The Pattern
The documented, nationwide record: the major datasets and investigations into abuse by clergy and church workers across every Christian tradition in the United States, the concealment playbook they uncovered, and the eight institutional behaviors that recur — with every figure kept in its source's own terms: accused, charged, and convicted never blurred.
Read case by case, an abuse scandal looks like one man’s crime or one institution’s failure. Read across the documented record — grand jury reports, denominational investigations, court-record databases built by newsrooms — the same institutional behaviors recur, in tradition after tradition, state after state, decade after decade. This page lays out that record: what the major investigations found, what the numbers actually count, and where the record runs out. The case files in this archive then document the behaviors one case at a time.
Sourcing note. Every figure on this page is given as published by its named source, with a link. The figures were compiled on October 18, 2026, from those publications; most were retrieved as the sources' published summaries of their own work. This archive re-verifies each figure against its primary source as the underlying cases become standalone files. Where a source's number counts accusations, this page says accusations. Where it counts charges or convictions, it says so. No figure on this page is rounded into a national total, because — as the Open Questions section states plainly — no verified national total exists.
The big datasets
There is no U.S. government database of clergy child-abuse offenders. What exists is a patchwork: church-compiled lists, state grand-jury and attorney-general investigations, and investigative-newsroom databases built from court records. These are the major ones, each in its own terms.
| Dataset | What it counts | Headline figures (as published) | Caveats — read before citing |
|---|---|---|---|
| John Jay Report (Catholic, 2004) — commissioned by the U.S. bishops from John Jay College | Allegations in diocesan and religious-order files, 1950–2002 | 4,392 priests accused (about 4% of the ~109,694 who served in the period); 10,667 people reported abuse | Based on diocesan self-reporting; described at publication as probably an undercount. “Accused” here means an allegation existed in a church file — not a charge. |
| Pennsylvania Grand Jury (Catholic, 2018) — statewide investigating grand jury under AG Josh Shapiro | Priests in 6 of Pennsylvania’s 8 dioceses since the 1940s, from the dioceses’ own “Secret Archives” | More than 300 “predator priests”; more than 1,000 identified victims; only 2 could be charged | The grand jury believed the true victim number was “in the thousands” — its stated belief, based on lost records and non-reporting, not a count. Charges were barred in nearly every case by limitations or death. |
| BishopAccountability.org (Catholic, 2003–present) — lay-run archive | Publicly accused U.S. priests (court records, news reports, church lists, released files) | ~7,500 publicly accused U.S. priests in its database | “Publicly accused” is not “charged” or “convicted.” The site itself notes the U.S. bishops have counted 6,721 accused bishops, priests and seminarians, while its own database held only 3,883 names in those categories at the time — i.e., by the bishops’ own count, ~42% of accused clergy were not publicly known. |
| ProPublica compilation of diocesan lists (Catholic, January 2020) | Clergy that dioceses and religious orders themselves listed as “credibly accused” | 178 lists (156 dioceses, 22 religious orders, covering 64.7 million Catholics); ~5,800 unique names (≈6,754 entries with duplicates) | There is no national standard for “credibly accused”; each diocese decides whom to include. At that date, 41 dioceses and dozens of orders — covering ~9 million Catholics — had published no list. |
| “Abuse of Faith” — Houston Chronicle / San Antonio Express-News (Southern Baptist, 2019) | SBC church leaders and volunteers, 1998–2018, from court, prison, sex-offender and news records | ~380 faced allegations; ~220 convicted or pleaded guilty; more than 700 victims | The 380 is not a conviction count: it includes people convicted, credibly accused, successfully sued, and who confessed or resigned. Nearly 100 were still in prison at publication; 100+ were registered sex offenders; some still worked in SBC churches. |
| SBC Executive Committee’s internal list, exposed by the Guidepost Solutions report (Southern Baptist, 2022) | Accused ministers tracked privately by EC staff for more than 10 years | 703 names on the most recent list; 409 believed SBC-affiliated; 9 still in active ministry when reviewed | Guidepost found no indication anyone took any action to ensure listed men were no longer in positions of power. The list was kept secret — not shared even with EC trustees — and released, redacted, only after the report. |
| Fort Worth Star-Telegram investigation (independent fundamental Baptist, 2018) | Allegations in IFB churches and affiliated institutions, 40 states and Canada, some dating to the 1970s | 412 allegations; 187 churches/institutions; 168 leaders accused or convicted of sex crimes against children; at least 45 continued in ministry after accusations reached authorities | The IFB movement has no headquarters and no registry; the paper built the picture from courts, news and survivors over eight months, newly uncovered 21 of the allegations itself, and reported victims believed the true number was much larger. “Accused or convicted” is the paper’s combined category — kept combined here. |
