In the nineteenth century, a frenzy of “measuromania” arose in the practices of imperial science, which was particularly enacted in colonized regions in Africa and beyond. Embarking on the idea of getting to know the world “scientifically”, colonial researchers used measuring and statistical data evaluation, and various measurement as well as description schemes, and extensively collected data from Indigenous people in order to determine “racial” and “sexual” differences. Besides collecting data on site, European colonizers appropriated – mostly without consent, often in clandestine ways or violently – thousands of skulls and other human body remains and transported them to European archives, museums and private collections, where most of them are still kept (Palladino and Worboys 1993; MacLeod 2000; Berner, Hoffmann and Lange 2011). These bodies have been stored in anthropological institutes, collections and museums; they have been presented at meetings and conferences of anthropological societies, in museum displays and exhibitions. Morphometric descriptions, measurements and visualizations derived from them have been discussed in scientific journals, lectures and popular science books.
Human remains and the measurement data collected from living bodies served in this way as comparative “race” collections throughout Europe. Huge amounts of data were extracted to form a comprehensive archive of dispersed analogue databases with source data for further anthropological processing. Hesitantly, European museums and scientific institutions have begun to investigate the “acquisition contexts” of their collections, to reflect their deep entanglement in colonial power and violence and to slowly engage in the restitution of human remains. A forerunner in Germany was the “Human Remains Project” at Charité – Berlin University of Medicine (Stoecker, Schnalke and Winkelmann 2013).
Some human remains from European collections have been returned since but thousands are yet waiting to be restituted. Although awareness has been rising about the ancestral remains in European archives, what I call the “metric-statistical” practices of knowing in this colonial discourse remain strangely untouched. On the contrary, legacies of colonial statistics
Within the digitalization paradigm in museums and in sciences, many researchers – sometimes using the catchphrase “virtual anthropology” – call for an automatization of anthropometric measurement, data processing and data sharing across the Internet (Weber 2014; Errickson and Thompson 2017; Hassett 2018). Wittwer-Backofen and Schläger (2013) even propose scanning colonial human remains before their restitution in order to keep the data for further scientific use.
Ethical debates about this “digital euphoria” started only recently (Squires, Errickson and Márquez-Grant 2020; Fforde et al. 2024). And in the current “AI euphoria” it might not be surprising that machine learning, neural networks and deep learning have been proposed for the evaluation of forensic and anthropometric data (Bell and Jantz 2002; Ousley 2016; Dong 2021). In these approaches, the biological concept of “race” is mostly either avoided or replaced, or synonymously used with terms like “ethnicity” or “ancestry”. An “absent presence of race” (M’charek, Schramm and Skinner 2014) is thus enacted with regard to the concept of “race” and lingers in still-circulating colonial data methods and data sets.
Since the relational entanglements of anthropometric modes of knowledge and colonialism have not yet been recognized fully (Chakrabarty 2000), this chapter proposes a critical engagement with the genealogies and legacies of measurement and data processing as colonial modes of knowledge-making and with their epistemic violence (Spivak [1988] 2010). With respect to un-disciplining African studies this chapter grapples with the “coloniality” (Maldonado-Torres 2007) of a specific knowledge mode that is genealogically bound to European colonization of the African continent. The chapter calls for an awareness of the lingering legacies in current anthropometric knowledge that still draws on African bodies.
The aim and endeavour of “race science” has been widely criticized and denounced since the twentieth century for its inherent racism. Statements
With respect to a critical discussion on the history of science, Stephen Jay Gould’s The Mismeasure of Man (1981) was seminal, carving out the racist prejudices and bias that underlay “race science”. The genealogy and aftermaths of this specific knowledge, its functionality for colonialism and racism, and its inscribed racist prejudices have been widely discussed in history of science (Stepan 1982; Stocking 1991; Kaupen-Haas and Saller 1999; Zimmermann 2001; Evans 2010; Weindling 2010; Müller-Wille 2014a, 2014b; Bauche 2017; Zack 2017). But these critical approaches have barely touched on the epistemology of the scientific methods, as if the problem were mainly a matter of wrongly (because prejudiced, racist) applied scientific methods – “biased science”.
Taking a closer, more detailed look at the material of 1900s “race anthropology”, I was struck by the appearance of “scientificity” in this discourse. Having a feminist and queer critical view on all essentialisms, and being trained in the scholarship of Derrida and Foucault (Hanke 1999), the performance of objectivity in this discourse seemed so persistent to me that unpacking its racist presumptions or detecting biased applied methods would not be enough. It was the very mode of knowing and its evidence effects of biological difference that unsettled and worried me, in addition to the often clearly articulated racism. Using numbers as its main medium, physical anthropology acts as neutral, objective and universalized in this mode of knowing. It operates as a “god trick of seeing everything from nowhere” (Haraway 1988: 581; Marcus 2020). Through measuring and data evaluation, physical anthropology claims to comprehensively grasp the world without being positioned or situated anywhere. This characterization holds not only for this discipline but haunts the dominance of metric-statistical approaches in contemporary sciences.
In the following, I add to discussions about the racist dimensions of “race science” by locating the problem of epistemic violence in the practice of metric-statistical procedures and by unpacking the epistemic infrastructures of the field of physical anthropology, with the intention of contributing to the decolonizing of research methodologies in African studies and beyond (Ndlovu-Gatsheni 2019). My contribution tries to uncover the specific positionality of this scientific mode of knowledge and its aftermath by situating
I take a closer look at the genealogies and predecessors of measurement in physical anthropology. Complementing colonial historiographies, my approach connects discourse analysis and deconstructive, epistemological and media theoretical methods with decolonizing approaches. This is to understand the epistemic violence and structural relationality to colonialism that I consider inherent in this scientific approach. I use Michel Foucault’s Archaeology of Knowledge (1972) methods of discourse analysis to unpack discursive regularities of European physical anthropology in given historical settings and to understand their continuities, stabilizations and disruptions.
