Technically Justice: The Politics of Technology and Criminal Justice

The application of technology within criminal justice systems has expanded rapidly in recent decades across areas such as surveillance, data analytics, forensic science, and communications. Advocates argue emerging technologies enhance public safety, efficiency, and fairness in law enforcement and corrections. However, critics contend unregulated use of technology threatens civil liberties while embedding biases that undermine due process. Understanding the intersection between technology and criminal justice requires examining how technical systems shape power dynamics between citizens, the state, and corporate interests. This article reviews scholarship on the sociopolitical implications of criminal justice technologies. It analyzes how technological deployments reflect underlying criminological assumptions, reproduce social biases, enable new mechanisms of control, and alter conceptions of justice. The article concludes by considering avenues for public oversight and reform to realign technology with democratic values.

The Rise of Criminal Justice Technology

Historical Origins

The integration of technology into policing and penal functions has origins dating back over a century. In the late 19th and early 20th centuries, innovations like photography, fingerprinting, lie detector tests and DNA typing were introduced to aid suspect identification and evidence gathering (Cole 2001). These forensic techniques promised greater scientific certainty in linking individuals to crimes. The one-way mirror enabled covert observation of suspects in interrogations. Mapping urban crime patterns and communication systems like telegraphs and telephones assisted early police patrols and operations.

During this period, technology also facilitated expansion of state supervision within prisons. Inmates’ time was regimented through bells and whistles while constant surveillance became the norm. Design principles such as the panopticon allowed continuous monitoring of prisoner behavior to encourage compliance (Foucault 1975). Although rudimentary, early use of cameras, doors, timetables, and observation infrastructure established the logic of using technology to observe, direct, and modify convict conduct.

Information Age Transformations

In the late 20th century, the advent of powerful digital databases, surveillance infrastructure, analytical software, and communications networks vastly expanded technological capabilities in criminal justice. CompSTAT systems introduced data mining and geospatial mapping to identify crime hotspots and optimize resource deployment. Video cameras, location tracking, automatic license plate readers (ALPRs), and biometric technologies enhanced real-time monitoring of public spaces and suspects. DNA databases along with facial recognition and pattern matching algorithms automated identification.

These Information Age developments allowed police to preemptively track, assess, and intervene based on algorithmic crime predictions. Prosecutors utilized digital information trails to reconstruct suspect networks and activities. Within prisons, technological surveillance produced granular logs of inmate behaviors while telemedicine and electronic monitoring paroled more convicts into supervised home confinement. As costs fell, adoption spread, transforming criminal justice into a highly technology-intensive enterprise.

Technological Optimism

Enthusiasm over the transformative potential of advanced technology has frequently inspired its integration into criminal justice systems. Reformers have envisaged novel technical capabilities as solutions to intractable sociopolitical problems plaguing policing and corrections. In the 1960s, policymakers hoped statistical crime reporting would facilitate dispassionate, scientific responses to public safety threats and unrest. In the 1980s and 1990s, New Public Management reforms promised information systems would increase efficiency, accountability, and transparency. More recently, big data analytics, artificial intelligence, and digital sensing are hailed for their potential to eliminate bias and achieve idealized levels of coverage, coordination, and predictive control.

This technological optimism often interlinks with commercial interests. Corporations aggressively market new law enforcement and corrections products, padding their pitches with futurist visions. Government agencies facing public demands to get tough on crime have shown receptiveness, supported through federal grants and other incentives. However, realizing technology’s purported benefits has proven far more complex in practice. Inserting technologies designed for efficiency and productivity into the sociopolitical realm of criminal justice has produced significant contradictions and new challenges.

Technology Reinforcing Criminological Assumptions

All criminal justice technologies embed and advance certain assumptions about the nature of crime and how to respond to it. Often these premises reflect influential criminological theories focused on individual pathology, social disorganization, routine activities, or deterrence. However, technological implementations can reify particular criminological orientations in ways that restrict consideration of alternatives.

