The Select Committee on the Armed Forces Bill will be taking evidence from two panels in our session this afternoon. I welcome our witnesses; thank you for taking the time to come and give evidence. I will ask you to introduce yourselves, and then we will get straight into our questions. Are first names okay? Thank you.

I am Zoë. I am the deputy chief executive of Aurora New Dawn. We are a charity that has a service working with serving personnel or their family members who are experiencing domestic abuse, sexual violence or stalking.

I am Emma Norton. I am the director of a charity called the Centre for Military Justice. We represent servicemen and women who have experienced sexual violence, sexual harassment and other forms of discrimination and abuse.

I am Ahmed. I am a partner and solicitor advocate at Bolt Burdon Kemp solicitors. We represent members of the armed forces and their families.

Good afternoon, everybody. I will ask you all the same question: what is your view of the service justice system’s progress on improving the treatment of victims of service offences?

I think it has improved. We have been working in the system for a few years now, since 2017. There have been some really positive developments such as the establishment of DSCC and the Victim Witness Care Unit. Having worked outside the service justice system for many years, we are starting to see some signs that collaboration is improving. We are starting to see referrals being made by the Victim Witness Care Unit and by DSCC to independent organisations for support for victims. Information sharing is better; overall, there has been an improvement in the skillset for investigations; and collaboration is improving. There is a way to go, but I have seen some positive changes.

I would second all of that. There have been a lot of very significant changes and improvements over recent years, particularly since the 2021 Defence Committee inquiry, and it is really important to recognise that and all the hard work that is going on within Defence. This Bill, to some extent, is a reflection of that: it recognises and builds on some of those improvements and changes, and perhaps we can talk about that a little more. I do think that some key areas are missing; I hope we will have the chance to talk a little more about that, too.

I largely agree. There have definitely been some improvements. The messaging from Defence has improved—the vocabulary has. Certainly, the messaging from the top brass down is that this is very important. There have been some improvements in the service justice system, but there is a long way to go, for reasons that we might explore today.

In recent years, there has been a lot of criticism of the culture of the armed forces and how they approach victims of sexual assault, in particular. Do you feel that this Bill addresses the need for cultural change?

A lot of the provisions in the Bill are at the sharp end. It is about how we respond once harm has occurred and how we support victims. It is really important that we have that in place, and really important that the messaging is there. Equally, there is the piece about culture change and early intervention. Changing culture is a complex process. It involves everything from the policies and processes we have in place to who we recruit, how we recruit, how we train people and how we deal early with unacceptable behaviour. It is important that we address this from both ends.

The answer to the question of whether the Bill is addressing cultural change has to be no. I am not sure how you can legislate for cultural change, but I do not think it does. There are some areas that it could have gone into but has not. When you think about sexual harassment inside the armed forces, how do women respond to that? Will they make formal complaints about it? What we know is that the complaints process is broken. The independent ombudsman for service complaints has said that it is not effective, efficient or fair. While there is all this wider work going on, below the waterline of criminal justice—below that threshold—there is still an awful lot of work to be done. The last sexual harassment survey was published in November 2025, and it still reveals very, very serious problems. I think 67% of women said that some sort of unacceptable behaviour had been directed towards them, and that was just in the preceding 12 months, so there is still a long way to go. Those kinds of experience are far, far worse in the armed forces than in any other workplace environment that I know of, and unfortunately this Bill does nothing about the service complaints process, for example. That is regrettable.

I agree with Emma. I do not think the Bill, in and of itself, will solve the cultural issues. As Al Carns said in a previous evidence session, this is a question of education and communication—I would say “over-communication”—but the Bill is definitely a step in the right direction.

Thank you all for being here. Clause 10 will introduce a new victims code of practice. Can you share your views on what the code should look like and how it should be implemented to make sure that it helps victims to navigate the service justice system?

It is really positive that it will be on a statutory footing. It is really important that we be open and transparent with victims about what they can expect, and this is an important step in doing that. It also helps us, if we are advocating from within the system, to be able to use that framework. How commanding officers can respond to victims will be really important. I recognise that a lot of what happens in the very early stages will be decided by those individuals. Having a really clear framework about how we approach that, to ensure that we are supporting victims appropriately from an early stage, is going to be really important. We certainly need some clear guidance; I believe that guidance is already in place, but it is just not known about, so this is really about making that clear in the first response to victims. That is possibly the most important thing.

