Navigation – Plan du site

AccueilNuméros34 (2)Articles thématiquesHuman, almost human: how many hum...

Articles thématiques

Human, almost human: how many human species are there?

Humain, Presque humain : combien y a-t-il d’espèces humaines ?
Thierry Hoquet


La question "combien y a-t-il d’espèces humaines" ouvre sans doute à d’interminables débats. Laissant de côté les aspects théologiques ou politiques de la question, nous analysons trois dimensions du problème avec une approche à la fois épistémologique et historique : métaphysique, biologique et technique. Sur le plan métaphysique, nous revenons aux travaux de Linné et Buffon pour explorer ce qui est en jeu dès lors que l’on suppose que "l’humain" désigne non pas une espèce mais un genre, et que ce genre comprend plus d’une espèce. Nous explorons également la difficulté qu’il y a à donner un critère physique (anatomique) de reconnaissance de l’humain : faut-il s’appuyer davantage sur la main, sur le cerveau ou sur l’usage d’outils ? De plus, le choix des divisions taxinomiques et les dénominations adoptées font aussi polémique. Elles traduisent souvent les croyances implicites des anthropologues, par exemple que l’être humain est le couronnement de l’évolution des primates. Sur le plan biologique, le dénombrement des espèces est problématique dans la mesure où le concept d’"espèce" ne désigne pas seulement des taxons, mais aussi un rang taxinomique (par opposition à "variété"). Les taxinomistes se partagent rituellement entre "agglomérants" (visant à réduire au maximum le nombre d’espèces) et "diviseurs" (n’hésitant pas à promouvoir certaines sous-espèces au rang d’espèces). Enfin, sur le plan technique, de nouvelles méthodes comme la cladistique ou la paléogénomique apportent des outils prometteurs et de nouveaux résultats. Il nous paraît cependant improbable qu’elles soient en mesure de résoudre toutes les difficultés liées à la classification des espèces humaines. Ainsi la question "combien y a-t-il d’espèces humaines ?", si passionnante soit-elle, nous paraît vouée à rester sans réponse : nécessairement ouverte et infiniment débattue.

Haut de page

Notes de la rédaction

Cet article fait suite à une communication invitée présentée lors des 1846es journées de la Société d’Anthropologie de Paris dans le cadre de la session "Humanité(s) : Définition(s), diversités et limites"

Texte intégral


  • 1 No, we have not discovered two new human species!

1During the autumn of 2021, the Cité des Sciences et de l’Industrie in Paris announced a lecture to be given on November 9 by Silvana Condemi, entitled "Denisova, Naledi, Flores: the Homo family is growing". The magazine Pour la Science published an interview with Jean-Jacques Hublin entitled "Non, on ne vient pas de découvrir deux nouvelles espèces humaines".1 These two events are indicative of both the significant excitement generated by recent discoveries, and the accompanying intense controversies over their interpretation. Questioning the number of human species is sometimes interpreted as a brutal undertaking, as it can be seen as challenging our notion of the unity of humankind. We must make it clear here that we are not seeking to undermine the old monogenist claim that all humans descend from the same strain or origin, in order to insidiously revive the kind of polygenism generally considered to be the source of racist hierarchies. In fact, striving to find out how many human species there are is not an anthropological question only: it touches just as much on metaphysics ("the essence of Man"); it also has to do with issues of general biology and bears in particular on more theoretical aspects related to the practice of taxonomy (what is a species? how do we distinguish them?). Our answer to the number-of-species question might ultimately be related to the invention of new methodologies that claim to be the key to solving (if not dissolving) the question. In this paper, we will avoid the theological and political overtones of the debate to focus on its metaphysical, biological and technical aspects. The "how many human species" debate is constantly recurring, especially as both the terms "human" and "species" have no single definition or meaning, as if their contours were doomed to be endlessly disputed.

Metaphysical issues

2Palaeoanthropology is deeply enmeshed in a host of metaphysical issues surrounding the definition of "humanness". This is what gives palaeoanthropological debates their particular flavour. In this section, I will discuss taxonomic debates from a historical perspective, starting with eighteenth-century taxonomic difficulties related to the question of species, and their metaphysical implications.

Homo: the scandalous notion of a multi-species genus

3When the Swedish naturalist Carl Linnaeus was putting together his general system of classification, he was desperately seeking an anatomical trait that could be the hallmark of humanity and would set our kind apart from other animal species. In the first edition of the Systema Naturae (1735), Linnaeus created the class Anthropomorpha divided into Homo, Simia and Bradypus (the sloths). While he characterized the latter two by physical features (their number of fingers), Homo was associated with three puzzling Latin words: "Nosce te ipsum". Faced with the attacks of his fellow naturalists such as J.T. Klein (1743), who claimed that humans were not "anthropomorphs", Linnaeus persistently replied that, despite all his efforts, he had been unable to identify any sound anatomical character (see the preface to Linnaeus 1746, and his letter to Johann Georg Gmelin, Linnaeus 1747). To the scientific beholder, "Nosce te ipsum" sounds like a riddle: not a biological trait but rather a mental attitude usually translated as "Know thyself" (we shall return to this below). Strikingly, the Linnaean criterion at once allowed the coining of the binominal species name Homo sapiens, and prohibited any practice of palaeoanthropology, because the stuff that makes humans human could never be evinced by fossil remains. Linnaeus’s formula allows at most for palaeoprimatology, insofar as human specificity is encapsulated in a mysterious enigma that cannot be solved by mere naturalist observation. A notable feature of the Linnean attempt at classification is that the human species is not identified by any definite anatomical or physiological character: it is as if any physical description of humans were doomed to miss the point.

4Likewise, Linnaeus’s contemporary, Georges-Louis Leclerc, Count of Buffon, the French director of the Jardin du Roy in Paris, considered that humans are defined by the possession of only two characters – thought on the inside, language on the outside (Buffon, 1749a) – which do not refer in the least to the anatomical structure of individuals. As a matter of fact, Buffon does not provide his readers with any "Adam-and-Eve"-like illustration of what a typical human pair would look like: his engravings present only peculiar monstrosities (a piebald child) and typical "passions of the soul" (Hoquet, 2007).

5However, Linnaeus and Buffon do not offer the same answer to the question of how many human species there are. Buffon was clearly obsessed with defining what species are (Sloan, 1987), and is even considered as a precursor of some modern species concepts (Gayon, 1996). In his view, humans were very much like dogs, both being species that come with a host of varieties but which remain unmistakably one and the same (see Buffon, 1749b and his "Table de l’ordre des chiens", in Buffon, 1755). However, humans differ from dogs as they are not only incredibly diverse in their shapes and looks (colour of the skin, texture of the hair…), but also anatomically similar to and even indistinguishable from great apes. Like any other animal species, the human species is, in Buffon’s eyes, defined by a biological trait: intraspecific fertility and interspecific sterility (on this, see below). But Buffon also exaggerates the physical, anatomical and morphological proximity of some ape species such as the "Pongo" and the "Jocko" (resp. an Orangutan and a Chimpanzee) to human varieties like the Hottentot. Claiming to see no difference at all between them, Buffon even stated that physical hybridization was actually taking place between humans and apes.

6For this, Buffon compensates by attributing to humans the possession of a metaphysical soul, of which apes and monkeys are deprived (Hoquet, 2014). Buffon thus made his anthropological distinction not on biological but on metaphysical grounds, by emphasizing the role played by the soul and its twofold manifestations (thought and language), which no other animal possessed.

7In contrast, Linnaeus did not consider Homo as a species but as a genus. In the extant literature of the time, he found another figure called Homo sylvestris or Orang-Utan. Faced with this ambiguous anthropomorphic species, Linnaeus failed to build a strong case based on physical difference. From a taxonomic perspective, several options were open to him: he could classify it with monkeys, in the genus Simia; he could make it a separate intermediate order between Homo and Simia – viz. Satyra; he could classify it as a species in the genus Homo. It was this latter alternative – undoubtedly the most scandalous one – that Linnaeus eventually chose. Thus, in the tenth edition of Systema naturae (1758), the genus Homo contains two human species: not only the famous Homo sapiens, "diurnal man," "varying according to the culture and according to the place", but also a second species, Homo troglodytes, or nocturnal. Sapiens therefore does not have the exclusive title of "human species." From the very moment it was coined, the genus Homo defined by "Nosce te ipsum", contained two types of "humans": those who are "sapiens" and those who are not.