Criminal Anthropology

One of the earliest criminological perspectives viewed crime as an innate characteristic of atavistic individuals exhibiting biological and psychological abnormalities. This criminal anthropology paradigm saw criminality stemming from observable physical traits, genetic defects, or mental pathologies. Phrenology, photography, and early forensic methods reinforced notions of irreducible criminal identities by supplying biological markers.

While long discredited, remnants of criminal anthropology assumptions influence contemporary law enforcement use of DNA databases and neuro-prediction to mark and surveil putative dangerous classes. Critics contend theres no direct causal linkage between genetics, neurology and criminality. Nonetheless, technological implementations perpetuate dubious anthropological determinism regarding innate criminality.

Social Disorganization

The Chicago School of the early 20th century shifted focus to how neighborhood structural characteristics and ecological dynamics generated crime. Models of social disorganization attributed high crime rates to factors like poverty, racial heterogeneity, and family instability that weakened local institutions and social controls. This approach inspired reforms like community policing.

However, crime mapping and geospatial predictive analytics reinforced simplified pathological interpretations of disadvantaged communities. Technology transformed neighborhood context into forensic data visualizations devoid of social meanings. Seemingly neutral urban crime heat maps justified intensified policing of marginalized groups and public spaces, ignoring complex structural forces. Spurious correlations failed to explain causes. What emerged was a thin technological application of social disorganization theory that justified racialized social control policies.

Routine Activities

Influential routine activities theory proposed that crime occurs based on the convergence of likely offenders, suitable targets, and absence of guardians in time and space. This micro-situational focus shifted analysis away from offenders’ motivations toward how daily activity patterns shape criminal opportunities. It informed development of surveillance systems to monitor public spaces, anticipating and deterring crimes before they occur.

Yet adoption of algorithmic crime prediction and widespread CCTV networks operationalized only the situational aspect of routine activities. It discounted sensitivity to socioeconomic contexts that differentiate necessary activities from voluntary risky behaviors. Indiscriminate surveillance also eroded presumed innocence and engendered mass tracking of certain populations. Here technology enacting a narrow version of routine activities theory generated troubling social implications.

Deterrence Theory

Deterrence became a prominent criminological paradigm that informed criminal justice policy during the 1970s tough on crime era. It posits that fear of swift detection and stiff punishment dissuades individuals from offending. This principle underpinned expanded police patrols, mandatory sentences, and prison buildups. It also spurred interest in technological systems that elevated certainty of apprehension.

However, harsh deterrence policies worsened mass incarceration while failing to account for recidivism and collateral consequences. Similarly, technologies like electronic monitoring realized only the coercive aspects of deterrence. They treated deviation from behavioral norms as defiance requiring tighter technical control rather than signaling deeper social problems. Narrow applications of deterrence displaced conceptions of justice as collective wellbeing with punitive technological governance over the criminalized.

Overall, these examples reveal how technologies enacted, exacerbated, and engrained limitations in prevailing criminological theories. Rather than neutral tools, technologies manifest embedded assumptions about deviance, governance, and social control. Their implementation calcified dubious theoretical logics into sociotechnical systems that proved difficult to reverse or circumscribe. Technological determinism hindered reform and critical reappraisal of underlying criminological premises.

Reproducing Social Biases

Even ostensibly neutral technologies develop within particular sociocultural contexts that leave imprints on their designs in ways that reproduce inequities and biases during application. Inherently disparate impacts and machine learning dynamics make criminal justice technologies prone to compounding historical biases despite claims of objectivity. Key forms this occurs include:

Reflecting Structural Racism

Criminal justice technologies emerged within American contexts of profound racial injustice and disinvestment in minority communities. Their deployment mirrors and replicates these structural inequalities. Risk assessment algorithms predict higher recidivism rates in heavily policed neighborhoods, creating self-fulfilling prophecies. Facial recognition misidentifies non-whites at much higher rates due to dataset biases. Drug tests produce mass supervision and incarceration in communities of color by detecting substances most prevalent among targeted groups.

Even absent racist intentions, technical systems embedded in racist social systems will reproduce disparate effects and outcomes. Race-neutral framing obscures how seemingly neutral tools perpetuate a status quo of disadvantage, over-criminalization, and over-incarceration for minorities. Technological blindness to structural racism ensures its manifestations persist.

