The German comparative anatomist Johann Friedrich Blumenbach (1752-1840) was a pivotal figure in the taxonomic and biological turns in the natural history of man and literally inscribed the changing import of the concept of a race. The oft-asserted primary motives for his 'favourite anthropological studies' were to prove the singularity of man's place in the animal kingdom — 'poles apart from the Orangutang' — and establish the membership of all human beings within 'the same common species'. But he also insisted that the normal process of classifying 'the races and degenerations' of animals and plants be applied to 'the varieties of mankind that had emerged from its common original stock'.^[18]

Accordingly, in the radically revised third edition of De Generis Humani Varietate Nativa, 'On the Natural Varieties of Mankind', Blumenbach (1795:284-7) formalized his long emergent classification of the 'five foremost varieties of mankind, one true species', and labelled them 'Caucasian', 'Mongolian', 'Ethiopian', 'American', and 'Malay' (Figure 4). Through the three Latin editions of this work (1775, 1781, 1795), the concrete nouns Blumenbach used most often to refer to units of collective human difference were gens, 'nation', 'race', 'people', and varietas, 'variety'. Neither connoted a race in the narrowly biological sense but whereas he saw gens as a real, 'natural division', varietas was a 'general division' in a taxonomy 'which we constituted'.^[19] In the penultimate section of the third edition of De Generis Humani, Blumenbach (1795:114‑283) drew extensively on empirical descriptions of actual gens to illustrate his deduction that collective human 'degeneration' or change — as indexed particularly in 'national differences in [skin] colour' — resulted from the operation of external physical causes on a single migrating human species rather than from an original plurality of species. In the final section (1795:284‑322), he moved from a consideration of abstract varietas, 'diversity', 'variety', in the major elements of human physical appearance to the formal classification of human beings into a small number of theoretically concrete varietas, 'varieties'. At this point, the empirical noun gens almost disappears, to be largely replaced by two botanical metaphors, stirps and stemma, both connoting descent from an ancestral stock.^[20]

Figure 4: Anon., 'Crania collectionis meae quina selectissima adumbrat, ad totidem generis humani varietatum principalium diversitatem demonstrandam: 1. Tungusae; 2. Caribaei; 3. Feminae juvenis Georgianae; 4. O-taheitae; 5. Aethiopissae Guineensis'.^[21]

Engraving. Photograph B. Douglas.

The publication in 1798 of a German translation by Johann Gottfried Gruber (1774-1851) of the third edition of De Generis Humani was identified by Timothy Lenoir (1980:93) as a key moment in the articulation of Blumenbach's taxonomic lexicon and more generally in the biologization of the term race.^[22] A close reading shows why. For much of the text, Gruber paralleled Blumenbach's Latin terminological mix and usually translated his concrete nouns gens and varietas respectively as ein Volk, 'a people', 'a nation', and eine Varietät, 'a variety'. Stemma and stirps are generally ein Stamm, 'a stock', 'a stem', but occasionally eine Rasse or Race, 'a race', always with reference to Blumenbach's problematic category of the Malay race — in these instances, the term had what Georg Forster called the 'undetermined' implication of a 'crowd' of people of 'idiosyncratic character' but 'unknown ancestry'.^[23] Otherwise, Rasse/Race scarcely figure in the translation until the final section (1798:203‑24). Here, Varietät is used initially for the taxonomic unit varietas but is abruptly supplanted by Race as the work climaxes in detailed characterization of Blumenbach's five Abarten, 'hereditary varieties', of mankind. The insertion of Race where the Latin text moves definitively into taxonomic mode was no mere whim but a deliberate semantic strategy by both author and translator, as Gruber (1798:259-61) made clear in a long appendix. In response to his own rhetorical complaint about the lack of a consistent classificatory vocabulary for the natural history of man, he lauded the precise Natureintheilung, 'natural classification', proposed by 'our great Kant'.

