilling prophecy: the more different objects one lumps together and defines as variants of the same thing, the greater becomes a presumed realm of variability, which then dilutes the systematic significance of difference and makes increasingly easy the justification of each subsequent addition to this construct of variability.


Approximately the same year as the publication of Man's Place in Nature, a professor of geology at Queen's College, Galway, Ireland, William King, read a now-lost paper at a meeting of the British Association in which he concluded that the differences between the Feldhofer Grotto Neanderthal and humans were sufficient to warrant recognition of the former in its own species, for which he coined the binomen Homo neanderthalensis. A year or so later, in an article published in 1864, King reviewed in detail the morphology of the Neanderthal remains and resoundingly rejected what he saw as Huxley's superficial analysis of them. As King (1864: p. 92) summarized:

“Huxley, while admitting that it is the most ape-like and most brutal of all human skulls yet discovered, states that it is 'closely approached' by some Australian forms, and 'even more closely affined to the skulls of certain ancient people, who inhabited Denmark during the Stone period.' I have no intention to deny that there are general features of resemblance between the Australian, Neanderthal, and ancient Danish crania; but it appears to me, judging from the figures … in the deeply philosophical work, 'Man's Place in Nature,' that a closer resemblance is assumed than really exists. No one would have any hesitation in admitting that the Borreby [Australian Aborigine] skull, represented under one of the figures, is strictly human,—nay, from what I have seen myself, I have no hesitation in saying that precisely the same cranial conformation is often repeated in the present day; but it has yet to be shown that any skulls hitherto found are more than approximately similar to the one under consideration.”

King (1864: p. 96) believed that the Neanderthal, while showing some features seen in humans during fetal development, was more chimpanzee-like and thus had its “closest affinities to the apes.” This alternative to Huxley's conclusion should not be surprising considering the fact that, aside from a smattering of not very ancient human skeletal remains, the primary comparisons available to King and to Huxley for the Feldhofer Grotto Neanderthal were with extant apes and humans. While Huxley thought that the large Neanderthal brain was less pithecoid than the shape of its skull might suggest, King (1864: p. 96) interpreted the supposed pithecoid-like cranial features as reflecting a brain whose “thoughts and desires … never soared beyond those of the brute.” Like Huxley, however, King (1864: p. 96) was not averse to ranking human races on the basis of a perceived “primitiveness” of skull and/or intellectual prowess (e.g., “the Andamaner … stand[s] next to brute beknightedness”). But whatever transpired in King's mind between his oral presentation, when he named Homo neanderthalensis, and his composition of the 1864 article, he came to believe that the Feldhofer Grotto Neanderthal should not even be included in the genus Homo. Regardless of their different taxonomic conclusions, it is clear that Huxley's and King's arguments came from the same mold: trying to decide where to draw the line between the lowest of human races and the highest of the brutes.


What is interesting about 20th-century paleoanthropologists is that many of them appear to have lost touch with the social and historical beliefs that shaped the interpretations of their heroes, Huxley and especially Darwin. Thus, as already mentioned above, Huxley is said to have demonstrated a close evolutionary relationship between humans and the African apes when in fact he did not. Rather, if (rightly or wrongly) we imbue Huxley's essay with a phylogenetic theme, we must conclude that he argued for the relatedness of humans to a great ape group. A similar misunderstanding exists with regard to the claim that Huxley demonstrated that Neanderthals were not a species distinct from Homo sapiens. Nevertheless, although neither of these inferences is accurate, Huxley at least discussed morphology. Darwin did not.

Beginning with Raymond Dart's (1925) discussion of the human-like Taung specimen, paleoanthropologists have continually praised Darwin (1871) on two points: his insights about humans being closely allied to the African apes, and originating in Africa. Yet Darwin did not provide one whit of morphology in The Descent of Man or any other discourse in support of either assertion (Schwartz,1987,1999b). Nevertheless, his supposed demonstration of a close human-African ape relationship is now so entrenched in paleoanthropological lore that one can only conclude that those who cite this assertion have not actually read The Descent. But even a casual reading of part 1 of this opus reveals that Darwin's statement is not based on morphology. Instead, it is based on a selectionist argument for why human evolution could not have occurred in the “danger-free” setting of southeast Asia (the home of the red ape, the orangutan) as well as on sustaining a hierarchically race-based notion of the evolution of civilized from primitive “man” (Schwartz,1987,1999b). Even though Darwin did not state this outright either in The Descent or in his notebooks or letters, it seems reasonable, given his Victorian mindset, to infer that his emphasis on Africa as the seat of human origins rested on the presence on that continent of both primitive humans (“Negroes” as in many of his examples) and black African apes.

