As MPs and peers examine the Draft Conversion Practices Bill, a report by the LGBT+ charity Galop has entered the Government’s evidential case for new criminal offences. Its published statistics contain errors, its accounts of religious activity lack essential context, and the official estimates of the problem rest upon assumptions which the Government itself acknowledges may be unreliable. Before Parliament legislates, these matters require answers.
On Thursday 29 October, representatives of Galop and Stonewall appeared before the parliamentary Joint Committee examining the Government’s Draft Conversion Practices Bill. Among those giving evidence were Kai O’Doherty, Galop’s Director of Policy and Research, and Professor Ilias Trispiotis of the University of Leeds, a member of the Ban Conversion Practices Coalition’s steering group. Galop’s research has already found its way into the Government’s official impact assessment, where it is cited as evidence that contemporary conversion practices involve physical and psychological abuse. The charity’s latest report, Still Not Illegal: Evidence of Modern-Day Conversion Practices from Galop’s Frontline Services, is consequently more than another campaigning publication. It belongs to the body of material upon which Parliament is being invited to create new criminal offences.¹ ²
That gives the report a significance which its authors, and those relying upon their conclusions, cannot reasonably evade. A charity providing confidential support to people reporting abuse must be able to receive accounts sympathetically, without demanding that callers prove their allegations before receiving assistance. Parliament has a different responsibility. It is considering legislation capable of exposing individuals to prosecution, restricting their activities through civil protection orders, and altering the manner in which parents, clergy, teachers and therapists discharge their responsibilities. Evidence offered for that purpose must be sufficiently reliable to distinguish what has been reported from what has been established, what is already prohibited from what remains lawful, and what constitutes genuine abuse from conduct which someone has interpreted as abusive because of a disagreement over religion, sexuality or identity.
Readers of Nuntiatoria will recognise the underlying concern. Our examination of the proposed legislation in May considered the danger of treating non-affirmation as evidence of abuse. When the draft Bill appeared in June, we questioned whether Parliament could safely legislate amid substantial clinical uncertainty. In July, our reporting on Tavistock, WellBN and safeguarding examined the practical consequences for parents and professionals who question the assumptions surrounding gender identity. Those arguments remain pertinent. What has changed is that the Government has now supplied a more substantial evidential record, and Parliament has begun testing it. The question is no longer confined to what the legislation might do. We can examine what its supporters say is actually happening, and whether the evidence supports the conclusions drawn from it.³
Galop’s report presents itself as a study of modern-day conversion practices drawn from its frontline services. Its researchers examined records relating to more than 13,500 clients and over 14,800 cases handled between November 2022 and November 2025. Using recorded categories of abuse and keyword searches, they identified 371 potentially relevant cases, eventually selecting 195 cases concerning 191 individuals. The final selection was made according to the depth and detail of the available case notes. Those records included correspondence, assessments and accounts gathered in the course of providing support. They were analysed using qualitative research software, with the findings subsequently checked by members of Galop’s own frontline teams.⁴
There is nothing inherently objectionable about such research. Specialist charities encounter experiences that may otherwise escape public scrutiny, and case-note analysis can reveal recurring difficulties in the support system. But the nature of the material places limits upon what may responsibly be claimed. Galop’s researchers did not undertake a nationally representative investigation. The report does not describe systematic independent corroboration of the allegations, interviews with those accused, or external investigation of the religious institutions concerned. Practitioner validation may help establish whether researchers understood their colleagues’ records correctly; it does not establish that every recorded allegation occurred as described, that the motive attributed to the alleged perpetrator was accurate, or that the researchers possessed the necessary expertise to identify a particular religious rite.
Galop acknowledges that its sample is not statistically representative and should not be used to estimate prevalence. Yet its accompanying publicity describes the research as providing clear evidence that conversion practices are widespread across the United Kingdom. The distinction is consequential. A deliberately selected collection of serious or detailed cases may demonstrate the existence of reported conduct, illustrate how it is experienced, or identify possible safeguarding failures. It cannot establish how frequently that conduct occurs throughout the population. Nor can a collection of accounts, however disturbing, establish which new statutory offences are necessary without examining whether the alleged conduct is already unlawful.