8Linnaeus and Buffon offer two different approaches to the same metaphysical conundrum of situating "humans" within a naturalistic framework that does not provide a clear answer to what the boundaries of the species are. Both men seem to refer to the immortal soul as the hallmark of what defines "humans", but they both leave us with a paradox. Linnaeus’s classification provides a metaphysical criterion to identify the genus Homo, but it prevents us from identifying the corresponding species on an anatomical basis. Buffon defines species not in terms of resemblances but as individuals performing fertile matings, but he also admits the occurrence of sexual intercourse between apes and humans – as if the immense metaphysical distance between them allowed for the most intricate biological intimacy.

Nosce te ipsum: "know yourself" or "find out for yourself"?

9When it comes to defining humans, biological aspects are not everything. We have seen that Linnaeus could not find any anatomically distinctive features of the genus Homo. However, he could have adopted Aristotle’s criterion (the upright position as the defining trait of bipedal humans) or Descartes and Buffon’s criteria (language as the external expression of inner thinking, or the metaphysical sign of an immortal and immaterial soul). I argue in this section that the existence of feral humans (ferus) forced Linnaeus to discard both criteria (bipedalism and language).

10The case of "feral" or "wild" children – raised by animals without any contact with humans – was an enduring concern in eighteenth-century philosophy (Tinland, 1968). It prompts us to take culture into account in our definition of what counts as "human". Linnaeus was especially struck by these individuals. In the tenth edition of his Systema (Linnaeus, 1758), he describes them as "tetrapus, mutus, hirsutus" (quadrupeds, mute, uneducated). In order to accommodate them in his classification, Linnaeus introduced a new variant within the species Homo sapiens that he called ferus. The case of "ferus" individuals shows that properties such as language or bipedalism are not natural properties of humans but depend on their actualisation in a social environment, in short, on receiving an education. This explains why neither language nor bipedalism can be the hallmarks of H. sapiens as a biological species, as they can be (accidentally) missing. A biological definition must be independent of context: it should make us capable of recognizing the specimen regardless of the context in which it is encountered. For instance, language is not a biological, species-defining trait, as some human individuals may not develop it. On the other hand, it is now known that, placed under certain conditions, individual chimpanzees can acquire a capacity for rudimentary language. But, language is not part of the biological definition of the chimpanzee species: if a chimpanzee spoke, it would simply be by accident. As Tim Ingold noted (1995, p. 194): "The biological definition of species depends upon the possibility of context-independent specification: thus a chimpanzee is a chimpanzee, Pan troglodytes, whether reared among other chimpanzees or among humans, whether in the forest or in the laboratory."

11From the ferus case, it seems obvious that Linnaeus could not adopt bipedalism or language as the biological hallmarks of the sapiens species, as they do not provide for any context-independent specification. As a result, Linnaeus was forced to embrace a vague and intangible character, nosce te ipsum, whereby humans must engage in an infinite project, the never-ending task of knowing themselves. We can even suppose that the Latin expression "Nosce Te Ipsum" means not so much "Know thyself" as "Find the knowledge for yourself", or "Find out for yourself", echoing the riddle of the sphinx who proposed to Œdipus a definition made up of various characters for which he had to find the corresponding species name: "a creature that is first four-footed, then two-footed and finally three-footed." Linnaeus challenges humans in the opposite way: he provided us only with a name "H. sapiens," but as to what this means, "yours to say, yours to find out: nosce te ipsum!" The Linnaean moment in 1758 is therefore essential to our question, "How many human species are there?" as it shows that, from a biological perspective, there has never been one human species. Linnaeus introduced humans as a genus within the zoological classification, but he also excluded them from the field of systematics in four different ways: by characterizing them first by a moral trait (sapiens) and not by a physical mark; by showing, then, that within sapiens, it is necessary to include forms like ferus that lack the most obvious characters of this class – bipedalism and language, but also culture and civilization; by giving the name Homo not to one single species but to a genus, thereby opening up the possibility of more than one human species; then, finally, by actually including a second species (troglodytes) in this genus even though he could have referred it to another genus (Simia), or made it a completely distinct genus (Satyra), or reduced it to merely a variety of sapiens (as he did with ferus or with the major racial types). Thus, troglodytes is the first form that pluralized the meaning of Homo: a form that haunts our definitions of sapiens like an eerie Doppelgänger. In short, the fundamental fact here is that Homo is a genus and that as such it can accommodate a plurality in which sapiens is only a particular form, without any claims to exclusivity. Since Linnaeus’s struggles with the case of wild (ferus) children, evolutionary palaeoanthropology has been consistently striving to identify the distinguishing feature of humans. Three traits are in competition: bipedalism, language, tools and material culture. These three different approaches also function as competing criteria to determine what really makes humans human, and to decide whether or not one species belongs to the human lineage. Palaeoanthropologists dream of a clear and precise criterion. Some turn to the skull or teeth, others to the foot, others to the hand. Le Gros Clark (1955) suggested a formal redefinition of the genus Homo that rests explicitly on a cranial capacity of at least 900 cc. Ashley Montagu (1961) spoke of a "Cerebral Rubicon" to encapsulate the notion of a simple and clearly identifiable threshold – a clear demarcation between human and non-human.

Humanizing Primates? An evolving Telos

12Between the first (1735) and the tenth (1758) editions of his Systema naturae, Linnaeus abandoned the concept of "Anthropomorpha" and replaced it with "Primates". It might be claimed that the term "Primates" also bears hints of anthropocentrism: after all, "primate" comes from the Latin "primus", designating that which has first rank. In France, for instance, the archbishop of Lyon bears the title of "Primate of the Gauls", which suggests his authority over the other bishops of France. In the case of the Linnean taxonomy, it seems undeniable that "Primates" rank highest in the animal kingdom as they include the human species. But in spite of this suspicion of lingering "anthropomorphism", the term is undeniably better than its rival Anthropomorpha. It is noticeable that Linnaeus’s contemporaries protested vehemently, not against his alleged anthropomorphism but, quite to the contrary, against the mixing of humans with other natural creatures: they attacked not only the concept of Anthropomorpha (Klein, 1743) but also the division of the genus Homo, with its duplication into sapiens and troglodytes. To name but a few: Jean-Jacques Rousseau (1755) summoned the naturalists to decide whether or not they included troglodytes within the human species; Thomas Pennant (1771) considered that Linnaeus’s Homo nocturnus should not be separated from Simia satyrus; Johann Friedrich Blumenbach (1775) claimed that the Systema naturae’s greatest fault was to confuse human and simian attributes and took up Buffon’s distinction between Bimanes and Quadrumanes – a division exemplifying his desire to correct Linnaeus. Against the attacks from these critics, the Linnaean inspiration prevailed on many accounts.

13Contemporary classifications are still in the throes of coining names for Primate taxa (species, genera, classes). One example is the systematic biology of the much debated "Homo longi" (Ji et al., 2021), which is defined as:

14- Order Primates Linnaeus, 1758;
- Suborder Anthropoidea Mivart, 1864;
- Superfamily Hominoidea Gray, 1825;
- Family Hominidae Gray, 1825;
- Tribe Hominini Gray, 1825;
- Genus 
Homo Linnaeus, 1758;
Homo longi sp. nov. Ji & Ni.

15It is striking that two Linnaean categories are still in use in modern anthropological taxonomy: the order Primates and the genus Homo. However, as we move up the taxonomic hierarchy, we find for instance the superfamily "Hominoidea" that includes gibbons, chimpanzees, gorillas, orangutans, and humans. This anthropomorphic taxon name strongly suggests that the (great or lesser) apes look hominoid, and thus that the typical form is human not simian or that apes are human-like. The taxonomic hierarchy does not go downwards with the wider term Primates: it goes upwards from Homo to Hominini to Hominidae, Hominoidea and Anthropoidea: all these taxonomic choices are closer to Linnaeus’s initial idea of ‘Anthropomorpha’ than to his category of ‘Primates’. This choice of terminology is also conceptual: it raises apes up to the anthropoid line rather than reminding humans of their simian nature. This classification therefore reinforces the impression that there is an anthropogenic trend, or telos, in the primate family (e.g. Simpson, 1945). It should also be noted that the term "Anthropoidea" is used by Ji et al. (2021) although it has been constantly criticized and rejected since Hoffstetter, who replaced it with "Simiiformes" (Hoffstetter, 1974; 1977; 1979; 1982).

16As for Julian Huxley, he coined the term Psychozoa (psychic animals) to designate humans, as an attempt to record the singularity of our mental existence in scientific nomenclature, thus triggering debates attempting to determine whether our species deserves such an emphatic term and whether human history has really evinced any "anagenetic" or progressive trend (Hartwig, 1958; Straus, 1958). Becoming "human" appears as a trend within the human lineage, a telos that must be attained, as if evolution had culminated in the appearance of sapiens (Gundling, 2020).