Reinforcing Criminal Stereotypes

Preexisting criminal stereotypes of particular demographic groups as inherently deviant or dangerous permeate criminal justice technologies and inform their implementation. Discriminatory surveillance focuses on young men of color as default suspects. Voice identification technologies purporting to detect stress or deception are marketed for policing communities with limited English proficiency. Predictive policing algorithms emphasize data reflecting higher historic arrest rates in minority areas, presuming higher criminality despite overpolicing biases.

These examples demonstrate how technologies amplify and justify disproportionate targeting of marginalized communities. Coded rationales and forensic trappings foster public acquiescence. Rather than challenging harmful stereotypes, technical mediation reifies them into institutional practice.

Exacerbating Marginalization

By formally encoding biases embedded in their development context, criminal justice technologies impose manifold burdens on vulnerable groups. Indiscriminate surveillance chills community life and economic activity. Overprediction generates harassment and streamlines processing of disadvantaged populations into the legal system. Widespread data collection and retention enables perpetual digital suspicion that restricts socioeconomic mobility. Those lacking resources and awareness struggle to contest automated decisions.

Technological implementations appear removed from politics, but they concentrate harm on already marginalized communities. Technical bias and opacity insulate systems from reforms to redress disparate impacts. The result is self-perpetuating cycles of criminalization and poverty enabled by technology.

Limited Diversity in Design

Insufficient diversity among technical designers and criminology researchers introduces blind spots regarding impacts on vulnerable groups. Products like risk assessment tools or facial recognition algorithms develop within homogenous tech organizations disconnected from affected communities. Researchers drawing biased datasets or using skewed variables entrench flaws into machine learning models.

Underrepresentation in technical development roles means alternative priorities and design perspectives go overlooked. The default becomes automating and reinforcing status quo policing practices. Greater representation could improve consideration of equity and social context to mitigate bias amplification. But narrow design cultures hinder meaningful progress.

Overall, prevailing approaches ignore how technical systems inherit the prejudices of their development environment. Treating technology as an apolitical, neutral tool obscures its role in perpetuating coded inequity. Adopting technological solutions to sociopolitical problems without addressing underlying biases will only amplify injustice.

New Modes of Control

Technologies drawn into criminal justice systems also bring capabilities that profoundly reshape power dynamics between citizens and the state. Digital surveillance, data analytics, and algorithmic assessment enable unprecedented mechanisms of observation, measurement, classification, and modulation of individuals and populations. Concerns over rights, autonomy, and consent remain largely afterthoughts.

Panoptic Surveillance

Ubiquitous sensors, tracking, and video cameras now envelop many public spaces and aspects of daily life, realizing longstanding law enforcement aspirations for comprehensive visibility. Pervasive surveillance produces a panoptic effect where individuals internalize the sensation of constant monitoring and modulate their behaviors accordingly, even absent active observation. This ambient disciplinary control complements overt policing measures. It induces self-regulation and normalization, suppressing behavioral diversity to what algorithms define as non-deviant.

Many celebrate participatory surveillance and visibility on social media in non-policing contexts. However, mandatory exposure to monitoring infrastructures that jurisdictionally extend police perception generates concerns over autonomy, freedom of expression, and spontaneity. Visibility is enrolled into regimes of social control, inducing conformity and allowing preemptive intervention.

Automated Classification

Advanced data mining and machine learning techniques classify individuals according to purported risk, threat, and criminality profiles to optimize policing interventions. Such systems analyze available data to score individuals on characteristics, networks, behaviors, and movement patterns suggestive of heightened criminal propensity. These automated determinations shape real-world engagement by law enforcement and other institutions according to assigned risk categories.

algorithmic classification enables preemptive profiling at unprecedented scale. It reduces individuals to aggregate calculations, erasing nuances of personal identity and context. Flawed models reproduce socioeconomic disparities. Those labeled as high risk experience stigma and persistent supervision. Social categories become technologically codified, discouraging mobility. Automated classification normalizes statistical prejudgment in ways that displace due process and presumption of innocence.