In order to grasp how this mode of knowing functions, what kinds of objects are constituted in this way and what knowledge and power effects are thus been generated, I look at examples of anthropological material – utterances and statements that were deemed legitimate within this discourse. Conceptually, my analytic process aims to begin deconstructing colonial anthropology from within and to understand the ways in which measurement functions as part of the production of knowledge.
In my contribution I focus on the emerging discipline of physical anthropology, the foundations of which lie in the development and application of new metric-statistical methods and practices to scientifically identify and determine “races” in the European colonies. In order to discuss the coloniality of this knowledge of the global North, regarding its genealogy and its epistemology, my approach interweaves two lines of thought. I situate the analyzed practices of knowing within the broader transformation of European knowledge production and its relationality to colonialism. And I reflect epistemologically on the specific approaches of measurement and statistics, by focusing on the nitty-gritty practices of metric-statistical knowledge production in colonial anthropology at the end of the nineteenth century, to expose the epistemic violence of this approach (and not only its faulty/biased and prejudiced application). My contribution argues from an interdisciplinary background, unpacking the power structures and knowledge effects of this colonial form(ation), to invite a reflexive critique of this hegemonic, seemingly self-evident, mode of knowing.
1 Historicity of Objectivity
The comprehensive datafication of nineteenth-century anthropology and the debates about identification were part of the search for a scientific grounding of European “race” theory. Data collection and statistical evaluation act as
In their historical study, Lorraine Daston and Peter Galison describe a fundamental shift in European concepts of scientific objectivity in the nineteenth century, when the principle of “truth-to-nature” (Daston and Galison 2007: 55–113) is replaced by a “moralized epistemology” of “mechanical objectivity” (Daston and Galison 2007: 39), in which self-discipline, restraint and asceticism are demanded of the scientist. Daston and Galison explore the close connection of mechanical objectivity and the increasing use of the new technical medium of photography in various scientific atlases of anatomy, botany, astronomy and other fields, where the medium appears to be an intermediary that “preserves” scientific knowledge from the subjectivity of the researcher, be it their scientific and aesthetic judgement or their dogmatic thought constructs and anthropomorphisms.
This approach seemed to guarantee that the epistemic objects were now being recorded independently of the researcher – as the title, The Pencil of Nature, by one of the inventors of photography, William Fox Talbot, had envisaged (Talbot 1844–1846). Thus, the apparatus of technical media (photography, the sphygmomanometer and other curve plotting tools) promised to mechanize scientific research processes through simple, mechanical, non-interventionist recordings of nature. Photography was implemented in nineteenth-century colonial anthropology hand in hand with the fundamental process of datafication through new instruments, standardized measurement procedures and statistical data analysis. The new mode of objectivity in anthropology can therefore be situated in an even broader scientific approach in which “mechanical objectivity” is only one component.
The movement towards the mechanical objectivity of a technical apparatus and away from the researcher’s subjectivity is part of a larger context, namely the comprehensive mathematization of the sciences in Europe from the middle of the nineteenth century. In The Rise of Statistical Thinking, Theodore Porter (1986) describes a development in which measurement and metric-statistical data processing were established as the new guarantors of scientific objectivity. Therefore, I expand on Daston and Galison’s concept and propose to coin the new mode of knowing that gained traction in imperial science in the nineteenth century, “metric-statistical objectivity”.1 Technical media approaches,
The measurement and statistical analysis of data are – as I discuss in the example of anthropology – a very specific mode of knowing. Measuring can be characterized as a complex epistemological process, in which numbers are assigned to phenomena or things and thus transform them. In perceiving things or phenomena as measurable, they are thus already subjected to a specific mode of knowledge. Statistics, in turn, is a method for analyzing large amounts of data. Its claim to validity as an objective scientific method is closely related to its mode of posteriority: “normalism” researcher Jürgen Link points out that “types” are now no longer normatively presupposed but “normalistically” produced ex posto (Link 1996: 2004a–c). Instead of measuring only those phenomena that appear to be “characteristic”, “ideal” or “typical”, the metric-statistical approach claims to collect data from samples that are as large or random as possible. Furthermore, statistical data evaluations “aggregate” and produce new objects from the data of individuals (Desrosières [1993] 2002: 77–115) – in the case of anthropology, racialized categories.
Measurement and statistical analyses as a mode of knowledge are closely related to the emergence of the modern European nation-state, capitalism and industrialization, underpinned by new infrastructures of communication and transportation within Europe and the colonized regions (the steam engine, shipping routes, railways, telegraphy) and the standardization of scales for space and time (Desrosières [1993] 2002: 36–40). Additionally, new forms of governmentality and population control emerged at that time, together with administrative compilations of death, birth and marriage lists, accident and illness data for monitoring workers’ productivity, and a new concept of risk, calculable for the newly implemented insurance systems. The transfer of the Gaussian error curve to phenomena in society and nature, comprehensive data collections and their statistical evaluation, and methodological debates, became an integral part of new form(at)s of knowledge that arose in the course of the nineteenth century. In all areas of European societies, measurements were taken, data was collected and statistical analysis was done. In this way, the machinery of metric-statistical evidence spread around the world and fundamentally transformed scientific knowledge production.
Colonial anthropology is an important mediator in this process, because its approach of racialized othering not only entails knowledge and power effects that are constitutive for white supremacy and the legitimization of Europe’s colonial endeavours but also generates knowledge that can be further used in colonial infrastructure and administration. Moreover, colonized spaces and people serve as distributed experimental and laboratory fields, where new modalities of knowledge are practised and tested, discussed and further developed. Therefore, historicizing scientific objectivity (Daston and Galison 2007), statistical thinking (Porter 1986) and the politics of large numbers (Desrosières [1993] 2000) has to take into account the intrinsic coloniality in these modes of knowledge.
