Mandatory minimums in the United States, France, and Italy have each been tried and then narrowed or repealed, usually citing prison overcrowding and cost rather than a change of view on proportionality doctrine itself. The clearest documented sentencing disparity anywhere in this comparison is the US crack and powder cocaine gap; no comparable dataset exists for continental Europe, a methodological gap this article treats as a finding in its own right rather than smoothing over with an unsupported comparison. Current reform across these jurisdictions traces substantially to prison capacity and cost rather than a renewed constitutional argument, from Britain's 2025 Sentencing Bill to Italy's narrowing of drug-trafficking floors. What follows traces that question, who decides how severely to punish and by what standard, across a human-rights court, two national doctrines, three guidelines architectures, and three mandatory-minimum jurisdictions that each tried the same instrument and walked it back differently.
The ECHR's Proportionality Floor: Vinter's "Right to Hope" and the Line Kafkaris Draws
The ECtHR does not tell states how severely to punish crimes. The Court tells them punishment cannot become so severe, or so irreversible, that it strips away human dignity entirely. In Vinter and Others v. the United Kingdom [GC], Applications nos. 66069/09, 130/10 and 3896/10, judgment of 9 July 2013, the Grand Chamber held that states retain a wide margin of appreciation in fixing sentence length for serious crimes, and a whole-life order is not itself a violation of Article 3, the Convention's prohibition of inhuman or degrading treatment. What violates Article 3 is irreducibility: a life sentence with no mechanism, at any point, for domestic authorities to review whether continued imprisonment remains justified. The Court reasoned that removing any prospect of release is incompatible with human dignity, and commentators have since given that reasoning its own name, the right to hope, meaning a whole-life sentence does not itself violate Article 3, but a whole-life sentence with no mechanism for review at any point does.
Kafkaris v. Cyprus [GC], Application no. 21906/04, judgment of 12 February 2008, establishes the floor beneath Vinter. Kafkaris, sentenced to mandatory life imprisonment for premeditated murder under Cypriot law that gave judges no sentencing discretion at all, argued the mandatory nature of his sentence was itself disproportionate. The Grand Chamber rejected the Article 3 claim: because Cyprus's life sentence was, in practice, reducible (a number of life prisoners had in fact been released under presidential powers, without first having to serve a fixed minimum term), the sentence did not cross the irreducibility line, even though the Court flagged procedural shortcomings in the release mechanism for future reform.
Read together, Vinter and Kafkaris draw the same line from opposite sides. Mandatory sentences and long fixed terms are constitutionally tolerable under Article 3. The total absence of any realistic release mechanism is not. That distinction, between a harsh sentence and an irreducible one, is the ECHR's contribution to the proportionality question this article traces across four more legal systems: not how long a state may punish, but whether it must leave a door open at all.
National Proportionality Doctrines: Germany's Culpability Principle and the US Eighth Amendment's Retreat
Germany takes the opposite institutional route to the ECtHR's supranational floor. There is no sentencing guidelines body and no binding grid. Statutory ranges are deliberately wide: manslaughter in its ordinary form carries five to fifteen years' imprisonment, and voluntary homicide ranges from a suspended six months up to life, and the Federal Constitutional Court of Germany (Bundesverfassungsgericht) treats that breadth as constitutionally compelled, not merely permitted. The doctrinal anchor: the Schuldprinzip, or culpability principle, an application of the broader Verhältnismäßigkeitsprinzip (proportionality principle) rooted in Article 20(1) of the Grundgesetz (Basic Law) and its Rechtsstaat, or rule-of-law, guarantee. Punishment must track the offender's individual guilt, and only a judge assessing the full case, not a legislative formula applied in the abstract, can make that determination.
Section 46 of the Strafgesetzbuch (StGB), the German Criminal Code, is the statutory anchor: it lists the factors a judge must weigh but stops short of saying how to weigh them, the "leeway theory," or Spielraumtheorie, that German scholarship both relies on and criticizes. Section 46 sits within the StGB's General Part, a structural feature the forthcoming article on criminal law codification examines in full. In September 2018, the biennial Deutscher Juristentag (German Jurists' Forum) debated whether Germany needs binding guidelines to fix the resulting sentencing disparities; the proposals were rejected, leaving broad judicial discretion the settled, if contested, German model.
The United States shows the most doctrinal instability of any jurisdiction surveyed here. Solem v. Helm, 463 U.S. 277 (1983), established a three-factor proportionality test, comparing the gravity of the offense against the harshness of the penalty, sentences for other crimes in the same jurisdiction, and sentences for the same crime in other jurisdictions, and on that basis struck down a life-without-parole sentence for a nonviolent recidivist as unconstitutionally disproportionate.
