When Policy and the Law Collide

When a court determines any question with respect to the upbringing of a child, the child’s welfare shall be the court’s paramount consideration. Section 1(1) of the Children Act 1989 could not be clearer. Section 1(3) sets out the particular circumstances that the court will have regard to when determining a question relating to a child’s upbringing. Within the welfare checklist are some very well-known circumstances such as harm, needs, capability of their parents and the child’s ascertainable wishes and feelings.

The Children’s and Family’s Act 2014 inserted s1(2A) which provides that ‘unless the contrary is shown, that involvement of that parent in the life of the child concerned will further the child’s welfare.’ This subsection has its critics, who argue that there shouldn’t be a presumption in favour of contact, particularly when there has been domestic abuse. Be that as it may, the statute is the statute and the rebuttable presumption exists.

The reality is that the statute brought in through the 2014 Act mirrors the position that had developed in case law. Sir James Munby in the case of Re C (Direct Contact: Suspension) [2011] EWCA Civ 521 set out the following principles when considering the issue of suspending contact to one parent:

a) Contact between parent and child is a fundamental element of family life and is almost always in the interests of the child;

b) Contact between parent and child is to be terminated only in exceptional circumstances, where there are cogent reasons for doing so and when there is no alternative. Contact is to be terminated only if it will be detrimental to the child's welfare;

c) There is a positive obligation on the State, and therefore on the judge, to take measures to maintain and to reconstitute the relationship between parent and child, in short, to maintain or restore contact. The judge has a positive duty to attempt to promote contact. The judge must grapple with all the available alternatives before abandoning hope of achieving some contact. He must be careful not to come to a premature decision, for contact is to be stopped only as a last resort and only once it has become clear that the child will not benefit from continuing the attempt;

d) The court should take both a medium-term and long-term view and not accord excessive weight to what appear likely to be short-term or transient problems;

e) The key question, which requires “stricter scrutiny”, is whether the judge has taken all necessary steps to facilitate contact as can reasonably be demanded in the circumstances of the particular case;

f) All that said, at the end of the day the welfare of the child is paramount; “the child's interest must have precedence over any other consideration.

This approach was reaffirmed in the case of Re J-M (a child) [2014] EWCA Civ 434 with a restatement (with slightly different wording) of the propositions as follows:

a) the welfare of the child is paramount;

b) it is almost always in the interests of a child whose parents are separated that he or she should have contact with the parent with whom he or she is not living;

c) there is a positive obligation on the state and therefore on the judge to take measures to promote contact, grappling with all available options and taking all necessary steps that can reasonably be demanded, before abandoning hope of achieving contact;

d) excessive weight should not be accorded to short-term problems and the court should take a medium- and long-term view; and

e) contact should be terminated only in exceptional circumstances where there are cogent reasons for doing so, as a last resort, when there is no alternative, and only if contact will be detrimental to the child's welfare.

This legal position appears to have been fairly settled for some time and was reaffirmed by the statutory addition within the 2014 Act.

Why then write an article on this topic? The answer to that question appears in the recently published ‘Domestic Abuse Practice Policy’ from CAFCASS[1]. The contents of this document appear to run contrary to the legal position above which may lead to a tension between the legal position and the advice being routinely provided to the court.

[1] https://www.cafcass.gov.uk/sit...

As readers will know, CAFCASS is the court’s independent advisory service on matters of welfare. Their own website describes their role as ‘CAFCASS advises the family courts about the welfare of children and what is in their best interests’. Predominantly CAFCASS gives recommendations within section 8 proceedings, public law proceedings and adoption proceedings. The advice provided by CAFCASS is often of central importance to the court in coming to a determination, ultimately they are the independent professionals who give recommendations.