The report’s own numerical presentation provides an immediate reason for caution.
On page 22, a table attributes religion-based conversion practices to 30 transgender cases and 22 cases involving cisgender lesbian, gay or bisexual clients, giving percentages of 30 and 27 respectively. The percentages do not correspond to the numbers. On page 35, the counts are reversed, producing the stated proportions: 22 transgender cases, representing approximately 30 per cent of that group, and 30 cisgender LGB cases, representing approximately 27 per cent. The report cannot be correct in both places.
The discrepancy concerning physical violence is more serious. Pages 22 and 40 record 30 transgender cases as 22 per cent, and 16 cisgender LGB cases as 27 per cent. Yet the respective group totals are 74 and 111. Thirty out of 74 is approximately 41 per cent; 16 out of 111 is approximately 14 per cent. Reversing the case counts would reconcile them with the published percentages. Until the underlying data are checked, however, the reader cannot know whether the numbers or the percentages were transposed. The published figures are plainly inconsistent, and the difference is large enough to alter the apparent distribution of alleged violence between the groups concerned.
Other mistakes follow. Page 27 reports 76 cases of psychological control and emotional manipulation as 33 per cent, although the preceding table records 64 cases, which is approximately 33 per cent. Sixteen cases of economic abuse become 12 per cent on page 33, although they represent approximately eight per cent of the sample, as the earlier table correctly states. Page 42 describes 25 cases involving psycho-medical practices as eight per cent, while the accompanying table correctly gives 13 per cent. Even the number of cases involving a single incident varies between pages 10 and 23, which report 37 and 36 respectively. These discrepancies may result from transcription, editing or the reversal of columns rather than mistakes in the original coding. They nevertheless require correction. This is a report being cited in the official evidential justification for criminal legislation, not a document whose figures can safely be left unresolved.
The treatment of religious activity warrants equally exacting examination, although the difficulties here are rather different. Galop identifies 52 cases, approximately 27 per cent of its selected sample, as involving religion-based practices. Twenty-three reportedly involved activity within homes, while 16 involved religious or faith-based settings; the categories can overlap. In 76 per cent of all cases, the alleged behaviour was initiated or carried out by family members, particularly parents. Consequently, the report does not describe 195 instances of institutional religious activity, nor even 52 established cases of misconduct by churches or religious ministers. The distinction matters because the report repeatedly refers to religious practice without establishing the relationship between the individuals involved and any recognised religious institution.⁴
On page 15, the researchers admit that information about clients’ religious affiliations was inconsistently recorded and that religion was therefore excluded from the demographic analysis. This is a remarkable limitation in a study which subsequently devotes an entire section to religion-based conversion practices. We are not told systematically how many cases involved Catholics, Anglicans, Baptists, Methodists, Pentecostals, independent evangelical fellowships, African-initiated churches, Muslim communities or other religious traditions. Nor does the published methodology describe independent theological or ecclesiastical review of the relevant accounts. The report’s advisory board included legal and academic expertise, but that is not the same as a documented assessment by specialists competent to distinguish the authorised practices of particular religious bodies.
An investigator examining alleged misconduct in a school would be expected to understand the difference between a headteacher, a governor, a visiting speaker and a parent acting privately. The same care is necessary when examining religious institutions. A Catholic priest, an Anglican priest, a Baptist minister and an independent Pentecostal pastor do not exercise interchangeable ministries. Their authority arises through different structures; their official teaching and ritual practices are not identical. There are Christians who observe elaborate ceremonial, others who reject it, and still others whose local practices combine elements drawn from several traditions. A description which merely says that a pastor performed a cleansing, or that church members conducted an exorcism, tells us remarkably little about whether the activity was authorised, improvised, prohibited by the denomination concerned, or even accurately identified.
