The Fifty-Year War

The War That Was Declared

On June 17, 1971, President Richard Nixon declared a “war on drugs” in remarks to the White House press corps, identifying drug abuse as “public enemy number one in the United States” and calling for a comprehensive federal response that would include law enforcement, interdiction, treatment, and prevention components.¹ The declaration was the operational beginning of the contemporary American drug-policy framework. The framework’s principal subsequent developments — the Drug Enforcement Administration’s establishment in 1973, the Comprehensive Drug Abuse Prevention and Control Act of 1970 that had immediately preceded the declaration, the federal scheduling system that classified substances by perceived addiction and medical-use potential, the progressive expansion of mandatory minimum sentences across the 1980s — were the operational mechanisms through which the declared war was conducted across the subsequent decades.²

The 1986 Anti-Drug Abuse Act, signed by President Reagan on October 27 of that year in response to the death of college basketball star Len Bias from cocaine overdose, established the most consequential single piece of contemporary drug-policy legislation.³ The Act created mandatory minimum sentences for federal drug offenses, established the structural disparity between crack and powder cocaine sentencing (the so-called 100:1 ratio, in which possession of five grams of crack cocaine triggered the same five-year mandatory minimum as possession of five hundred grams of powder cocaine), and substantially expanded federal funding for drug enforcement. The Act’s mandatory minimums and the 100:1 disparity produced the operational mechanism by which federal drug enforcement, across the subsequent two decades, disproportionately targeted Black and Hispanic Americans whose crack-cocaine usage and trafficking patterns differed from the powder-cocaine usage and trafficking patterns of the predominantly white population that the powder-side enforcement affected. The disparate-impact pattern was documented across two decades of empirical research before the partial correction of the Fair Sentencing Act of 2010, which reduced the disparity from 100:1 to 18:1 — a substantial improvement that nonetheless preserved the principle of differential sentencing for the same underlying drug.⁴

The 1994 Violent Crime Control and Law Enforcement Act, signed by President Clinton on September 13 of that year and known thereafter as the Crime Bill, was the most consequential single piece of contemporary American criminal-justice legislation across both the drug-enforcement and broader-criminal-justice axes.⁵ The Act provided substantial federal funding for state and local prison construction (the Truth-in-Sentencing Incentive Grants), established new federal criminal offenses with substantial sentences, expanded the federal death penalty, banned certain semiautomatic firearms (the Federal Assault Weapons Ban that lapsed in 2004), and provided funding for community policing through the Community Oriented Policing Services (COPS) program. The Act was passed with substantial bipartisan support in both chambers; the Democratic-coalition support was substantially provided by Senator Joe Biden, who had been the Senate Judiciary Committee chairman during the legislative process. The Act’s effects across the post-1994 period have been the subject of substantial subsequent debate, with the empirical literature substantially supporting the conclusion that the Act contributed materially to the substantial expansion of American incarceration across the late 1990s and 2000s while contributing only modestly to the simultaneous decline in crime rates that has been the principal political defense of the Act.⁶

The cumulative effect of the 1971-2010 sequence of statutes was the construction of the carceral system whose contemporary outcomes the empirical literature has documented at length. The American incarcerated population grew from approximately three hundred thousand in 1970 to approximately two million three hundred thousand at the 2008 peak, before declining modestly to approximately one million eight hundred thousand by 2024.⁷ The American incarceration rate per one hundred thousand population is, as of 2024, approximately five hundred sixty-five — the highest of any country in the world for which reliable data is available, substantially higher than the rate of any comparable developed country (the contemporary German rate is approximately seventy per hundred thousand; the Canadian rate is approximately one hundred; the United Kingdom rate is approximately one hundred forty).⁸ The American carceral system, in operational terms, holds approximately twenty percent of the world’s incarcerated population while the United States holds approximately four percent of the world’s total population. The disparity is the principal empirical feature of the American criminal-justice landscape that the comparative-criminology literature has identified as anomalous.