Almost-humans and other ambiguous beings

17There seems to be no way out of the metaphysical (and, for that matter, terminological) conundrum of palaeoanthropology. Describing both non-sapiens humans and within-sapiens varieties, the phrase "almost-humans" (Hoquet, 2021) aims to capture the feeling shared by many naturalists that there exists, or has existed, a number of intermediate beings between humans and apes. The identities of almost-humans are blurred, and they disturb well-established dichotomies like the one between apes and humans. Conceiving what it is to be "almost-human" is complex but two previous uses of the phrase might be relevant in a palaeoanthropogical context. First, in the 1895 issue of the Bulletins de la Société d’anthropologie de Paris, Léonce Manouvrier described Pithecanthropus erectus as a "pre-human anthropoid race [...] remarkable for a combination of various almost human characters that we observe scattered in the various modern anthropoid genera" (1895, p. 633, emphasis added). The second occurrence belongs to popular culture. In May 1970, Roger Lécureux and André Chéret, the authors of the Rahan comic book series, released a volume in which, marginal to Ceux qui marchent debout ("Those who walk upright", who are the humans), they feature a (rather disquieting) depiction of Ceux qui marchent presque debout – "Those who walk almost upright." What are they: humans or apes? They would seem to be more accurately described as intermediate.

18Ancient zoological literature readily echoed Latin editions of Aristotle’s De animalibus, where one reads that some creatures are partim hominem partim quadrupedem imitentur, having an intermediate nature between humans and quadrupeds. The phrase was echoed by Tyson in his classic description of the Orangutan (1699) and it can also be found in Daubenton’s Leçons sur l’homme (1795).

19Palaeoanthropology is replete with such ambiguous beings, part human and part quadruped. It is haunted by the question of the "almost" and that of the "hybrid". Darwin (1871) referred on several occasions to "half-human" or "semi-human" living beings. New palaeontological depictions of the pre-human emerged, functioning as a "missing link", a hypothetical type between two forms of life, an unknown step in the scala naturæ, which led to an unrelenting search for intermediate entities. New entities are constantly being introduced inside or outside the genus Homo, or into the subfamily Homininae. When coining the term Pithecanthropus (literally: Simian-Human), Ernst Haeckel powerfully suggested the existence of an intermediate creature (1868; 1876). This term triggered Eugène Dubois’s own discoveries (Theunissen, 1989; Lequin, 2019). In 1892, Dubois first described the newly-found fossils as Anthropopithecus erectus (upright human-like Simian), but then adopted (Dubois, 1894) the reversed term Pithecanthropus erectus (upright simian-like Human) as a move from Man-Ape to Ape-Man (Le Gros Clark, 1967; Wood, 2020). Haeckel also suggested that his Pithecanthropus was alalus (deprived of language). But what could be the meaning of such an ancestral non-speaking arch-human? Is it possible to imagine a species named "Homo alalus", "rude forefather of Nature’s Masterpiece, progenitor of man, tho’ not yet man" (Rice, 1900)?

20The analysis of fossils often tends to replay the Aristotelian concern with ambiguous beings, "partim hominem partim quadrupedem imitentur". Raymond Dart’s discovery of Australopithecus africanus in 1925 offers a good example of this. Although the name Australopithecus emphasizes the pithecus, the simian nature of this being, it was first described not as "an ape-like man, a caricature of precocious hominid failure, but a as creature well advanced beyond modern anthropoids in just those characters, facial and cerebral, which are to be anticipated in an extinct link between man and his simian ancestor" (Dart, 1925, p. 128, emphasis added). Robert Broom commented on the discovery in these terms: "The study of the teeth renders it pretty certain that Australopithecus stands somewhere between the chimpanzee and man." (Broom, 1925, p. 414, emphasis added). This quote is striking on two accounts: first, in that it literally takes up the notion of an intermediate being between chimpanzee and human; secondly, it bluntly abandons the evolutionary notion of an "unknown common ancestor," and places the new fossil between two contemporary forms – the human and the chimpanzee. Let us now consider another quote, from William King Gregory (1927, p. 232): "In other words, even if we did not have the chimpanzee, we should have to infer its existence as a sort of half-way station in the long road of ascent from the primitive Eocene primates." Here again we notice that the anthropomorphic ape is not only "almost-human", but is, in Darwin’s words (1871), "semi-human": perceived as a "half-way" stage in a series culminating in the emergence of the human. "Almost-human" also suggests the possibility of failure, of unsuccessful twigs on the Tree of Life, but, more radically, a view of primates as an unfulfilled promise (a point well evinced in Gundling, 2020).

21Along with teeth, skulls, and other bones, feet also play a role in the search for and identification of intermediate beings. The use of the great ape’s foot as an analogon of the pre-human foot leads on the one hand to conceiving of the great ape’s foot as an imperfectly human foot, which has failed to become fully human; on the other hand, the typically human foot is suddenly propelled into a long story, haunted by traces of an arboreal and prehensile past, also making up a subtext that needs to be deciphered (Lequin, 2019).

Biological aspects of the argument: systematics and the definition of species

22If there is a dispute over the number of human species, this is due to several causes. Some of them are general issues in the field of systematics, such as multiple definitions of the species category, the debate between "lumpers" and "splitters" and the phenomenon of taxonomic inflation. They raise questions such as the plurality of taxonomic ranks, or the nature of hybridization events.

"Species": a complex level in taxonomy

23Since Linnaeus (1758), species have been conventionally named using Latin binomials comprising a generic substantive and a specific epithet. The term "species" usually refers to a kind or a form. It is sometimes associated with the notion of "essence" – to the point that the term "species" is at times associated with a Platonic or Aristotelian legacy or seen as a matter for the theory of universals (e.g. Hull, 1965; Ereshefsky, 2001). It is well known that more than twenty different rival definitions of "species" are in competition (Mayden, 1997), which often conveys a sense of "the failure of modern species concepts" (Hey, 2006). Methodologically, such a bewildering profusion of definitions suggests a notion of "pluralism" and a feeling of "anything goes". The debate often bears especially on properties such as reproductive isolation or monophyly – generally treated as necessary properties of species although they can be viewed as contingent. Note that Darwin (1859) refused to take part in the definition of species, clearly embracing definitional vagueness: "Nor shall I here discuss the various definitions which have been given of the term species. No one definition has as yet satisfied all naturalists; yet every naturalist knows vaguely what he means when he speaks of a species." Darwin also admitted that for most naturalists of his time, the term "species" referred "to the unknown element of a distinct act of creation." What mattered most to Darwin was the fluctuation between taxonomic ranks, and especially the species and the varieties conceived of as incipient species. Darwin was astonished that the botanists of his time were diverging widely over the actual number of oak species existing in Britain. He concluded that we do not have any characteristic criterion that would identify a form as a full-fledged species rather than as a simple variety. Even a "dwarfed condition" could "in some cases be inherited for at least some few generations" (Darwin, 1859).

24Besides, definitional vagueness is a multi-dimensional problem because there may be confusion between several concepts of species: species as a classificatory unit or taxon bearing a binomial name; species as a systematic rank, a class occupying a given level in the nomenclatural hierarchy; species as an evolutionary concept, a twig in the Tree of Life or the basic unit of evolution (Dubois, 2011). Although different definitions of species qua taxa are in competition, Darwin challenges our ability to identify species qua rank or taxonomic level (as opposed to, say, variety). Darwin might have raised doubts about the nature of species qua taxonomic rank while acknowledging the relevance of species qua taxa (Stamos, 1996; 2007). Discussions on species concepts are liable to trigger heated debate, as these concepts may have both an ideological and an explanatory role. Some species concepts are of the mechanistic type: they focus on speciation, the process by which species originate by divergence, incompatibility and reproductive isolation. Such is the biological species concept (BSC): genetic and ecological mechanisms ensure cohesion in a population, allowing interbreeding within a population and simultaneously preventing breeding with members of different species. Other species concepts are more historical: they consider pattern (rather than process). Two historical species concepts, whose names sometimes vary, are the most prominent: the phylogenetic or evolutionary concept (ESC) is based on the temporal dimension of the species, the ancestor-descendant sequence of populations, and the notion of monophyly; the phenetic concept (PSC) is based on the external or morphological resemblance of species members, and is more compatible with the method of character analysis.