Ramon Amaro (2023) examines the relation of data science, statistics and inherent racism in European sciences with respect to Leibniz, in the seventeenth and eighteenth century, and draws a line of continuity from then up to contemporary machine learning and AI. With this in mind, I focus on a specific turn of knowledge and its research methodology in physical anthropology around 1900 – the practice of metric-statistical objectivity. Since this mode of knowing is characterized by a mathematization of knowing and statistical data processing, I propose to understand its use in colonial anthropology as a genealogical precursor to datafication, data repositories (databases), Big Data, surveillance, forensics and other digital data processing (Benjamin 2019; Bender et al. 2021; Sharma and Singh 2022). Not only were methods of data collection and evaluation widely tested and enhanced in colonial “laboratories”, but the epistemology of this very mode of knowing can be characterized as epistemically violent and colonial.
2 Physical Anthropology
Form(at)s of knowledges that operate in modes of what I call “metric-statistical objectivity” are a core and intrinsic part of European colonialism, not only in
Colonial anthropology has multiple effects. Besides legitimizing racism and colonialism, it is a field of experimentation and profiling for the professionalization of the discipline of anthropology, in which physical anthropological, ethnographic and prehistoric approaches were closely interwoven. Moreover, colonial anthropology plays a central role in the development, testing and implementation of the new modes of European science. It operates as a distributed methodological laboratory (Tilley 2011). Strong debates revolve around data recording, standardizations of data collection and statistical evaluation procedures, which colonial anthropology tests on concrete data, only rudimentarily accompanied by mathematical theory; debate also rages around imaging processes and descriptive recordings of bodies.
In what follows, I undertake a close reading of colonial anthropometric methods (Hanke 2007), with the aim of understanding their specific knowledge and power effects and to render visible the epistemic violence that lies in this metric-statistic modality. In order to unpack the “god trick” of these new scientific modes of knowing, which are still used today, my problematization draws on one of the discipline’s most important discussion platforms in the German language, the Archiv für Anthropologie (Archive of Anthropology), which was the scientific journal of the Deutsche Gesellschaft für Anthropologie, Ethnologie und Urgeschichte (German Society for Anthropology, Ethnology and Prehistory) from 1890 to 1914.
This journal is replete with data and visualizations extracted from bodies subjected to measurement. It can be characterized as an archive that provides anthropology with the working objects on which this emergent scientific discipline carries out its identification procedures. In these identifications, “racial” and “sexual” differentiations are relationally bound to each other and categories of “race” are often sexed, while categories of “sex” are often racialized. Anthropology’s objective to determine “races” (and “sexes”) is pursued by scanning bodies in terms of shapes (morphology), measures and proportions (anthropometry) and colours (“complexion”). This scanning not only involves a specific gaze by the anthropologist but also positions and displays the people to be identified by the anthropologist’s eye and measurement tools in literal
This approach is characterized by a fundamental epistemic violence. The anthropological gaze converts bodies into “rasterized”, fragmented objects, which are located in a mediatized realm of words (standardized catchwords/categories) and numbers/numerical indexes. From this data extracted from single individuals, physical anthropology constructs collective bodies – “races” and “sexes” – that are then assigned to these metric-statistical constructions. In this way, thousands of people are transferred as data into the order of European physical anthropological discourse. Both the individual body and the “population” body appear as fragmented, taken apart and put together in the form of aggregated objects/categories (Foucault [1967] 1991).
With my close reading, the practice of rasterization, or scanning, of bodies can be understood as a specific media technology that transforms bodies into data. Recent digital 2D and 3D scanning procedures of bodies in “virtual anthropology” and beyond can be situated genealogically in these processes of colonial body scans.
By taking individual bodies as a starting point for comprehensive data extraction, physical anthropology transforms and transfers them fundamentally into the medial realm of data – their living body and individuality are dehumanized and turned into numbers. It is this erasure of context that impedes the restitution of data and ancestral remains (Stoecker et al. 2013; Förster and Fründt 2017) and hinders the memory and mourning of the societies of origin.
It is this transformation of living bodies into data that I call out as epistemic violence. The epistemic violence and coloniality of these physical anthropological procedures therefore is not only a problem of violent appropriation and missing consent within colonial power regimes. Epistemologically, it can already be situated in the approach towards world and people practised here, in this scanning and converting of living bodies into numbers.
This transformation of bodies into data is by no means self-evident – bodies do not spill out data as if it was intrinsic to them. Since the flesh of human bodies belongs to a completely different realm from the realm of numbers, the very act of measuring it becomes recognizable as practising a specific epistemological chiasm and violence. Furthermore, the anthropological approach enacts a
3 Modes of Metric-Statistical Knowing: Tables, Frequency Schemes and Curves
The specific performativity of the number medium consists in its “practice of schematization, arrangement and formation [my translation]”.4 Numbers “number, sort, divide, create registers, lists and hierarchies, or fabricate chronologies [my translation]”, as the media philosopher, Dieter Mersch, argues5 (Mersch 2002: 210). The ordering structure of this medium produces discrete patterns with certain “step sizes”: 1, 2, 3, 4, or 1.0, 1.1., 1.2., etc. This organization also implies a decision logic – a logic which is only fully realized in the binary logic of the computer. Numbers, as a medium, correspond to the metric-statistic objectivity paradigm of the modern sciences because in their abstractness they act indifferent or neutral – they perform the very “god trick of science”. Their intrinsic logic opens a field of general comparability because they allow “universal interchangeability” (Adorno and Horkheimer [1944] 1989: 10)6 – thus offering infinite, arbitrary possibilities to relate to and correlate with other data. The correlation of data distributions is a highly productive statistical procedure used to generate new relationalities and new knowledge from multiple data collections. It is practised in machine learning, AI, large language models (LLM), public health and the medical sciences.