The drug-policy component of the carceral system is the component the war-on-drugs declaration of 1971 most directly produced. Approximately one in five American incarcerated persons are held for drug offenses; in federal prisons, the share is approximately forty-five percent.⁹ The substantial majority of drug-offense incarcerations are for non-violent offenses — possession, low-level distribution, or non-violent trafficking. The empirical record on the relationship between the drug-enforcement effort and actual drug consumption is, as the next section describes, substantially unfavorable to the framework’s continued operation.

The Empirical Record

The principal empirical question about the American drug-policy framework is whether the framework has, across fifty-four years of operation, achieved its stated objective of reducing drug consumption and the harms drug consumption produces. The empirical literature on this question is, by the standards of the policy-evaluation research, unusually unambiguous. The framework has not achieved the objective.

American drug consumption, across the post-1971 period, has not measurably declined relative to the consumption rates that would have prevailed in the absence of the war-on-drugs framework. The National Survey on Drug Use and Health, the principal federal data series on American drug consumption, has tracked roughly stable rates of past-month illicit drug use across the entire period, with the specific drug-of-choice pattern shifting (the 1980s crack epidemic, the 2000s prescription opioid epidemic, the post-2013 fentanyl epidemic) but the overall consumption rate remaining substantially constant.¹⁰ The drug-overdose death rate, which is the principal measure of the most serious drug-consumption harm, has risen substantially across the period: from approximately five thousand annual overdose deaths in 1970 to approximately one hundred seven thousand in 2023, a more-than-twentyfold increase.¹¹ The increase has been driven principally by the post-2013 fentanyl epidemic, which has produced the steepest single increase in overdose mortality in American history. The framework that has been operating for the entire period, at substantial fiscal cost and with substantial criminal-justice consequences, has not prevented the overdose-mortality outcome that is the principal harm the framework was constructed to prevent.

The cost of the framework, across the same period, has been substantial. The cumulative federal expenditure on drug enforcement, treatment, prevention, and incarceration across the 1971-2024 period has been estimated at approximately one trillion dollars in inflation-adjusted terms.¹² The state and local expenditure has been comparable. The combined expenditure has produced the carceral system the previous section described, with its associated humanitarian and economic costs (approximately seventy-six billion dollars per year in direct incarceration costs as of 2024, plus the indirect costs of removed labor force participation, family disruption, post-incarceration employment effects, and broader community impacts that the empirical literature has estimated at multiples of the direct costs).¹³ The combined direct and indirect cost of the framework, on the most extensive contemporary cost-benefit analyses, substantially exceeds any plausible measure of the framework’s benefits.

The disparate-impact dimension of the framework’s operation is the empirical feature that the policy debate has been most contested over. Black Americans constitute approximately thirteen percent of the American population and approximately thirty-three percent of the American incarcerated population; Hispanic Americans constitute approximately nineteen percent of the population and approximately twenty-three percent of the incarcerated population.¹⁴ The disparate-impact pattern is not principally explained by differential drug-use rates: the National Survey on Drug Use and Health has consistently found that white and Black Americans use illicit drugs at approximately the same rates, with white Americans using slightly higher rates of some specific drugs.¹⁵ The disparate-impact pattern is principally explained by differential enforcement: Black and Hispanic Americans have, across the post-1971 period, been arrested, prosecuted, convicted, and incarcerated for drug offenses at rates substantially higher than their proportional share of the drug-using population. The empirical record of this differential enforcement is the subject of extensive academic literature; the literature is substantially uncontested in its principal finding.¹⁶

The Comparative Record

The comparative record on drug-policy frameworks is the second empirical input that the contemporary policy debate has been slow to integrate. The comparator countries that have, across the post-1971 period, adopted policy frameworks substantially different from the American war-on-drugs model have produced empirical outcomes substantially different from the American outcomes.

