13 The Urban Public School as Cybernetic Apparatus

A Universal Science?

Among the significant impacts of the Josiah Macy Foundation’s conferences on the American postwar sciences was the codification of “cybernetics”—so called by Norbert Weiner as science of “control and communication in the animal and machine”—a discourse that later prompted a wealth of media theory studying the shifts in human agency and cognition nascent in these once emerging communications theories. Iconic in this regard is Hayles’s account of the cybernetic deconstruction of the “liberal humanist subject” who, she writes, “shares with its predecessor an emphasis on cognition rather than embodiment” (Hayles 1999, 5). As Hayles shows, cybernetic ontologies developed between fields diverse as engineering, biology, linguistics, and mathematics in the twentieth century gestated informatic notions of learning as a goal-seeking behavior understood as separable from biological substrates or lived experience. However, a more complete historiography of the “posthuman” nascent in post–World War II sciences should include ecological and affect-based discourses of psychology and human development that challenge the supposed posthumanism of cybernetics. While feminist critiques of cybernetic abstractions call attention to the irreducibility of embodiment to information and code, there is a limit to theorizations that, in Weheliye’s terms, “yoke humanity to the limited possessive individualism of Man” (2014, 10) without accounting for how the subjugation of nonhuman others underwrites the scientific project of Man. Such oversight, I add, also limits the critique of racializing technologies of learning in the context of global neoliberal capitalism and data-driven education reforms.

This memo intervenes in these debates by framing the urban public school within the cybernetic context. Interwar communications research was directed in large part by the urgent need to reconcile liberal democratic doctrines of equality, freedom of thought, and scientific optimism with the rampant social inequalities of an increasingly complex capitalist society (see Geoghegan 2022). Here, informatic notions of society produced supposedly value-neutral and generalizable theoretical insight into the “human” condition. Geoghegan discusses how the anthropologists Margaret Mead and Gregory Bateson’s influential studies of the nuclear family as a cybernetic machine, for instance, were conducted in response to emergent crises around mental illness and family instability (Geoghegan 2022, 57). Contemporaneous treatments in sociobiology also harbored this therapeutic kernel in seeking to identify and treat psychological and social issues as measurable patterns (see Haraway 1981–82).

Certain informatic concepts shared a universal appeal, providing the context for the broad-based rethinking of communication in the fields of development psychology and pedagogy. By the late 1960s, ecological, holistic, experiential-based, and, importantly, cybernetic, concepts of learning would emerge in response to the structural disparities hindering low-income school districts. Following Haraway, these cybernetics of learning provided a “consistent understanding of the social without supposedly jarring philosophical categories like consciousness” (Haraway 1981–82, 252). For a fuller understanding of the application of cybernetics to pedagogy, one would include key thinkers such as psychologist Urie Brofenbrenner, whose widely cited Ecology of Human Development outlines principles implemented in his role in instituting the US Department of Health and Human Services’ Head Start program (Bronfenbrenner 1981). Bronfenbrenner’s views on human development as a process constituted through sustained interactions between an individual and its environment directly influenced the psychologist Edmund Wyatt Gordon, whose expansive work on the “achievement gap” scaffolds his dynamic, multiscaled, and environmental concept of learning to address how structural disparities shape education outcomes (Yazuv and Haynes 2017). I take up Haraway’s suggestion that the social mechanism of the cybernetic apparatus set the material context for nascent ideas of learning (1981–82, 252), ideas that assume new functional forms with twenty-first-century computing technologies and a “datalogical turn” in humanities and the social sciences. To elaborate, I focus specifically on how the Comer School Development Program’s (SDP’s) systems ecological pedagogy contributed to cybernetic perspectives on how race, gender, and class-based inequity should be understood and addressed in public schools (Comer 1988).