| NBC News, “Pastors and Prey” (Assemblies of God, 2025) | AG pastors, employees and volunteers accused over ~50 years, from lawsuits, criminal records and news archives | Nearly 200 accused; more than 475 people allegedly abused, mostly children; ~30 placed or restored into authority after accusation; ~40 cases of alleged cover-up or dismissal of reports | An accusation dataset built from records, not a conviction count. Of the accused, 123 were ministers and nearly half of those were youth pastors. Restoration records were kept confidential at the denomination’s General Council office. |
| FLOODLIT.org (Latter-day Saints, ongoing) — independent nonprofit database | Documented reports involving LDS leaders, members and others tied to the institution | 4,000+ individual reports catalogued | These are FLOODLIT’s counts of reports, mixing accused, charged and convicted — not a verified conviction count. The LDS Church publishes no list of accused leaders. |
| Watchtower internal database (Jehovah’s Witnesses, 1997–present) | Reports of suspected child molestation sent to headquarters by all U.S. congregations | Contents never produced in any U.S. case. Existence and structure confirmed under oath; size unknown | Outside estimates of “tens of thousands” — and a leaked figure of 23,720 sometimes circulated — are estimates, not verified counts, and this page does not present them as counts. What is documented is the litigation behavior around the database (see below). |
A companion finding belongs beside the Catholic rows: the Associated Press (October 2019) reviewed the nearly 2,000 credibly accused clergy still living and found almost 1,700 living largely unsupervised — some teaching, counseling, nursing, fostering children, or living near schools and daycare; 76 could not be located; and dozens of the ~1,700 had since committed crimes, including sexual assault and possessing child pornography. Most had never been criminally charged.
For mainline Protestant traditions — United Methodist, Episcopal, Presbyterian (PCUSA), ELCA and others — no comparable national dataset exists. No mainline denomination was found to have published a comprehensive accused-clergy list or faced a comparable grand-jury investigation. That absence is itself a finding: for most traditions there is no central registry at all, so no cross-tradition total can be compiled from denominational sources. Mainline cases in this archive are documented at case level (see Cases).
The state reports
Since 2018, the Catholic rows of that patchwork have been joined by a second kind of record: statewide investigations by attorneys general (and, in Colorado, an independent special master) that opened diocesan files under legal compulsion or negotiated access. They are the closest thing the United States has to an official Catholic record — and no two use the same standard, which is why this archive takes each one separately, in its own terms, in The State Reports. The roster of them, as each office states its own result:
| State report | What it is | Its headline, in its own terms |
|---|---|---|
| Illinois — Attorney General (2023) | All six dioceses; names compiled from Catholic sources | 451 clerics and brothers who abused at least 1,997 children — substantiations compiled by the AG, not court findings |
| Maryland — Attorney General (2023) | Archdiocese of Baltimore | 156 people; over 600 children — accusations that “do not constitute findings of guilt”; two entries remain redacted by court order |
| Massachusetts — Attorney General (2026) | Fall River, Springfield, Worcester | 275 accused clergy; 944 survivors — aggregate counts only, “not filtered by any assessment of credibility”; the report publishes no names |
| Michigan — Attorney General (2022–2026) | Six of seven dioceses, one report each | Allegation files — inclusion “does not reflect a determination… that the allegations are credible”; the Detroit report has not been released |
| Rhode Island — Attorney General (2026) | Diocese of Providence | 72 men in Appendix A, determined credibly accused by the AG himself under a “more likely than not” standard |
| Nebraska — Attorney General (2021) | All three dioceses | 57 officials credibly accused; 258 documented victims — published as named case summaries; no prosecution was possible |
| Colorado — Special Master (2019–2020) | All three dioceses | 52 priests; 212 children (cumulative) — a review of the dioceses’ own files, with no subpoena power; “substantiated” means more likely than not |
| Florida — Office of Statewide Prosecution (2020) | All seven dioceses | 97 priests identified, prosecution barred in every case — plus 80 names as printed on a second list: accused priests the Church relocated to Florida |
The funnel: what happens to an accusation
The John Jay data are the best-documented pipeline from allegation to imprisonment in any tradition. Of every 100 priests accused in U.S. diocesan files between 1950 and 2002, roughly 6 were convicted.