The case of W v W (Custody of child) [1998] 1 FCR 640 makes it clear that judges are not entitled to depart from the recommendation of an experienced court welfare officer without at least reasoning the reasons for any departure. The case of Re A (a minor) [1998] 2 FCR 633 emphasises that any misgivings that the court may have with the written report are tested with officer giving oral evidence. The court may then voice such misgivings in the form of questions and understand the view given by the CAFCASS officer. In the case of Re W (Residence) [1999] 3 FCR 274 Thorpe LJ emphasises the views taken in the cases above before saying the following:

In relation to the role of the court welfare officer, it cannot be too strongly emphasised that in private law proceedings the court welfare service is the principal support service available to the judge in the determination of these difficult cases. It is of the utmost importance that there should be free cooperation between the skilled investigator, with the primary task of assessing not only factual situations but also attachments, and the judge with the ultimate responsibility of making the decision. Judges are hugely dependent upon the contribution that can be made by the welfare officer, who has the opportunity to visit the home and to see the grown ups and the children in much less artificial circumstances than the judge can ever do.

It is notable that all of these cases are now over 24 years old. It is a settled legal position.

The court therefore must (and clearly should) place a large amount of reliance on the recommendations of CAFCASS in coming to decisions in cases. It is open to the court in these cases to come to a different conclusion, provided that the reason for the departure can be clearly explained and results realistically having heard evidence on the point to test any misgivings about the written report. It remains unusual for the views of CAFCASS to be departed from.

That then leads to the policy published in early October 2024. I would recommend that all practitioners read this document in its entirety as it will likely frame CAFCASS recommendations in public and private law work moving forwards. I have already heard of examples of it being used to justify proposals being made.

The policy document seeks to set out ‘what must be done’ by employees of CAFCASS in the written work that they produce for courts. It describes the policy setting out practice requirements to support practitioners and managers. Surprisingly given this statement, it still suggests that the policy does not seek to ‘Supplant the professional independence and judgment of FCAs and Guardians’. It goes on however to specifically accept that if this policy is not adhered to:

Cafcass and individual Family Court Advisors and managers can be subject to challenge through complaints, the Parliamentary & Health Services Ombudsman, referral to Social Work England, or a Judicial Review. A decision not to adhere to a policy must be supported by a compelling rationale, endorsed by a manager, and recorded.

The apparent central purpose of the policy is to set out the actions that CAFCASS practitioners must undertake when working with children and adults who have experienced domestic abuse, and who are therefore victims of domestic abuse under the Domestic Abuse Act 2021. It specifically provides that:

Any departure from the starting points set out within this policy must be supported by a compelling rationale, discussed with a manager and recorded contemporaneously on the child’s case record.

Given this clear statement, what are the ‘starting points’? They can be summarised as follows:

a) If a parent is being investigated by the police for a sexual offence, has a conviction for a sexual offence, and/or has served a prison sentence for a sexual offence there is a ‘clear starting point to inform a recommendation for a child not to spend time with that parent due to the significant risks that exist’.

b) Practitioners must not support or recommend any contact (direct or otherwise) where the resident parent is currently living in a refuge, having disclosed domestic abuse by the other parent. If contact has not yet been suspended, then a referral should be made to social care to recommend suspension of any interim arrangements;

c) If a child does not want to see another parent following separation, particularly where the non-resident parent alleges ‘parental alienation’ the CAFCASS officer must first consider whether the cause of the refusal is because the child is a victim of domestic abuse and harmful parenting, regardless it would appear as to whether this has ever been alleged;

d) When there is a finding (or presumably conviction) that someone has been domestically abusive, CAFCASS officers should not recommend a child spend time with the parent who has inflicted the harm on a child, without clear evidence that the perpetrator recognises the harm, has taken responsibility for the harm, has taken action to sustain their change in attitude and the changes have resulted in an assessment that the risk of them perpetrating that behaviour ‘has been removed to the point of enabling a recommendation.’

e) If a police investigation concludes with there being no further action for a sexual allegation, the starting point must be to consider the risk of harm is significant and there needs to be a fact-finding hearing with no direct contact until those allegations are determined.