To return to Mayr's description of hominid taxonomy as a “bewildering diversity of names,” it is certainly true that the years following King's recognition of Neanderthal as a species of Homo witnessed the naming of a plethora of hominid genera and species. This should not, however, be surprising. The history of paleoanthropology did not begin in 1669, when Steno infused fossils with biological reality. Rather, it was not until the 1886 publication of the Spy Neanderthals that humans were broadly recognized as having a fossil record. Consequently, when Mayr inserted himself into paleoanthropology, the science of seeking, discovering, interpreting, and potentially naming extinct human taxa was still quite young. Furthermore, in contrast to the centuries-older discipline of nonhuman paleontology, whose practitioners were comparative anatomists and geologists, the study of human fossils was dominated by human anatomists (e.g., beginning with Schaaffhausen), whose focus typically was and still is on the minutiae of human variation. Thus, rather than being incorporated into the broader arena of paleontology and systematics, from its very beginnings paleoanthropology remained a distinctly separate discipline, no doubt torn between Huxley's legacy of the ancientness of human variability (and the appeal of that to human anatomists), the tradition of paleontology, systematics, and taxonomy, and a history of centuries of debate on the nature of “being human.” The issue of what it meant to be “human” was clearly as relevant to Huxley's and King's discussions of the Feldhofer Grotto Neanderthal as it was to earlier philosophers (e.g., Rousseau and Monboddo) and naturalists (e.g., Linnaeus and Buffon), who endeavored to reflect qualities of humanness versus brutishness in their classifications (Schwartz,1987).

Another element of the history of human paleontology appears to derive from the legitimization of human antiquity. First and foremost, this accreditation made possible exploration that was motivated by the goal of finding human fossils. Indeed, in the 1890s the Dutch paleontologist, geologist, comparative anatomist, and hydrologist Eugène Dubois was the first to seek purposefully and discover human fossils, which he initially referred to the taxon Anthropopithecus erectus but subsequently changed to Pithecanthropus erectus (Dubois,1892,1894). By the third decade of the 20th century, additional Neanderthal specimens as well as the Mauer or Heidelberg mandible (the holotype of Homo heidelbergensis) had been discovered in Europe, a skull from what is now Zambia had had bestowed on it the species name Homo rhodesiensis, Sinanthropus pekinensis had been based on remains excavated from northern China, and a handful of species, divided between three different genera (Australopithecus, Plesianthropus, and Paranthropus), had been recognized from a few sites in South Africa. During the 1940s, as excavations for human fossils proceeded apace in the Old World, the unearthing of new and often unexpectedly different-looking specimens led to their being given new species or even genus and species names [e.g., a somewhat crushed mandible with some teeth from the South African site of Swartkrans was named Telanthropus capensis because it was smaller and less robust than specimens from the same site allocated to Paranthropus crassidens (Tattersall and Schwartz,2000)].


When Mayr published his 1950 article, he was confronted both by a history of seeking human fossils that had not yet spanned 60 years and by a discipline that had barely had time to confront a sampling of human fossils that was only then becoming large enough for substantial critical comparative study and reflection. As such, and in contrast to his proclaimed interest in the history of evolutionary biology, Mayr's antipathy toward the then-current state of paleoanthropology seems insensitive to the discipline's particular history. Nevertheless, with characteristic self-assurance. Mayr assumed the role of the systematist who, although unfamiliar with the specimens themselves, was going to clean up the mess.

Mayr was also clearly unappreciative of perhaps the most significant consequence of human evolution: almost uniquely among organisms, Homo sapiens is the only surviving species of its immediate clade. Consequently, in contrast to the taxic world in which Mayr typically operated, and from which systematists gained knowledge that they applied to the study of the fossil record, one cannot achieve insight into human evolution by studying living humans alone. Thus, if one adopts Huxley's and subsequently Mayr's notion of a purported vastness of human variation in both the present and the past, one will see human fossils in this context and blur the distinction between morphological variation within species and differences among taxa. If, however, one embraces the general pattern of diversity among extant nonhuman taxa as a model for interpreting any fossil record, human or otherwise, one will recognize fossil taxa on the basis of kinds (versus degrees) of difference. With regard to either focus, the biggest mental and visual hurdle one must overcome when engaging in hominid systematics is that, no matter which ape is the closest living human relative, it does not look very much like Homo sapiens.