To explain the principles and the specific productivity of metric-statistic modes of knowledge, I zoom into the nitty-gritty practices of colonial anthropology and unpack its data processing. In the following it will become clear
In the Archiv für Anthropologie, the numerical data extracted from living and dead bodies by anthropologists, ethnologists, medical doctors and missionaries, among others, is stored and presented in vast numbers of tables. Traces of the measured bodies as living (or dead) individuals are generally eliminated in these data collections, although sometimes individual names appear in the tables, before they become mere numbers in the process of data evaluation. The mediality of the table makes data extracted from bodies available for analysis, which is why I understand these anthropological tables as pre-computer databases, as data archives before the time of the computer.
The specific performativity of the table format consists in the principle that it provides an ordering system. Data is arranged according to the individuals who were measured and by measured variables. Tables arrange the collected data along two coordinates, their ordering structures of rows and columns enable horizontal and vertical reading and comparisons. Rows and columns of cells can be flexibly rearranged according to various aspects and directions. They map the measurement data on a two-dimensional surface and can be understood as a media hybrid between images or graphics and numbers or text. Tables are readable in lines and columns, the digits are countable in discrete steps, tables plot data in a coordinate system – a grid stretched between the x- and y-axis – and the (data) space of a table is structured according to repetition and difference (Deleuze 1968). Moreover, in tables, the grid structure of measuring – that is, the scanning and rasterization of bodies – comes to the fore: the rasterization of bodies is itself (co-)visualized in tables. The specific mediality of the table does not hide but enters the picture; it displays, it shows itself (see Figure 11.1).
The mediality of the table
SOURCE: VOLZ (1899: 723, EXTRACT)
A study of tables and their use in anthropology shows how tables store data and keep it at hand but also how they work by enabling structured ways of reordering. The metric-statistical identification of “race” and “sex” is undertaken using the likewise accumulated, plotted numerical data. Statistics then “aggregate” new things from the data extracted from individuals. It does this by focusing on data clusters – that is, those areas of the measurement scale in which most of the measured values are located. From such data accumulations, “typical” or “normal” values are determined. Statistical processing is described as a procedure that “[die] Eigenschaften des [Daten-]Aggregats
For the anthropometric purpose of clustering data in order to determine “typical values”, the collected data is arranged in specific ways. The arrangement of data in a coordinate system forms the basis for any statistical evaluation, both in visual and metric terms. One example of measurement data might explain and demonstrate this better (see Figure 11.2).
The stacked curve pre-form
SOURCE: FALKENBURGER (1913: 92)
Figure 11.2. displays a table that functions according to the principle of the bar chart, a new table format that arises at the beginning of the twentieth century. Extracted data (which is first inserted into in a table similar to Figure 11.1) is rearranged according to the steps of the measurement scale, to show how many measurements were assigned the one value. The numbers that are noted under each measurement relate to the individuals who were assigned to this angle in the course of the measurement.
Differentiating between the two media – number and image (Mersch 2002) – this table can be characterized as follows. With regard to the ordering structure of the number medium, a field is rasterized according to two dimensions which determine the imaging structure for the visualization. Horizontally, at the top, the values of the discrete measurement scale of the
Describing this table as if it were a picture enables a reflection about the epistemic power of these kind of data visualizations. Numbers (and a few letters) are distributed on a plane. Through this rather impenetrable crowd of digits, a vertical and horizontal structure can be discerned. Schematically, a shape stands out: an accumulation, a cluster of digits. Out of the indeterminate field of numbers, a bulge with a strange internal structure emerges – an accumulation of data. The anthropological gaze takes from this example a cluster of measured values in a spectrum of angles between 86° and 92° and concludes that the “normal” “craniofacial angle” of the measured skull group (which has been ascribed to a “racial variety” beforehand) lies somewhere between these values. The perception of this accumulation of measured values is an effect of the purposeful reordering of the data and occurs from the interaction of the two media – numbers (order, readability) and image (density, shape). In anthropological and other statistical data processing, distributional “landscapes” with “hills” and “valleys” become visible via specific orderings of the data.
In the history of data visualizations, this example shows the epistemic transition from tables to graphs or curves. Although it is presented as a table, it is already well on its way to becoming a curve. Visually, such tables function like curves without a curve line: an imaginary contour line along the respective lowest entries in this table results in a curve that resembles an upside-down Gaussian curve (see Figure 11.3).
This process of accumulation of data through reordering is ideally embodied – or better, visualized – by the Gaussian bell curve of normal distribution: The “belly” of the curve shows the field of the “typical”/“normal”. It is by means of such Gaussian fields that “race” and “sex types” are identified in 1900s anthropology. But the Gaussian bell curve has been productive in many
Interestingly, where exactly the margins of these accumulations lie, where the deviations outside of the “typical”/“normal” begin, remain arbitrary in this form of data processing and visualization – as in its metric-statistical predecessors and variants. This can be characterized as one of the performative effects of such statistical data evaluations, which runs against the aim of “sorting things out” (Bowker and Star 1999). Here, a specific effect of rasterizing grids in the statistical knowledge mode becomes visible. Statistical grids do not produce clearly delineated or separable categories; rather, they fundamentally undermine category formation and a constitutive indeterminacy creeps in. My research shows that in the course of statistical data evaluations category boundaries are not determined qualitatively but quantitatively. There is no clear, qualitative and absolute demarcation between “race” and “sex types”, and thus the identification of “race” and “sex” is potentially undermined. This phenomenon can be observed throughout the discourse of physical anthropology.