The policy goes on to restate that if there is to be a departure from these starting points there needs to be a ‘compelling rationale’ which are discussed with a manager, is recorded on the case record and the parent who the child lives with is made fully aware of the proposed advice to the court including the reasons for departure from the ‘starting points’.

There are a number of other elements to the policy which may be of relevance to legal practitioners which can be summarised as follows:

a) In long running or repeat proceedings CAFCASS officers must reflect and take account of previous history and patterns of behaviour, reports of known domestic abuse, safeguarding checks and criminal history;

b) CAFCASS officers must not ‘dismiss or minimise domestic abuse as historical or as a one off incident’. In trauma-informed practice ‘there is no such thing as historical abuse’.

c) CAFCASS officers are to use the person’s own words to describe what has happened rather than reinterpreting or rewording the experience;

d) CAFCASS officers must not use language such as ‘claims or alleges’ when domestic abuse is reported. It is said that to do so minimises and diminishes the experience of the adult and child.

e) CAFCASS officers must ‘provide a clear, unequivocal and compelling rationale’ in court reports if they seek to discount domestic abuse as a risk to a child when ‘abuse and harm has been shared with the practitioner by the child or by one or both parents’.

f) Parents can never be recommended to supervise contact time if the supervising parent has disclosed domestic abuse by the other parent, this applies even if they offer to do so.

It goes without saying (but probably never hurts to restate) that domestic abuse causes harm, often enduring harm to its victims and to children who are direct or indirect victims of it. This article is not designed to minimise or challenge that core principle. Indeed, the Domestic Abuse Act 2021 specifically sets these matters out and there is statutory and widespread social acceptance of the impact. It may be right or wrong for the children who have been born into, or lived through, domestically abusive relationships to either see or never see those parents again. Each case needs to be decided on its own merits.

This article merely seeks to highlight the apparent gulf between the role that the court needs to undertake, and the advice that it appears now likely to receive from CAFCASS because of this policy. Many of the aims of the policy appear entirely appropriate to the legal process. There is a clear reminder that it is the court that will determine disputed factual issues. There is an understandable reminder that the accounts given by parents should be written up as accurately as possible, without reinterpreting what they said. This type of reinterpretation often risks the fairness of a fact-finding process, when differing accounts may be used as an argument against any findings being made.

However the strength of the language around ‘starting points’ is an anathema to the court process. The ‘starting point’ for all of the elements of this policy is that there should be no direct contact in certain situations. That is the opposite of the language of the statute and of the caselaw in the area. How can the starting point be that there should be no contact, when the statute says, ‘unless the contrary is shown, that involvement of that parent in the life of the child concerned will further the child’s welfare.’ The presumption is clearly a rebuttable one (as it must be) however the policy rather implies that the burden for CAFCASS is now the other way round. In that, unless the contrary is shown, and ‘supported by a compelling rationale’ there should be a recommendation for no contact in a large number of situations.

The language about departing from these starting points is also particularly stark. Whilst the introduction to the policy seeks to emphasise that its purpose is to not to ‘supplant the professional independence of FCASs and Guardians’ it requires those professionals to work within this strict regime. It requires those practitioners to have specific starting points at the opposite end of the statutory presumption. Further to this it requires those practitioners to discuss with their manager and record any reasons for going against the standard starting point. They also must explain their rationale to the resident parent. This is all before recommending something that is in their professional and independent judgment correct for a child, either on an interim or a long-term basis.

It is unclear from the policy whether they are entitled to go against their manager’s views (after they have their discussion) or what the ramifications of any disagreement with their manager may be. Whilst it is, of course, CAFCASS must have some form of management structure, the need to speak to a manager about an independent decision, taken within a professional role, could add a layer of limitation on the independence of any individual. Particularly when inevitably that manager will be case managing several different practitioners, will have other roles and crucially will have not met the parties within the case. The repeated reference to the involvement of others within an independent decision-making process may cause difficulties.