25Many other definitions exist of course, but they might be considered as mostly redundant and reduced to mere variants of the three main ones (BSC, ESC, PSC). Examples are the "Recognition" species concept (Paterson, 1985) that emphasizes the specific mate recognition system, or the "Cohesion" species concept (Templeton, 1989) that insists on the mechanisms of phenotypic cohesion within the species: both could easily be subsumed under the BSC. "Chronospecies" are included in the ESC while "morphospecies" pertain to the PSC.

26However, general biology has failed to provide a clear hierarchy between these various conceptions of what defines a species. Each has its advantages and its faults, and no species concept has ever been unanimously adopted. For instance, the BSC seems to exclude inter-specific hybridization, but even its proponents have varied over its definition and operability. For instance, in 1753, Buffon changed the definition of species from a phenetic concept to a biological one based on reproductive isolation and a criterion of interfertility (Buffon, 1753). But Buffon himself hesitated as to the sterility of inter-specific hybrids (Buffon, 1776; Sloan, 1987). An important breakthrough occurred when the ornithologist Ernst Mayr took up the Buffonian criterion, making it a central feature of his populational approach distinguishing groups that actually or potentially interbreed (Mayr, 1942; 1963; De Queiroz, 2005). On the other hand, the PSC is based on resemblance, and is the most operational for palaeontological entities where interfertility cannot be tested. The ESC emphasizes the historical aspects of species evolution: from this perspective, species are understood as descending from each other, described as lineages or "space-time worms" (Hull, 1976; Gayon, 1996). The BSC is sometimes difficult to apply, due to intermediate cases such as Rassenkreis or hybrids that are neither fertile nor sterile, but may have slightly or greatly reduced fecundity. Neontologists can find it difficult to agree on parsing and classifying the lineages of living species, but the problem is further aggravated in palaeosystematics, where identifications are made from fossils, i.e. often isolated and fragmentary dead specimens (i.e. not evincing any behavioural or reproductive patterns).

"Humans" at the above-species level

27As mentioned above (see the "Humanizing Primates?" section), palaeoanthropologists use various names, with neighbouring terms that can cause confusion. "Hominid" for instance should not be confused with "hominin" or "hominine". "Hominin" refers to the tribe Hominini, grouping together taxa that are more closely related to extant humans than to chimpanzees; "Hominine" refers to the subfamily Homininae (which includes chimpanzees) while "Hominid" refers to the members of the "Hominidae" family, which includes "great apes" (see Wood and Richmond, 2000). None of these concepts puts a clear divide between what counts as "human" and what does not. "Human" does not seem to be a taxonomic category. Higher taxa have always raised numerous difficulties (Hoffstetter, 1973), to such a degree that some have declared "the end of higher taxa" (Cartmill, 2013).

28But first of all, the plurality of taxonomic rankings for the term "human" is obscured by the fact that Homo is not the name of a species, but that of a genus. The nature of the taxonomic rank called "genus" is even more obscure than that of the species (Dubois, 1982; 1983; 1985). When we speak of "a new human species," the need that arises is to understand under what conditions this form may or may not belong to the genus Homo. The difficulties met at the species level also occur at the genus level (Wood and Collard, 1999; Collard and Wood, 2015).

"Lumpers" and "splitters"

29Linnaeus’s Systema naturae (1758) takes the concept of species as its taxonomic basis. His position is often associated with fixism, the idea that species do not change through time. However, he was fully aware of the perplexing case of the Peloria, a Greek word designating a monstrosity identified in a plant of Linaria vulgaris. This suggested to Linnaeus that new forms could perhaps originate by natural means. Linnaeus also claimed that species were as numerous today as were originally created (Linnaeus, 1736; 1751, §157). This discretism (the idea that there exists a discrete and finite number of categorical entities) is often linked to the topic of essentialism. The notion of Linnaeus’s "essentialism" – the idea that there is a limited number of fixed and unchanging essences – was a clear target for the ornithologist Ernst Mayr (1982) and the philosopher David Hull (1965) who were opposing "typological" to "populational" thinking. Note, however, that this historical narrative has been sharply criticized since by Winsor (2006a; 2006b) and Müller Wille (2007). Mayr and Hull’s idea was to prove not only that species were changing and evolving through time, but also that there was no ontological reality associated with the species category (Eldredge, 2013). In this perspective, the number of species becomes a debatable matter: it might be too great to be mastered by our limited brains; more importantly, it is dependent on epistemological decisions and on consensus on what counts as a species.

30As Davies (2015) put it: "When quantifying the number of species, systematists are subjected to contrary incentives, leading to increasing or reducing the number of taxa." This difference in appreciation has been described as distinguishing "lumpers" from "splitters" – a divide that has constantly bedevilled palaeoanthropology. A good example of such "contrary incentives" is provided by the case of Ernst Mayr. As previously mentioned, Mayr was an ornithologist and a great reformer of animal systematics, and is especially well known for formulating the Biological Species Concept based on the criterion of fertility of matings. When dealing with the neontological genera he studied, Mayr was rather on the "splitters" side as he admitted great taxonomic diversity. But when it came to the study of human fossils, he moved to the "lumpers" side, denouncing a "bewildering diversity" that he proposed to reduce to three species within the genus Homo (Mayr, 1950). I would say that, because of his anti-essentialism and his focus on populations, Mayr’s approach makes it possible to split populations that do not actually interbreed. But, on the other hand, the BSC is a good way of lumping populations together by looking at whether they have a possibility (at least theoretically) of producing offspring together, as in the case of the North American wapiti (Cervus canadiensis) and the European red deer (C. elaphus). Mayr’s conclusions on palaeoanthropology may be taken as sound parsimony, healthy epistemological prudence or, on the contrary, as a sign of metaphysical panic. When it comes to the biological definition of humans, any attempt at taxonomic revision can be suspected of being motivated by metaphysical considerations.

31But it should be noted that "lumpers" and "splitters" are essentially evaluative terms: 1/ even though the "gold standard" or "litmus test" of what defines a species is still in debate, species are treated as facts; 2/ naturalists (palaeoanthropologists included) seem to agree that species are separate (discrete) entities and that there is a "good number" of species – not too few, and not too many; 3/ in a sound classification, the number of species should reflect the level of perceived differences. As Maciej Henneberg (1995) put it: "If it differs as much from us as we differ from living apes, it cannot be identified as ancestral only to ourselves. Since we identify our fossil ancestors by similarity to ourselves the variation within the fossil hominid assemblage cannot be too great." However, what matters in counting two different entities as two different species differs with the species concept adopted: for the BSC, interfertility or barriers to the gene flow; for the PSC, synchronous discontinuity of morphological characters; for the ESC, diachronic variation or alteration in the lineage.

32The divide between "lumpers" and "splitters" has an impact not only on the number of species, but also on the way species are taxonomically organized: between those who prefer linear sequences of speciation, and those who prefer a many-branched (hierarchical) tree of evolution. De Silva (2004) provides a comparison of two "family trees": one grouping fossils into relatively few species and rejecting the idea that the human family tree is shaped like a bush (associated with the name of Tim White) and one that is particularly "bushy", using measurable differences as evidence for prolific speciation in the past (associated with the work of Ian Tattersall, 2002).

Splitting tendencies and taxonomic inflation

33Taxonomic inflation has two main causes: splitting of species and new findings. Splitting of species is a process by which subspecies are raised to the species level. This is especially true for neontological taxa, where taxonomic inflation is mostly due to a change in the definition or concept of species, rather than to new discoveries (Isaac et al., 2004). Species inflation is driven by the rise of conservation biology, but also by the academic career reward system. But it is also true in palaeoanthropology, for instance in the case of H. erectus (e.g. Kramer, 1993).

34New findings are also a major source of taxonomic inflation, especially as the field of palaeoanthropology is rather recent. In the 18th century, no fossil human remains were known. "Sacred physics" (the idea that the Bible gave the actual description of the history of nature) interpreted fossil remains as "Diluvium reliquiae" (remnants of the biblical Deluge). It was a perplexing puzzle that no human bones were found in quarries until, in 1725, parts of a skeleton were discovered by Johann Jakob Scheuchzer in the quarry of Öhningen, Germany. Those bones, however, were not described as a natural phenomenon, but as a miraculous sign: the irrevocable evidence of the existence of the accursed human race that was annihilated by the Deluge. Scheuchzer (1732) took it as a Homo diluvii testis – a human witness to the Deluge. However, these remains were later reinterpreted by Georges Cuvier (1811) as a fossilized kind of salamander of unknown species. Discoveries of human remains only started later in the 19th century, with the Engis child in 1830, the Forbes Quarry adult in 1848, and the discovery of the skeleton from the Neander valley in 1856. In comparison with these humble beginnings, the rate of discoveries in the field of palaeoanthropology has clearly been increasing. The data now available might be seen as scarce and fragmentary, providing a poor sample of the full range of variability, but on the other hand, a relative profusion of data has to be acknowledged. This, in turn, might have favoured the "splitters" tendency and led to some "taxonomic inflation."