| Stage | Priests | Share of the 4,392 accused |
|---|---|---|
| Accused (allegation in a church file) | 4,392 | 100% |
| Reported to police | 1,021 | ~23% |
| Charged | 384 | ~9% |
| Convicted | 252 | ~6% |
| Imprisoned | 100 | ~2% |
About 3,300 of the accused were never investigated because they had already died. A count of “clergy convicted of child abuse” therefore measures the criminal-justice pipeline — reporting, limitations law, survival, evidence — not how much abuse occurred. Every dataset above has a version of this funnel inside it; John Jay’s is simply the one with published numbers at each stage.
The playbook, and the Secret Archives
Pennsylvania’s grand jury worked from the dioceses’ own “Secret Archives” — the canon-law files, accessible only to the bishop, in which abuse records were kept. To analyze how the files showed abuse being handled, the grand jury used a behavioral analysis prepared by FBI National Center for the Analysis of Violent Crime agents. It named a seven-step “playbook for concealing the truth”:
- Use euphemisms. Never say “rape.” Say “inappropriate contact” or “boundary issues.”
- Don’t conduct genuine investigations. Assign fellow clergy to ask inadequate questions and judge colleagues’ credibility.
- Send priests for “evaluation” at church-run psychiatric centers, based largely on the priest’s self-report.
- When a priest must be removed, don’t say why. “Sick leave,” “nervous exhaustion,” or nothing.
- Continue to pay abusers’ housing and living expenses.
- Transfer problem priests to new assignments.
- Avoid reporting to police.
The grand jury also found officials “routinely and purposefully described the abuse as horseplay and wrestling” and “inappropriate conduct.” Steps 2, 3, 6 and 7 are not Catholic inventions, as the behaviors below document — they are the same moves, made by institutions with no shared hierarchy at all.
“And have no fellowship with the unfruitful works of darkness, but rather reprove them.”
Ephesians 5:11
The counter-text to concealment as a system. Reprove means expose — the opposite of the playbook above.
The eight recurring behaviors
Each behavior below is documented in at least two traditions, with the strongest documented examples. This is a synthesis of the investigations above and the cases in this archive — not a claim that every congregation behaved identically in every case. (Contrast Erik Hughes, where the church said it removed the man immediately and cooperated fully with police; this archive records that too.)
1. Transfer without disclosure
The accused is moved to a new congregation, school, or state, and the receiving community is not told. Catholic: playbook step 6; Geoghan reassigned repeatedly in Boston after treatment; Shanley transferred to California with only partial disclosure. IFB: Dave Hyles moved from Hammond to Garland after accusations, with no charges; the Star-Telegram documented a pastor, Bruce Goddard, who — per his own former staff’s account — rented an accused man (Paul Fox) a U-Haul and told him to leave the state, after which Fox was working with teens at First Baptist Hammond within a year; Fox was not charged in that account, and it is recorded as internal-handling evidence only. SBC: Gilyard moved serially across Texas, Oklahoma and Florida churches despite allegations known to senior leaders; the Chronicle found 35 convicted or accused men who continued in churches, sometimes re-offending in new communities.
2. Internal discipline, therapy, prayer or forgiveness — instead of police
Catholic: playbook steps 2, 3 and 7 — clergy-run inquiries, church-run “evaluation,” and no police report. Old Apostolic Lutheran: leaders sent Clint Massie to a therapist and held forgiveness sessions with child victims, never reporting despite a Minnesota legal duty to report, and the abuse continued. Assemblies of God: NBC found accused preachers sent to therapy rather than police and a confidential denomination-run “restoration” process. Jehovah’s Witnesses: the two-witness rule and confidentiality instructions channeled allegations into internal judicial committees; in Lancaster, Pennsylvania, leaders who received two victims’ reports never called police (Aviles-Garriga).