Consider the account attributed to Client U4 on page 39. The individual reportedly spent two days in a church, fasting, reading Scripture and holding candles, and was allegedly struck with a Bible by a pastor at home to expel a demon. Being struck with a Bible is not an authorised rite of the Catholic Church, the Church of England or mainstream Protestant worship. If the allegation is substantiated, it concerns physical violence. But what is the religious identity of the church? Was it an independent charismatic congregation, an established denomination or an informal religious group? Two-day prayer vigils and fasting are familiar in some Christian communities, although their precise customs vary. Candles feature prominently in certain traditions and far less frequently in others. None of these details makes the account impossible. They make denominational identification essential if the researchers intend to draw conclusions about the religious character of the events.
Another account, Client B6, concerns church members allegedly pinning down a woman while Scripture was read over her. Galop classifies the incidents as forced exorcisms. Physical restraint against someone’s will is a serious allegation, but the term exorcism requires examination. In the Catholic Church a solemn exorcism is subject to ecclesiastical authority and cannot legitimately be undertaken by anyone who simply considers himself capable of performing it. Canon 1172 of the 1983 Code requires a priest to possess special and express permission from the local Ordinary, preserving a discipline already familiar in earlier Catholic canon law. The Church of England likewise regulates deliverance ministry, requiring appropriate authorisation and safeguarding. Its published guidance explicitly prohibits deliverance intended to alter sexual orientation or gender identity.⁵ ⁶
If members of either communion carried out the intervention described, it would be necessary to establish whether they were acting in defiance of their own Church’s rules. If the event occurred in an independent fellowship, different questions would arise about authority and accountability. A person’s description of an incident as exorcism does not automatically make it an example of an authorised ecclesiastical rite. Galop’s analysis does not supply the information needed to resolve these distinctions.
Other descriptions are similarly uncertain. One client says that a parent involved various pastors and priests in attempts at spiritual cleansing. Were these ministers of different denominations? Did they participate in a formal rite, offer informal prayers, or simply speak with the family? Another account describes an imam reciting the Quran over an individual and a separate incident involving needles and the extraction of blood by a woman. The latter description might correspond to a form of traditional wet cupping, but the evidence presented does not establish the procedure or show that the imam performed or authorised it. There are legitimate questions here about religious practice, traditional medicine, coercion and physical safety. They cannot be answered satisfactorily by collecting the events under a general heading of religious cleansing.
More troubling is the ease with which ordinary prayer and scriptural discipline appear within the same analytical category as alleged violence. Case J1 describes someone questioning sexuality and gender identity who consulted a pastor and subsequently received what the notes characterise as conversion practices, identified simply as prayer. The extract does not explain whether the individual requested prayer, what was said, or what made the encounter abusive. Case X3 describes a church encouraging someone to replace unwanted sexual thoughts with biblical verses. The method may have been coercively imposed; the reported words of those involved certainly invite closer questioning. Yet meditation upon Scripture, the discipline of thoughts and recourse to prayer in moments of temptation are familiar elements of Christian spiritual life. It is necessary to establish how the practice was undertaken before concluding that the spiritual technique itself constituted abuse.
This is particularly important from a Catholic perspective. The Church teaches that every person possesses an intrinsic dignity, that moral freedom entails responsibility, and that chastity is required according to one’s state of life. She distinguishes temptation from deliberate consent, inclination from action, and the person from the moral quality of particular conduct. A priest who explains these distinctions in confession or spiritual direction is exercising an ordinary pastoral office. A Catholic who voluntarily asks for help in living chastely is not thereby consenting to mistreatment. The Church’s teaching does not depend upon an individual changing his sexual orientation; nor does a minister’s refusal to affirm every claimed identity establish an intention to subject that individual to psychological coercion. Such distinctions must not disappear merely because the moral teaching involved is contested.