35In palaeoanthropology, even more than in general biology, an interpretive bias favours the promotion of each finding as a new taxonomic entity. New hominin fossils are routinely given new species names (e.g. Ardipithecus ramidus, Australopithecus anamensis, Homo antecessor, etc.) – raising concerns that the eagerness with which palaeontologists emphasize hominid species diversity is rather an artefact or a distortion of facts. Tim White for instance has been a fertile creator of new hominin taxa (the genus Ardipithecus, and the species kadabba, ramidus, afarensis, garhi) but he has constantly urged his colleagues not to multiply species names in order to accommodate or isolate every species lineage (e.g. White, 2003; 2014). Of course, while differences between fossil findings should not automatically lead to the coining of a different species, it remains true that there is an indisputable bonus associated with declaring that a specimen belongs to the human lineage, and even more so, that it represents a new species.

36It is also important to realize that human remains are not interpreted only in an isolated manner. Interpretations also take place in the context of a higher-level narrative on the origins of humanity (Balzeau, 2021), in which the unity of the monogenist paradigm (the "out of Africa" model – see Hublin et al., 2017) may be challenged by new findings. In 2021, controversy raged when a skull found in Harbin (China) was described as a new human species (Ji et al., 2021). In contrast, Hublin (2021) declared that two new human species had not been discovered, denouncing a taxonomic frivolousness that jeopardized the unity of the interpretive paradigm in place. He also suspected that nationalist interests might be at stake, in recent attempts to claim the non-African origin of some humans. Questions of nomenclature should not be invaded and perverted by ideologies.

37Human taxonomy today is littered with "species names" that may not designate "species". When the names Homo heidelbergensis (Rightmire, 1990) and Homo ergaster (Wood, 1992) were resurrected, they were perceived as distinct from Homo sapiens and Homo erectus. But this resulted in a recent human fossil record that was almost as specious as that documenting the earlier history of the family Hominidae. This also suggested that human taxonomic diversity had been significantly pruned or slashed over the later periods. However, these names have been disputed, and the number of taxa has been deemed to be seriously overestimated (Kramer, 1993). The validity and unity of the term Homo erectus is also in question. Sometimes described as the common ancestor of humans and Neanderthals, H. erectus is a polytypic palaeospecies that encompasses the entire range of hominid spatio-temporal variation over the two million years or so of its existence. With a biological or phenetical concept of species in mind, H. erectus seems to be a practical designation encompassing all hominin fossils found in Old World geological deposits ranging from 2 to 0.1 million years ago. Palaeoanthropologists wonder what they should do with H. erectus: consider it as polytypic species, or merge it with H. sapiens? Should Meganthropus, for instance, be considered as a genus of its own, or should it be merged with H. erectus? It is often claimed that H. erectus was promoted in the wake of Mayr 1950 in order to bury a taxonomic nomenclature that was obscuring the evolutionary process – Pithecanthropus, Sinanthropus, Meganthropus, Atlanthropus (see e.g. Wolpoff, 1994) – but this is not necessarily so (Zanolli, 2019). In other words, anyone who ventures into palaeoanthropology should be wary of names. Some names (such as H. erectus) are used so broadly that they may designate not species but temporal, geographical or morphological perimeters.

Lumping together: to what extremes?

38Since Scheuchzer’s so-called Homo diluvii testis, many new hominin remains have been discovered, named and published in recent years. However, it is not always clear how they relate to Homo sapiens. For a long time, the predominant view on Neanderthal remains was that they were too primitive to have played any role in the evolution of our own lineage. As a result, they were placed as a separate species, Homo neanderthalensis, and denied any ancestral role (e.g., Boule, 1911; 1913). At the same time, some peculiarities of their anatomy supported the idea that they were simply ordinary but undernourished humans with a Vitamin-D deficiency and suffering from rickets (see also the debate between Ivanhoe (1970) and Mayr and Campbell (1971), leading the latter to the paradoxical conclusion: "In spite of rickets and arthritis, the morphology of Neanderthal man can be clearly recognized and interpreted as a harmonious functional adaptation to the environment in which he lived").

39More recently, similar debates have arisen with respect to the hominin remains discovered on the island of Flores, Indonesia in October 2004. The combination of primitive and derived features led to the description of a new species called Homo floresiensis (Brown et al., 2004). In the wake of this discovery, there has been an ongoing debate about the status of this species and how it is related to sapiens. Some considered it a late surviving species of primitive Homo, or simply a ‘hobbit’, a modern individual with a growth disorder and various syndromes such as microcephaly. The question was raised as to whether its existence on Flores in conditions of long-term isolation might be responsible for endemic dwarfing (Tucci et al., 2018).

40These conflicting interpretations in the cases of Homo neanderthalensis and Homo floresiensis suggest that a radical "lumping" tendency is tacitly at work in palaeoanthropology. "Lumping" interpretations could be more than welcome in some Creationists’ circles, with all human diversity ultimately reduced to Adam’s lineage and its monogenetic heirs. But "lumping" tendencies are often advanced on scientific grounds also, when the feeling dominates that "too many" taxa have been recognised. Some have raised questions such as "why bother to distinguish Australopithecus africanus from A. boisei, or Homo habilis from H. erectus?" Mayr and his "single species hypothesis" pertaining to the Pleistocene hominids (1950) had followers such as Wolpoff (1971). Henneberg and Thackeray (1995) considered that "before a claim of evolutionary branching of a lineage is laid, the null hypothesis of the lack of speciation must be falsified". Accordingly, they formulated a "single-lineage hypothesis of hominid evolution": a suggestion "also compatible with the uniformitarian postulate – recent human evolution occurred, and occurs, within one lineage consisting of widely dispersed but interacting populations". A suggestion that Henneberg himself identified later as that of an "extreme lumper" (see his comment to Holliday, 2003).

Technical artefacts and the role played by new methodologies

41The general epistemological question here is the following: can new advanced techniques provide a definitive answer to taxonomic questions? And do they affect our answer to the "how many human species" question?

Two taxonomic innovations

42Two different taxonomic innovations are currently transforming the field: palaeosystematics (Hennigian cladistics), and palaeogenetics and palaeogenomics. The first methodological innovation is the broader adoption of cladistics as the proper method to produce phylogenies. Palaeosystematics mostly rests on the analysis of various morphological characters observed on bones and teeth (except for a few rare fossils containing sufficient DNA). Is cladistics (based on morphological or genetic characters) likely to bring more clarity or to create more confusion in the field of palaeohuman taxonomy? The rise of Hennigian cladistics tends to encourage the tendency to multiply species in the field of palaeoanthropology (White, 2014).

43The second technical innovation is related to the fact that, since the discovery of PCR (Polymerase Chain Reaction, a method used to amplify old DNA), genetics have shaken up the genealogy of different human species giving rise to palaeogenetics and palaeogenomics. These new methodologies have delineated a very dynamic and promising field. In fact, a striking result of these new genetic methods is that they have enhanced our perception of hybridization phenomena. The newly acknowledged existence of hybrid specimens (Neanderthal with sapiens, Denisovans with sapiens, Denisovans with Neanderthal) exacerbates the sense among researchers of facing taxonomic chaos, and almost every single paper in the field refers to "admixture" or "mixed ancestry". The existence of high levels of hybridization has two main consequences: 1/ it is used to challenge traditional species concepts (such as the BSC) and replace them with a vaguer concept (such as "syngameon"); 2/ they also support a reticulate view of human evolution, instead of the hierarchical "tree of life" view (e.g. Holliday, 2003). Ultimately, there may no longer seem to be any human species, biologically speaking, since Homo sapiens might appear as the product of a hybridization between various strains: an amalgam of Neanderthal, Denisovans and other human variants (Carroll, 2003). The progress of genome deciphering techniques is boosting the cladistic method in systematics, as it makes character analysis easier. Another striking result of the introduction of genetic traits in the cladistics of human extinct species is that it blurs the boundary between palaeoanthropology and neontology. These used to be distinguished by the nature of their objects (fossil vs. living), by their concept of species (PSC vs. BSC), and the way they approached the "intermediary" notion (morphology vs. interfertility and hybridity), but the study of ancient fossil DNA tends to bring them closer together. However, while palaeogenetics allows a neater study of hybridism, the persistent use of the term "introgression" is revealing of the fact that scholars still consider the three groups (sapiens, Denisovans, Neanderthal) as different. "Hyper-lumping" stances like Henneberg’s "single-lineage hypothesis" are clearly not the dominant model of hominid evolution.

