Summary

  • The New Orleans Archdiocese’s “removed from ministry” designation functioned across four decades as a canonical representation that the Guardian and WWL Louisiana reporting establishes as divorced from the material reality of continued financial support for Gerard Joseph “Jerry” Howell.
  • The archdiocese of New Orleans kept Howell as a credibly accused priest in active or retired ministry from 1980 through 2020, retaining him after reporting him to then-District Attorney Harry Connick Sr, after a 1992 public exposure by WWL Louisiana, and after a 2010 psychological assessment that labeled him “high risk” with at least 24 child victims.
  • Survivor Kathleen Austin’s account, captured in the 2025 documentary “God As My Witness,” places the divergence on the survivor record: then-Archbishop Philip Hannan directed her to “kick him in the balls” the next time she saw Howell, an instruction Austin described as outside the protective categories the institution claimed to apply.
  • The 2020 federal bankruptcy-court cancellation of Howell’s retirement benefits — which then-Archbishop Gregory Aymond characterized in correspondence to a Vatican official as “unjust and painful” — is the terminal expression of a $305 million settlement now dividing among approximately 740 clergy molestation claimants.

Howell, 87, died on 8 September at his Texas senior-living apartment after an apparent fall, according to Austin police; the medical examiner’s office was still working to determine cause of death as of Tuesday and police said the death was “not being investigated as a homicide.” The disclosures accompanying that death make visible an institutional vocabulary whose operational meaning diverged from its stated protective purpose: across four decades, the canonical categories the New Orleans Archdiocese used to describe priestly discipline — “credibly accused,” “removed from ministry,” “retired,” “active ministry” — coexisted in Howell’s case with monthly stipends, insurance, and living expenses the archdiocese continued to pay. The $305 million bankruptcy settlement now dividing among approximately 740 clergy molestation claimants is the eventual expression of that divergence.

How the categories were meant to operate

According to the Guardian and WWL Louisiana reporting, the archdiocese of New Orleans managed a disclosure page of “dozens of priests and deacons who were removed from ministry for abuse of a minor.” The archdiocese’s own chosen term had a stated ameliorative purpose: communicating to laity, alleged victims, and the public that the named clergyman is no longer trusted in active ministerial roles. Descriptively, however, the reporting shows that in Howell’s case the label coexisted for decades with monthly stipends, insurance, and living expenses paid by the archdiocese; with Vatican correspondence defending that support as “full financial benefits” owed to a priest; and with internal correspondence from then-Archbishop Gregory Aymond characterizing the federal bankruptcy court’s cancellation as “unjust and painful.” Read as an internal category, “removed from ministry” named Howell’s parochial assignment while preserving intact the financial and reputational relationship with the institution.

The Guardian’s reporting also notes that many of the approximately 740 claimants in the broader settlement “name abusers whom the archdiocese has not included in its clergy removal list,” indicating that the conceptual category operates institution-wide rather than as a Howell-specific artifact.

Alfred Korzybski’s foundational formulation — that the map is not the territory — names the structural pattern the disclosures document: the categories used to describe priestly discipline (“credibly accused,” “removed from ministry,” “retired,” “active ministry”) functioned across four decades as a representation divorced from the underlying reality of material support, financial benefits, and continued institutional backing.

The documented record, 1980 to 2026

According to records obtained by the Guardian and reported jointly with WWL Louisiana, the documented conduct runs as follows.

In 1980, then-Archbishop Philip Hannan reported Howell to then-District Attorney Harry Connick Sr; the archdiocese nevertheless kept Howell in active ministry, and Connick never charged Howell. Twelve years later, WWL Louisiana confronted Howell at his Texas home in 1992. Then-Archbishop Francis Schulte responded by telling Howell to “quit and get a job.” Vatican officials then ordered Schulte to let him “continue as a priest with full financial benefits.” In 2010, the archdiocese sent Howell for mental-health treatment, where a psychologist diagnosed him as “high risk” and reported the church was aware of at least 24 child victims; the archdiocese nevertheless continued quietly supporting his retirement. In 2020, a federal bankruptcy court halted Howell’s retirement benefits — living expenses, insurance, and a $650 monthly stipend — after the archdiocese filed for protection amid the financial fallout of the worldwide Catholic clergy abuse scandal. Aymond wrote to a Vatican official in 2020 calling the bankruptcy court’s cancellation “unjust and painful” and directed Howell to seek aid from Men of Melchizedek, a non-profit dedicated to helping priests prohibited from ministering whose leader had previously headed a separate organization “shut down for misusing donations and misleading donors about its purpose,” per the Guardian and WWL Louisiana reporting. A spokesperson for the archdiocese said it never restarted Howell’s retirement benefits after the bankruptcy settlement.