These detailed analyses of anthropological procedures to determine “racial” differences show that statistics can be understood as an epistemic instrument of a specific gaze, which brings with it specific indeterminacies. In the accumulation of data, this approach spots the “normal”, the “typical”. Within the metric-mechanical objectivity of anthropology, the intermediary instrument of statistics is applied to see the “racial”. The statistically operating anthropological gaze is now directed to numbers and data arrangements, to the new imagery of statistics. Curves and accumulation schemes visualize metric data and identify fields of the “typical” in the accumulations – but at the same time they can undermine these fields from the margins and visualize their dispersions.7
In this discussion I have demonstrated the specific productivity and performativity of the knowledge mode that is practised in colonial anthropology. Measurement and statistics have to be characterized as complex procedures to process data and to bring forth racialized entities by ordering and reordering within a specific will to difference. The analysis above has shown that statistics
This is precisely what drives the multiplication of approaches to determine “races”, to develop methodologies further, to secure results that dissolve at same hand. At the same time, this demonstrates the complexity and difficulties in metric-statistically objective and scientific determinations of “races” that seemed so evident to colonial researchers. The use of statistics testifies to an attempt to secure the stability and boundaries of the categories scientifically, which are flexible in their procedure. The necessity of dealing with the deconstructive side-effects of the metric-statistical mode of knowing does not result in discarding the approach but in methodological discussions and negotiations of solutions within this new mode of knowing. This shows the persistence of this mode of scientific knowledge, which is enacted in the spirit of a new form of objectivity that gained traction in the nineteenth century.
Drawing on the epistemology and mediality of metric-statistic objectivity, a ubiquitous and taken-for-granted scientific practice becomes visible as a very specific and situated approach, which, starting from Europe, became established in modern natural science and empirical social science. This mode of knowledge is characterized by a specific detachment from the things examined. Although the approach claims to capture and record neutrally and objectively – without interfering – it undertakes a radical transformation of living bodies into numbers and from there into statistically constructed objects. It is in these transformations that this scientific approach enacts systemic epistemic violence, which is well hidden in its metric-statistical objectivity.
4 Conclusion: Colonial Legacies
Colonial anthropology’s modes of knowing linger to date in anthropological, medical, forensic and archaeological knowledge as well as in approaches to
In order to situate this mode of knowledge, I have combined science and technology studies, epistemology, media theory and decolonial studies to unpack the legacies and epistemic violence of what I call metric-statistical modes of knowledge. I have contextualized this form(at) of knowing within the broader European transformation by capitalism, industrialization and modernization and its close relationality to colonialism. Metric-statistical knowledge – as I have argued – is an intrinsic part of modernity, as its “darker side” (Mignolo 2011). While colonial anthropology contributed to legitimizing colonialism through its metric-statistical, objective, evidence of racialized differences, it also served as an experimental laboratory for the testing and development of these methodologies, which are practised – even being developed further – today.
Supplementing research that has focused on the fraudulent and biased collection or evaluation of research data, I have located coloniality in the scientific epistemologies and procedures themselves and demonstrated their intrinsic epistemic violence. These analyses contribute to critical analytical tools for discussing the measurement and datafication of the body regardless of the disciplines involved.
Considering the colonial legacies of these metric-statistical modes raises serious questions with regard to emergent form(at)s in current knowledge production and future knowledges. Although metric-statistical knowledges are immensely productive, they are haunted by epistemic violence.
As coloniality is intrinsically interwoven with the genealogy of contemporary methodologies, and since colonial anthropometric data has left traces in archaeological, forensic, anthropometric and medical knowledges, a strategy of unknowing seems quite impossible. Against the backdrop of my analysis, I would propose three tactics towards decolonial reflexivity:
Unpack colonial legacies! More research is needed, that follows anthropometric data by asking: Which data from colonial contexts has made its way into which contemporary knowledges? How has this data been transformed and incorporated? Through which papers and handbooks
has it been circulated? Which knowledge and power effects has colonial data enacted in current knowledges? Restitute colonial data! While unknowing or deleting colonial data from contemporary knowledge archives is hardly possible, a call for the restitution of colonial data would draw attention to the problem of colonial data extraction. Discussions around the restitution of cultural heritage and appropriated ancestral remains have to be complemented by reflection on how to deal with data appropriated in colonial situations. Thus, Indigenous data sovereignty (Kukutai and Taylor 2016; Davies et al. 2019; Walter et al. 2021) is not just to be enacted with newly collected data, but communities of origin are to be included in discussions about how to deal with colonial data. How might a restitution be possible if this data still circulates decontextualized in databases?
Challenge measurement! When measurement, datafication and statistical data analysis are called upon as empiricist methods in African studies and beyond, we need to remember that we are enacting approaches that were conceived of and shaped in colonial situations. Acknowledging their situatedness in colonial genealogies might lead to a more critical and hesitant use of these seemingly neutral approaches. Taking seriously the epistemic violence inherent in the transformation of people and things into data calls also for pausing before producing digital scans of human remains and cultural objects appropriated in colonialism.
These proposals could be taken further in tracing the coloniality of data practices in other disciplinary fields. Big Data, machine learning, AI, digital humanities, LLM, chatbots and other statistically operating modes of knowing perform as current laboratories that design innovative methodologies of data processing and evaluation. How these experimental fields are genealogically interwoven with colonial laboratories of imperial science and how epistemic violence lingers in these fields are subjects for future research.
References
Adorno, Theodor W. and Max Horkheimer. [1944] 1989. Dialectic of Enlightenment, new ed. Translated by John Cumming. New York: Continuum.
Amaro, Ramon. 2023. The Black Technical Object. On Machine Learning and the Aspiration of Black Being. Cambridge, MA and London: MIT.
Bauche, Manuela. 2017. Medizin und Herrschaft. Malariabekämpfung in Kamerun, Ostafrika und Ostfriesland (1890–1919). Frankfurt am Main and New York: Campus.
Bell, Susanne and Richard Jantz. 2002. “Neural Network Classification of Skeletal Remains”. In Archaeological Informatics: Pushing The Envelope. CAA 2001. Computer Applications and Quantitative Methods in Archaeology. Proceedings of the 29th Conference, Gotland, April 2001 (BAR International Series 1016), edited by Göran Burenhult and Johan Arvidsson, 205–11. Oxford: Archaeopress.
Bender, Emily, Timnit Gebru, Angela McMillan-Major and Shmargaret Shmitchell. 2021. “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?” FAccT 21, March 3–10: 610–23, Virtual Event, Canada, https://doi.org/10.1145/3442188.3445922.
