In the nineteenth and early twentieth centuries, the cranial or cephalic index was a widely used calculation for racial classification. This particular measurement, which could be applied both to skulls and to the heads of living people, allowed the comparison of members of ancient and biblical civilizations to modern inhabitants of the same territories. Human remains excavated from archaeological sites across the Middle East prompted transregional interest in racial origins: who were the closest living descendants of (and therefore legitimate political-cultural heirs to) the Phoenicians, Indo-Aryans, and other celebrated pre-Islamic civilizations? Here, I analyze anthropometric studies in Lebanon and Iran in the first half of the twentieth century, showing how this preoccupation with ancient origins collided with intersectional and contested meanings of race and nation. In both countries, nationalist intellectuals and politicians used the cephalic index as a scientific tool, both to bolster the international legitimacy of their sovereignty claims and to promote particular narratives of national history. In Lebanon, anatomists and archaeologists argued over the racial classification of different Christian and Muslim sects as part of a highly politicized debate about Phoenician versus Arab ancestry. Meanwhile, Iranian scholars exhumed the remains of national heroes like Avicenna, measuring their skulls to prove their “Aryan” racial identity and reconstruct their physiognomy for sculptural monuments and portraits. Phoenicianism and Aryanism remain powerful racial-national discourses in contemporary Lebanon and Iran, where they continue to shape scientific interpretations of recent ancient DNA studies and forensic facial reconstructions of human remains.

Historians often assume that physical anthropology before 1945 relied on a simple typological, descriptive method to analyze skulls and classify races, which was only successfully challenged by populational genetics after World War II. This paper revisits and complicates this history by turning our attention to a fundamental attack on the typological tradition before 1945: by Karl Pearson, his introduction and development of statistical methods in anthropology, and the racial research his Biometric Laboratory produced between 1900-1938. The application of complex statistical formulae to the study of skulls and race unsettled long standing anthropological methods and theories. Whereas anthropologists had long studied the skull by itself, identifying racially-representative “types,” biometricians turned crania into means, standard deviations, and probable errors fit for statistical analysis. “Pearsonian anthropology” greatly expanded a geometric approach to craniometry which was already present in older anthropological practices. This paper argues that Pearson’s approach to craniometry set the stage for a durable relationship between biometry, geometry, and the skull that continues to live on in present-day biometric practices and technologies. At the same time, the paper discusses how anthropologists questioned Pearson’s approach and only partially adopted statistical methods, suggesting that the relationship between skulls and statistics was not sturdy but shaky and not fully trusted. The history of Pearson’s interventions in physical anthropology thus reveals deep divisions concerning the methods of classifying races well before 1945.

Since the publication of the paper “On lines and planes of closest fit to systems of points in space” by Karl Pearson in 1918 principal component analysis (PCA) has become an important statistical method in multiple research fields from the natural sciences (i.e. archeology, atmospheric sciences, psychology and physical anthropology) where big datasets of observations are collected. In studies of human facial difference, PCA works by producing statistical description of these differences that are later used to support common sense racial distinctions. In doing so, it establishes standards of normality for different races and, by comparing these normal faces, naturalizes racial difference. The present paper explores the influence of Pearson’s PCA in the theory and development of applications for face perception and recognition. Therefore, it focuses on three central cases in the development of facial recognition technologies (FRT): (1) ‘Eigenvector’ algorithms developed, among others, by Turk and Pentland (1991), (2) Valentine’s (1991) influential “Face Space” theory of face perception, and (3) Recent FRT such as DeepFace from Facebook (2014 to present). As shown by these cases, Pearson’s technique has deeply shaped contemporary FRT as PCA guides the way how computer scientists, forensic scientists and psychologists understand human facial difference as well as the perception of these differences. More generally, telling the story of PCA shows why racial categorization remains central in contemporary identification technologies and practices.

Forensic DNA Phenotyping (FDP) technologies aim at reconstructing the face of a suspect from samples of DNA left at a crime scene. Law enforcement agencies employ FDP-generated “DNA Snapshots” of suspects in their criminal investigations, and share these with the media. Scholars expressed skepticism towards the “science” behind FDP. Clinical researchers argued that the methods upon which FDP are based on are hardly replicable and do not meet the scientific standards for validity and reliability (Hallgrimsson et al., 2014). Anthropologists pointed out that FDP-generated portrays are racially biased, and warned against the ethical issues related to their rapid diffusion (M’charek, 2017). Meanwhile, novel approaches to reconstructing faces from DNA samples keep emerging. In February 2018, an international team of physical anthropologists and computer engineers published on Nature Genetics a novel methodology that aims at addressing past criticism (Claes et al., 2018). Central to this novel methodology is the use of phenotypic and genotypic data from genome-wide association studies (GWAS), and of machine-learning algorithms for the calculation of facial phenotypes. Drawing on ethnographic research and document analysis from early 2000s to present days, this paper narrates the emergence of data-driven methodologies for DNA-based facial reconstruction, and examines the rationales behind their adoption as the new standard for replicable research on DNA-based facial reconstruction. Most importantly, the paper highlights the persistence of arbitrary choices made by the researchers in defying facial phenotypes over the years and throughout different methods, including novel data-driven approaches.