Aymond retired in February 2026 and was succeeded as New Orleans archbishop by James Checchio. Howell died on 8 September 2026 at age 87 after an apparent fall at his Texas senior-living apartment, according to Austin police. Officers responding to a welfare check found Howell about 5 p.m.; local emergency medical personnel arrived to provide aid but he stopped breathing and was pronounced dead at the scene. The Austin medical examiner’s office was still working to determine cause of death as of the Guardian’s 16 September 2026 reporting, and police said the death was “not being investigated as a homicide.” A church spokesperson said the archdiocese learned of Howell’s death later that same week, and the archdiocese-managed disclosure page was quietly updated to list him as deceased.

At the time of his death, the archdiocese’s disclosure page listed nine bankruptcy claims having named Howell as the offender — “the second most among eight living clergymen” on the list of those removed from ministry for abuse of a minor, per the Guardian’s reporting. Howell’s late brother and fellow priest Rodney is considered by church officials to be credibly accused of child molestation.

Survivor record on film and on the record

The documentary film “God As My Witness” (2025), directed by Lindsay Quinn Pitre and Michael Brandner Sr, records a survivor-level instance of the same divergence. According to the Guardian’s reporting on the film, Kathleen Austin — one of Howell’s most prominent survivors, who grew up with parents and a sister who were hearing impaired — recounted a meeting with Hannan about Howell’s abuse during which she was directed to “kick him in the balls” the next time she saw him. Austin, quoted in the film, said: “I remember looking at him thinking, ‘Really? This is how you handle this?’” Austin added in the film: “That’s not my job. This is your job.” The exchange places the canonical-divergence argument on the survivor record: the institutional figure holding the categories that signal protective purpose advised a survivor in language the survivor herself recognized as outside those categories.

Austin, asked by reporters after Howell’s death, said that it was a relief to learn of his death. “He will no longer be able to harm another innocent child,” Austin said. “I can now relax, not worrying about where he is [or] what he’s doing.” Austin characterized Howell’s apparent fall as “a miserable way to go,” describing him as having “apparently becoming one of the more than 43,000 people in the U.S. who die from falls annually on average” — her characterization, not an independently verified figure. His death “puts me at rest,” she said, adding that when he meets “our maker” she hopes “his punishment far exceeds what he did to children while he was here on the planet.” Austin expressed empathy for relatives of Howell, who “had to live with two pedophile priests and the terror that … they inflicted.”

How the canonical map drifted from the territory of protection

Herman Cappelen and David Plunkett, with Alexis Burgess, in the 2020 Oxford volume “Conceptual Engineering and Conceptual Ethics,” and Sally Haslanger in parallel work on conceptual ethics, ask what work a concept is meant to do and what work it actually does. The diagnostic paraphrase applied here — that the inquiry runs in two steps — is the analyst’s own formulation rather than the authors’. Applied to “removed from ministry” as the New Orleans Archdiocese applied it, the diagnosis the disclosures support is sharp. The category was meant to signal that a credibly accused priest had been separated from access to children. As applied to Howell, the reporting establishes, the category permitted his continued receipt of full financial benefits under Vatican direction, his continued residence in a senior-living community while on a list of “dozens of priests and deacons” the archdiocese had removed, and his continued quiet receipt of living expenses, insurance, and stipend for decades after the 1980 report.

Two candidate revisions of the category admit themselves. The first ties material support directly to canonical status, so that the receipt of a $650 monthly stipend or of living-expense payments is itself an audit-visible signal of the underlying classification — making the divergence visible at every annual reporting cycle rather than once a generation. The second couples “removed from ministry” to geographic and vocational restrictions in such a way that the Vatican override pathway the reporting documents is no longer available to preserve the prior arrangement. The category as currently engineered admits both moves; the canonical vocabulary as applied resisted them.

Operational machinery to enforce a stricter operational definition would include a written cessation order from the metropolitan or archbishop with a documented date and signatory; independent audit by a review body with published annual compliance reports; and a victim-notification protocol entailing written notice through a designated advocate with verification of receipt and an established grievance channel. The coordination problem adoption would face is the one the reporting already shows. The category is partially defined in Rome, and a 1992 Vatican order indicates that at least one U.S. archbishop’s attempt to make the term operationally meaningful has been overridden. A diocese-level revision would therefore require Vatican cooperation; the Guardian’s 16 September 2026 reporting carries no record of operational change in the archdiocese’s pattern after Aymond’s 2020 correspondence, and successor Archbishop James Checchio is not described in that reporting as having altered the disclosure approach.

