Bankruptcy as shield, liability as conversion, and the Supreme Court's 2024 ruling
On September 15, 2019, Purdue Pharma LP filed for Chapter 11 bankruptcy protection in the Southern District of New York. The filing halted approximately 2,600 lawsuits from states, municipalities, hospitals, and individuals across the United States. Under the automatic stay provisions of Chapter 11, all civil litigation against the debtor — Purdue Pharma — was suspended the moment the petition was filed.
What made the Purdue bankruptcy unusual — and what eventually reached the Supreme Court — was the scope of the liability protection being sought. The settlement plan proposed by Purdue and the Sackler family sought not only to resolve Purdue's corporate liability through the bankruptcy process, but to extend liability releases to members of the Sackler family who had never themselves filed for bankruptcy. The family proposed to contribute approximately $4.5 billion (later increased to approximately $6 billion) to the settlement in exchange for permanent releases from all civil opioid-related claims against them personally.
This mechanism — using the bankruptcy court's authority to extinguish claims against a non-debtor third party — was the novel legal architecture that the case introduced to American bankruptcy law. The Bankruptcy as Shield is the tobacco Liability-to-Revenue Conversion (TB-006) executed through Chapter 11 rather than through direct settlement, with the additional innovation of seeking liability protection for parties who had not themselves become insolvent.
Purdue Pharma's bankruptcy was not economically inevitable. The company, by the time of its filing, had settled the 2007 federal misbranding charges, continued profitable operations, and maintained a manageable balance sheet. The bankruptcy was a legal strategy, not a financial necessity — a mechanism for consolidating all opioid litigation in a single federal proceeding where the resolution could be negotiated comprehensively rather than fought piecemeal across thousands of separate cases.
The automatic stay was the immediate legal benefit. The moment Purdue filed, the entire litigation landscape stopped. States that had been preparing for trial, counties that had developed their damages evidence over years, individuals whose claims had survived motion practice — all were frozen in place. The litigation leverage that plaintiffs had accumulated was suspended while the bankruptcy proceeding allowed Purdue and the Sacklers to negotiate from a consolidated position.
The legal innovation in the Purdue settlement — and the issue that reached the Supreme Court — was the use of Chapter 11 to provide liability releases to the Sackler family members who had not filed for bankruptcy themselves. Under ordinary bankruptcy law, the automatic stay and eventual discharge of claims applies to the debtor (the entity or individual that files). Non-debtors — parties who have not themselves sought bankruptcy protection — are not entitled to have claims against them discharged through someone else's bankruptcy proceeding.
The Purdue plan proposed an exception. In exchange for the family's contribution to the settlement fund, the plan would permanently enjoin any civil opioid claims against family members — even though they had not filed for bankruptcy and had not subjected their personal assets to the bankruptcy court's jurisdiction. The mechanism relied on provisions of Chapter 11 that allow courts to approve releases of third-party claims "in connection with" a reorganization, when those releases are necessary to the plan's feasibility.
The argument for the mechanism: without the family releases, the Sacklers would not contribute to the settlement, and without their contribution, the settlement fund would be too small to compensate creditors and claimants. The family contribution was therefore necessary to the plan — making the releases a condition of plan feasibility that the bankruptcy court had authority to approve.
The Purdue settlement, as ultimately structured, allocated the settlement fund across multiple claim categories: states and municipalities that had sued for economic damages, individuals with addiction claims, healthcare providers, and litigation costs. The distribution structure prioritized abatement — funds for public health programs to address the epidemic — over compensation to individual victims, a prioritization that drew criticism from victims' advocates.
The Sackler family's contribution — approximately $6 billion over 18 years, with the initial commitment of $4.5 billion increased under pressure — represented a significant portion of the publicly reported family wealth. It also represented substantially less than the estimated $10 billion or more distributed to the family from Purdue in the decade before bankruptcy. The settlement was asking plaintiffs to accept, in full satisfaction of claims, a contribution smaller than the wealth that had already been moved out of the company's reach.
