Editorial Policy

How CLCI Hub decides what to publish, what we will not publish, the terminology rules our writers apply, and the editorial principles that govern every group profile, blog post, and glossary entry.

Mission

CLCI Hub exists to help survivors, families, researchers, journalists, students, clinicians, and policymakers understand coercive-control mechanics in religious, spiritual, wellness, ideological, and political organisations. We score groups on a transparent 0–40 spectrum derived from Steven Hassan's BITE model rather than applying binary “cult” / “not cult” labels.

Terminology policy

We do not call any group “a cult” as a definitive label. The word is contested, theologically loaded, and legally fraught. Instead we use:

• high-control group or high-demand group when describing organisations whose documented operational mechanics fit the BITE pattern;
• cult-like control patterns or coercive-control patterns when describing specific behaviours;
• new religious movement (NRM) for academic-register references to recently-founded religious traditions, regardless of CLCI score;
• mainstream tradition when distinguishing the broader religious tradition from a specific high-control sub-branch.

We distinguish mainstream traditions from high-control sub-branches in every applicable profile. Mainstream Catholicism and a specific religious order under Vatican apostolic visitation receive different entries with different scores; the Catholic Church as a whole is not graded as a single entity.

Public-source-only policy

Every factual claim in every profile is sourced from material that is publicly accessible: court records, government investigations, peer-reviewed academic work, reputable journalism, organisational public statements, public ex-member testimony, documentaries, and books. We do not publish private allegations, leaked private documents we cannot independently verify, or material from sealed proceedings.

We do not fabricate URLs. Where a source is well-documented in print or court records but does not have a stable public URL, we cite it by name and outlet and offer a Google Scholar search link as a verify-path for academic sources.

Source hierarchy

Source weight follows the hierarchy detailed on the Source Policy page. Briefly: court records and government investigations rank highest; peer-reviewed academic work and reputable journalism rank above public ex-member testimony; and all of these rank above anonymous claims or single-source allegations.

Low-confidence ratings are marked explicitly. A score may be assigned with Medium or Low confidence when documentation is sparse; readers should treat such ratings as provisional pending further public-record evidence.

How contested claims are handled

Where a group disputes a specific claim or where reasonable observers disagree, we present the dispute itself rather than picking a side. A profile may note: “X is alleged by Y; the organisation denies Z; the matter remains unresolved in public record.” Allegations under active litigation are described with neutral language and dated qualifiers.

How CLCI avoids blanket labels

Two structural features of the CLCI prevent inappropriate blanket labelling. First, every score is a sum of four independently-measured BITE axes plus a signed modifier — readers can see which axis drives a high score and which does not. Second, every score carries an explicit confidence rating from High (court records + academic work + multiple journalistic investigations) to Low (fragmented anecdotal reports). Low-confidence ratings are provisional and clearly marked as such.

How corrections work

We accept correction requests from anyone — current members, former members, researchers, family members, journalists, and named organisations. The Corrections page details the process, what evidence is accepted, and the public changelog policy. Organisations that wish to respond to specific characterisations may use the Right of Reply process.

How AI is and is not used

AI tools (large language models) are used to summarise public material at scale and to draft initial outlines. Every published claim is human-reviewed; we do not publish unreviewed AI output, and we do not allow AI tools to generate citations that we have not independently verified. We do not run user-facing AI tools that incur ongoing API costs. The methodology page documents the review pipeline in detail.

Conflicts of interest

CLCI Hub is independently operated. We do not accept advertising from religious organisations, ex-member-recovery for-profit therapy networks, or media outlets that we cite. Where the operator's personal history intersects with a group covered by the dataset, that connection is disclosed in the relevant profile's editorial notes.

No paid removals

We do not accept payment, advertising, sponsorship, donations, or any other consideration in exchange for removing a profile, lowering a score, removing a specific claim, or softening editorial language. There is no rate card and no back channel. Offers of payment in exchange for editorial changes are treated as material relevant to the Transparency Report and may be cited in the profile of the offering organisation.