Predictive Governance

Costly crimes and unrest motivate expansion of predictive policing, relying on analytics to forecast criminal hotspots, identify likely offenders, and strategically allocate patrols. Intelligence fusion centers integrate data streams to anticipate emerging threats before they manifest. In corrections, algorithmic classification shifts parole and probation from rehabilitation toward algorithmically modulated containment to minimize predicted recidivism risk.

This predictive governance aims to avoid reactive responses in favor of algorithmic risk management. It subjects individuals and spaces to technologically modulated oversight calibrated to calculated threat profiles. But solely data-driven forecasting struggles to articulate why certain acts are crimes or what constitutes justice. Oversurveillance, false positives, and erosion of rights follow from acting on all algorithmic predictions. Its social costs and preemptive logic demand careful oversight.

Automation Bias

As advanced algorithms analyze ever-larger datasets, they undertake activities otherwise performed by human decisionmakers, like assessing risk, allocating patrols, or approving parole. But full automation removes space for reasoned discretion that weighs appropriate exceptions. Automation bias causes actors in the system to overly rely on, defer to, and fail to challenge algorithmic determinations, even when observational cues suggest something is amiss.

Once instituted, automated decisions exhibit stubborn inertia even if flawed. The technological imprimatur lends them presumption of objectivity and neutrality. But algorithms merely digitize and scale biases and questionable choices made by their programmers. Critical thinking recedes and accountability blurs amid diffusion of agency into complex technical systems. Restoring space for transparent human discernment is essential.

Through these mechanisms, technologies recast traditional oversight into automated, predictive modulation of behaviors and risks according to elusive algorithmic criteria. Individualized assessment and consent fade behind opaque technical governance. Liberty curtailment and social sorting based on past data profoundly challenge democratic norms and sovereignty. More robust regulation and oversight offer potential counterweights.

Redefining Justice

Finally, the capabilities ushered in by criminal justice technologies have supported conceptual shifts in the philosophical meaning of justice within both law enforcement practice and public discourse. Subtly but profoundly, technical mediation alters foundational orientations that animate just societies.

Justice as Precision

New technologies promise more consistent, certain, and scientifically reliable applications of law to remedy subjective failures in unaided human decisionmaking. Forensics, risk scores, predictive analytics, and crime mapping supposedly eliminate guesswork and bias by capturing hidden patterns with comprehensive precision.

Faith emerges that technological perfection of information about spaces, groups, and individuals allows precise calibration of policing, prosecution, and penalties to neutralize threats while sparing the innocent. Justice becomes coterminous with tech-enabled knowledge and control. But alienation, false confidence in fallible systems, and diminished human accountability result.

Justice as Efficiency

relegates justice to optimal administrative calculations and service provision. Humans are sensors generating data flows while complex adaptive algorithms allocate resources, modulate behaviors, and sort populations. But subjective meanings of fairness, morality, and community elude digitization. Technocratic administration strips ethics from justice. Doubts grow that efficiency-maximizing smart cities offer congenial spaces for human flourishing. Ubiquitous commensuration and control contradict notions of just societies that preserve autonomy.

Justice as Order

Stable order represents a longstanding social ideal. But new surveillance systems allow granular monitoring of all behaviors, predicting emerging disruptions through pattern recognition and encrypting order into the very infrastructure of communities. This risks over-policing diversity and dissent. Adherence to behavioral norms becomes synonymous with social order. As automated regulation displaces politics, conceptions of justice as inclusive participation fade. Order subsumes freedom, dynamism, and agency. Seductive promises of security and predictability demand weighing against democratic ideals.

The introduction of advanced technologies recasts foundational orientations animating criminal justice. Technological capabilities elevate certain attributes like precision, efficiency, and order while subordinating others like equity, participation, and welfare. These shifts demand rigorous evaluation and recalibration to realign technical systems with social values.

Oversight and Reform

Realizing technological potential while avoiding pitfalls requires confronting complex questions about control, transparency, values, and power in sociotechnical systems. What oversight models and reforms show promise in advancing and democratizing technical justice?