Many individual victims and their families supported the Sackler release, because they preferred a certain significant payment to uncertain individual litigation against a family with substantial legal resources. The bankruptcy settlement offered a defined payment in a near-term timeframe; direct litigation against the Sacklers would have taken years and might have produced smaller individual recoveries. The settlement's appeal to individual claimants — who rationally preferred certainty — gave it support that the states and the US Trustee opposed. The victims' dilemma — individual rationality producing a collectively problematic outcome — is the settlement's ethical core.
The United States Trustee — the Justice Department's bankruptcy watchdog — challenged the Sackler releases as exceeding the bankruptcy court's authority. The challenge worked its way through the federal courts and reached the Supreme Court in 2023 (Harrington v. Purdue Pharma L.P.). In June 2024, the Supreme Court ruled 5–4 that the bankruptcy court lacked authority to grant liability releases to the Sackler family members who had not themselves filed for bankruptcy and had not consented to the court's jurisdiction over their personal assets.
The majority opinion held that the Bankruptcy Code does not authorize a reorganization plan to discharge the personal liability of non-debtors in the absence of their consent to jurisdiction. The plan's mechanism — using the automatic stay and injunction authority to protect family members who had not filed for bankruptcy — exceeded the statutory limits of Chapter 11.
"The bankruptcy code does not authorize a release and injunction that, as part of a plan of reorganization under Chapter 11, effectively seeks to discharge the liabilities of non-debtors without the consent of affected claimants." — Supreme Court majority, Harrington v. Purdue Pharma, 2024
The ruling returned the case to the lower courts for further proceedings consistent with the opinion. The practical consequence was that the Sackler family remained exposed to direct civil litigation — the protection they had sought through the bankruptcy mechanism was not available in the form the plan proposed. The family and remaining creditors began negotiating a revised settlement structure that could comply with the Supreme Court's ruling.
The OA series has documented the opioid architecture through six papers. OA-001 established the epistemic foundation: the Porter-Jick letter and its citation engineering — the Five-Sentence Foundation. OA-002 documented the KOL Network: the physician influence system that produced the appearance of independent clinical consensus. OA-003 documented the Metric Engineering: the pain-as-vital-sign campaign that created institutional pressure toward opioid prescribing. OA-004 documented the Family Office Defense: the Sacklers' knowledge gap claim and what the documents show. OA-005 documented the Supply Chain Bystanderism: the DEA's failure to exercise available enforcement authority against wholesale distributors.
OA-006 closes the series by documenting the accountability architecture — the settlement mechanisms through which the accumulated liability of a $50+ billion epidemic was resolved, or attempted to be resolved, at costs substantially lower than the wealth generated and with legal protections that the Supreme Court eventually found exceeded statutory authority. The resolution is incomplete; the litigation continues.
The Opioid Architecture closes Saga VII's evidentiary argument. Four series. Four archives. One mechanism observed across seventy years and four industries.
The Institutional Capture Record (IC series) established the five-element Red Flag Diagnostic in contemporary, legally verified cases — demonstrating that the mechanism operates in plain sight, in publicly accountable institutions, in cases with full documentary records. The Tobacco Record (TB series) proved the five mechanisms operated for forty years in the most document-rich industrial case in history, and documented the template that subsequent industries would adapt. The Lead Record (LD series) established that the playbook predates tobacco by thirty years, produced the most geographically comprehensive cognitive harm in history, and demonstrated that the mechanism leaves physical infrastructure residue that persists after the primary source is eliminated. The Opioid Architecture (OA series) proved the playbook was consciously adapted and redeployed in 1996 — within living memory, in a heavily regulated industry, with a legal record still being written.
The mechanism is not historical. The archive documents it as active, transferable, and predictable. The next deployment — in whatever industry, with whatever harmful product, using whatever version of the five elements — will be recognizable to anyone who has read the series. That recognizability is the archive's purpose. The record exists so that the next regulatory window does not close before the pattern is named.
Internal: This paper is part of The Opioid Architecture (OA series), Saga VII. It draws on and contributes to the argument documented across 69 papers in 13 series.
External references for this paper are in development. The Institute’s reference program is adding formal academic citations across the corpus. Priority papers (P0/P1) have complete references sections.