The same standard applies to in-kind offers: free legal services, speaking-circuit invitations, access in exchange for editorial veto, and comparable benefits are not accepted. Legitimate access (interview invitations, on-record statements for inclusion in profiles, document disclosures) is welcome, separate from any editorial consideration.

No score manipulation

We do not adjust scores to drive traffic, to align with partner organisations, to please survivor networks, to penalise critics of our methodology, or to match the prevailing direction of public sentiment about a group. Every score change requires a documented evidentiary basis under the scoring-appeals framework, logged in the public changelog.

We do not run A/B tests on scores, do not adjust scores in response to social-media campaigns either supporting or attacking a group, and do not accept score-change requests from anyone — including the operator's personal contacts — outside the documented appeals process. Where a former member, current member, family member, journalist, researcher, or representative of the organisation believes a score should change, the appeals route is the same for all parties.

Sensitive allegations policy

Some categories of allegation — child sexual abuse, sexual assault of adults, forced labour, financial exploitation of vulnerable members, domestic violence within group leadership, retaliation against critics — carry both higher evidentiary scrutiny and higher editorial care. Our standard for publishing claims in these categories:

• We report court findings and government-inquiry findings in our own voice with case name, jurisdiction, and disposition. We do not soften the language of a documented finding to spare the organisation.
• We report allegations as allegations, attributed and dated, with the organisation's public response where one exists. We do not treat ex-member testimony of serious abuse as automatically false; we also do not treat it as automatically true.
• We do not identify alleged victims who have not chosen to be named publicly themselves. Where the alleged victim has chosen pseudonymous or anonymous routes, we respect that choice even when their name appears elsewhere.
• We do not name alleged minor victims beyond what the formal court record requires, and typically use less detail than the court record does.
• We do not publish single-source allegations against named living individuals. A single anonymous online account, a single deposition not tested by cross-examination, or a single contested journalistic claim is insufficient on its own for a serious allegation against a named living person. See the Living Persons policy for the full source-strength requirements.
• Where allegations are under active litigation, we use neutral, dated qualifiers and link to the public-record filings rather than restating allegations in our editorial voice.

Sensitive-allegations editorial decisions are documented internally with the reviewing editor's notes; the final published language is the only public surface, but the underlying reasoning is auditable on request.

Religious neutrality

We score operational mechanics, not religious truth. We have no editorial position on whether a doctrine is true, a tradition is salvific, or a practice is sacred. A high CLCI score does not imply that the group's underlying theology is false; a low CLCI score does not imply that the group's theology is endorsed.

We deliberately distinguish mainstream traditions from high-control sub-branches within those traditions. We score specific organisational entities — a particular order, a particular branch, a particular era — and not entire religions. We use the same evidentiary framework for groups in Christianity, Islam, Judaism, Hinduism, Buddhism, Sikh, Bahá'í, Jain, Taoist, Shinto, pagan, new-religious-movement, and secular categories. We are an educational site, not a missionary or counter-missionary platform, and we welcome readers from across the religious spectrum, including readers from within the groups we cover.

Political neutrality

We apply the same framework to political and ideological organisations that we apply to religious ones. Political movements appear in the dataset when their documented operational practice fits the BITE-control pattern; they are absent when it does not. Mainstream political participation, ordinary partisan organising, and routine ideological debate are not material for the dataset.

We do not hold an editorial position on the policy correctness of any political tradition we cover. We do not score conservative groups differently from progressive ones, religious-right groups differently from revolutionary-left ones, mainstream parties differently from outsider-fringe ones, or movements aligned with the operator's own political preferences differently from those opposed. Where political organisations engage in the same coercive-control mechanics that high- control religious groups do, they are scored under the same framework.

Compassionate framing

We write for readers who may include current members of the groups we cover. Profiles avoid sensational, graphic, or unnecessarily distressing detail; we prioritise educational clarity over dramatic effect. We never imply that every member of a group is a perpetrator or victim — the spectrum of member experience is wide, and our editorial language reflects that.

This site is educational and not legal, medical, or clinical advice. See the Legal Disclaimer for the full statement.