The SDP or “Comer model” was an outcome of research conducted by psychologist James P. Comer on underachieving schools in New Haven, Connecticut, in the later part of the 1960s. Resourced by a grant from the Ford Foundation, Yale-based researchers sought a method for analyzing the school district through the lens of public health (Comer School Development Program 2020). Working as part of an interdisciplinary team involving developmental and behavioral psychologists, educators, mental health professionals, and student parents, Comer used a systems theory approach to social behavior to bear on the structurally mediated injustices obstructing learning in the Martin Luther King and Baldwin elementary schools. The aim was to utilize the school system to ameliorate racialized issues such as disproportionately distributed drop-out rates and exposure to drug use (Comer and Edelman 1995).

While many critics sought to address the role of race and class within the emergent challenges faced by urban schooling in the 1960s, the SDP’s novel contribution to educational reform was to apply clinical concepts to schooling problems. In doing so, the project reconceptualized both the subjects and environments of learning as components in complex social systems—systems understood to be shot through with continuously shifting fluxes of power (Comer and Edelman 1995, 9). The SDP was widely considered to be a windfall success in improving school climate. This contributed to its institutionalization as a set of practices and technologies associated with socioemotional paradigms of learning popularized as part of contemporary social justice–oriented education reform (Elias et al. 1997). However, as I later discuss, these contemporary paradigms tend to erase the causal conclusions drawn by the SDP, linking them to emergent modes of global racialized management and data-driven education reform. Thus, while SDP’s holistic notions of the learner were ambivalently informatic—ambivalent in that they reimagined approaches to educational disparities by linking maligned learners to malignant environments—the project also acted as an important precursor to contemporary concerns with emotional intelligence animated by an analytic of racialized control.

Before detailing the SDP’s socio-cybernetic framework, it is useful to consider how this type of communications research functions as part of what Geoghegan describes as a “cybernetic apparatus.” The term encapsulates how philanthropic investments lent veracity to the application of new telecommunication instruments, electronic media, and computing to social problems, thus transcending disciplinary constraints. The conceptual blur between “instruments on one hand, a strategic convention of heterogeneous actors on the other—resulted in a new epistemic machinery” (Geoghegan 2022, 14). Perhaps the greatest consequence of rendering the public school as cybernetic apparatus is the production of a knowledge-power that produces a racializing view of emotion compatible with the demands of globalized informatic labor. Cybernetic learning is both an epistemology and technology that demands a critical treatment of embodiment and affect in data ontologies.

A Socio-cybernetic Learning Model

The SDP was organized around a hypothesis that “the application of social and behavioral science principles to every aspect of a school program will improve the climate of relationships among all involved and facilitate significant academic growth of students” (Comer and Edelman 1995, 60). SDP selected Baldwin and King elementary schools, two public schools grappling with issues of student attendance and behavior. Since New Haven’s systematizable scale lent itself to scientific and social analysis, it had often acted as model city for informing government welfare programs (Comer and Edelman 1995, 42, 60).

The informational analytic of the Comer model encompasses the factors contributing to New Haven’s disparities in education—among them, mass immigration of poor Black people from the South in the 1950s, a housing crisis, and racial discrimination in living wage job opportunities. “Schools,” Comer observed, “are complex networks of human interaction” (Comer and Edelman 1995, 221). While the problems rocking low-income school districts were seen as shaped by forces that transcended the school, the school became a key site of intervention for configuring a broader urban learning environment (1995, 54).

For Comer, challenging the poor outcomes of education and lack of opportunities for African Americans and other minority subjects required reenvisioning the aims and goals of public education in rapidly changing post–World War II American life. Learning, a key aspect of social reproduction, must become adaptable to rapid scientific and technological developments, urbanization, accelerated transportation, and the rapid visual communication of information (1995, 12–14). In an increasingly complex society, the school must become an environment responsive to shifting dynamics of power, authority, and societal organization. The general disaffection percolating in minoritarian urban schools would be transposed into a scientific, informatic, and cybernetic research problem linked to a new ethical and political program for educational development centered on the management of the environments of postwar society.