Why each year of non-action deepened the institutional exposure

Nassim Taleb, with Raphael Douady in a 2012 paper that provided a mathematical definition, has developed a framework distinguishing systems that gain from volatility (convex) from those that lose from it (concave). Under that framework, the archdiocese’s pattern of continued support for credibly accused clergy is a concave-exposure system. Each year of non-action forestalled the immediate cost of scandal, public disclosure, civil litigation, and Vatican negotiation. Each year also increased the cumulative tail the institution was carrying: survivors’ memories, statutes-of-limitations exceptions relaxed over time, and additional allegations accumulating against a single priest whose clinical file already read “high risk” with at least 24 known child victims. The 2020 bankruptcy filing and the $305 million settlement are the eventual expression of that tail. The 2010 “high risk” assessment that produced no operational change is the structural mechanism by which the gap between documented risk and institutional response was preserved. The decisions in 1980 to route the matter through Connick’s office, and in 1992 to permit Vatican override of Schulte, are the documented record by which the responsibility for any given year of non-action was carried by bodies — a Catholic district attorney’s local discretion; a Curial office distant from U.S. parishes — whose relationship to the protection of Howell’s alleged victims the reporting does not document.

Diane Vaughan, in her 1996 study of the Challenger launch decision, has named this dynamic “normalization of deviance”: repeated departures from accepted practice that produce no immediate adverse consequence gradually redefine the boundary of acceptable risk. Each year in which a reported allegation produced no canonical consequence became, within the institution, evidence that the institutional practice was sufficient. The reported risk profile of the underlying conduct did not change; only the perception of it did. The categories on the canonical map — robust enough to satisfy outside auditors and Vatican officials — drifted further from the territory of protection with every year of continued support.

Three fragilities in the system warrant separate naming because they predict different things. First, interface fragility: the joint between Vatican canonical authority and local diocesan practice proved to be the failure surface rather than either institution alone, with the 1992 Vatican directive that overrode Schulte’s instruction as the documented instance. Second, key-person dependency with a structural counter-reading: the institutional pattern required that specific actors — Hannan, Schulte, Aymond — each maintain the practice through their tenures, and the 2020 bankruptcy-court cancellation produced Aymond’s documented correspondence characterizing the cancellation as “unjust and painful.” The Vatican override of Schulte in the 1990s indicates that the institutional structure — diocesan administration answering to Roman authority — actively sustained the practice independent of any individual holder’s preferences, with Aymond’s 2020 letter to a Vatican official the same structural pattern at a later moment. The dependency may sit with the office sequence rather than the individuals occupying the offices, with the fragility real either way but the remedies differing. Third, durability inversion: the older canonical disciplinary procedures and the older canonical vocabulary proved more durable than the underlying protective purpose they were designed to serve, with the same vocabulary, applied with the same formal apparatus, having also to absorb Howell’s late brother and fellow priest Rodney, whom church officials consider credibly accused of child molestation — a two-cleric household of credibly accused priests whose canonical coverage the institution’s enumeration of offenders documented without naming a protective outcome.

What the bankruptcy settlement has not yet resolved

Gary Klein’s pre-mortem method asks which specific interface yielded first and under what load condition. Applied to these disclosures, the interface that failed earliest was the 1980 DA referral: a complaint channel constituted solely by the office of a Catholic district attorney produced no prosecution and no institutional change, leaving the archdiocese in sole management of a credibly accused priest. The second interface that failed was the 1992 Vatican reversal, which converted a partial removal directive into a full benefits reinstatement. The third was the 2010 psychological assessment, where a clinical “high risk” finding generated no recorded change in Howell’s support status; there is no reported operational gate between the psychologist’s diagnostic output and the archdiocese’s benefits-continuation decision. The 2020 bankruptcy-court order is not a fresh failure of the same system so much as its terminal output: once 740 claims and a $305 million settlement expressed the cumulative tail, the financial continuation that had defined “removed from ministry” became unsustainable. The pre-mortem also surfaces that the bankruptcy court’s intervention operated as a non-institutional lever on the same vocabulary problem, completing the work that no institutional body — the 1980 DA referral, the 1992 Vatican reversal, or the 2010 clinical finding — was structurally positioned to perform.