Earlier technical artefacts

44Are they able to solve all the old questions of taxonomy? Research on ancient DNA is undoubtedly an exciting field with promising research avenues. But one can infer from previous technical artefacts how doubtful it is that advanced genetics will provide the solution to all questions of classification. In 1958, Francis Crick declared: "Biologists should realize that before long we shall have a subject which might be called ‘protein taxonomy’– the study of amino acid sequences of proteins of an organism and the comparison of them between species. It can be argued that these sequences are the most delicate expression possible of the phenotype of an organism and that vast amounts of evolutionary information may be hidden away within them." In the wake of Crick’s project, Emile Zuckerkandl put forward the concept of "molecular anthropology" (1963). This proposal followed on from research carried out with Linus Pauling on a molecular clock allowing evolution to be measured by comparing certain molecules such as haemoglobin (Zuckerkandl et al., 1960). The question is, can we distinguish species, reconstruct phylogenies and measure the time of evolution by simply comparing the structure of certain molecules? As the microbiologists Olsen and Woese (1993) put it, paying homage to the work of Zuckerkandl and Pauling, "When dealing with molecular sequences an evolutionist feels a sense of liberation." However, palaeontologist Simpson (1964) steadfastly criticized Zuckerkandl’s molecular approach and, in particular, disparaged the thesis that "from the standpoint of haemoglobin, man is just an abnormal gorilla": if that is what the haemoglobin says, then either it is only valid from the point of view of haemoglobin, or haemoglobin lies, or at the very least, it strongly prompts us to take a relative view by considering other complementary standpoints. The use and misuse of molecular clocks is a warning against technical artefacts giving the illusion that a new methodology can solve – or rather, dissolve – all conceptual difficulties (Marks, 2009).

The promises of fossil DNA

  • 2 Abductive reasoning may be described as seeking the law L which accounts for phenomena A and B; if (...)

45There is something fascinating about genetic methods and their traditional cohort of metaphors (codes, programmes): they suggest that a secret lurks deep within organic molecules and that we only need to learn how to decipher it in order to understand everything. As a result, since the prognostications of physicist Erwin Schrödinger in the 1940s, much of molecular biology has been devoted to cracking the code of life (Kay, 2000). Some suggest that even without fossil DNA, traces of archaic DNA could be found by comparing current genomes (Vernot and Akey, 2014). But this triggers new difficulties, such as the emergence of "ghost species" (or "ghost populations"), identified neither by fossil remains nor by traces of old DNA, but by genetic echoes left in current populations (David Reich in Callaway, 2014). For example, Patterson and colleagues (2012) assume a now extinct population in northern Eurasia that is thought to have descended from ancestors of Europeans and a Siberian group that later migrated to the Americas. "Ghost populations" are identified using statistical models when genetic data from fossils are absent. This may lead to a sort of "Le Verrier effect": in 1845, Urbain Le Verrier famously predicted the existence of an as yet unknown planet from the observation of irregularities in the motion of the planet Uranus. Le Verrier’s bold a priori deduction was later confirmed by the a posteriori discovery of Neptune. "Le Verrier effects" thus describe the case of theoretical predictions that are later confirmed by empirical observations. Le Verrier effects give a sense of perfect demonstration, but we can never be sure that our hypothesis indeed refers to an actual biological reality and that we are not reifying a statistical artefact. Besides, in the case of palaeoanthropology, predictions are made neither by deduction (from premises to consequences), nor by induction (generalization from a certain number of particular cases), but by abduction2, another kind of scientific reasoning, but it is far removed from the standards of Platt’s strong inference (1964).


46Counting the number of "human" species is not an easy task. Like other biological fields, palaeoanthropology has become pervaded with "lumping" and "splitting" tendencies – so much so that it has sometimes been described as being at the crossroads between two contradictory trends (e.g. Schwartz, 2006): on the one hand, it strives towards constant taxonomic enrichment, which leads to naming every fossil even to the point of suggesting that each represents a new species; on the other hand, it evinces a desire to maintain the unity of the species, which leads to thinking that all forms discovered are ultimately only variations within one and the same species. How might these opposite trends be reconciled?

47Today’s palaeoanthropologists may be tempted to look down on the ancient naturalists, the scarcity of their data and their rudimentary tools. But can we solve the taxonomic issues at stake and get rid of all forms of "almost-human" beings simply because we now have significant collections of fossils and formidable new techniques for determining the hierarchy of relationships of resemblances and differences between different specimens? There is tremendous hope as genetic material is only beginning to be exploited. Cladistic techniques are renewing the phylogenetic approach – bringing a sense that palaeoanthropological research is now able to distinguish humans not only from present-day non-human primates, but also from all their "almost-human" ancestors, hominids or hominins. However, the question of what makes "true", or "full-fledged" "humans" remains unanswered – partly because this question might not be solvable by applying epistemic criteria only.

48If we are to answer the question raised in the title of this paper – how many human species are there? – we need to realize that two different questions are closely entangled: first, we have to define the overall circle that defines the sphere of the human (what Linnaeus called the genus Homo, defining an inside and an outside); secondly, we need to define the number of parts into which this sphere will be divided (the number of species). The first section of this paper, dealing with metaphysical issues, shows that it is difficult to determine what makes a human human, at least from a purely zoological point of view, partly because the topic is loaded with a strong metaphysical charge. Our analysis of the biological concepts of species insists again on the fact that metaphysical decisions have to be taken. This leads to the final section on technical aspects: while new tools really do change our views on different issues, it is hard to believe that instruments have the magical power of closing metaphysical questions. Thus, it seems that instead of being able to count how many human species there actually are, we are facing a totally different question: what is a human, an anthropos?

49From today’s perspective, it might be tempting to say that Linnaeus and Buffon shied away from the scientific reality of humankind; they stepped back to a purely metaphysical stance, and for that reason, their thinking seems totally out-of-date, irrelevant. However, we should keep in mind that they had absolutely no idea of the diversity of human fossils. They were dealing only with living humans – their contemporaries – and were reluctant to characterize them in a purely anatomical or morphological way. Both undoubtedly exaggerated the physical closeness between modern humans and the great apes; they also gave enormous credit to ethnographic descriptions of the huge variety existing among humans. Both also admitted the possibility of a (primate) being who could be the exact anatomical double or Doppelgänger of extant humans without, however, being human. In the 19th century, this Doppelgänger took on the aspect of various ape-men and other pithecanthropes. At the beginning of the 20th century, a novel by Stéphane Servant (1906) coined the term "Morphè-Anthropos" – the "morphè" (i.e. form) being what distinguishes a human being. But if the form (morphè) is not enough to make the human (anthropos), what then do we require from the entity in order to admit it as – fully or partially – "human"? What distinguishes the anthropomorph from the full-fledged anthropos? But what makes an anthropos, if not its form?

Remerciements : This paper was first presented as a plenary talk during the Congress of the Société d’Anthropologie de Paris in January 2021. My gratitude goes to François Marchal, its Secretary General, for his invitation to present a philosophical and epistemological perspective on these fascinating questions, and for his wise guidance in a complex and developing field. I also thank Mathilde Lequin, who introduced me to the history and philosophy of palaeoanthropology. I am indebted to both François and Mathilde for our friendly discussions, for all the "food for thought" they offered to me together with their witty advice and friendly recommendations. Two anonymous reviewers read my draft: they made many useful suggestions and shared important insights with me. I thank them for their time, and for helping me to improve the paper. I am personally responsible for all remaining mistakes.