SPD rejected top-down discipline in favor of three main principles derived from human ecology and behavioral science: decision-making by consensus, no-fault problem-solving, and collaboration (Comer School Development Program 2020). Success was not quantified based on students’ capacities to master a discipline but viewed as a flexible aim to produce socially viable subjects in a changing society organized around relational, emotional, and adaptable notions of citizenship and labor. Hybrid and interdisciplinary researchers envisaged a holistic notion of the child as a learner responsive to the complex patterning of information across various levels of society. As a result, students’ behavioral problems were not to be treated as individual failures, but as socioenvironmental challenges that potentially manifest in the school, family, neighborhood, and home (Comer and Edelman 1995, 20). Behavioral issues thus could not be isolated instances. They demanded an approach to learning as a process emerging from the relationship between students, teachers, administrators, and researchers.

A key point was that the SDP would not reproduce hierarchies between scientific concepts and teachers engaged in day-to-day classroom activities. As Comer writes, aspects of the SDP “grew out of school practice needs and insights and not from the fertile but removed minds of academic theorists” (Comer and Edelman 1995, 209). The program would instead privilege process-based principles of adaptability capable of bringing together theory and practice in a reciprocally transformative relationship. “The perception of a school as a dynamic, continuously changing system which clinical, naturalistic, systems analysis and case study research can provide” would emphasize “principles to manage complex forces” rather than envision a top-down “all-powerful solution to school problems” (1995, 24). Indeed, a primary implication drawn from the Baldwin-King program’s success rate is that “no particular model, technology, method or person” is as significant as “a process that places highest priority on flexibility, accountability, shared expertise, open communication, trust and respect” (234). Mutual understanding requires a common language established through a practice in which participants express weaknesses and strengths to develop a unified curriculum. The process model thus understands the failure of most programs to improve schooling as lapses in communication between the various sub-systems comprising the expanded field of the learning environment. This informatic view enabled the SDP to readdress social inequities as lapses in communication, thus challenging an overdetermined view of racialized crisis within public schooling. It is not fully intuitive, then, that the cybernetic framework of the project could create the conditions for new metrics of racialized control, but the key point is that the resulting epistemic machinery rendered affect legible in the developing languages of complex systems theory and information.

Emotional Learners

Making the leap from the equity-oriented ambitions driving the turn to informatics in public education to emerging racializing techniques of education-based control does require that we reposition the turn to emotion and affect in learning within in the wider trajectory of systems theory and cybernetic governmentality. As part of a broader focus on public school safety and psychological climate, socioemotional learning (SEL) paradigms have emerged as a global neoliberal policy response that is formalized by technologies, practices, and metrics or “psychodata” methods. Empathy, relationship building, and responsible decision-making are key facets of SEL approaches to promoting social equity in learning environments (How Does SEL Support Educational Equity and Excellence?). Here, the child is no longer necessarily the subjective interpreter of information patterns as with the SDP. The learner is reconfigured as a component within a constantly expanding learning ecosystem comprised of schools, technologies, corporations, international policies, think tanks, and scientists. The Collaborative for Academic, Social, and Emotional Learning (CASEL) describes an emerging policy agenda promoting flexibility and “sustained interaction” between “researchers and practitioners,” where problem-driven approaches to optimizing pedagogical policy would involve the ongoing co-constitution of objectives gleaned through multiscalar data collection (Jagers 2020, 7–8). This recursive approach entails the production of an epistemic machinery that encompasses a broad funding coalition (i.e., the Bill and Melinda Gates Foundation, the Chan-Zuckerberg Initiative), SEL-positive think tanks (such as the conservative John Templeton Foundation and the Aspen Institute), and other state funding mechanisms (like DARPA), alongside an emerging set of technologies for measuring actionable insights into the social benefit and monetary value of SEL paradigms (Williamson 2019).