The Portugal case is the most thoroughly documented and the most directly relevant.¹⁷ Portugal, in 2001, decriminalized personal possession of all illicit drugs (cannabis, cocaine, heroin, methamphetamine, MDMA, all others) — not legalizing supply, but removing criminal penalties for personal-use quantities and replacing the criminal penalties with administrative penalties (fines, mandatory treatment referrals, community-service requirements) administered by Commissions for the Dissuasion of Drug Addiction. The legal change was paired with substantial expansion of treatment capacity, harm-reduction infrastructure (needle exchanges, supervised consumption facilities), and reintegration support for drug users. The reform was, at the time of enactment, controversial and was predicted by opponents to produce substantial increases in drug consumption.

The empirical record across two decades of operation has substantially supported the reform. Portuguese drug consumption rates have remained stable or declined modestly across the post-2001 period; Portuguese overdose mortality has declined substantially; Portuguese HIV transmission rates among drug users have declined substantially; Portuguese incarceration for drug offenses has declined substantially.¹⁸ The Portuguese model has been studied extensively by the European Monitoring Centre for Drugs and Drug Addiction, by the Cato Institute (whose 2009 evaluation, conducted by Glenn Greenwald, was the principal English-language introduction of the Portuguese case to the American policy debate), and by subsequent academic researchers across multiple disciplines.¹⁹ The comparative case has not been incorporated into the operational American policy framework despite the substantial empirical record supporting it.

The Swiss heroin-assisted treatment program, established across the 1990s in response to the open-air drug scene at Zurich’s Platzspitz Park, is the second comparator case. The program provides medically-supervised heroin to long-term opioid-dependent patients for whom conventional treatment has been ineffective; the empirical record of the program has demonstrated substantial reductions in overdose mortality, in HIV and hepatitis transmission, in property-crime rates among program participants, and in broader public-health and public-order metrics in the affected communities.²⁰ The Swiss model has been adopted, with variations, in several European countries (Germany, the Netherlands, Denmark, the United Kingdom in selected pilots). It has not been adopted in the United States despite the substantial empirical record.

The Oregon Measure 110 case is the most directly relevant American case and is the case that has most informed the contemporary policy debate. Measure 110, passed by Oregon voters in November 2020 with approximately fifty-eight percent support, decriminalized personal possession of all illicit drugs in Oregon (replacing criminal penalties with one-hundred-dollar civil fines that could be waived through treatment-screening contact) and committed substantial cannabis-tax revenue to expand drug-treatment capacity.²¹ The implementation across 2021-2024 was, by the principal evaluations, substantially defective: the treatment-capacity expansion was slower than the legislation had anticipated; the public-health response to the simultaneous fentanyl epidemic was inadequate; the visible homelessness and drug-use patterns in Portland and other Oregon cities, which were principally driven by structural factors that the decriminalization legislation did not address, were politically attributed to the decriminalization. The Oregon legislature, in March 2024, partially reversed Measure 110, restoring criminal penalties for personal-use possession (as a misdemeanor) while preserving expanded treatment funding.²² The Oregon case has been frequently cited as evidence against decriminalization; the evidence the case actually provides is more nuanced — that decriminalization paired with inadequate treatment-capacity buildout does not produce the Portuguese outcomes — but the political effect of the case has been to substantially constrain the contemporary American decriminalization debate.

The marijuana decriminalization and legalization wave that has, across the 2012-2024 period, produced legal recreational marijuana in approximately two dozen American states is the third American comparator case. The empirical record of the marijuana legalization across the implementing states has not produced the catastrophic outcomes that legalization opponents had predicted.²³ The structural change has produced substantial state-level tax revenue, has reduced marijuana-related arrests substantially, has not measurably increased youth marijuana consumption, and has had modest measurable effects on traffic safety. The marijuana legalization has been, in operational terms, the most successful contemporary American drug-policy reform; its successful implementation has been the principal political asset for the broader decriminalization debate the present essay describes.