The report itself briefly acknowledges that religious and cultural beliefs do not automatically amount to conversion practices. That acknowledgement is welcome. Unfortunately, it is not accompanied by a consistent method for deciding when an ordinary religious conversation becomes an attempt to alter identity, when parental guidance becomes coercion, or when spiritual counsel becomes an abusive intervention. The concern is especially acute in the therapeutic examples. On page 44, one client reports that a therapist suggested healing trauma might result in the person no longer identifying as transgender. That may have been an attempt to steer a patient towards a predetermined conclusion. Alternatively, the remark may have arisen during exploratory discussion of the relationship between trauma, distress and identity. The published extract does not establish which. It gives no treatment plan, transcript, professional explanation or independent clinical assessment. Yet the event is included as evidence of psychology-based conversion practices.
A report seeking to influence clinical law and policy ought to recognise that such distinctions are central to responsible psychotherapy. The Cass Review and subsequent changes to NHS gender services have reinforced the importance of holistic assessment, particularly for children and young people presenting with complex psychological, developmental and social difficulties. An exploratory clinical hypothesis is not necessarily correct; it must be evaluated according to evidence and professional standards. But neither should it become evidence of professional abuse merely because its possible implications are unwelcome. Our earlier examination of Tavistock and WellBN raised precisely this danger: the possibility that the legal language of conversion might be used against the sort of careful assessment that modern safeguarding increasingly requires.³
There is a further conceptual mismatch between Galop’s research and the Bill to which it is being applied. The charity employs a broad definition encompassing attempts to change, cure or suppress sexual orientation or gender identity. The Government’s current draft, by contrast, is expressly directed at acts intended to cause a person to have or not have a particular sexual orientation or transgender identity, or to believe that this is the case. Its equality impact assessment says that the measure is directed at change rather than suppression, partly to reduce interference with freedom of religion and expression.⁷ That difference is of considerable importance. Conduct counted as conversion practice by Galop is not necessarily conduct caught by the proposed statutory definition, still less conduct satisfying the separate tests of abuse and serious harm. The report does not provide a case-by-case assessment establishing how many of its 195 examples would actually meet those requirements.
The comparison with official criminal statistics exposes another evidential difficulty. The Government acknowledges in its own impact assessment that conversion practices are not captured within a dedicated official statistical series or administrative dataset. There is therefore no national police-recorded total of conversion-practice offences against which Galop’s findings or the campaigning organisations’ prevalence estimates can be directly checked. The absence of such figures does not prove that the conduct is rare; much abuse goes unreported. But it also means that Parliament must not confuse estimates derived from surveys with established numbers of criminal incidents.²
The Home Office does maintain records relating to forms of conduct which Galop identifies within its case notes. In the year ending March 2025, police in England and Wales recorded 2,949 offences associated with so-called honour-based abuse, including 125 forced-marriage offences. There were also 2,047 reported honour-based-abuse incidents which, following assessment, did not amount to recorded crimes. The Home Office cautions that changes in reporting and recording practices affect these figures and that police data do not capture every offence occurring in private. Nevertheless, the statistics demonstrate that relevant forms of familial coercion are already recognised within the criminal justice system.⁸
Galop records 60 cases involving efforts to arrange or compel marriage, of which 20 involved planned, attempted or completed forced marriages. That is potentially significant. But an account of family pressure to marry, an actual attempt to force a marriage and a completed forced marriage represent different factual and legal circumstances. The 60 cases cover three years, whereas the Home Office figures concern a single reporting year. The figures cannot simply be placed alongside one another as though they measure identical conduct. Their proper use is to identify what Galop has classified more broadly, what the police already investigate, and where any demonstrable gap remains.
The Forced Marriage Unit’s figures are illuminating in the same limited but important sense. During 2025, the joint Home Office and Foreign, Commonwealth and Development Office unit received 1,295 contacts concerning possible forced marriage or female genital mutilation. It provided tailored assistance in 406 cases, of which 391 involved forced marriage and 15 female genital mutilation. Five cases were recorded in which the victim voluntarily identified as LGBT+ and this was a driving factor in the forced marriage. The Unit explicitly warns that this information is incompletely collected and that the true number may be higher.⁹ There is no legitimate basis for concluding that Galop’s accounts are exaggerated merely because the official figure is smaller. But neither may Galop’s broader classification be used to imply that every reported parental effort to encourage marriage represents a completed criminal offence. Parliamentary scrutiny must establish what the conduct actually involved.