Haut de page


Balzeau A (2021) Brève histoire des origines de l’humanité. Taillandier, Paris, 317 p

Blumenbach JF (1775) On the Natural Variety of Mankind. In: Bendyshe T (ed) (1865) The Anthropological treatises of Johann Friedrich Blumenbach. Longman, London

Boule M (1911-1913) L’homme fossile de la Chapelle-aux-Saints. Annales de paléontologie 6:11-172; 7:21-56; 8:1-70

Broom R (1925) On the newly discovered South African Man-Ape. Natural History 25(4):409-418

Brown P, Sutikna T, Morwood MJ et al (2004) A new small-bodied hominin from the Late Pleistocene of Flores, Indonesia. Nature 431:1055-1061

Buffon GL Leclerc de (1749a) Histoire naturelle de l’homme. In: Histoire naturelle, générale et particulière. Imprimerie Royale, Paris, vol. 2

Buffon GL Leclerc de (1749b) Variétés dans l’espèce humaine. In: Histoire naturelle, générale et particulière. Imprimerie Royale, Paris, vol. 3

Buffon GL Leclerc de (1753) L’Âne. In: Histoire naturelle, générale et particulière. Imprimerie Royale, Paris, vol. 4

Buffon GL Leclerc de (1755) Le Chien avec ses variétés. In: Histoire naturelle, générale et particulière. Imprimerie Royale, Paris, vol. 5

Buffon GL Leclerc de (1776) Des Mulets. In: Histoire Naturelle générale et particulière, servant de suite à l’histoire des animaux quadrupèdes. Imprimerie Royale, Paris, vol. 23 (Suppléments, tome III)

Callaway E (2014) The Neanderthal in the family. Nature 507:414-416

Carroll SB (2003) Genetics and the making of Homo sapiens. Nature 422:849-857

Cartmill M (2013) The end of higher taxa: a reply to Tattersall. Evolutionary Anthropology 2(4):172-173

Collard M, Wood B (2015) Defining the genus Homo. In: Henke W, Tattersall I (eds) Handbook of Paleoanthropology. Springer Verlag, Berlin-Heidelberg, pp 2107-2144

Crick FH (1958) On protein synthesis. Symposium of the Society for Experimental Biology 12:138-163

Cuvier G (1812) Recherches sur les ossements fossiles de quadrupèdes, où l’on rétablit les caractères de plusieurs espèces d’animaux que les révolutions du globe paraissent avoir détruites. Déterville, Paris

Dart RA (1925) Australopithecus africanus: the Man-Ape of South Africa. Nature 115:195-199

Darwin CR (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London

Darwin CR (1871) The descent of man, and selection in relation to sex. John Murray, London

Daubenton LJM (2006) Leçon sur l’homme (26 Floréal An III, 16 mai 1795). In: Guyon É (ed) L’École Normale de l’An III. Leçons de physique, de chimie, d’histoire naturelle. Éditions Rue dUlm, Paris

Davies W (2015) Human evolution: How we misread our own story. Nature 522:154-155

De Silva J (2004) Interpreting Evidence: An Approach to Teaching Human Evolution in the Classroom. The American Biology Teacher 66(4):257-260+262-267

Dubois A (1982) Les notions de genre, sous-genre et groupe d’espèces en zoologie à la lumière de la systématique évolutive. Monitore zoologico italiano (N.S.) 16:9-65

Dubois A (1983) Hybridation interspécifique, similarité génétique, parenté phylogénétique et classification supraspécifique en zoologie. Année biologique 22:37-68

Dubois A (1985) Le genre en zoologie : essai de systématique théorique. PhD thesis, Université des Sciences et Techniques du Languedoc Montpellier II

Dubois A (2011) Species and "strange species" in zoology: do we need a "unified concept of species"? Comptes Rendus Palevol 10:77-94

Dubois E (1894) Pithecanthropus erectus: eine menschenähnliche Uebergangsform aus Java. Landesdrückerei, Batavia

Eldredge N (1993) What, If Anything, Is a Species? In: Kimbel WH, Marti LB (eds) Species, Species Concepts and Primate Evolution. Advances in Primatology. Springer, Boston (MA), pp 3-20

Ereshefsky M (2001) The Poverty of the Linnaean hierarchy. A philosophical study of biological taxonomy. University Press, Cambridge

Gayon J (1996) The individuality of the species: A Darwinian theory? – from Buffon to Ghiselin, and back to Darwin. Biology and Philosophy 11:215-244

Gregory WK (1927) Dawn-man or ape? Scientific American 137(3):230-232

Gundling T (2020) Ineluctably us: early hominid discoveries, mass media, and the reification of human ancestors. History and Philosophy of the Life Sciences 42(41) []

Haeckel E (1868) Natürliche Schöpfungsgeschichte. Gemeinverständliche wissenschaftliche Vorträge über die Entwickelungslehre im Allgemeinen und diejenige von Darwin, Göthe und Lamarck im Besonderen, über die Anwendung derselben auf den Ursprung des Menschen und andern damit zusammenhängende Gründfragen der Natur-Wissenschaft. Reimer, Berlin

Haeckel E (1876) The history of creation: or, the development of the earth and its inhabitants by the action of natural causes. A popular exposition of the doctrine of evolution in general, and of that of Darwin, Goethe and Lamarck in particular. Translated from the German by E Ray Lankester. HS King & Co, London

Hartwig CW (1958) Psychozoa. Science 127(3301):765-766

Henneberg MM, Thackeray JF (1995) A single-lineage hypothesis of hominid evolution. Evolutionary Theory 11:31-38

Hey J (2006) On the failure of modern species concepts. Trends in Ecology & Evolution 21:447-450

Hoffstetter R (1973) Origine, compréhension et signification des taxons de rang supérieur : quelques enseignements tirés de l’histoire des Mammifères. Annales de Paléontologie (Vertébrés) 59(2):135-169

Hoffstetter R (1974) Apidium et l’origine des Simiiformes (= Anthropoidea). Comptes rendus hebdomadaires de l’Académie des Sciences, Paris. ser. D. 278:1715-1717

Hoffstetter R (1977) Phylogénie des Primates. Confrontation des résultats obtenus par les diverses voies d’approche du problème. Bulletins et Mémoires de la Société d’Anthropologie de Paris 4(4):327-346

Hoffstetter R (1979) Controverses actuelles sur la phylogénie et la classification des Primates. Bulletins et Mémoires de la Société d’Anthropologie de Paris 6(3):305-332

Hoffstetter R (1982) Les Primates Simiiformes (=Anthropoidea) (compréhension, phylogénie, histoire biogéographique). Annales de paléontologie 68(3):241-290

Holliday TW (2003) Species Concepts, Reticulation, and Human Evolution. Current Anthropology 44(5):653-673

Hoquet T (2007) Buffon illustré. Les gravures de l’Histoire naturelle (1749-1767). Éditions du Muséum national d’histoire naturelle, Paris, 816 p

Hoquet T (2014) Biologization of race and racialization of the human: Bernier, Buffon, Linnaeus. In: Bancel N, David T, Thomas D (eds) The invention of race. Scientific and popular representations. Routledge, New York, pp 17-32

Hoquet T (2021) Les Presque-Humains. Mutants, Cyborgs, Robots, Zombies, … et nous. Le Seuil, Paris

Hublin JJ (2021) Non, on ne vient pas de découvrir deux nouvelles espèces humaines ! Pour la Science 527:54-57

Hublin JJ, Ben-Ncer A, Bailey SE et al (2017) New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiens. Nature 546:289-292

Hull DL (1965) The effect of essentialism on taxonomy: Two thousand years of stasis (I). The British Journal for the Philosophy of Science 15:314-326

Hull DL (1976) Are species really individuals? Systematic Biology 25(2):174-191

Ingold T (1995) ‘People like us’: the concept of the anatomically modern human. Cultural Dynamics 7(2):187-214

Isaac NJB, Mallet J, Mace GM (2004) Taxonomic inflation: its influence on macroecology and conservation. Trends in Ecology & Evolution 19(9):464-469

Ivanhoe F (1970) Was Virchow right about Neandertal? Neandertals living in early Würm times may have suffered from a vitamin D deficiency. Nature 227:577-579

Ji Q, Wu W, Ji Y et al (2021) Late Middle Pleistocene Harbin cranium represents a new Homo species. Innovation (Camb) 28;2(3):100132 []

Kay LE (2000) Who Wrote the Book of Life? A History of the Genetic Code. University Press, Stanford, CA, 441 p

Klein JT (1743) Summa dubiorum circa classes quadrupedum et amphibiorum in celebris domini Caroli Linnaei Systemate naturae: sive naturalis Quadrupedum historiae promovendae prodromus cum praeludio de Crustatis. Leipzig

Kramer A (1993) Human taxonomic diversity in the Pleistocene: Does Homo erectus represent multiple hominid species? American Journal of Physical Anthropology 91:161-171

Le Gros Clark W (1955) The fossil evidence for human evolution: An introduction to the study of paleoanthropology. University of Chicago Press, Chicago (The Scientist’s Library: Biology and Medicine)

