With race theories, you can prove and disprove anything you want.

Max Weber footnote^＊

Race is arguably the single most troublesome and volatile category of the social sciences in the early twenty-first century—as Zora Neale Hurston put it, it is ‘like fire on the tongues of men’. Do you put it in scare quotes or not? Do you pair it with ethnicity to specify its scope or extend its reach? Do you use it as a substantive (as if it were a ‘thing’ out there in the world) or as an adjective (racial, racialized, racialist or the accusatory racist) attached to a perception, belief, action or institution? Is race premised on descent, phenotype, or skin tone? But what of such varied social properties as legal status, region, language, migration and religion that have also long served as vectors of racialization?footnote¹

What is the relationship between the social understanding of race and its putative genetic and neurological designation? Is race a self-propelled social force or does it derive from other causal powers (for instance, class or nationality)? A historical construct of utility in certain societies, such as imperial powers and their colonies, or an abstract construct of universal reach? Most urgently still, is it a ‘sin of the West’ (linked to chattel slavery), as loudly proclaimed by many race scholars and activists, or does it operate across civilizations? The principles guiding the conceptual autopsy of the ‘underclass’ as racialized category—elaborated in my latest book, The Invention of the ‘Underclass’—may help us gain some clarity and traction on these issues, allowing us to see how notions which have gained wide currency such as ‘structural racism’ and ‘systemic racism’ create more trouble than they resolve.footnote²

First principle: historicize. The trouble with ‘race’ in the West did not start in the 20th or 21st century. It is co-extensive with the life of the notion, which, from its coalescence in the mid-18th century, has constantly trafficked in the complicity between common sense and science. The naturalists of that era, who concocted the idea that humanity could be divided into biophysical categories (Linnaeus’s four races, white, black, yellow, red, corresponding to the four humours of the body and to the four continents of the earth, which survive under various guises to this day), which would later be decreed inherently unequal by Gobineau and his followers, were both codifying an extensive array of ordinary pre-modern perceptions and participating in a scientific revolution that was posing, for the first time, the question of how to fit together human diversity and hierarchy.footnote³

That originary confusion between common sense and scholarship has continued into the present and is embedded in the conventional coupling of ‘race and ethnicity’. Whenever social scientists take up this doxic duo, they endorse and amplify the defining symbolic effect of race, which is, precisely, the ideological belief that it is fundamentally different from ethnicity. The same applies to the pairing of ‘race and racism’: what is race if not a figment of the collective belief in its autonomous existence, that is, racism? So why the duplication? And the pluralization of the category, as in the seemingly self-evident and endlessly multiplying ‘racisms’, only compounds the trouble. This dubious commerce between common sense and science has gone on uninterrupted for three centuries, so that countless pre-sociological tenets about ‘race’ survive, indeed thrive, in contemporary social science. Inside too many racial constructivists, there is a racial essentialist struggling to get out.footnote⁴

Second principle: expand the geographic scope to decentre the discussion. This entails three moves. The first is to bring East and West together to escape continental parochialism. It is a curiously Eurocentric vision of history to believe that race as an essentialist principle of classification and stratification is a monopoly of Western nations and empires. The Japanese, to take but one example, did not wait for Commodore Perry’s arrival in 1853 to racialize the medieval caste of the Eta (meaning ‘filth abundant’) and the criminal class of the Hinin (‘non-human’) into the ‘invisible race’ of the Burakumin (‘hamlet people’), believed to be innately different, inferior and defiling, and to treat them as such across the centuries, including after their emancipation in 1871, even as no phenotypical property marked them out. And the Japanese penetration of Korea in the early 20th century was a colonial project steeped in racial thinking and action, even as the Japanese cloaked this capture in the language of amalgamation and assimilation grounded in the paradox of common ancestry.footnote⁵

The next move consists in linking the colonial and the metropolitan domains to track down the similarities and differences in the treatment of the subaltern of the interior (peasants, working class, ethnic minorities on grounds of region and religion) and the subaltern of the exterior (colonial subjects), as well as the two-way transfer of racialized representations, subjectivities and techniques of government between the imperial centre and its periphery. This is the task of a new generation of scholars promising to produce a colonial and postcolonial sociology whose work bears directly on theories of race (and group-making) in the global North of the contemporary era.footnote⁶