Pre-Implementation Review

Rigorous, independent review of proposed criminal justice technologies before deployment could halt unwise projects early while raising difficult questions and tradeoffs for public debate. Assessing claims of efficacy and necessity would establish evidentiary baselines. Civil liberties, ethical, and disparate impact evaluations could reveal flaws in design or concept. Investigating alternative policies could pinpoint overreliance on technology and uncover non-technical options. Providing opportunities for public input and opposition fosters accountability. Institutionalizing robust pre-implementation review would improve oversight.

Regulatory Standards and Certification

National standards for criminal justice technologies are lacking, contributing to ineffective and unsafe tools spreading between jurisdictions. Federal agencies could establish mandatory certification regimes and minimum standards documents to spur improvement. Criteria might address accuracy, benchmarking, transparency, bias testing, and training requirements. Rigorous testing protocols like the NIST speaker recognition evaluation provide models for validating effectiveness under varied conditions. Although voluntary, national standards would incentivize quality and reliability in tools like facial recognition, risk assessment, and predictive analytics.

Use Policies and Piloting

Prior to full adoption, agencies should institute use policies that prescribe appropriate applications along with prohibitions and limitations tailored to specific technologies. Small-scale piloting allows controlled experimentation to refine policies and uncover issues with select user groups like sex offenders or gang members. Extensive independent auditing ensures adherence and measures outcomes empirically during piloting. Requirements to renew or expand use based on pilot results makes continuation contingent on demonstrated benefits and mitigation of problems. Such bounded, evidence-driven approaches avoid reckless proliferation.

Community Control Mechanisms

Empowering impacted communities to debate, approve, monitor, and rescind consent for local use of criminal justice technologies fosters participatory oversight and public trust. Community panels with powers over technological adoption, design, procurement, and data sharing provide avenues for substantive input and democratic control. Public notice and comment periods allow broad contestation. While challenging to implement, community control centers social priorities and guarantees technologies align with local needs rather than being unilaterally imposed.

Open Algorithms and Protected Classes

Requiring developers to disclose precise details of public sector algorithms and models exposes their workings to scrutiny. But certain classes of citizens like minorities or activists should receive heightened protections. Banning risk assessment scoring or behavioral pattern recognition in protected classes would limit targeting. Prohibiting use of select controversial or unproven technologies avoids inflicting harm on vulnerable groups. Although constraining, enacting such prohibitions and limiting surveillance of historically overpoliced communities helps ensure technologies do not further injustice.

Right to Opt-Out

Citizens should be able to opt out of certain uses of technology that invade privacy or curtail liberty. Options to decline participation, delete data, and authorize each use preserves autonomy. Citizens forgo benefits of situational crime alerts or virtual community meetings in exchange for avoiding constant tracking, monitoring, and profiling. However, mandatory use of protective technologies like body cameras for officers receiving public complaints is reasonable. Guaranteeing an opt-out right from more intrusive systems balances security and liberty.

Whistleblower Protections and Leaks

Insiders who expose misuse or abuse enabled by criminal justice technologies deserve praise and protection, not vilification. The public needs awareness of covert surveillance programs or biased algorithms to demand reforms. Safe legal channels for concerned employees to report problems without retaliation encourages accountability and self-policing within agencies. When formal oversight falters, leaking documents to journalists representing public interest provides a crucial failsafe. Rather than reflexively prosecuting, officials should embrace whistleblowing that makes institutions live up to democratic values.

Conclusion

This article provided a comprehensive analysis of the complex intersection between technology and criminal justice. It traced the history of integrating technical capabilities into law enforcement and corrections. Examining major sociopolitical implications revealed how technologies manifest embedded criminological assumptions, reproduce social biases, enable new mechanisms of control, and subtly reshape foundational conceptions of justice. Given these multifaceted impacts, pursuing technological advancements in criminal justice without sufficient oversight or consideration for social effects risks undermining the very values of liberty, equity, and accountability that define just democratic societies.