Gruber was alluding to a series of papers in which the German philosopher Immanuel Kant (1724-1804) had addressed a paradox at the core of natural history: the presence of radical, seemingly permanent physical diversity in a single human species with a common ancestral stock.^[24] Kant's solution was to yoke teleology to genealogy in order to explain present human Verschiedenheit, 'variety' in the abstract, in terms of the triggering in different environments of pre-existent Keime, 'germs', or Anlagen, adaptive natural 'predispositions', within the single original Menschenstamm, 'human stock'. In the process, he formally differentiated Racen, 'races', from Arten, 'species', on the one hand, and from Varietäten, 'varieties', on the other: individuals of different races of the same Stamm could interbreed and produce fertile hybrid offspring, unlike those belonging to different species; while races, unlike varieties, 'remain constant over prolonged generation' when transplanted and could engender stable hybrids. Accordingly, a race was 'inevitably hereditary' and the demonstrated capacity to propagate a Mittelschlag, 'blended character', was a prime determinant of Kant's human classification.^[25] This partial epigenetic theory differed from that of the mature Buffon in a crucial respect: for Kant, human races were structurally distinct because the original Stamm was predisposed to be permanently and irreversibly adaptive to different external conditions, making skin colour the paramount outward sign of 'natural' inner organic differences and capacities; whereas for Buffon, the human 'germ' was everywhere the same, degeneration was externally induced and theoretically reversible, and variations in skin, hair, and eye colour were 'superficial' products of 'the influence of climate only'.^[26]

Lenoir (1980:92-5) linked Blumenbach's formal endorsement of Kant's biological terminology in 1798 to a recent metamorphosis in Blumenbach's thinking: his acknowledgment of reproductive criteria as critical signifiers of human diversity alongside his longstanding emphasis on morphology. In 1797, Blumenbach had modified his earlier insistence on external causes of morphological differences, especially in skin colour, by defining race along generative Kantian lines and invoking empirically the diagnostic significance of racial mixing: 'the word race indicates a character born of degeneration which necessarily and inevitably becomes hereditary through reproduction, as for example when whites engender mulattos with negroes, or métis with American indians'. He then acknowledged Kant as the first to identify heritability as the main 'difference between races and varieties'.^[27] The imprint of Kantian terminology is patent in key changes between the 1790 and 1806 editions of Blumenbach's Beyträge zur Naturgeschichte, 'Contributions to Natural History'. In the first edition (1790:79-83), he briefly outlined but did not name the pentad of human 'varieties' that he had already sketched, also unnamed, in the second edition of De Generis Humani (1781:51‑2), translating his own Latin phraseology directly into German as fünf Spielarten, 'five varieties/sports'.^[28] The second edition of Beyträge zur Naturgeschichte (1806:55‑66) includes a new section extolling the value of 'anthropological collections', notably Blumenbach's own, for an empirically based natural history of mankind. His unequalled assemblage of the 'skulls of foreign nations' rendered corporeal the paradox of human unity in diversity (that Kant's biology had resolved deductively): the collection displayed the 'identity of mankind as a whole' and the 'boundless transitions' linking its physical 'extremes'; but concurrently it provided 'proof of the natural division of the whole species into the five principal races [Hauptrassen]'. In the following section (1806:67‑72), rehearsing his now named classification, he retained Spielarten as a general term for 'the varieties of mankind within its common original stock' but systematically substituted Rassen for Spielarten when referring to the particular divisions — his five Hauptrassen — suggested by his reading of the 'open book of nature'.

The insertion of Race or Rasse into Blumenbach's taxonomic vocabulary between 1795 and 1806, where he had previously used Varietät or Spielart, is a textual marker of the precipitation of a narrower, biological connotation of a race from a much older semantic slurry. It also signals the incipient normalization of hereditarian ideas of human difference in conjunction with new anatomical and physiological knowledge that challenged climatic and humoral explanations. In France, the altered usage slid easily into the technical lexicon of the natural history of man and the term race was duly redefined in the sixth edition of the Dictionnaire de l'Académie (1835). For instance, a prospectus for the shortlived Société des Observateurs de l'Homme issued in 1801 by the society's perpetual secretary and Cuvier's ally, the pedagogue and publicist Louis-François Jauffret (1770-1840), called for a 'methodical classification of the different races' grounded in a 'complete work on the comparative anatomy of peoples' (1978:74). On the cusp of this racialization of human difference, Cuvier exercised considerable practical influence in the emerging science of race, belying the relatively little he published on the subject.^[29] If in the early 1790s he had refused to attribute supposed Negro shortcomings to their anatomy, by the end of the century (1978:173-4), he had clearly imbibed Kant's and Blumenbach's reconfiguration of Rassen as organic and hereditary.