The latest police-recorded hate-crime statistics provide further context. On 8 October, the Home Office reported that, in the year ending March 2026, police in England and Wales recorded 21,407 hate crimes involving sexual orientation, a decrease of one per cent from the previous year, and 3,955 involving transgender identity, a decrease of four per cent. Overall recorded hate crime increased by seven per cent to 146,825 offences, with increases particularly in racial and religious hate crime.¹⁰ None of these figures supplies a direct estimate of conversion practices. Hate-crime recording depends upon an offence being identified and the relevant hostility characteristic being recorded, while some alleged conversion practices might not be motivated by the hostility contemplated in hate-crime recording rules. The figures do, however, provide an independently maintained picture of police activity that should form part of any serious account of how existing offences are being used.
The most arresting numerical claim in the Government’s own case concerns neither police-recorded offences nor Galop’s case notes. The official impact assessment estimates that approximately 75,000 to 93,000 people experience conversion practices annually in England and Wales. A reader might reasonably suppose that a figure of such magnitude derives from extensive administrative records or a representative investigation identifying tens of thousands of alleged victims. It does not. The Government obtained the estimate by extrapolating answers to two questions in an Opinium survey commissioned by Stonewall: whether respondents had been told to ingest purportedly purifying substances and whether they had received what was described as pseudo-scientific counselling. The survey concerned experiences reported within the preceding twelve months and used weighting intended to produce a representative sample. The underlying methods deserve to be examined on their own terms. The Government’s estimate is nonetheless an extrapolation from self-reported answers, not a count of verified incidents.²
The calculation is straightforward enough to reproduce. The Government combined the reported proportions for those two activities to produce a central estimate of 4.7 per cent. It then applied an uncertainty allowance of half a percentage point in either direction, yielding a range of 4.2 to 5.2 per cent. Those percentages were multiplied by an estimated LGBT+ population of 1,784,842 adults in England and Wales, producing approximately 75,000 to 93,000 people annually. The numerical multiplication is not the central difficulty. The crucial assumptions concern what respondents understood by the terms purifying substances and pseudo-scientific counselling, whether their experiences met the proposed legal definition, how far the two categories overlapped, and whether the survey proportions can reliably be projected onto the wider population.¹¹
The Government itself admits that some of the experiences included in its estimate may fall outside the proposed offence. It also acknowledges that the number of providers of abusive conversion practices is unknown. For comparison, the earlier National LGBT Survey produced an unweighted lifetime estimate of 2.4 per cent having undergone conversion therapy, subsequently revised through weighting to 2.9 per cent, with five per cent reporting that it had been offered. The survey was self-selecting and employed a different approach to defining the conduct.¹² These figures cannot establish a rising or falling trend without first resolving the differences between the research instruments. A lifetime measure from one survey cannot be compared with a twelve-month measure from another as though the two were testing the same proposition.
There is a further problem with the apparent precision of the Government’s estimates. The official impact assessment expressly acknowledges criticism of an earlier survey question concerning exorcism, for which no definition was supplied. Respondents may have understood the term differently, and the Government notes published concerns about the resulting figures. This is particularly relevant to Galop’s own religious classifications. If researchers do not establish what a participant means by exorcism, and cannot identify the religious institution allegedly responsible, their count may combine very different events. A formal rite, informal deliverance prayer, a family member reading Scripture and a violent restraint incident cannot responsibly be treated as equivalent merely because the same word has been applied to them.