Le Gros Clark W (1967) Man-Apes or Ape-Men? The Story of Discoveries in Africa. Holt, Rinehart & Winston, New York

Lequin M (2019) Bipédie et origines de l’humanité : philosophie et histoire de la paléoanthropologie. Hermann, Paris

Linnaeus C (1736) Fundamenta botanica quae majorum operum prodromi instar theoriam scientiae botanices per breves aphorismos tradunt. Salomon Schouten, Amsterdam

Linnaeus C (1746) Fauna Suecica, sistens animalia Sueciae regni… distributa per classes et ordines, genera et species. L. Salvius, Stockholm

Linnaeus C (1747) Manuscript Letter to Johann Georg Gmelin, St Petersburg 25 February []

Linnaeus C (1751) Philosophia botanica in qua explicantur Fundamenta Botanica cum definitionibus partium, exemplis terminorum, observationibus rariorum. Kiesewetter, Stockholm

Linnaeus C (1758) Systema naturæ per regna tria naturæ, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Editio decima reformata. Laurentii Salvii, Stockholm

Manouvrier L (1895) Deuxième étude sur le Pithecanthropus erectus, comme précurseur présumé de l’homme. Bulletins de la Société d’anthropologie de Paris 6:553-651

Marks J (2009) What is the Viewpoint of Hemoglobin, and Does It Matter? History and Philosophy of the Life Sciences 31:241-262

Mayden RL (1997) A hierarchy of species concepts: the denouement in the saga of the species problem. In: Claridge MF, Dawah HA, Wilson MR (eds) Species: The units of diversity. Chapman & Hall, New York, pp 381-423

Mayr E (1942) Systematics and the Origin of Species. Columbia Univ. Press, New York

Mayr E (1963) Animal species and evolution. Belknap Press, Cambridge

Mayr E (1950) Taxonomic categories in fossil hominids. Cold Spring Harbor Symposium on Quantitative Biology 15:109-118

Mayr E (1982) The Growth of Biological Thought. Belknap Press, Cambridge/London

Mayr E, Campbell B (1971) Was Virchow Right about Neandertal? Nature 229:253-254

Montagu A (1961) The "Cerebral Rubicon": brain size and the achievement of hominid status. American Anthropologist 63:377-378

Müller Wille S (2007) Collection and collation: theory and practice of Linnaean botany. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences 38(3):541-562

Olsen GJ, Woese CR (1993) Ribosomal RNA: a key to phylogeny. The FASEB Journal 7:113-123

Patterson N, Moorjani P, Luo Y et al (2012) Ancient Admixture in Human History. Genetics 192:1065-1093

Paterson HEA (1985) The recognition concept of species. In: Vrba E (ed.) Species and speciation. Pretoria, Transvaal Museum Monograph 4:21-29

Pennant T (1771) Synopsis of quadrupeds. J Monk, Chester

Platt JR (1964) Strong Inference: Certain systematic methods of scientific thinking may produce much more rapid progress than others. Science 146(3642):347-353

de Queiroz K (2005) Ernst Mayr and the modern concept of species. Proceedings of the National Academy of Sciences 102(suppl 1):6600-6607

Rice LL (1900) Homo alaus. The Open Court 9:512

Rightmire GP (1990) The Evolution of Homo Erectus. Comparative Anatomical Studies of an Extinct Human Species. Cambridge University Press, Cambridge

Rousseau JJ (1755) Discours sur l’origine et les fondements de l’inégalité parmi les hommes. Marc-Michel Rey, Amsterdam

Scheuchzer JJ (1732-1737) Physique sacrée ou Histoire naturelle de la Bible. Schenk, Amsterdam

Schwartz JH (2006) Race and the odd history of human palaeontology. The Anatomical Record (Part B) 289:225-240

Servant S (1906) La légende humaine. La Revue intellectuelle des faits et des œuvres (Organe rationaliste) 1(1)

Simpson GG (1945) The Principles of Classification and a Classification of Mammals. Bulletin of the American Museum of Natural History (New York) 85

Simpson GG (1964) Organisms and Molecules in Evolution. Science 146:1535-1538

Sloan PR (1987) From logical universals to historical individuals: Buffon’s idea of biological species. In: Histoire du concept d’espèce dans les sciences de la vie (Colloque international, Paris, 1985). Éditions de la Fondation Singer-Polignac, Paris, pp 101-140

Stamos DN (1996) Was Darwin really a species nominalist? Journal of the History of Biology 29:127-144

Stamos DN (2007) Darwin and the nature of species. State University of New York Press, Albany, New York, 273 p

Straus WL (1958) Response: "Psychozoa". Science 127(3301):766

Tattersall I (2002) Monkey in the Mirror. Harcourt, New York

Templeton AR (1989) The meaning of species and speciation: A genetic perspective. In: Otto D, Endler JA (eds) Speciation and its consequences. Sinauer Associates, Sunderland (MA), pp 3-27

Theunissen B (1989) Eugène Dubois and the Ape-Man from Java: the history of the first missing link and its discoverer. (Translated from the Dutch by Enid Perlin-West). Kluwer Academic, Dordrecht, Boston, London

Tinland F (1968) L’homme sauvage : homo ferus et homo sylvestris : de l’animal à l’homme. Payot, Paris

Tucci S, Vohr SH, McCoy RC et al (2018) Evolutionary history and adaptation of a human pygmy population of Flores Island, Indonesia. Science 361(6401):511-516

Tyson E (1699) Orang-Outang, sive Homo Sylvestris: or, the Anatomy of a Pygmie Compared with that of a Monkey, an Ape, and a Man. T. Bennett & D. Brown, London

Vernot B, Akey JM (2014) Resurrecting surviving Neandertal lineages from modern human genomes. Science 343:1017-1021

White TD (2003) Early Hominids: Diversity or Distortion? Science, New Series 299(5615):1994-1995+199

White TD (2014) Delimitating Species in Paleoanthropology. Evolutionary Anthropology 23:30-32

Winsor MP (2006a) Linnaeus’ biology was not essentialist. Annals of the Missouri Botanical Garden 93:2-7

Winsor MP (2006b) The Creation of the Essentialism Story: An Exercise in Metahistory. History and Philosophy of the Life Sciences 28:149-174

Wolpoff MH (1971) Competitive Exclusion Among Lower Pleistocene Hominids: The Single Species Hypothesis. Man 6(4):601-614

Wolpoff MH, Thorne AG, Brno JJ et al (1994) The case for sinking Homo erectus: 100 years of Pithecanthropus is enough. Courier Forschungsinstitut Senckenberg 171:341-361

Wood B (1992) Origin and evolution of the genus Homo. Nature 355:783-790

Wood B (2020) Birth of Homo erectus. Evolutionary anthropology 29(6):293-298

Wood B, Collard M (1999) The Human Genus. Science 284:65-71

Wood B, Richmond BG (2000) Human evolution: Taxonomy and paleobiology. Journal of Anatomy 196:19-60

Zanolli C, Kullmer O, Kelley J et al (2019) Evidence for increased hominid diversity in the Early to Middle Pleistocene of Indonesia. Nature Ecology & Evolution 3:755-764

Zuckerkandl E (1963) Perspectives in Molecular Anthropology. In: Sherwood Washburn (ed) Classification and Evolution. Aldine, Chicago, pp 243-272

Zuckerkandl E, Jones RT, Pauling L (1960) A Comparison of Animal Hemoglobins by Tryptic Peptide Pattern Analysis. Proc Natl Acad Sci USA 46:1349-1360

Haut de page


1 No, we have not discovered two new human species!

2 Abductive reasoning may be described as seeking the law L which accounts for phenomena A and B; if this law is valid, we can predict C; C is attested, therefore law L is certainly the one that accounts for A, B and C.

Haut de page

Pour citer cet article

Référence électronique

Thierry Hoquet, « Human, almost human: how many human species are there? »Bulletins et mémoires de la Société d’Anthropologie de Paris [En ligne], 34 (2) | 2022, mis en ligne le 12 octobre 2022, consulté le 27 novembre 2022. URL : ; DOI :

Haut de page


Thierry Hoquet

Département de philosophie, Université Paris Nanterre, Nanterre, France ; t.hoquet[at] ;

Articles du même auteur

Haut de page

Droits d’auteur


Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

Haut de page
  • Logo Société d'Anthropologie de Paris
  • Logo Fonds National pour la Science Ouverte
  • Revue soutenue par l’Institut des sciences humaines et sociales du CNRS
    CNRS - Institut national des sciences humaines et sociales
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search