Benjamin, Ruha. 2019. Race After Technology: Abolitionist Tools for the New Jim Code. Cambridge and Medford, MA: Polity Press.
Berner, Margit, Anette Hoffmann and Britta Lange. 2011. Sensible Sammlungen. Anthropologische Objekte im Depot. Hamburg: Philo Fine Arts.
Bowker, Geoffrey C. and Susan Leigh Star. 1999. Sorting Things Out. Classification and Its Consequences. Cambridge, MA and London: MIT.
Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial Thought and Historical Difference, new ed. Princeton, NJ and Oxford: Princeton Press.
Czekanowski, Jan. 1907. “Untersuchungen über das Verhältnis der Kopfmaße zu den Schädelmaßen”. Archiv für Anthropologie 34 (N.F. 6): 42–89.
Daston, Lorraine and Peter Galison. 2007. Objectivity. Cambridge, MA and London: MIT.
Davies, Tim, Stephen B. Walker, Mor Rubinstein and Fernando Perini, eds. 2019. The State of Open Data: Histories and Horizons. Cape Town and Ottawa: African Minds and International Development Research Centre (IDRC).
Deleuze, Gilles. 1968. Différence et Répétition. Paris: PUF.
Desrosières, Alain. [1993] 2002. The Politics of Large Numbers. A History of Statistical Reasoning. Translated by Camille Naish. Cambridge, MA: Harvard University Press.
Dong, Yibo et al.2021. “A Deep Learning Model for Ancestry Estimation with Craniometric Measurements”. 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Houston. https:/doi.org/10.1109/BIBM52615.2021.9669742.
Errickson, David and Tim Thompson, eds. 2017. Human Remains: Another Dimension. The Application of Imaging to the Study of Human Remains. London, San Diego, CA, Cambridge: Academic Press.
Evans, Andrew D. 2010. Anthropology at War: World War I and the Science of Race in Germany. Chicago, IL: University of Chicago Press.
Falkenburger, Fritz. 1913. “Diagraphische Untersuchungen an normalen und deformierten Rassenschädeln”. Archiv für Anthropologie, 40 (N.F. 12): 81–95.
Fforde, Cressida, Alexandra Roginski, Alan Goodman, Paul Turnbull and Hilary Howes. 2024. “Craniometry and Indigenous Repatriation”. In Repatriation, Science and Identity, edited by Cressida Fforde, Hilary Howes, Gareth Knapman and Lyndon Ormond-Parker, 74–101. London and New York: Routledge. https:/doi.org/10.4324/9781003144953-11.
Fischer, Martin. 2019. “Jena Declaration. The Concept of Race is the Result of Racism, Not Its Prerequisite”. Jena: Friedrich Schiller University. https://www.uni-jena.de/190910-jenaererklaerung.
Förster, Larissa and Sarah Fründt, eds. 2017. Human Remains in Museums and Collections. A Critical Engagement with the “Recommendations for the Care of Humans Remains in Museums and Collections” of the German Museums Association Historisches Forum. Bd. 21. Berlin: Clio-online and Humboldt University, Berlin. https://edoc.hu-berlin.de/handle/18452/20147.
Foucault, Michel. [1967] 1991. “Leben machen und sterben lassen. Die Geburt des Rassismus”. In Bio-Macht. DISS-Texte 25, 27–50. 2nd ed. Duisburg: DISS.
Foucault, Michel. [1969] 1972. The Archaeology of Knowledge. New York: Pantheon Books.
Foucault, Michel. [1976] 1988. The Will to Knowledge. History of Sexuality 1. Harmondsworth: Penguin.
Galison, Peter. 1998. “Judgment Against Objectivity”. In Picturing Science, Producing Art, edited by Caroline A. Jones and Peter Galison, 327–59. New York: Routledge.
Gould, Stephen Jay. 1981. The Mismeasure of Man. New York: W.W. Norton & Company.
Grupe, Gisela, Kerrin Christiansen, Inge Schröder and Ursula Wittwer-Backofen, eds. 2015. Anthropologie. Einführendes Lehrbuch. 2nd ed. Berlin/Heidelberg: Springer.
Hanke, Christine. 1999. “Kohärenz versus Ereignishaftigkeit? Ein Experiment im Spannungsfeld der foucaultschen Konzepte Diskurs und Aussage”. In Das Wuchern der Diskurse. Perspektiven der Diskursanalyse Foucaults, edited by Hannelore Bublitz, Andrea D. Bührmann, Christine Hanke and Andrea Seier, 109–18. Frankfurt am Main/New York: Campus.
Hanke, Christine. 2005. “Visualisierungen der physischen Anthropologie”. In Sichtbarkeit und Medium: Austausch, Verknüpfung und Differenz naturwissenschaftlicher und ästhetischer Bildstrategien, edited by Anja Zimmermann, 129–50. Hamburg: Hamburg University Press. https:/doi.org/10.15460/hup.209.1545.
Hanke, Christine. 2007. Zwischen Auflösung und Fixierung. Zur Konstitution von “Rasse” und “Geschlecht” in der Physischen Anthropologie um 1900. Bielefeld: Transcript. https://doi.org/10.14361/9783839406267-003.
Hanke, Christine. 2019. “Restituting Colonial Body Archives?” In Handle with Care: Post-Colonial Object Matters (Iwalewabooks zine #01), edited by Katharina Schramm and Nadine Siegert. Bayreuth: iwalewabooks.
Haraway, Donna. 1988. “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective”. Feminist Studies 14 (3) (Autumn): 575–99. https://doi.org/10.2307/3178066.
Hassett, Brenna R., ed. 2014. “Digital Bioarchaeology: New Dimensions, New Methods, New Ethics”. Archaeologies, Journal of the World Archaeological Congress (Special Issue) 14 (2). https://link.springer.com/journal/11759/volumes-and-issues/14-2.
Hefner, Joseph T. 2018. “The Macromorphoscopic Databank”. American Journal of Physical Anthropology 166: 994–1004. https:/doi.org/10.1002/ajpa.23492.