Cuvier's most authoritative pronouncement on human variation comprises a ten-page segment of his magnum opus Le règne animal, 'The Animal Kingdom' (1817a), concluding his discussion of the 'first order' of mammals, the 'Bimana or Man'. He first sketched a standard four-stage Enlightenment theory of human progress:^[30] 'man's development' was 'retarded' or 'advanced' at very different 'degrees' according to 'circumstances' such as climate, soil, and vegetation. But two ominous provisos qualified this universal schema and underwrote a rigid racial taxonomy: that the human species showed 'certain hereditary conformations which constitute what we call races'; and that 'intrinsic causes' appeared to 'halt the progress of certain races, even in the most favourable circumstances'. Earlier in the text, Cuvier had expressed strong doubt that all the characteristic differences between 'organized beings' could be produced 'by circumstances'. He now identified three 'eminently distinct' major human races characterized by congenital somatic features: the 'white, or Caucasic', ('to which we belong'), was typified by the 'beauty' of its 'oval head form'; the 'yellow, or Mongolic', by its 'prominent cheek bones', 'flat face', and 'narrow, slanting eyes'; and the 'negro, or Ethiopic' by its 'black' complexion, 'compressed skull', and 'squashed nose' while its 'projecting snout [museau] and thick lips put it visibly close to the apes'.^[31] Faithful to the genre, the prose of these passages is purportedly scientific and definitive but nonetheless shot through with ill-disguised racialist presumptions. Yet the argument at this point follows logically from the seemingly objective principles of the science of animal 'organization' outlined in the book's introduction.^[32] There, Cuvier had asserted a functional relationship between the extent of development of an animal's nervous system, the 'relative size of the brain', and its 'intelligence'. In conjunction, these factors determined the 'degree of animality', Cuvier's core criterion for the hierarchical grading of animals, implicit in his ranking of human races, and the ultimate source of the 'intrinsic causes' that allegedly stymied the 'progress of certain races' (Figlio 1976:24-5).

The novel signified of race as an hereditary natural category percolated more slowly into English, kept at bay by Evangelical philanthropic values — personified in the physician ethnologist James Cowles Prichard (1786-1848) — which retained ideological and moral ascendancy in the natural history of man in Britain until the mid-nineteenth century (Stocking 1973). The term race occurs relatively seldom in the first edition of Prichard's Researches into the Physical History of Man(kind) (1813) and more often in the second (1826), but in both is used in the loose eighteenth-century sense. The earlier text (1813:233-9) broadly differentiates 'savage' from 'civilized' but the logic of Prichard's speculative history of mankind made it 'probable' that the 'fairest races of white people in Europe' were ultimately descended from 'Negroes'. Yet by 1850, the language applied by British humanitarians to non-whites, particularly Negroes and Aboriginal Australians, was often as racialized as that used earlier in the century by uncompromisingly physicalist French naturalists (Hall 1991, 2002). Prichard had rapidly shelved his early thesis that 'the primitive stock of men were Negroes'. In the third edition of Researches (1836-47), he reinscribed without comment the scurrilous racial terminology and discriminations of his (often French) sources and in the process essentialized the characters of certain races in very negative terms: 'the Australians', for example, were 'squalid', 'miserable hordes', 'repulsive', 'disgusting', and 'ferocious'. His conventional distaste for stereotyped Negro anatomy, muted in 1813, was now palpable: the corollary of 'black' skin and 'crisp' or 'woolly' hair was 'features of a corresponding ugliness'.^[33]