The Policing Component

The drug-policy framework operates against the broader American policing apparatus, whose own contemporary structural features have been the subject of substantial debate across the post-2014 period (the period defined principally by the Ferguson, Missouri, killing of Michael Brown and the subsequent police-reform movement). The structural questions about American policing — use-of-force standards, accountability mechanisms, qualified immunity, federal oversight, training requirements — interact with the drug-policy framework at the operational level: the substantial majority of American police-citizen encounters that escalate to use-of-force outcomes involve drug enforcement or drug-adjacent activities, and the federal funding mechanisms that have funded the militarization of American policing across the post-1980s period have been principally drug-enforcement-oriented.²⁴

The qualified immunity doctrine is the principal contemporary federal-jurisprudential feature that the policing-reform debate has focused on. Qualified immunity, as developed by the Supreme Court across the post-1967 jurisprudence (principally Pierson v. Ray in 1967 and Harlow v. Fitzgerald in 1982), provides government officials performing discretionary functions with immunity from civil liability unless their conduct violates “clearly established” statutory or constitutional rights of which a reasonable person would have known.²⁵ The doctrine, as applied by the federal courts across the post-1982 period, has produced the operational outcome that police officers who violate constitutional rights through use of force are substantially insulated from civil liability for the violations, because the “clearly established” standard requires nearly identical prior precedent that the federal courts have, in practice, rarely found to exist. The doctrine’s structural effect is the substantial elimination of civil-liability accountability for police use-of-force violations, in the absence of comparable criminal-prosecution accountability that has historically been rare. The combined effect is the absence of meaningful legal accountability for the categories of police misconduct that the post-2014 reform movement has documented at scale.

Federal standards on policing have been advanced periodically across the post-2014 period and have not been enacted at the federal level. The George Floyd Justice in Policing Act, advanced in the 116th and 117th Congresses, would have established federal standards on use of force, body-worn cameras, racial-profiling prohibition, and qualified immunity reform; the Act passed the House in 2021 but did not advance in the Senate.²⁶ The structural pattern follows the broader pattern the previous essays have described: legislative reform attempts pass the House under Democratic-coalition majorities and fail in the Senate to overcome the filibuster.

The Two Parties’ Two Failures

The Democratic-coalition position on criminal justice and drug policy across the post-2010 period has been, in operational terms, the substantial reversal of the post-1994-Crime-Bill Democratic-coalition framework, with the contemporary Democratic-coalition position substantially aligned with the structural reforms the previous sections have described. The reversal has been incomplete and has been internally contested across the coalition; the operational outcome has been the gradual but substantial shift in Democratic-coalition policy away from the post-1994 carceral expansion toward the contemporary harm-reduction-and-treatment framework.

The Democratic-coalition’s specific failures on the structural reforms have followed the pattern the broader series has described. Federal sentencing reform has been advanced by Democratic-coalition legislative majorities (the First Step Act of 2018, passed with substantial bipartisan support; the broader sentencing-reform proposals of subsequent Congresses) but has not produced the structural reform that the empirical literature identifies as required.²⁷ Federal marijuana legalization with comprehensive expungement, despite substantial Democratic-coalition support and majority-public-opinion endorsement, has not advanced in the Senate. Federal policing reform, including qualified immunity reform, has been advanced and has not been enacted. The pattern is the pattern the broader series has described: substantive reform legislation passes the House and dies in the Senate, against the structural barriers the campaign-finance and electoral lock-ins have institutionalized.

The Republican-coalition position has been structurally different and has been internally inconsistent across the period. The Republican-coalition’s traditional law-and-order framing has, across the post-2010 period, been substantially complicated by the state-level Republican experimentation with criminal-justice reform — the Texas Right on Crime initiative, the Georgia and Mississippi sentencing reforms, the broader red-state reform pattern that has, in operational terms, often advanced sentencing-reform legislation more aggressively than blue states.²⁸ The Republican-coalition’s federal position, however, has remained substantially aligned with the traditional law-and-order framework, opposing federal marijuana legalization, opposing qualified immunity reform, opposing structural sentencing reform that goes beyond the First Step Act framework, and opposing federal policing standards. The internal inconsistency between the state-level reform pattern and the federal-level traditional position has been a feature of the Republican-coalition framework across the post-2010 period.

