The other week, it was revealed that the consumer goods giant Unilever was going to start using facial recognition technology in job interviews. Developed by a U.S.-based company called Hirevue, the interview technology is designed to “analyze the language, tone and facial expressions of candidates when they are asked a set of identical job questions which they film on their mobile phone or laptop,” selecting the best applicants by “assessing their performances in the videos against about 25,000 pieces of facial and linguistic information compiled from previous interviews of those who have gone on to prove to be good at the job.”
Ostensibly, the technology is supposed to allow firms to interview more candidates in the initial stage rather than simply rely on CVs. In practice, as a number of critics on social media pointed out, Hirevue really just seems to have re-invented phrenology.
love that we’re always just circling back to phrenology https://t.co/TcIWBOL3O4— b-boy bouiebaisse (@jbouie) October 1, 2019
Techbros keep reinventing phrenology. They should, instead, reinvent a wheel to fuck themselves sideways https://t.co/U5N1ftPVWc— بوكيبلينكي (@pookleblinky) September 30, 2019
Phrenology 2: Electric Boogaloo https://t.co/gkBsQDkkep— the severed left hand of jon rosenberg ✋👻🎃 (@jonrosenberg) September 30, 2019
But then, as a number of these tweets also pointed out, phrenology — the practice of using the human skull to measure certain intellectual or moral qualities — really does seem to be enjoying a bit of a “moment.” Though notoriously discredited as a pseudoscience, people just can't help but keep on trying to to bring phrenology back — but why?
Phrenology first flourished towards the beginning of the 19th century, taking its cues from the work of the Swiss physiologist Franz Josef Gall. Operating on the assumption that the brain is the organ of the mind, and that the skull crystallizes around the brain during infancy, phrenologists set about measuring the bumps on people's skulls in order to assess their mental traits. You might be familiar with those numbered phrenological busts, which purport to show the regions of the skull corresponding to affection, metaphysics, religiosity, and so forth. Along with the closely related science of physiognomy — the study of character on the basis of facial features like jaw length and the height of the forehead — phrenology was at the forefront of scientific racism, as well as fantasies of innate criminality. Thus for instance, phrenologists claimed that African skulls had large regions corresponding to “veneration” and “cautiousness,” making them easily “tameable” — in short, natural slaves. Meanwhile, criminals were supposed to have large regions for “acquisitiveness.”
Disciples were enamored of phrenology’s promise to help them understand the mind not through introspection but through what looked like objective, testable scientific methods.
Phrenology was one of the first popular sciences. By the 1840s, there were almost 30 “phrenological societies” in London alone, and phrenology was the subject of stage shows and public lectures; as a young man Charles Darwin attended phrenological talks while studying in Edinburgh, then a center of phrenological research; Edgar Allen Poe was allegedly influenced by phrenology in his fiction. An 1841 book, Coombe's Popular Phrenology, advised bachelors that measuring young ladies’ skulls was a good way to find a wife: “One of the first requisites in a good wife,” the book claimed, “is to ascertain that she has a good head.”
Disciples were enamored of phrenology’s promise to help them understand the mind not through introspection but through what looked like objective, testable scientific methods. Of course, those very same methods quickly demonstrated that there was basically no correspondence between skull bumps and mental traits at all. By the end of the 1840s, phrenology was widely considered to be on the same level as astrology or palmistry.
Since then, the pseudoscience has lived on as the go-to comedic reference for old-timey nonsense — in one scene from The Simpsons episode “Mother Simpson,” for example, Mr. Burns is depicted extending his skull-measuring callipers over a phrenological bust, as he declares that Homer's mother has “the sloping brow and cranial bumpage of the career criminal,” and that Smithers (when he voices his skepticism) has “the brainpan of a stagecoach tilter.”
And yet phrenology — or something like it — has often seemed poised on the verge of a comeback. There were repeated revivals of interest in it from the late 1800s to the early 20th century; in the 1930s, Belgian colonial authorities in Rwanda used phrenology to justify the alleged superiority of the Tutsi over the Hutu.
Lately, the term “phrenology” most readily invokes the so-called “Intellectual Dark Web” — that loose network of hack popular scientists and race realists who consider themselves, bizarrely, to be in the business of “questioning orthodoxies” with the reactionary propaganda they promote to a blinkered audience of acolytes for profit. Indeed, one often sees phrenology being used almost as a cipher for the IDW’s views: a neat, evocative phrase one can use both to characterize — and caricature — their pompous performance of rationality and wink-wink “maybe-I'm-defending-it, maybe-I'm-just-defending-the-idea-that-someone-might-defend-it” scientific racism. Mention the IDW, and more often than not someone will make a joke about callipers.