Most consequential of all is the Government’s attempt to translate the supposed prevalence of conversion practices into anticipated health benefits. Annex C of the impact assessment relies upon a United States review which found depression reported among 65 per cent of young people exposed to sexual orientation or gender identity change efforts, compared with 27 per cent among those not exposed. The Government treats the difference of 38 percentage points as an estimate of additional depression attributable to those practices. Applying that figure to its projected annual population, it calculates that the proposed legislation would prevent between 28,486 and 35,268 cases of depression every year. It subsequently models the monetary benefits of avoiding depression, arriving at a best estimate of £334.7 million annually before reducing the value by 75 per cent to £83.7 million to account for uncertainty.¹³
That construction deserves much closer examination than it has received. A statistical association between an experience and depression does not, without further analysis, establish how much depression the experience caused. It is still less capable, by itself, of establishing how much depression a new criminal statute will prevent. The American research concerns a different population and setting, and the Government extrapolates its findings to LGBT+ adults in England and Wales irrespective of age. Behavioural responses to legislation are uncertain. Some perpetrators might desist; others might continue secretly or employ different forms of coercion. The likelihood of reporting, investigation and conviction remains unknown.
The extraordinary feature is that the Government’s own impact assessment acknowledges these problems. It states that there is no robust evidence establishing causality between conversion practices and depression, but that causation has been assumed for modelling purposes. It recognises that the American study did not adequately control for confounding factors, that its external validity is uncertain, and that no reliable UK figure exists for the proportion of people experiencing depression as a result of conversion practices. The 75 per cent reduction in monetised benefits is intended to reflect these uncertainties. Yet reducing a financial projection does not establish the causal relationship upon which that projection depends.¹³
No one needs to deny the possibility of profound psychological harm from coercive or degrading treatment to recognise this weakness. A person subjected to threats, humiliation, isolation or violence may experience serious and lasting distress. The medical and social consequences deserve careful investigation and effective support. But where a Government proposes to quantify the annual benefits of legislation in tens of thousands of prevented cases of depression, Parliament is entitled to demand evidence that the law itself can reasonably be expected to produce those outcomes. The same rigour applied to the calculation of public expenditure must be applied to the assumptions by which supposed benefits are justified.
There is also a curious contrast between the confidence of the projected benefits and the uncertainty surrounding the basic incidence of the alleged conduct. The Government does not know how many organisations are undertaking abusive conversion practices. It possesses no comprehensive official recording system for such incidents. Its chosen survey measures only two forms of reported experience, not all of which would necessarily meet the statutory threshold. Yet the modelling proceeds to estimate substantial health and economic benefits from prohibiting them. These estimates may be useful as preliminary policy scenarios, provided their uncertainty is understood. They cannot be treated as independently established facts demonstrating the necessity or likely effectiveness of the Bill.
The legal position is more complicated than either the campaign to ban conversion practices or some of its opponents have sometimes acknowledged. Assault, sexual offences, forced marriage, unlawful confinement, stalking and harassment can already be prosecuted where the relevant elements are proved. Section 76 of the Serious Crime Act 2015 addresses repeated or continuous controlling or coercive behaviour in qualifying personal relationships, and changes introduced through the Domestic Abuse Act 2021 removed the previous requirement for relevant family members or former partners to be living together. The offence nevertheless has defined limits, including an exclusion concerning a person responsible for a child under sixteen, where other child-protection offences may apply. It does not follow that every non-violent act outside those relationships is adequately covered by existing law.¹⁴
The Government therefore has a legitimate question to put before Parliament: whether some seriously abusive conduct escapes existing offences because of the relationship between perpetrator and victim, the nature of the conduct, or the evidential requirements of current law. That question deserves an answer supported by concrete cases. If a complainant reported repeated intimidation by an unrelated religious leader, did the police conclude that no existing offence applied? Was the difficulty an absence of criminal prohibition, insufficient evidence of the alleged conduct, or unwillingness or inability to pursue the case? Did professional regulation or safeguarding procedures fail? These circumstances would point towards different remedies. Galop’s report does not systematically identify which cases demonstrate an actual deficiency in the law rather than a failure to use powers already available.