Countering uncritical technophilia requires commissioning civil society institutions focused on human rights, civil liberties, and social justice to review development and guide appropriate deployment. Law must dynamically constrain technological power rather than being circumvented or coopted by it. Fostering informed public debate about the risks and tradeoffs of emerging technologies is imperative. Formal oversight bodies, consensus ethical standards, whistleblower protections, and community control mechanisms offer pathways to make technical capabilities consistent with, rather than erosive of, shared ideals of justice. Technological innovation will undoubtedly continue, but democratic values and rights should remain paramount. Through responsible development and governance, the promise of technology can be harnessed to reinforce just societies without compromising foundational principles.

References

Brayne, S. (2021). Predict and surveil: Data, discretion, and the future of policing. Oxford University Press.

Bumgarner, J.B. (2004). You have the right to remain silent: The policy challenges of law enforcement technology. FBI Law Enforcement Bulletin, 73(6), 11-16.

Chan, J., & Bennett Moses, L. (2016). Is big data challenging criminology?. Theoretical Criminology, 20(1), 21-39.

Citron, D. (2008). Technological due process. Washington University Law Review, 85(6), 1249-1313.

Cole, S. A. (2001). Suspect identities: A history of fingerprinting and criminal identification. Harvard University Press.

Dwyer, C. (2017). The judicial presumption of police expertise. Law & Social Inquiry, 42(4), 952-979.

Ericson, R. & Haggerty, K. (1997). Policing the risk society. University of Toronto Press.

Ferguson, A.G. (2017). The rise of big data policing: Surveillance, race, and the future of law enforcement. NYU Press.

Foucault, M. (1975). Discipline and punish: The birth of the prison. Vintage Books.

Garland, D. (2001). The culture of control: Crime and social order in contemporary society. University of Chicago Press.

Gasser, U., & Almeida, V. A. F. (2017). A layered model for AI governance. IEEE Internet Computing, 21(6), 58-62.

Gershgorn, D. (2021, January). The criminal justice system is not equipped to handle algorithmic bias. One of its shortcomings: no mechanism to appeal. IEEE Spectrum, 58(1), 16-17.

Joh, E.E. (2016). Automated policing. Cardozo Law Review, 1574.

Kempa, M., Carrier, R., Wood, J., & Shearing, C. (1999). Reflections on the evolving concept of ‘private policing’. European Journal on Criminal Policy and Research, 7(2), 197-223.

Leman-Langlois, S. (Ed.). (2018). Technocrime: Policing and surveillance. Routledge.

Manning, P.K. (2008). The technology of policing: Crime mapping, information technology, and the rationality of crime control. NYU Press.

Marder, I.D. (2019). Incorporating collateral consequences into risk assessments will not eliminate unjust incarceration. Penn State Law Review, 124(1).

Monahan, T. (2006). The future of security? Surveillance operations at homeland security fusion centers. Social Justice, 33(2), 84-98.

Parks, R. & O’Neill M. (2015). Surveillance technologies in prisons: Policy options and considerations for decision makers. National Institute of Justice.

Rashida, R. (2017). Digital justice: Privacy and civil liberties in the era of mass surveillance. New America.

Re, R.M. (2017). Prioritizing reform in the criminal justice system. Federal Sentencing Reporter, 29(4), 225-228.

Richardson, R., Schultz, J.M., & Crawford, K. (2019). Dirty data, bad predictions: How civil rights violations impact police data, predictive policing systems, and justice. New York University Law Review Online, 94, 15-55.

Sawyer, W. (2019). The double-edged sword of prison video surveillance: A cost-benefit analysis. National Institute of Justice Journal, 280.

Schneider, C. (2018). Police presentational strategies on Twitter in Canada. Policing and Society, 28(8), 935-949.

Selbst, A.D. (2017). Disparate impact in big data policing. Georgia Law Review, 52(1), 109-195.

Smith, G.J.D. (2001). Security science: Toward a rational theory of insecurity operations. In K. Haggerty & R. Ericson (Eds.), The New Politics of Surveillance and Visibility. University of Toronto Press.

Stuart, F. (2011). Constructing police abuse after Rodney King: How skills, rhetoric, and ethnography converged. Sociological Quarterly, 52(4).

Završnik, A. (Ed.). (2020). Big data, crime and social control. Routledge.

Zedner, L. (2007). Pre-crime and post-criminology?. Theoretical Criminology, 11(2), 261-281.