The convergent failure of the two parties has been the absence of any serious legislative framework for the structural reforms — marijuana legalization with expungement, decriminalization paired with treatment expansion, federal policing standards, qualified immunity reform — that the empirical and comparative records validate. Both coalitions’ competitive positions are protected by aspects of the existing framework; both coalitions have, in operational terms, declined to enact the structural reforms that would substantially modify the framework.

The Cascade That Closes

The criminal-justice and drug-policy framework described in the previous sections operates, considered in isolation, as the structural mechanism through which the substantial American carceral apparatus continues to operate against the empirical record demonstrating its inadequacy. Considered against the broader cascades the previous essays have described, the framework becomes the structural element that compounds the household-extraction mechanisms the broader series has documented.

The household whose income is being compressed by the AI cascade and whose carrying costs are being escalated by the housing, healthcare, and student-debt mechanisms is, in the contemporary American context, also the household whose interaction with the criminal-justice system is most likely to occur and to produce the structural disabilities that the carceral apparatus produces. The post-incarceration employment penalties, the felony-disenfranchisement mechanisms, the housing-eligibility constraints, the federal-benefits restrictions, and the broader collateral consequences of criminal conviction operate against precisely the population the broader cascades are most affecting. The criminal-justice framework is, in this respect, the carceral mechanism through which the broader cascades’ losers are excluded from the political and economic recovery the broader fourth-settlement framework would otherwise require.

What’s at Stake

The Intelligent Party’s policy framework on criminal justice and drugs, articulated in the platform’s justice position, is the seven-component proposal that addresses the structural features the previous sections diagnosed.²⁹ The components are: federal marijuana legalization with automatic expungement; decriminalization of personal possession of other drugs; federal funding expansion for addiction treatment and harm reduction; federal sentencing reform with mandatory-minimum reconsideration; end of cash bail for non-violent offenses; federal standards on police use-of-force, body-worn cameras, and accountability; and qualified-immunity reform.

The seven components, taken together, address the structural problem the previous sections identified: the war-on-drugs framework whose empirical inadequacy is substantially documented, the carceral apparatus whose comparative anomaly is substantially documented, and the policing-accountability framework whose structural inadequacy the post-2014 reform movement has documented. The components do not, taken individually, constitute the full structural reform the comparative record validates; the Portuguese model includes substantial structural treatment-capacity buildout that the platform’s framework substantially incorporates and that the Oregon case demonstrates is required for the decriminalization component to operate as intended. The platform’s framework, in operational terms, is closer to the Portuguese model than to the Oregon model; the principal operational difference is the substantial commitment to treatment-capacity expansion that the platform’s framework includes.

The political coalition required to enact the seven components does not currently exist. The conditions under which it might assemble are the conditions the broader fourth-settlement framework requires.

The comparative record is, in this case as in the others, instructive. Portugal’s decriminalization model has produced substantially better empirical outcomes than the American war-on-drugs framework across two decades of operation. The European harm-reduction frameworks have produced substantially better empirical outcomes than the American enforcement-oriented frameworks. The marijuana legalization wave across the implementing American states has produced substantially better outcomes than the marijuana-prohibition framework that preceded it. The structural alternative is operating; the American refusal to adopt the comparative model has been, across fifty-four years, bipartisan in its operational effect.

The realism the previous essays have called for, applied to criminal justice and drug policy, is the realism that the structural reform is achievable on the specific architecture the platform has proposed, against the structural barriers the broader fourth-settlement framework would address. The conditions under which the barriers break are the conditions the broader settlement requires.

The fifty-year war is the war that has not been won, that has not been lost, and that has been continued. The structural ending of the war is the assembly the broader fourth-settlement framework requires. The realism is the realism the comparator countries have demonstrated and that the marijuana-legalization wave has begun to demonstrate at the American state level: the structural alternative works on the empirical record, against the political coalitions that have, across fifty-four years, declined to adopt it.

---

Notes