The draft Bill itself deserves to be read carefully. Its principal criminal offence does not simply prohibit religious disagreement, parental caution or a conversation that causes offence. It requires an intentional conversion practice amounting to abuse and causing serious physical or mental harm, or serious alarm or distress substantially affecting ordinary daily life. It proposes a maximum sentence of five years’ imprisonment on conviction on indictment, alongside offences concerning certain overseas conduct and a civil protection-order regime. The Government has expressly stated that legitimate healthcare, ordinary religious expression and responsible parenting should remain protected. Its provisions are intended to be symmetrical, applying to abusive attempts to alter identity or orientation in either direction.¹⁵
Those safeguards are relevant, and it would weaken the criticism to pretend that the Bill expressly criminalises a priest preaching chastity or a parent declining to affirm a child’s stated identity. The concern lies instead in the interpretation and application of the proposed tests, especially where psychological or emotional pressure is alleged. More immediately, the Bill provides for civil protection orders which may be imposed to prevent anticipated abusive conversion practices, including without advance notice to the person affected in specified circumstances. There are provisions for representations and appeals, and breaches of orders may become criminal offences. Such powers require close scrutiny because their practical consequences may precede any criminal conviction of the conduct originally alleged.
The implications for religious communities are particularly important. Ministers may sincerely intend to preserve lawful pastoral care. But statutes are administered by police officers, social workers, schools, safeguarding professionals, regulators and courts, whose understanding of religious teaching and practice may vary considerably. An imprecise classification of prayer or spiritual direction as abusive conversion activity could result in complaints or investigations even where the evidence would ultimately fall well short of the criminal threshold. The existence of such a possibility does not mean every complaint would succeed. It means that the clarity of definitions, procedural safeguards and professional guidance will matter greatly.
For Catholics, the principles are not difficult to state. No person should be physically intimidated, unlawfully restrained, sexually assaulted, threatened or manipulated under the guise of spiritual assistance. Clerical authority does not excuse cruelty. Where misconduct is alleged, it should be investigated in accordance with civil law and the Church’s own discipline. Equally, the freedom to teach Catholic moral doctrine, preach repentance, counsel chastity, hear confessions and provide spiritual direction is not a concession granted by the State to be withdrawn whenever religious teaching becomes unpopular. Parents likewise possess genuine responsibilities in the moral formation and protection of their children. Such responsibilities must be exercised with charity and prudence, but they cannot be reduced to a duty of affirmation.
What Parliament now needs from Galop is not another assertion that the reported experiences were distressing. It needs a defensible account of the research. The statistical errors should be corrected publicly. The coding criteria should be available for independent scrutiny, with appropriate protection for clients’ confidentiality. The researchers should explain how they distinguished completed conduct from threats and proposals, and how they separated allegations of coercion from ordinary religious teaching or voluntary therapeutic exploration. In the religious cases, the broad denominational identity, the status of the alleged minister, and the nature and authorisation of the purported ritual should be established wherever the records permit. Where the information is unavailable, that limitation should be stated rather than silently bridged by interpretation.
The Committee should also ask whether any relevant cases were independently corroborated, whether the available records contain police or safeguarding outcomes, and how the alleged conduct compares with the elements of existing criminal offences. Galop’s involvement in support for vulnerable individuals may well have given its staff access to important accounts of serious wrongdoing. It has not, however, relieved the researchers of the obligation to distinguish the purpose of a case note from the evidential standard necessary to justify criminal law. Confidentiality can be protected without abandoning methodological accountability.
Nor should scrutiny stop at Galop. The Government should explain why the Opinium survey provides a sufficiently reliable basis for its annual estimate, how much of the projected population would satisfy the actual statutory definition, and what evidence supports its assumptions about prevented depression. Its impact assessment discloses a number of these uncertainties with commendable candour. Parliament must ensure they are not subsequently lost beneath the apparent certainty of headline figures.
The Joint Committee’s call for written evidence closes at 5 pm on Monday 16 November 2026. Its chairman, Alex Barros-Curtis MP, has specifically invited submissions addressing the clarity, scope and definitions of the proposed offences. Those are precisely the matters raised by the Galop report and the Government’s impact assessment. The Committee is due to report by 22 January 2027, giving Parliament an opportunity to examine the evidence before any final legislation is introduced.¹⁶
The people whose experiences are described by Galop deserve to have serious allegations investigated properly. Those who have endured violence or coercive treatment should not be left without protection because of arguments about terminology or ideology. But the existence of genuine victims cannot remove the obligation to scrutinise evidence, and the seriousness of an allegation cannot establish that every interpretation subsequently placed upon it is correct. Victims have an interest in laws that work, as well as in laws that express sympathy.