Howells, William White. 1996. “Howells’ Craniometric Data on the Internet”. American Journal of Physical Anthropology 101 (4): 441–42. https://doi.org/10.1002/ajpa.1331010302.
Kallenberger, Lauren and Varsha Pilbrow. 2012. “Using CRANID to Test the Population Affinity of Known Crania”. Journal of Anatomy 221 (5): 459–64. https:/doi.org/10.1111/j.1469-7580.2012.01558.x.
Kaupen-Haas, Heidrun and Christian Saller, ed. 1999. Wissenschaftlicher Rassismus. Analysen einer Kontinuität in den Human– und Naturwissenschaften. Frankfurt am Main and New York: Campus.
Kukutai, Tahu and John Taylor, eds. 2016. Indigenous Data Sovereignty. Toward An Agenda. Canberra: ANU Press. https://www.jstor.org/stable/j.ctt1q1crgf.
Lévi-Strauss, Claude. 1952. Race and History. Paris: UNESCO.
Link, Jürgen. 1996. Versuch über den Normalismus. Wie Normalität produziert wird. Opladen: Westfälisches Dampfboot.
Link, Jürgen. 2004a. “From the ‘Power of the Norm’ to ‘Flexible Normalism’: Considerations after Foucault”. Translated by Mirko M. Hall. Cultural Critique 57 (Spring): 14–32. https://doi.org/10.1353/cul.2004.0008.
Link, Jürgen. 2004b: “On the Contribution of Normalism to Modernity and Postmodernity”. Translated by Mirko M. Hall. Cultural Critique 57 (Spring): 33–46. https://doi.org/10.1353/cul.2004.0009.
Link, Jürgen. 2004c. “The Normalistic Subject and Its Curves: On the Symbolic Visualization of Orienteering Data”. Translated by Mirko M. Hall. Cultural Critique 57 (Spring): 47–67. https://doi.org/10.1353/cul.2004.0010.
MacLeod, Roy, ed. 2000. Osiris 15: Nature and Empire: Science and the Colonial Enterprise. Chicago, IL: University of Chicago Press.
Maldonado-Torres, Nelson. 2007. “On the Coloniality of Being: Contributions to the Development of a Concept”. Cultural Studies 21 (2–3) (March/May): 240–70. https://doi.org/10.1080/09502380601162548.
Manthey, Laura and Richard L. Jantz. 2020. “Fordisc: Anthropological Software for Estimation of Ancestry, Sex, Time Period and Stature”. In Statistics and Probability in Forensic Anthropology, edited by Zuzana Obertová, Alistair Stewart and Cristina Cattaneo, 275–87. London, San Diego, CA, Cambridge, MA: Academic Press. https://doi.org/10.1016/B978-0-12-815764-0.00022-8.
Marcus, Miranda. 2020. “The God Trick In Data And Design: How We’re Replicating Colonial Ideas In Tech”. Towards Data Science, 17.11.2020, https://towardsdatascience.com/the-god-trick-in-data-and-design-4ec71e19811.
Mbembe, Achille. [2016] 2019. Necropolitics. Translated by Steven Corcoran. Durham, NC and London: Duke University Press.
M’charek, Amade, Katharina Schramm and David Skinner. 2014. “Technologies of Belonging: The Absent Presence of Race in Europe”. Science, Technology, & Human Values 39 (4): 459–67. https://doi.org/10.1177/0162243914531149.
Mersch, Dieter. 2002. “Wort, Bild, Ton, Zahl. Eine Einleitung in die Medienphilosophie”. Kunst und Medium. Zwei Vorlesungen, 131–253. Kiel: Kunsthochschule Kiel.
Mignolo, Walter. 2011. The Darker Side of Western Modernity: Global Futures, Decolonial Options. Durham, NC and London: Duke University Press.
Montagu, M.F. Ashley. 1942. Man’s Most Dangerous Myth: The Fallacy of Race. New York: Columbia University Press.
Mudimbe, V.Y. 1988. The Invention of Africa: Gnosis, Philosophy and the Order of Knowledge. Bloomington, IN and London: Indiana University Press and James Currey.
Müller-Wille, Staffan. 2014a. “Race and History: Comments from an Epistemological Point of View”. Science, Technology, & Human Values, 39 (4): 597–606. https://doi.org/10.1177/0162243913517759.
Müller-Wille, Staffan. 2014b. “Reproducing Difference: Race and Heredity from a Longue Durée Perspective”. In Race, Gender and Reproduction: Philosophy and the Early Life Sciences in Context, edited by Susanne Lettow, 217–35. New York: SUNY Press.
Ndlovu-Gatsheni, Sabelo J. 2013. “The Entrapment of Africa Within the Global Colonial Matrices of Power. Eurocentrism, Coloniality and Deimperialization in the Twenty-first Century”. Journal of Developing Societies 29 (4): 331–53. https://doi.org/10.1177/0169796X13503195.
Ndlovu-Gatsheni, Sabelo J. 2019. “Provisional Notes on Decolonizing Research Methodology and Undoing Its Dirty History”. Journal of Developing Societies 35 (4): 481–92. https://doi.org/10.1177/0169796X19880417.
Ousley, Stephen D. 2016. “Forensic classification and biodistance in the 21st Century: The rise of learning machines”. In Biological Distance Analysis: Forensic and Bioarchaeological Perspectives, edited by Marin A. Pilloud and Joseph T. Hefner, 197–212. New York: Elsevier. https:/doi.org/10.1016/B978-0-12-801966-5.00010-X.
Ousley, Stephen, Richard Jantz and Donna Freid. 2009. “Understanding Race and Human Variation: Why Forensic Anthropologists are Good at Identifying Race”. American Journal of Physical Anthropology 139 (1): 68–76. https://doi.org/10.1002/ajpa.21006.