A pre-mortem projection that treats the bankruptcy settlement itself as the failure scenario and traces causes backward surfaces three current decision points. According to a bankruptcy-court filing referenced in the Guardian’s reporting, fourteen of approximately 740 claimants have requested reconsideration of their allocated amounts, and about 220 have indicated they will not contest them. The reconsideration rubric, designed to assess “factors such as the nature and duration of the abuse,” appears to produce enough variance in outcome that a non-trivial number of claimants find the result worth contesting. The asymmetry the filing represents — fourteen reconsiderations against 220 non-contestations, with the remainder unstated — indicates that a rubric built on categorical descriptors of abuse type and duration, rather than continuous measures with documented comparator sets, leaves each individual allocation under-determined at the categorical boundaries. A claimant whose abuse crosses two descriptors, or whose duration lies at a rubric-tier edge, has grounds to ask why the boundary falls where it does.

The clergy removal list, which the reporting explicitly identifies as incomplete relative to the claimant universe, is a second decision point: a list that excludes named offenders produces both continued material exposure and continued canonical-coverage gaps. The alternative-support pathway, into which Aymond directed Howell and which the reporting documents as having been led by someone whose prior organization was “shut down for misusing donations and misleading donors about its purpose,” is a third: the question of what happens to credibly accused priests whose canonical benefits have been terminated is currently being answered by referral to outside organizations whose own integrity is itself an open question.

Flow of documented support across four decades

The reporting traces the flow of institutional support over four decades. Howell received full financial benefits, including the $650 monthly stipend, living expenses, and insurance, throughout the documented period. Men of Melchizedek, the non-profit to which Aymond directed Howell after the bankruptcy settlement ended direct payments, stands to receive donations and referrals under the arrangement. The Vatican officials who overrode Schulte’s 1992 directive preserved the canonical category of “full financial benefits” as a category the Curial office could defend. Diocesan administration retained the documented appearance of action — a disclosure page, a removal listing, a bankruptcy filing — while the underlying institutional practice continued.

The Guardian’s reporting, sourced to information compiled by Penn State University’s law school, indicates that forty-four U.S. Catholic religious institutions have filed for bankruptcy protection in the course of the clergy-abuse scandal and twenty-nine have settled such proceedings. The New Orleans archdiocese is in both groups. The cost of not adopting a stricter revision of “removed from ministry” is the cost the archdiocese has already paid in the bankruptcy settlement; the cost of the canonical divergence has been borne by Howell’s survivors across the span the reporting locates in 1980, 1992, 2010, and the years since.

Scope and inheritance

The Guardian’s 16 September 2026 article originated the reporting underlying this analysis; the joint investigation with WWL Louisiana extends the documentary record. The archdiocese’s broader bankruptcy settlement — $305 million to be divided among about 740 clergy molestation claimants, with compensation amounts based on a rubric meant to assess factors such as the nature and duration of the abuse — is the terminal expression of the same pattern. As of the day after Howell’s death, 14 claimants had requested reconsideration of their allocated amounts and about 220 had indicated they would not contest them, according to a bankruptcy-court filing.

The 43,000-deaths-from-falls figure cited by Austin is her characterization, not an independently verified statistic. The 44/29 diocesan-bankruptcy counts are sourced to information compiled by Penn State University’s law school, as cited in the Guardian’s reporting; independent verification of the underlying tracker was not produced for this analysis. The conceptual-engineering attribution to Cappelen and Plunkett (with Burgess) and to Haslanger’s parallel work reflects the 2020 Oxford volume and Haslanger’s separate conceptual-ethics project; the diagnostic paraphrase applied here is the analyst’s own formulation. The Taleb/Douady fragility-attribution is anchored to a 2012 paper that provided a mathematical definition of fragility and antifragility.

The canonical map after the listing

The canonical map has been quietly updated to list Howell as deceased. The underlying territory — the categories that permitted the divergence between canonical status and material support, the institutional exposure the bankruptcy settlement is still working through, the alternative-support pathway into which other credibly accused priests have been directed — survives the listing. The settlement process inherits both the categories and the institutional practice they describe, and the inheritance includes a reconsideration rubric generating a non-trivial rate of contested allocations, a clergy removal list shorter than the documented claimant universe, and a referral network whose own integrity is itself in question. Whether the next iteration of canonical vocabulary is engineered to perform the protective work its predecessors have now been documented not to do is a question the inheritance has not yet answered.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Conceptual Engineering
Asks not just what a concept means but what it should mean, and re-engineers it.
Fragility / Antifragility Audit
Asks whether a system gains or loses from volatility, shocks, and disorder (Taleb).
Pre-Mortem (Fragility)
Imagines a system has already broken and traces the structural fragilities that let it.