Together, these socioeconomic, commercial, and institutional processes and learning technologies establish a moving target for education reform focused on the cultivation of “desirable” learning traits determined by the Organization for Economic Cooperation and Development’s (OECD) Study on Social and Emotional Skills (SSES). In uncertain times when cognitive tasks have become easy to automate with AI, the OECD reports, findings gleaned from the SSES suggest that social and emotional skills “drive innovation and resilience in our economy and the cohesion between our communities in a world of increasing polarization” (OECD 2024, 24). If data shows that socioeconomic gaps manifest as disparities in socioemotional growth, the focus should be on optimizing school climate as a “virtuous circle” for fostering noncognitive attributes like resilience and preservation (80).

Under this logic, developing technologies for accumulating statistical data on emotional intelligence is essential to the production of “policy-relevant knowledge” that sees emotion, behavioral response, and other noncognitive processes as yielding measurable insight into students’ long-term success (Williamson 2019). The focus is not with disciplining the individual, but with generating value and insight from the measurement of differences that then circulate as data used to calculate future potentials. Accordingly, the OECD locates the effects of racism not at the individual level but at the level of populations that are “statistically organized and manipulated as groupings of features, characteristics, or parts” (Clough and Willse 2011, 52). For Clough and Willse, this is a type of technical governmentality deploying a “population racism” that calculates potential values as “racial probabilities,” which do not necessarily map onto existing groupings that are organized, for instance, by proximity or structural factors (2011, 52–53). While initially institutionalized in US urban reform contexts, the logic of resilience now shapes global education policy from OECD standards to World Bank–sponsored reforms in the Global South. The public school as cybernetic apparatus, then, links not only to the development of emotional intelligence as a nascent metric for emerging pedagogical practices, but also to an imperial and paternalistic dimension of racial governance that is not readily separable from a global digital economy characterized by uncertainty and incalculable risk.

The rise of such neoliberal models of control tracks with the paradoxical outcomes of the SDP: A “holistic” approach to educational reform must establish analytic models capable of denaturalizing the frameworks of domination shaped by and augmenting socioeconomic disparities. And yet, such approaches facilitate practices of racialized governmentality acting to transvalue populations both in terms of productive potentials and as the raw material from which ever more data may be extracted. It is perhaps perverse and paradoxical that conservative-led policy attacks in US public education target SEL as a vehicle for ideological indoctrination given the broader tendency of these reforms to depoliticize structural inequality. But these developments also reveal an unstable ideological terrain shaping data-driven pedagogical reform more generally. Once set in motion, the epistemic machinery of the cybernetic apparatus facilitates these constant slippages between eliminating racist attitudes and modulating the racial form itself as technical solution for educational reform.

It would not be a stretch to link SEL with what Clough and colleagues (2018) identify as a “datalogical turn” in social analysis, where the demands of computing technology set research agendas and methods even in fields as outwardly averse to technocratic approaches to knowledge as humanistic inquiry. “Big data,” they write, is “animating an emergent conception of sociality for which bodies, selves, contexts, or environments are being reconfigured in relationship to politics and economy” (2018, 148). What do the methods used to train algorithmic systems today inherit from the datalogical turn toward learning environments? The history of the SDP shows us that informatic and environmental concepts of the learner anticipated datalogical perspectives assumed by disciplines such as computational social sciences and the digital humanities, in which methods in the study of culture and society engage with technology as an objective means for legitimating theoretical inquiries. From this historical perspective, the machine learner is not particularly machinic. The synchronic drives to implement social and emotional learning at the cutting edge of education reform and the rapid integration of smart devices into educational settings (de Freitas et al. 2020) suggests, furthermore, that learning environments are key political sites for understanding the entanglements of noncognitive processes, race, and computing algorithms. There is much to be gleaned from the shifting position of learning environments in the cybernetic apparatus for refusing reductive battles between humanistic speculation and technocratic solutionism, theory and empirical experience.

References