Eight years later, Harmelin v. Michigan, 501 U.S. 957 (1991), gutted Solem without formally overruling it. Justice Scalia, writing with the Chief Justice, argued the Eighth Amendment contains no freestanding proportionality guarantee at all; Justice Kennedy's separate opinion, the controlling standard, narrowed proportionality review to a single gross disproportionality threshold, whether the sentence is grossly disproportionate, collapsing Solem's three factors into one and reserving comparative analysis for the rare case that clears that threshold. Harmelin upheld a mandatory life-without-parole sentence for possession of 672 grams of cocaine.
Graham v. Florida, 560 U.S. 48 (2010), shows the doctrine's remaining life, but only in a carved-out category: juvenile non-homicide offenders. The Court held life without parole categorically unconstitutional for juveniles who did not kill, reasoning that such an offender must be given "some meaningful opportunity" for release based on demonstrated maturity and rehabilitation, not a guarantee of release, but a realistic chance at consideration. the article on youth criminal responsibility and juvenile justice reform covers Graham as the second entry in its own five-case progression from Roper through Jones v. Mississippi and the current state-by-state count of juvenile life-without-parole bans; that fuller arc belongs there, not here.
Germany treats proportionality as a question only a judge assessing the full case can answer. The US Supreme Court has spent four decades disagreeing with itself about whether the Eighth Amendment even contains a proportionality guarantee at all, a genuine doctrinal instability the guidelines comparison in the next section inherits directly.
From Mandatory to Advisory: Sentencing Guidelines in the United States, the United Kingdom, and Germany
United States v. Booker, 543 U.S. 220 (2005), reshaped the mechanics of federal sentencing without touching Eighth Amendment doctrine at all. The Sixth Amendment holding was narrow: any fact, other than a prior conviction, that increases a sentence beyond the otherwise-applicable statutory maximum must be admitted by the defendant or found by a jury beyond reasonable doubt, and judges cannot find such facts themselves under a mandatory guidelines regime. The remedial opinion then severed the provision making the US Federal Sentencing Guidelines mandatory, along with the provision limiting appellate review of below-guideline sentences. The practical result: federal judges must still calculate the applicable guideline range and consider it, but are no longer bound by it. The guidelines went from mandatory to advisory overnight, and every subsequent federal reform, the Fair Sentencing Act's later retroactivity and the First Step Act's safety-valve expansion among them, has operated within this advisory baseline since.
The US Federal Sentencing Guidelines also include a distinct chapter governing corporate and other organizational defendants, compliance-program credit, and deferred prosecution agreements, a separate architecture the forthcoming article on corporate criminal liability will address in full.
The Sentencing Council (England and Wales) sits at the opposite structural pole from Germany's model: an independent body issuing guidelines that are binding unless a court finds it contrary to the interests of justice to depart from them, a middle position between the now-advisory-only US federal regime and Germany's guideline-free discretion. Within any of these three systems, a judge's discretion, or a guideline departure, may point toward an alternative to incarceration entirely: probation, a suspended sentence, or a restorative disposition. Restorative dispositions are one category of alternative sentencing available within a discretion-based system, and the article on restorative justice models worldwide covers that model's theory and its European statutory forms in full; this article notes only the connection to judicial discretion.
The same underlying question, whether a legislature or a judge should control sentence severity, produces three different institutional answers: mandatory-then-advisory guidelines in the US federal system, binding-unless-unjust guidelines in England and Wales, and no guidelines body at all in Germany. Does every legal system that limits mandatory minimums replace them with binding sentencing guidelines? No. Germany is the counter-example: it rejects both mandatory minimums as a general matter and guidelines entirely, relying instead on judicial discretion bounded by the culpability principle.
Mandatory Minimums Tried and Undone: Three Jurisdictions, Three Timelines
California's 1994 "Three Strikes and You're Out" law imposed a mandatory 25-years-to-life sentence on a third felony conviction regardless of severity, driving prison-overcrowding litigation for two decades. California Proposition 36, "Changes to the Three Strikes Sentencing Initiative" (2012), narrowed the 25-to-life trigger to cases where the third felony is itself serious or violent, and, as the first US ballot measure of its kind, created a resentencing pathway for people already serving third-strike life terms for non-serious, non-violent offenses. It passed with roughly 70% support in every one of California's 58 counties. Reported recidivism among those resentenced under Prop 36, about 25% reconvicted within three years, mostly for misdemeanors, is sometimes compared favorably to a roughly 42% general-release recidivism rate cited in the same reporting, but base-population comparability between the two figures has not been independently verified: different releases and follow-up windows are not always measured on the same terms, and the comparison should not be read as a controlled study or as proof that resentencing itself reduces reoffending.