In the earlier editions, the concept of races was sufficiently inconsequential to be left undefined but by the 1830s Prichard was prepared to naturalize 'those varieties in complexion, form, and habits, which distinguish from each other the several races of men'. The discursive dimensions of this shift are clear: the reification of human physical variation is manifest in the 'analogical' sections of the work which address 'the most strongly marked anatomical diversities of human races'; in contrast, qualifications, exceptions, and great diversity within races are ruling themes in the 'historical or ethnographical' sections which seek to delineate 'actual' changes in the 'physical characters' of nations or races.^[34] The greater salience of the term race in the final edition of Researches parallels Prichard's increased resort to taxonomy and comparative anatomy. He had made little attempt to classify human groupings in the first edition but contextually identified a shifting set of descriptors with respect to 'varieties of form and colour': seven 'varieties of colour' but an indeterminate five 'Races' or varieties in the overall 'structure of the parts in which the variety of colour subsists'. At this stage, Prichard privileged colour as a more 'general' and more 'permanent' discrimination than 'peculiarities of figure'. In the second edition, he distributed 'the human family' into unnamed, geographically defined 'departments' marked at once by 'important physical diversities' and conversely by 'remarkable approximations to the characters prevalent in other tribes'. In the third edition, racial taxonomy and anatomy loomed larger still. Exhaustive comparison of 'the principal varieties of form and structure which distinguish the inhabitants of different countries' saw him identify seven 'classes of nations' which differed 'strikingly from each other' and were 'separated' by 'strongly marked lines', especially 'peculiar forms of the skull'.^[35]

Yet Prichard always remained ambivalent about the racialization of human variation and at times tried to subvert the growing contemporary hegemony of the term race itself. His modern editor George Stocking, Jr. (1973:lxxi, lxxvi) argued that Prichard's nominalist distribution of human varieties into 'classes of nations' and his rejection of a higher level classification into a few racial types served to deny his classes 'the assurance of affinity [common descent] that alone would justify their designation as "races'''. There was surely an oppositional politics to Prichard's caveat, 'the various human races, if such exist'; to his consigning 'varieties' to 'the external and less essential parts' of 'the animal economy'; to his avowal of the 'common psychical nature' of mankind; and to his insistence that 'there are differences equally great, and even greater, between individuals and families of the same nation' as between different races.^[36] In the end, Prichard was prepared to normalize races and embrace 'diversification and differentiation' in order to turn them against a greater peril — the increasingly fashionable doctrine of 'an original diversity of races'. By defining 'races' as 'properly successions of individuals propagated from any given stock' but insisting that the term not imply 'that such a progeny or stock has always possessed a particular character', he explicitly refuted 'writers on anthropology' who took for granted that racial distinctions were 'primordial' and transmitted in 'unbroken' succession. Such a race 'would be a species in the strict meaning of the word' — a position Prichard consistently rejected, as did most of his British colleagues until after 1850.^[37]

That year, in The Races of Men, the Scottish anatomist Robert Knox (1791-1862) assailed the Prichardian creed and pronounced his notorious dictum: 'Race is everything: literature, science, art, in a word, civilization, depend on it'. The book was a collection of lectures delivered five years previously in provincial cities. At the time, Knox recalled, his views had been ignored by the London press; but since the outbreak of 'the war of race' in continental Europe — he meant the social and political upheavals of 1848 — the word race was in 'daily use' and his ideas had been appropriated by a 'leading journal'.^[38] Prichard (1850:147) also noted the sudden 'importance in public attention' assumed by the 'subject of human races, and their division' within Europe. By 1860, the primacy of race in the vocabulary of human difference in Britain was consistent, prosaic, and empirical. For instance, the word persistently infiltrated the 1865 English translation by Thomas Bendyshe (1827-1886) of the third edition of Blumenbach's De Generis Humani (1795): varietas and gens are sometimes 'race'; adjectival inflections of gens are usually 'racial'; and even stemma and stirps are frequently 'race'.^[39]

The biological notion of race emerged and gained potency in a complex historical conjuncture. Intellectually, the information about non-white people pouring into Europe from around the globe both enabled and seemed to require the demarcation of new scientific disciplines — notably biology and anthropology — which classed human beings as natural objects. Publicly, the escalation of European encounters with non-Europeans provoked fear and revulsion about supposed 'savages', not least in Oceania where several famous navigators met violent deaths at indigenous hands. Morally, the intensifying battle over slavery pushed abolitionists and defenders of slavery to adopt opposed scientific positions on the humanity or otherwise of Negroes. Imperially, a new phase of colonialism sought a philosophical basis for suppressing or governing indigenous people. Politically, revolution in France triggered dark imaginings about savages at home and abroad while its reactionary aftermath domesticated racial thinking by representing internal conflicts as the clash of a 'Gallic' third estate and a 'Germanic' nobility. By the mid-nineteenth century, the propensity to racialize local disputes had gripped much of Europe.^[40]