Predatory data Eugenics in big tech and our fight for an independent future

In Predatory Data, Anita Say Chan (media studies and information science, Univ. of Illinois) examines our information past and data-driven present, explicating a through line between the eugenics movement and contemporary data collection and operationalization. Just as eugenicists championed their data collection practices to justify their beliefs and practices as "evidence-based," today's tech giants employ data and algorithms that harm minority groups under the guise of technological impartiality and the promise of an optimized future. The first chapter opens with a description of a ledger containing photographs and details about the Chinese residents of Downieville, CA, recorded in the aftermath of the 1882 passage of the Chinese Exclusion Act. The archive serves to document the population and track its members' movements, demonstrating how surveillance and data collection methods were used to regulate groups that the powerful elite considered to be physically or mentally deficient, dangerous, or prone to immorality. In San Francisco, government-sponsored projects to map and systematically surveil Chinese residents across the city were commissioned as early as 1854 to document "the growing dangers to the physical, moral, and genetic health of the city" (p. 128). Data science and eugenics are inextricably linked. Sir Francis Galton, the founder of eugenics, is also credited with developing the statistical phenomena of correlation and regression. Data science methods were used to legitimize eugenics and led to the creation of the United States Eugenics Record Office and an explosion of literature espousing eugenic claims. As Chan explains, The Passing of the Great Race, published in 1916, popularized scientific racism and employed visual data to advance white genocide conspiracy theories, which were widely embraced by academics and the political elite. Despite the misinformation the book contained, Scribner praised it as one of the most successful books of the year, and its maps were used to successfully argue for restricted immigration. A century later, xenophobic rhetoric has returned to popularity, increasingly served to social media users through predictive algorithms. Throughout the volume, Chan demonstrates how eugenics thinking continues to be mirrored in contemporary technology. In the early 20th century, the development of intelligence assessments and IQ tests ultimately served to disenfranchise immigrants and minoritized populations, with intelligence and mental fitness considered a "politically neutral" way to categorize populations. As a particularly egregious example, Chan recounts how psychologist Henry H. Goddard began a study in 1913 that assessed immigrants arriving at Ellis Island, which included the question "What is Crisco?"--a product that had been introduced only two years earlier in the United States. Over 80 percent failed, and Goddard proudly shared that deportations of "mentally defective populations" rose substantially. A century later, data collection and assessments continue to derive human value. Chan provides the example of venture philanthropists funding initiatives that "filter deserving beneficiaries out from the rest" while using data and algorithmic software to dispense resources (p. 86). The first three chapters of the volume trace eugenics-based data practices, weaving threads between the past and the present with illustrative case studies. In chapters 4 through 6, Chan turns her attention to community-based alternatives for collecting and using data, which similarly have a long legacy. In chapter 4, she describes how the residents of Hull House engaged in community-based data practices focused on transforming social structures rather than profiling populations or generating profits. The authors of the Hull-House Maps and Papers (1895), which included feminist researchers and amateurs, advanced new methods in social scientific data collection, including the social survey. They used the collected data for local organizing and community action, advocating for social reform while prioritizing context and accountability. Chan contrasts this with the "hyper-detached, contextless mode of seeing from 'nowhere'" (p. 138) that attempts to understand the world through the scale and volume of big data. Chapters 5 and 6 shift the reader's attention to more recent examples of community-focused data practices, opening with Latin American feminist reproductive rights activists' grassroots campaigns that led to the development of new data resources, including Argentina's National Registry of Femicides. In collecting these data, activists emphasized that scale was not the goal; rather, their data collection allowed for "missing bodies" to be recovered, amending "official" datasets to account for the lived experiences that the tech elite has otherwise ignored. To combat big data's erasure and reductive practices, Chan suggests following Donna Haraway's approach of "seeing from below," grounding data practices in feminist and decolonial perspectives. In her conclusion, Chan articulately retraces the connection between eugenics and today's algorithmic systems with an added urgency. In November 2023, Elon Musk endorsed white genocide conspiracy theories on his social media site, X; in 2024, President Donald Trump publicly espoused racist rhetoric about immigrants who "poison the blood of our country" (p. 192). The United States' techno-elites, including Musk, Peter Thiel, and Marc Andreessen, claim that their technological advancements are the "ultimate engine of progress" (p. 194) and that regulation will only bring destruction. Chan urges the reader to push back, sharing a playbook for resisting datafication and prediction systems that recreate our biased past and reinforce majority voices. Predatory Data is thoroughly researched--with the author explicitly stating her intentional and inclusive citation practice--and makes compelling arguments. One criticism is that the narrative jumps back and forth while discussing case studies, which can make it challenging to follow historical timelines. Ultimately, it is a timely monograph that illuminates the history of technological advancement and advocates for "improbable futures" that resist big data's predicted outcomes. Summing Up: Recommended. Undergraduates through faculty; professionals. --Lindsey Skaggs, Illinois State University

In this troubling study, tech scholar Chan (Networking Peripheries) argues that the contemporary data economy, rather than being "inescapably evolutionary and progress driven" (as Big Tech would have it), is instead a direct product of the eugenics movement. The earliest population monitored via data collection, according to Chan, were Chinese residents of the California mining town of Downieville, who were surveilled from 1890 to 1930 because eugenicists in charge of public policy believed the community was "defined by hereditary vices." She tracks how data surveillance and eugenics became inextricably linked at elite institutions like Harvard, Northwestern, and Berkeley, where eugenics developed into an authoritative field of study that rationalized immigration bans and forced sterilizations of so-called "dysgenic" populations. Chan connects this academic nexus to the same policies that inspire concepts like "smart cities" today, showing how eugenics was all about designing "purified" lifestyles for elites by removing "anomalies." She also finds a connection between the eugenics movement's zeal for IQ tests as an indicator that public education was a useless government expenditure--since the tests supposedly proved that low intelligence was an inherited trait--and similar anti-education rhetoric espoused today by Silicon Valley billionaires like Peter Thiel and Elon Musk. It's an illuminating and unsettling depiction of Big Tech as deeply enmeshed in an ethically compromised brand of social science. (Jan.)