“For he is the minister of God to thee for good. But if thou do that which is evil, be afraid; for he beareth not the sword in vain: for he is the minister of God, a revenger to execute wrath upon him that doeth evil.”
Romans 13:4
In the text these institutions profess, punishment of crime belongs to the civil authority — not to internal discipline. Handling it in-house is not the biblical pattern; it is the departure from it.
3. Shaming, disbelieving or discouraging the people who report
SBC: Guidepost found reporters “ignored, disbelieved,” called “opportunists,” and abuse advocacy described internally as a “satanic scheme”; women reporting Gilyard were reportedly shamed, and one was told disclosure would embarrass her. Debbie Vasquez was made to apologize publicly for a pregnancy she said her pastor caused, without naming him. Assemblies of God: NBC documented a survivor who reported abuse by an elder being prayed over for “lying, demonic spirits.” IFB: the Star-Telegram documented accusers “ignored, doubted or blamed,” in a system where questioning the “man of God” was itself treated as wrongdoing.
“Take heed that ye despise not one of these little ones; for I say unto you, That in heaven their angels do always behold the face of my Father which is in heaven.”
Matthew 18:10
The people these systems shamed for speaking are, in the text's own terms, the ones no one is to despise.
4. NDAs and confidential settlements
LDS: AP’s recordings documented a $300,000 offer conditioned on a confidentiality agreement and destruction of the family’s recordings (Goodrich); the Jensen settlement (2018) was confidential. Evangelical camp: Kanakuk offered Newman’s victims’ families settlements with NDAs after his arrest (Newman). One survivor who signed later died by suicide; Texas’s 2025 “Trey’s Law,” named for him, now bans NDAs in child-sex-abuse civil settlements. Catholic: the Boston archdiocese secretly settled several Shanley lawsuits years before his arrest. Gateway: the survivor testified she rejected a $25,000 offer tied to an NDA (Morris).
5. Re-hiring or restoring access after warnings — or convictions
Assemblies of God: Timothy Scarr was restored to ministry five days after leaving prison, by his father the senior pastor, and re-offended; Shawn Butler was hired as a youth minister despite a 1990 child-sex-assault conviction; NBC counted ~30 post-accusation placements. SBC: convicted pastors Mason and Jones returned to or founded pulpits; Gilyard preached again after prison; 9 men on the Executive Committee’s own secret list remained in ministry. Jehovah’s Witnesses: Campos was promoted twice after elders knew of his confession. Catholic: AP’s ~1,700 unsupervised credibly accused former clergy (above) is the post-removal form of the same access problem.
6. Record destruction, secret files, and refusal to produce records
Catholic: Pennsylvania’s files sat in bishop-only “Secret Archives”; at Msgr. Lynn’s sentencing, the judge found that a 1994 list of accused priests Lynn drafted was ordered destroyed by Cardinal Bevilacqua (see Lynn). Jehovah’s Witnesses: Watchtower paid $4,000-per-day sanctions rather than produce database materials — the tab was reported at ~$2 million at one point — saw a trial court enter a $4 million default for non-production in 2015, and settled the demanding cases confidentially in early 2018. LDS: AP reported help-line records were regularly destroyed (the church disputes aspects of AP’s help-line characterization); the Goodrich offer was conditioned on the family destroying their recordings. SBC: the Executive Committee’s list was secret for more than 10 years and released, redacted, only after Guidepost exposed it.
“For there is nothing covered, that shall not be revealed; neither hid, that shall not be known.”
Luke 12:2
The archive's working text — and the reason it keeps copies of everything, with dates and hashes, whether the original survives or not.
7. No central registry
SBC: leaders resisted a registry for decades, citing local-church autonomy and liability — while privately keeping the secret list; the public Ministry Check database its messengers mandated after 2022 never published a single name — in February 2025 the Executive Committee’s president, Jeff Iorg, said it was “not a focus for us,” and when this archive checked the SBC’s abuse-prevention site on October 25, 2026, it offered churches a “Ministry Toolbox” and no searchable database of names at all. IFB: by design there is no headquarters; offenders move through informal fellowship and Bible-college networks. Assemblies of God: a voluntary fellowship; mandatory background checks, mandatory reporting, and a credential bar for convicted child abusers were debated (1997, 1999, 2019, 2021) and not adopted, with liability cited. Catholic contrast: hierarchy exists and diocesan lists were published after 2018 — but with no common “credibly accused” standard, dozens of dioceses initially not publishing, and documented omissions. Centralization alone did not produce a complete registry.