With choices available to him—diversity versus variation—why did Mayr (1950), the systematist of the synthesis, opt for collapsing the bulk of the then-known human fossil record into a mere three species within the same genus, Homo? He began with an example from the fruit fly genus Drosophila, in which at least 600 species had then been recognized and which, he claimed, “are probably more different from each other than are the various primates and certainly more than the species of the suborder Anthropoidea” (Mayr,1950: p. 109). This is clearly an overstatement inasmuch as the morphological criteria on which Drosophila species were typically based were often little more than differences in thoracic bristle number or wing or antenna shape (Morgan et al.,1926). In fact, many fruit fly species, including one Dobzhansky (1935) named, D. miranda, were distinguished not so much on the basis of morphology, but on the basis of different aspects of their chromosomes or karyotypes (see also Spencer,1963).

Mayr's (1950: p. 110) next argument rests on the claim that “recent studies indicate that the genus is not merely a morphological concept but that it has a very distinct biological meaning. Species that are united in a given genus occupy an ecological situation which is different from that occupied by the species of another genus, or, to use the terminology of Sewall Wright, they occupy a different adaptive plateau. It is part of the task of the taxonomist to determine the adaptive zones occupied by the various genera. The adaptive plateau of the genus is based on a more fundamental difference in ecology than that between the ecological niches of species.”

Aside from the circularity of this statement, it is important to note that Mayr is focusing on ecological situations—niches, not morphology—in making taxonomic judgments. This might seem logical, inasmuch as the concept of the ecological niche is the crux of the biological species concept that Mayr (1942) proposed 8 years earlier: in order for a new species to arise (i.e., for speciation to begin), there must first be a vacant econiche into which a subspecies of an existent species can invade. Once there, as Mayr envisioned the process, geographic isolation in conjunction with the different circumstances of variation and selection will gradually alter this incipient species until it also becomes reproductively isolated from the parent species. If, sometime later, a member of this new species were in a position to mate with individuals of its parental species, the offspring would be infertile and/or not viable.

With these assumptions in place, Mayr (1950: p. 110) then declares: “After due consideration of the many differences between Modern man, Java man, and the South African ape-man, I did not find any morphological characters that would necessitate separating them into several genera. Not even Australopithecus has unequivocal claims for separation. This form appears to possess what might be considered the principal generic character of Homo, namely, upright posture with its shift to a terrestrial mode of living and the freeing of the anterior extremity for new functions which, in turn, have stimulated brain evolution.”

This statement alone should lead one to question Mayr's knowledge of hominid comparative morphology. Perhaps less comprehensible, however, is why any paleoanthropologist should have taken or should still take Mayr's argument for taxonomic minimalism seriously (e.g., see White,2003). Indeed, it is difficult to understand how and/or why Mayr would gloss over and diminish differences between various hominid fossils and modern humans that are so striking they make the criteria on which different species of equid were at the time distinguished pale by comparison (e.g., see Simpson,1944). I shall return to this point in a moment. But to continue, having embraced a notion of presumed widespread geographical and “racial” variation among living humans, Mayr (1950: p. 112) makes the following unsubstantiated assertions:

“I interpret the available literature to indicate that primitive man showed more geographical as well as individual variation than modern man.

“Why primitive man should have been more variable than modern man is not entirely clear…. Whatever the reasons, we should not use the variability within populations of modern man as a yardstick by which to judge the probable variability of extinct populations.

“This point is important because it bears on the question whether or not more than one species of hominid has ever existed on the earth at any one time. Indeed, all the now available evidence can be interpreted as indicating that, in spite of much geographical variation, never more than one species of man existed on the earth at any one time.”

How can Mayr state that there has never been at one time more than one species of hominid? Because the degree of inferred variation (inferred a posteriori after first grouping specimens biochronologically) is used to suggest that there are not now and never were empty econiches that hominids could invade (Mayr's requirement for speciation). To bolster his claim, Mayr cites an argument formulated by Dobzhansky (1944), the fruit fly population geneticist who authored the first monograph in the series that provoked the evolutionary synthesis, Genetics and the Origin of Species (Dobzhansky,1937). As Dobzhansky (1944) imagined it, as culture-bearing organisms hominids exempted themselves from natural selection because they could produce, manipulate, and carry with them their own “environmental” circumstances. Along with their widespread geographic variation, hominids thus become even more unique and further separated from nature. From this perspective, Mayr then concludes that there was only one genus, Homo, and, in ascending order of succession, the species transvaalensis, erectus, and sapiens, which, of course, because of the transformational nature of this sequence, are only arbitrary constructs.