Further, and most concerningly, is the inevitable consequence that none of this policy notes the balance that must be taken in cases. As has been discussed in other articles, future welfare decisions are ultimately a balance. There is no right answer to many cases. Often the welfare checklist comes down to a balance of the harms that may be caused by two potential outcomes. The historic case law emphasises that harm will be caused to a child to have no contact with a parent, whether that harm is justified is then a balancing exercise considering the risks of allowing such contact.

Despite this, the focus of this policy can be read to deal in absolutes. If someone has a conviction for a sexual offence then the starting position should be that they have no contact. The starting point shows the absence of analysis. Whilst for some sexual offences no contact may be entirely justifiable, for other sexual offences the balance of harm test may shift the other way. The policy gives no description of how broadly ‘a sexual offence’ should be drawn. It provides no description of how long the prison sentence might have been for. There are no exemptions for offences that occurred historically prior to the onset of the relationship. There is no demarcation between sexual offences against adults or children. There is no detail of any sort, save for the view:

For example, a parent being investigated by the police for a sexual offence has a conviction for a sexual offence and/or has served a prison sentence for a sexual offence, provides a clear starting point to inform a recommendation for a child not to spend time with that parent due to the significant risks that exist.

It is entirely accepted that it is impossible to provide a rigid form of guidance about when contact will and won’t be appropriate. As above, each case needs to be decided on its own facts, but this isn’t what the policy advocates. It specifically gives outcomes depending on inputs, regardless of the nature of those inputs.

Similarly, the view that in every case if the resident parent living in a refuge this should automatically lead to no contact. Again in some cases this may be correct, but in others surely on an objective balance it might not be. This part of the policy reads as an absolute, without any consideration of the other elements of the welfare balance, such as a child’s wishes, their existing relationship and the circumstances that led to the admission into the refuge. Arrangements have historically been able to work with a child living in a refuge whilst having contact, to simply say that they can’t demonstrate the absence of any analysis.

Similarly the policy about when there is a conviction. The policy specifically provides that direct contact should not be recommended unless several criteria are met. The second is that they have taken action to address their actions, despite it often being very difficult to get such work set up either at all or in a timely manner. The last in the series of requirements is a need that ‘the risk is removed to the point of enabling a recommendation’. This again emphasises the need to show either the absence of any risk or a very low level of risk. The whole series of requirements appears to encourage that contact should only be recommended when all steps have been taken, regardless of the quantification of the initial event that was the result of the conviction or finding. This again could be viewed as the absence of balanced analysis.

The wording of the document appears to lead to the very real risk that CAFCASS officers' views will be simply a set of pre-determined outcomes from a policy document. This is particularly true when the only way to go against those pre-determined outcomes is to reach a very high threshold of a ‘compelling rationale’ which they can be accountable through a whole series of complaints (as set out in the policy).

The policy itself has no doubt gone through much detailed consideration before being endorsed. It may be that professional independence from CAFCASS officers continues to be robust and many seek to depart regularly from the specific starting points. However, that will require particularly robust decision-making from those officers, against the general tenor of a policy that they are required by their employer to act within.

The risk within proceedings is that this policy, if followed literally, runs contrary to the very principles upon which the court must make decisions on. It runs contrary to the clear case law in the area, particularly at the conclusions of proceedings. Whilst much of that caselaw was determined prior to the Domestic Abuse Act 2021 it has not been reversed. Most crucially the policy creates a linear or simplistic type of non-analysis to create a specific outcome. It removes from the professionals, that the court relies upon most, the ability to balance the various competing welfare issues. It will no doubt create situations that must lead the court to have misgivings upon the reading of a policy-driven report, which will need to be analysed through oral evidence at final hearings or DRA’s with CAFCASS in attendance.

Legal practitioners and Judges will no doubt need to be acutely aware in every case of when recommendations have had to have been made on a policy basis, rather than on a more global holistic analysis. Ironically the court may need to be more robust in challenging an analysis that relies upon simplified ‘starting points’, rather than an analysis that considers all the competing areas of the welfare checklist. This is ultimately because it is more difficult to rely on something that hasn’t used the same methodology as the court is required to use.