Palladino, Paolo and Michael Worboys. 1993. “Science and Imperialism”. Isis 84 (1) (March): 91–102. https://doi.org/10.1086/356375.
Porter, Theodore M. 1986. The Rise of Statistical Thinking. 1820–1900. Princeton, NJ: Princeton University Press.
Preedy, Victor R., ed. 2012. Handbook of Anthropometry. Physical Measures of Human Form in Health and Disease. Volume 1, Parts 1–6. New York, Dordrecht, Heidelberg, London: Springer.
Seth, Suman. 2009. “Putting Knowledge In Its Place: Science, Colonialism and the Postcolonial”. Postcolonial Studies 12 (4): 373–88. https://doi.org/10.1080/13688790903350633.
Sharma, Sarah and Rianka Singh. 2022. “Scanning for Black Data: A Conversation with Nasma Ahmed and Ladan Siad”. In Re-Understanding Media: Feminist Extensions of Marshall McLuhan, edited by Sarah Sharma and Rianka Singh, 179–90. Durham, NC: Duke University Press. https://doi.org/10.1515/9781478022497-013.
Slice, Dennis E. 2005. Modern Morphometrics in Physical Anthropology. New York: Kluwer Academic/Plenum Publishers.
Spivak, Gayatri Chakravorty. [1988] 2010. “Can the Subaltern Speak? From Marxism and the Interpretation of History”. In Can the Subaltern Speak? Reflections on the History of an Idea, edited by Rosalind C. Morris, 237–91. New York: Columbia University Press.
Squires, Kirsty, David Errickson and Nicholas Márquez-Grant, eds. 2020. Ethical Approaches to Human Remains. A Global Challenge in Bioarchaeology and Forensic Anthropology. Cham: Springer Nature.
Star, Susan Leigh and James Griesemer. 1989. “Institutional Ecology, ‘Translations’ and Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39”. Social Studies of Science 19 (3) (August): 387–420. https://doi.org/10.1177/030631289019003001.
Stepan, Nancy. 1982. The Idea of Race in Science: Great Britain, 1800–1960. Hamden, CT: Archon Books.
Stephan, Carl N. and Ellie K. Simpson. 2008. “Facial Soft Tissue Depths in Craniofacial Identification (Part I+II): An Analytical Review of the Published Adult Data”. Journal of Forensic Sciences 53 (6): 1257–79. https://doi.org/10.1111/j.1556-4029.2008.00853.x.
Stocking, George W. 1991. Colonial Situations: Essays on the Contextualization of Ethnographic Knowledge. Madison, WI: University of Wisconsin Press.
Stoecker, Holger, Thomas Schnalke and Andreas Winkelmann, eds. 2013. Sammeln, Erforschen, Zurückgeben? Menschliche Gebeine aus der Kolonialzeit in akademischen und musealen Sammlungen. Berlin: Ch. Links.
Talbot, William Henry Fox. 1844–46. The Pencil of Nature. London: Longman, Brown, Green and Longmans.
Tilley, Helen. 2011. Africa as a Living Laboratory. Empire, Development, and the Problem of Scientific Knowledge, 1870–1950. Chicago, IL: Chicago University Press.
UNESCO. 1950. The Race Question. Paris: UNESCO.
Volz, Wilhelm. 1899. “Zur somatischen Anthropologie der Battaker in Nord-Sumatra”. Archiv für Anthropologie 26: 717–32.
Walter, Maggie, Tahu Kukutai, Stephanie Russo Carroll and Desi Rodriguez-Lonebear, eds. 2021. Indigenous Data Sovereignty and Policy. London and New York: Routledge. https://doi.org/10.4324/9780429273957.
Weber, Gerhard W. 2014. “Another link Between Archaeology and Anthropology: Virtual Anthropology”. Digital Applications in Archaeology and Cultural Heritage 1: 3–11. https://doi.org/10.1016/j.daach.2013.04.001.
Weindling, Paul. 2010. “German Eugenics and the Wider World: Beyond the Racial State”. In The Oxford Handbook of the History of Eugenics, edited by Alison Bashford and Philippa Levine, 315–31. Oxford: Oxford University Press.
Wittwer-Backofen, Ursula and Stefan Schläger. 2013. “Anthropologische Zugänge zur Provenienzklärung menschlicher Skelettüberreste in Sammlungen”. In Sammeln, Erforschen, Zurückgeben? Menschliche Gebeine aus der Kolonialzeit in akademischen und musealen Sammlungen, edited by Holger Stoecker, Thomas Schnalke and Andreas Winkelmann, 224–43, Berlin: Ch. Links.
Zack, Naomie, ed. 2017. The Oxford Handbook of Philosophy and Race. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780190236953.001.0001.
Zimmermann, Andrew. 2001. Anthropology and Antihumanism in Imperial Germany. Chicago, IL: University of Chicago Press.
When Peter Galison (1998), with respect to anthropological debates about measurement vs aesthetics, uses physical anthropology, of all things, as a model for illustrating judgement rather than mechanical objectivity, he underestimates what I call the metric-statistical objectivity of 19th-century anthropology and the internal anthropological conflicts between perspectives that are valued as “objective” and those that are considered “judgemental” (Hanke 2007: 137–149).
For a comparable gaze and positioning, not of humans, but of Karakul sheep, see Schaumann in chapter 12 of this volume.
In reference to the “will to knowledge” formulated by Foucault ([1976] 1988), following Nietzsche.
“Praktik der Schematisierung, der Anordnung und Formierung” (Mersch 2002: 210).
“Zahlen […] nummerieren, sortieren, teilen ein, erstellen Register, Listen und Hierarchien oder verfertigen Chronologien” (Mersch 2002: 210).
In the European Enlightenment, the use of numbers was detached from numerological or mystical contexts and paradigmatically came to be the medium for the “disenchantment” of nature (Adorno and Horkheimer ([1944] 1989: 3), empowering the human mind “to hold sway over a disenchanted nature” via modern sciences (ibid.: 4).