The present debate has already acquired too much of the atmosphere of moral emergency, in which doubts about the proposed legislative remedy are too readily confused with indifference towards abuse. That atmosphere is ill-suited to the careful drafting of criminal offences. Parliament’s task is to identify conduct that can be defined, proved and punished with justice. It is entitled to hear from campaigning organisations, but it must retain sufficient independence to question their methods, their figures and their conclusions.
There is something troubling about the prospect of an individual being investigated under legislation justified partly by research whose own authors have not consistently reconciled their tables, identified the religious institutions involved, or established whether all the practices they describe would fall within the proposed offence. The Government’s elaborate forecasts of annual prevalence and prevented depression make the problem larger, not smaller, because the uncertainties multiply as the estimates move further from directly observed events.
Galop has submitted a body of accounts deserving examination. It has not supplied a series of judicial findings. Its statistical mistakes are real; its denominational ambiguities are substantial; and its conclusions cannot be transferred wholesale into the criminal law without examining the differences between its definition of conversion practices and the Government’s proposed offence.
The proper response is not to discard the report. It is to subject it to the scrutiny which its new parliamentary significance demands. That scrutiny must be no less rigorous because the subject is sensitive, the alleged victims vulnerable, or the advocates convinced of the righteousness of their cause.
Parliament is entitled to legislate against proven abuse. Before it legislates on the strength of this evidence, it should require those presenting it to answer for what their figures, their case notes and their classifications actually establish.
- UK Parliament, Draft Conversion Practices Bill: Oral Evidence, 29 October 2026.
- Office for Equality and Opportunity, Draft Final Stage Impact Assessment: Draft Conversion Practices Bill, updated 1 September 2026.
- Nuntiatoria, When Protection Becomes Compulsion: The Draft Conversion Practices Bill and the Threat to Conscience, Therapy, and Truth, 15 May 2026; Before the Evidence Settles: Labour’s Conversion Practices Bill and the Limits of Legislative Certainty, 26 June 2026; When Safeguarding Becomes “Conversion”: Tavistock, WellBN and the Legal Intimidation of Parents, 10 July 2026.
- Erin Carlisle, Still Not Illegal: Evidence of Modern-Day Conversion Practices from Galop’s Frontline Services, Galop, 2026, pp. 8–9, 14–15, 19–25, 34–44.
- Catholic Church, Codex Iuris Canonici, canon 1172, 1983.
- Church of England, Deliverance Ministry: Guidelines and Safeguarding Requirements, updated September 2023.
- Office for Equality and Opportunity, Draft Equality Impact Assessment: Conversion Practices Bill, 2026.
- Home Office, Statistics on So-Called ‘Honour-Based’ Abuse Offences, England and Wales, Year Ending March 2025, 6 November 2025.
- Home Office and Foreign, Commonwealth and Development Office, Forced Marriage Unit Statistics 2025, 28 April 2026.
- Home Office, Hate Crime, England and Wales, Year Ending March 2026, 8 October 2026.
- Office for Equality and Opportunity, Draft Final Stage Impact Assessment: Annex A, Estimating Annual Numbers Experiencing Conversion Practices, 2026.
- Government Equalities Office, The Prevalence of Conversion Therapy in the UK, 29 October 2021.
- Office for Equality and Opportunity, Draft Final Stage Impact Assessment: Annex C, Depression Prevention Benefits Calculation, 2026.
- Home Office, Controlling or Coercive Behaviour: Statutory Guidance Framework, 2023.
- HM Government, Conversion Practices Draft Bill, 25 June 2026; Office for Equality and Opportunity, Explanatory Notes, updated 2026.
- UK Parliament, Joint Committee on the Draft Conversion Practices Bill: Call for Evidence, 26 October 2026.
Leave a Reply