France experimented with its own mandatory minimums for repeat offenders, peines plancher, literally "floor sentences," introduced by the law of 10 August 2007 (loi n. 2007-1198) on combating recidivism, using a one-third rule: for a misdemeanor punishable by three years, the floor was set at one year. The LOPSSI 2 law of 14 March 2011 extended the floor-sentence regime to aggravated violence from the first offense. The mechanism was abolished on 1 October 2014 by the law of 15 August 2014 (loi n. 2014-896) on the individualization of sentences, under French Minister of Justice Christiane Taubira, reflecting a policy shift toward individualized, rehabilitation-oriented sentencing. Judicial resistance during the mandatory period was substantial: only about 38% of cases eligible for a floor sentence actually received one, suggesting French judges routed around the rule wherever they could.
Italy's path is structurally different from both the US and France: constitutional review of an existing floor, not legislative repeal. The Italian Constitutional Court cannot simply strike a mandatory minimum and leave a gap. It can only replace an unconstitutional floor with another figure already present elsewhere in the legal order, on the theory that inventing a new penalty from nothing would usurp the legislature's reserved constitutional role, the riserva di legge in materia penale. The Italian Constitutional Court (Corte Costituzionale), sentenza n. 40/2019 (rapporteur Marta Cartabia, later President of the Corte Costituzionale, deposited 8 March 2019), declared unconstitutional the eight-year minimum sentence in Article 73, first paragraph, of the Consolidated Text on Narcotics (D.P.R. n. 309/1990) for non-serious drug trafficking, on proportionality grounds, and substituted the six-year minimum already used elsewhere in the same statute for borderline cases.
An earlier warning shot, sentenza n. 179/2017, had already urged the legislature to close the gap between the penalty bands for serious and non-serious trafficking before the Court intervened directly. The maximum penalty, twenty years, was untouched; only the floor moved. That correction addresses how severely Italy punishes conduct that remains criminal, a distinct question from whether such conduct should be criminal at all, the question the forthcoming article on European decriminalization trends takes up directly, drug policy among its six tracks.
Jurisdictions that adopted rigid, legislature-set mandatory minimums have all walked them back within a decade or two, usually citing overcrowding and disparity rather than a change of view on proportionality doctrine itself, but through three structurally different mechanisms: a ballot initiative in California, ordinary legislative repeal in France, and constitutional-court substitution, not repeal, in Italy.
Racial and Socioeconomic Disparity: A Documented US Story, an Undocumented European Question
The clearest documented disparity in sentencing anywhere in this comparison is the US crack and powder cocaine gap. The Anti-Drug Abuse Act of 1986 set a 100-to-1 sentencing ratio: 5 kilograms of powder cocaine or 50 grams of crack triggered the same 10-year mandatory minimum, and 500 grams of powder or 5 grams of crack triggered the same 5-year minimum. Because crack offenses disproportionately involved Black defendants, the ratio produced one of the most heavily documented racial disparities in modern American sentencing. The Fair Sentencing Act of 2010, signed by President Barack Obama on August 3, 2010, reduced the ratio from 100:1 to 18:1, raising the crack trigger thresholds to 28 grams and 280 grams respectively, and eliminated the 5-year mandatory minimum for simple crack possession, but did so prospectively only. What actually reached people already sentenced under the old 100:1 ratio was the First Step Act of 2018's retroactivity provision (the Act's full mechanics are covered in the next section); the United States Sentencing Commission (USSC) credits that retroactivity fix with benefiting roughly 2,000 or more incarcerated people in its first year alone.
No comparable dataset exists for continental Europe. Reporting from the Open Society Justice Initiative and academic sentencing scholarship both note that most EU member states do not collect criminal-justice statistics disaggregated by race or ethnicity, either as a matter of practice or, in some jurisdictions, as a matter of law. The United Kingdom is the notable exception, systematically collecting and publishing ethnicity-disaggregated sentencing data, feeding directly into the Independent Sentencing Review's 2024-2025 work, described further in the next section. Where continental data does exist, it typically measures over-representation of non-nationals or minority groups in arrest, pretrial detention, and prison-population statistics rather than sentence length itself, a proxy rather than a direct measure, and one that conflates disparate treatment with a different underlying case mix.
Any comparative claim about the scale of European sentencing disparity relative to the United States should be sourced to a specific named study, not asserted generally, because the underlying data simply does not exist to support a precise comparison. Is the well-documented US racial sentencing disparity matched by comparable European data? No, and the absence of comparable European data is itself a documented methodological gap, a finding in its own right rather than evidence that no disparity exists on the European side of the comparison.