In part, this is because Quillette, the online magazine prominently associated with the intellectual dark web, has literally defended phrenology. Earlier this year, two Quillette authors — one of whom was the sociologist Noah Carl, best known for winning and losing a prestigious research fellowship at the University of Cambridge for his work defending the validity of racial stereotypes — wrote a review of a book called Superior: The Return of Race Science in which they defended not only race realism, but also the classification, by race scientists, of human skulls. Meanwhile, in 2018 Quillette published an article entitled “Biosocial Criminology and the Lombrosian Paradox” in which the author, an undergraduate psychologist named Samuel Forster, praised the work of the 19th century Italian phrenologist-cum-criminologist Cesare Lombroso as containing “an important glint of truth,” due to his being “one of the first people to conceptualize criminality in biological terms.”
Of course, on a certain level there is room for dispute here, since neither of these examples defend phrenology wholesale. On the basis of the literally-stated, surface-level meaning alone, it is perhaps more accurate to suggest that these evidence Quillette's willingness to publish pieces which speak well of certain aspects of phrenology. But true meaning always goes deeper than mere surface. And while they might all personally, as individuals, deny it: In the most accurate sense possible, when you boil things down to the fundamentals of their logic, these people really are just doing phrenology.
A phrenological logic is lurking in any intellectual discipline that attempts, whether deliberately or otherwise, to depoliticize the human world.
Because the truth is that phrenology has never really been discredited. Granted people don’t typically go around measuring the bumps on people's skulls any more, although who knows how the Quillette lads are currently planning to find genetically optimal brides. But when phrenology was first debunked, its foundational assumptions did not simply go away. Rather, they were dispersed across other disciplines. Actual bumpology might have been a scientific non-starter, but phrenology was nevertheless deeply influential on the development of modern anthropology, criminology, and evolutionary biology — as well as eugenics. The phrenological account of the mind continued to be influential, throughout the Victorian era, in popular psychiatry. It is in these assumptions that the contemporary heritage of phrenology consists.
This heritage remains shockingly widespread. When I was an undergraduate studying philosophy, I remember learning things in philosophy of mind classes that may as well just have been phrenology without the bumps: in analytic philosophy of mind, it is common to identify the mind with the brain — an identification that phrenology pioneered. This identification has led some philosophers to view the mind as simply another part of an unbroken chain of physical causality, dating back to the Big Bang. The idea that parts of the brain have discrete, localized functions remains common in contemporary neuroscience, in which the equivalent of callipers is the only somewhat more accurate MRI scanner. It is common for news articles to report that neuroscientists have discovered, based on MRIs, “which part of the brain” is responsible for a certain mental activity — but neuroimaging studies have long suffered from small sample sizes, low statistical power, and a lack of replicability. Meanwhile, another favorite discipline of Quillette, evolutionary psychology, has sought to revive the sort of essentialist racial categorization that allowed phrenologists to make claims about “tameable” African skulls.
A phrenological logic is lurking in any intellectual discipline that attempts, whether deliberately or otherwise, to depoliticize the human world: to stitch the contingent injustices of our present form of life into the fabric of the universe; to turn them from something we might hope to change, to something that we can at best merely describe. This sort of depoliticization (which it is not at all paradoxical to suppose forms part of a right-wing political project, since it stands opposed to all that is progressive) will always be bad science: the human world, after all, is something we ourselves participate in, so it is only possible to accurately anticipate it as somehow plastic.
But that has not stopped its contemporary advocates from claiming — as Victorian phrenologists did, I suppose — that they have rationality on their side. Note the IDW’s hostility to “social constructionism,” the “blank slate” view of the mind, and what they call “grievance studies,” as most publicly expressed through last year's cack-handed “Sokal 2” hoax, in which a group of reactionary intellectuals attempted to disprove Foucault by writing a bunch of gender studies papers with their fingers crossed.
And the use of “phrenological” AI technology will inscribe injustice into that process too (even if this particular form of contemporary phrenology perhaps more closely resembles physiognomy). Clearly Hirevue has not learned the lesson from Amazon’s human-resources algorithm, which the company scrapped last year after realizing that, having been trained on the data of past hires, it was penalizing resumes that — through contextual clues — it could identify as having been submitted by women. Based on the Telegraph article about Hirevue, there is no reason to think that the use of facial recognition technology in interviews won’t be equally sexist (and racist, and classist, and probably homophobic).
The logic of phrenology works to encode the past into the present — to make old injustices impossible to overcome. Far from assuming that phrenology has been “long-since” discredited, we must do all we can to oppose it today.