8. Liability protection as the stated internal priority
SBC: Guidepost found Executive Committee leaders’ and lawyers’ “primary concern was to avoid any potential liability for the SBC,” and database and background-check proposals were blocked or softened on liability grounds. Assemblies of God: NBC reported mandatory safeguards were not adopted because leaders argued they would increase legal risk. LDS: AP documented abuse reports routed through a Risk Management division and the church’s litigation law firm. Jehovah’s Witnesses: the 1997 letter that created the database was prompted, per the letter’s own framing as reported, by concern about the legal risk posed by child molesters in the ranks — and the resulting records were kept from police.
How this archive records it
A pattern claimed on the strength of one case is an anecdote. A pattern documented across cases, denominations and decades is a finding about how institutions behave. Each case file in this archive records the behaviors above — transfer instead of report, internal handling instead of police, non-response to records requests, sealed settlements — but only where that case’s own documentation supports recording them. The definitions live on this page; the evidence lives in the case files.
Open questions
Tier 3 · Open questions
There is no verified national total, and this site does not offer one. No DOJ or FBI series counts “clergy offenders” as a category, and no academic consensus estimate of total church-related child sexual abuse in the United States was found in this research. The datasets above overlap in method and differ in definitions, periods and duplication; adding them together would manufacture a number, not discover one. Any single “total” for all U.S. churches would be an extrapolation.
Charges understate incidence — structurally, not marginally. Four documented reasons:
- Statutes of limitation. Pennsylvania, 2018: of more than 300 accused priests, only 2 could be charged; more than 100 were dead. Andy Savage’s acknowledged 1998 incident was not prosecuted in 2018 for the same reason (see Savage).
- Non-reporting and delayed disclosure. Child sexual abuse is typically disclosed years or decades later, if at all. As background across all settings (not churches specifically), the CDC estimates 1 in 4 girls and 1 in 13 boys in the U.S. experience child sexual abuse (CDC 2024, via the National Children’s Alliance). Church-specific non-reporting is documented in behaviors 2 and 3 above.
- Death, flight and plea reduction. Accused men died before investigation (John Jay: ~3,300), died in custody before trial (Adams), fled the country (Campos), or pleaded to non-sex-offense misdemeanors that leave no registry entry (Aderholt) — each removes a case from “convicted sex offender” counts.
- No denominator and no registry. See behavior 7.
What happened when time bars were lifted. Lookback windows show what the charge record hides — with a caveat: civil claims are allegations tested, if at all, under civil standards, often resolved by global settlement without individual adjudication. They are not criminal findings.
- New York Child Victims Act (window August 2019–August 2021): across all institutions — schools, churches, camps and others — court-filing counts found 10,787 cases filed by 14,592 plaintiffs. Against Catholic institutions specifically: six of New York’s eight dioceses filed Chapter 11, those dioceses were together named in at least 2,800 abuse claims, and bankruptcy plans agreed to pay at least $1 billion. Buffalo alone reported ~920 claims.
- California AB 218 (three-year window, 2020–2022): produced ~4,000 claims against California Catholic dioceses. In October 2024 the Archdiocese of Los Angeles agreed to pay $880 million to settle 1,353 claims — on top of more than $740 million paid in earlier settlements — most alleging abuse more than 50 years ago that had been time-barred.
Figures this archive treats as contested or approximate, and labels wherever they appear: the Jehovah’s Witness database’s size (“tens of thousands”; 23,720) — estimate and leak-based, never court-verified. FLOODLIT’s 4,000+ — reports catalogued, statuses mixed. BishopAccountability’s 7,500 versus the bishops’ own 6,721 — two counts of the accused, compiled differently; the gap is itself documented. Prosecutors’ estimate of “hundreds” of Newman victims, and the “at least 22” alleged Jensen victims — estimates and allegations beyond the counts proved at trial. Pennsylvania’s “real number in the thousands” — the grand jury’s stated belief, not a count.
Compiled October 18, 2026, from the pattern research file (sources linked throughout; research notes retained by the archive). Updated as case files are verified against primary sources.