Recent Reforms: The First Step Act, Britain's Capacity Crisis, and Italy's Unfinished Docket
At the federal level, the First Step Act of 2018 (Pub. L. No. 115-391), signed by President Trump on December 21, 2018, reduced the 20-year mandatory minimum for drug offenders with one prior qualifying conviction to 15 years, and the mandatory-life minimum for two or more priors to 25 years. It expanded the statutory safety valve, the provision letting judges sentence below a mandatory minimum for qualifying nonviolent drug offenses, and made the Fair Sentencing Act of 2010 retroactive, reaching people sentenced under the old crack/powder ratio described above. Uptake was measurable: 41.8% of eligible federal drug traffickers received safety-valve relief in the year following the Act, compared with 35.7% in fiscal year 2018.
The Independent Sentencing Review 2024-2025, commissioned by the Ministry of Justice in October 2024 and led by David Gauke, published its final report in May 2025, recommending earned early release in place of automatic early release, greater use of community sentences over short custodial terms, and more intensive post-release supervision, feeding into the Sentencing Bill 2025. This reform is driven explicitly by prison capacity: adult prison population in England and Wales exceeded 87,000 as of April 2025, with demand projected to outstrip supply by 9,500 places in early 2028, not by proportionality doctrine, unlike the doctrinal arguments traced earlier in this article.
In Italy, the Constitutional Court's incremental proportionality review of narcotics sentencing has continued since sentenza n. 40/2019, though specific 2024-2025 holdings beyond that case have not been independently verified for this article and are not cited here as confirmed; any claim about a specific recent docket outcome would require a direct pull from the Court's own published record before publication.
Executive clemency is structurally distinct from the reforms above: it corrects individual cases at a head of state's discretion rather than changing rules going forward. The outgoing Biden administration commuted approximately 1,500 sentences and pardoned 39 people on December 12, 2024, described by the White House as the largest single-day clemency action in modern presidential history, and separately commuted 37 of 40 federal death-row sentences on December 23, 2024. Reported figures on the Trump administration's 2025-2026 clemency actions, including 1,500 or more pardons and a blanket pardon covering roughly 1,600 January 6 defendants, trace to advocacy and opposition-aligned sources rather than the Department of Justice's own published grant list, and are stated as reported, not independently confirmed. Is prison capacity or constitutional doctrine the primary driver of current sentencing reform? Capacity, most clearly in the British example, though the two are not always separable in practice.
Frequently Asked Questions About Sentencing Reform and Proportionality
Does the US Eighth Amendment guarantee a proportional sentence? The doctrine is contested rather than settled. Harmelin v. Michigan (1991) split the Court: Justice Scalia argued the Eighth Amendment contains no freestanding proportionality guarantee at all, while Justice Kennedy's controlling concurrence narrowed review to whether a sentence is grossly disproportionate, a threshold cleared only rarely.
Are mandatory minimum sentences disappearing worldwide? Narrowing in some places, not disappearing. The US federal system and California have narrowed specific mandatory minimums, and France abolished its peines plancher regime outright in 2014, but Italy's Constitutional Court can only substitute an existing statutory figure for an unconstitutional floor rather than eliminate mandatory minimums as a category, and mandatory minimums remain common in all three systems for other offenses.
Does California's Proposition 36 prove that resentencing reduces reoffending? The reported figures point in that direction, but the comparison is not a fully controlled study. Recidivism among Prop 36 resentenced individuals is often measured against general-release recidivism drawn from different releases and follow-up windows, a comparability gap the underlying data has not resolved.
Can a European constitutional court simply strike down a disproportionate mandatory sentence? Not in the way US courts can. Italy's Constitutional Court can only replace an unconstitutional statutory floor with another figure already present elsewhere in the legal order, not invent a new penalty or eliminate the floor entirely, reflecting the principle that only the legislature may define criminal penalties in the first instance.
Related Articles
This article has traced how five legal systems decide how severely to punish conduct that remains criminal, from a supranational human-rights floor through two national culpability doctrines, three guidelines architectures, and three mandatory-minimum jurisdictions that each tried the same instrument and walked it back differently. The remaining articles take up adjacent questions left aside here. A forthcoming article on criminal law codification examines how criminal codes are structured, including where provisions like Germany's Section 46 sit within a code's General Part, and why some jurisdictions never finish replacing an outdated one. A forthcoming article on European decriminalization trends asks the same proportionality question from its other side: whether conduct should be criminal at all, not how severely to punish conduct that remains so. A forthcoming article on corporate criminal liability extends this comparison to organizational defendants under separate frameworks. A final article on statute of limitations in comparative perspective addresses proportionality's temporal dimension: how long after an offense the state's punitive interest remains enforceable at all.