But another—probably the ultimate—reason underlying Mayr's (1950: p. 116) unsubstantiated assertion about humans being the only known nonspeciating metazoan is revealed in this passage:

“If fossils of Congo pygmies and of Watusi were to be found in the same deposit by a paleontologist, a million years hence, he might well think that they belonged to two different species. As stated previously, the known diversity of fossil man can be interpreted as being the result of geographic variation within a single species of Homo…. What is the cause for this puzzling trait of the human stock to stop speciating in spite of its eminent evolutionary success? It seems to me that the reason is man's great ecological diversity. Man has, so to speak, specialized in despecialization. Man occupies more different ecological niches than any known animal. If the single species man occupies successfully all the niches that are open for a Homo-like creature, it is obvious that he cannot speciate.”

The real crux of Mayr's argument is thus the following: if systematists accorded species status to Neanderthals and other human-like fossils, or, more dangerously, recognized some of these fossil hominids as separate genera, this would leave the door open for those who wished to do so to classify human races at a taxonomic level above the subspecies, as some earlier taxonomists, such as Bory St. Vincent, had done (Schwartz,1987) and even contemporaneous taxonomists still advocated [e.g., Gates (1944), who, in the same issue of the American Journal of Physical Anthropology in which Dobzhansky argued that humans were a wildly polytypic species, recognized five living species of human]. If, however, systematists minimized the significance of difference between fossil hominids at any point in time and between fossil and extant hominids—differences that, under other, non-human-focused systematic circumstances, would be given taxonomic valence—then whatever differences (real or the result of prejudice) that might be perceived to exist among living humans were further diminished.

As a systematist who has spent a career analyzing human skeletal morphology and also nearly a decade intensively studying virtually the entire human fossil record, I must reject Mayr's dramatic oversimplification of the differences between “Congo pygmies” and Watusi. Theirs are differences in degree of expression in, not in kind of, morphology (Schwartz,2003,2006). But if we put morphological accuracy aside for the moment and try to appreciate the above quote in its historical context—the immediately post-World War II years of Nazi Germany—we may better understand the raison d'être, the underlying, unspoken motivation, behind Mayr's seemingly systematically neutral essay: if supposedly primitive-looking fossils such as Neanderthals and the Zambian Kabwe skull and many other more or less contemporaneous fossils (all of which differ markedly from any living human and among which are clearly defined morphological groupings) are collapsed into the same species, it then becomes ludicrous even to pretend that any perceived differences between living humans are biologically, and thus evolutionarily, significant (see discussion by Müller-Willie,2007). But the two points of Mayr's essay are not as connected as one might think. On the basis of his biological species concept alone, one can argue that all living humans, as far as biology is concerned, are members of the same species.

Nevertheless, the effect on paleoanthropology of Mayr's (1950) uninformed sense both of human variability and of the human fossil record, together with Dobzhansky's (1944,1955,1962) notion that humans never speciated because the advent of culture removed them from the processes of natural selection, was a stifling of intellectual curiosity that for over half a century prevented this discipline from embracing the more rigorous theoretical and methodological underpinnings of paleontology and systematics in general. Given the clout Mayr and Dobzhansky had as “fathers” of the evolutionary synthesis—and which Mayr maintained until his recent death—they should have exhibited greater restraint in imposing their scenarios about human uniqueness onto the interpretation of the human fossil record. But they did not, and the unfortunate consequence was that paleoanthropology became mired in a systematic “dark age” from which alternative thinking was largely excluded. Regrettably there was some consistency in Mayr and Dobzhansky's actions, because this is precisely the effect the evolutionary synthesis had on evolutionary biology in general: to squelch debate, and to brand as “antievolution” those who offered alternative ideas (Mayr,1949).


To return to the beginning of this contribution, with the references to White and Wilford, is it viable to claim that a surge in multiculturalism underlies those recent analyses of the human fossil record that have recognized taxic diversity in it? If true, this would suggest that cultures and species are of equal biological significance. Yet no one, including these individuals, would accept the assertion that different present-day cultures are the products of different human species. So maybe it is just coincidental that with a rise in multiculturalism, there has been a growing appreciation among paleoanthropologists that the taxic blindfold that first Huxley, and then Mayr and Dobzhansky, imposed on hominid systematics has long outlived whatever sociological relevance it may once have had. The history of paleoanthropology may have gotten off to a rough start, but it is certainly about time that it took its scientific place in the wider intellectual domains of paleontology and systematics.


M. Sommer's invitation to participate in a session on race at the 2003 meeting of the International Society for the History, Philosophy and Social Studies of Biology was the impetus for this contribution. I thank her, as well as I. Tattersall and J. Marks for providing useful criticism of earlier versions, and also S. Müller-Willie for sharing his unpublished manuscript.