I completely agree. There is already an armed forces code for victims of crime. This is going to be placed on a statutory footing, which is obviously a good thing. You hope that that will give greater confidence to people, for example in the Victim Witness Care Unit, to promote it and ensure that people are aware of it. I wanted to say a couple of things about this. One is that I think it will be very important that parts of the code be permitted to apply to people even if their case is being heard in the civilian justice system. For example, there is a very important proviso in the existing code about markers. A marker is really important: if you are a victim of crime and your perpetrator is serving alongside you, you can ask that a marker be placed on your file and on his or her file, so that it can be considered when thinking about whether you are posted and where you are deployed. It has been said to us multiple times—I think it is contained in the existing code—that the armed forces code cannot apply if your case is being dealt with in the civilian justice system. That seems to make no sense, because exactly the same questions of risk and concern arise. That is important. I also want to make a plea. If the pilot for independent legal advice for victims of crime goes ahead—there are lots of reasons to think that it will—can we ensure that that is embedded in the code, or at least that there is a link in the code so that when people have a concern or a reason to consult the code, they know that the option is there to get some independent advice about it?

I echo that. Without sounding too broad, a framework setting out what is going to happen and when, who has responsibility and how communications are going to be managed is a must. Expanding on what Emma said, I would encourage you to think of the victim and their unique position as a serviceperson. Service personnel have far fewer legal rights than civilians. Imagine a young woman who has perhaps suffered a service offence—perhaps a sexual crime—and is dealing with very serious issues in that respect. She needs to be provided with clear advice on a number of areas: her employment law rights, her civil rights, the criminal process and the issue about jurisdiction, which we may or may not touch on. A comprehensive signposting service will be really important.

Zoë, Emma and Ahmed, thank you very much for everything that you are doing to champion this cause and push for a meaningful change. Zoë, I want to get your reflections on the work that you are doing on the violence against women and girls taskforce that we are setting up and trialling with your organisation, and on what you think about the tri-service complaints system that we are developing to do exactly what Ahmed has just said: pull some of the reporting chain outside and allow people to have an independent complaints system.

I had probably best address the work with the taskforce and the collaborative element of that. One thing I will say is that since this service started, I have not been in a position in which there has been as much collaboration and consultation as there is at the moment. I think that is really positive. When we first started, there was no communication at all, and we really struggled to get a free flow of information between ourselves and the institution. That has changed noticeably, particularly over the past 18 months to two years. We have spoken to the taskforce; we are linked in with them, and as I have already explained, we are linked in closely with DSCC and the Victim Witness Care Unit. That is really positive, and I hope it continues. There are other services coming into this space, doing similar work to ours and developing an armed forces specialism, so the ability to consult is getting broader as well. I hope that is taken into account.

I echo everything Zoë says. There are improvements, but in relation to the service complaints process, this has been called for since 2019 and the Wigston review. You are probably aware that there has long been enormous concern that the vast majority of people who experience serious, unacceptable behaviours, including sexual behaviours, do not report them. They do not complain, and the reason they give for not complaining is that they think there will be implications for their career and that they will suffer consequences. That is exactly our experience at CMJ. Multiple women would say that exactly that has happened to them. This recent announcement is enormously welcome. It follows the Jaysley Beck case. Obviously, we are not going to get into that case in any detail, but it was very important because, following the inquest, her family undertook wider advocacy around this issue. It is a testament to them that this is happening. It is more than a year on from the Beck inquest, and we still do not have this in place. I understand that it will follow later this year, so we will be watching that very closely.

I commend the tri-service approach. All the reviews that have happened year on year—look at the Lyons report and the inquiry into servicewomen’s lives—have indicated that the more we can show true independence in these processes, the better. That is something. Taking it away from the chain of command is the feature. In respect of giving a warning, the MOD’s response to the concerns over the service complaints process since that inquiry has been to focus on efficiencies and not necessarily on the quality of outcomes, so trust within those systems remains very low. Until you tackle the real issues there, and not just the speed of the system, we will continue to have problems.

Good afternoon to you all. You have just mentioned the Lyons report. In a previous session, Sarah Atherton described this Bill as a missed opportunity because it does not take forward the review’s recommendation that serious sexual assaults be tried solely in the civilian jurisdiction. I am keen to hear all your reflections on that, but perhaps we could start with Emma’s.

You have mentioned that that is where this all comes from: Lyons’s recommendation in 2020 that the most serious sexual offences be taken outside the service justice system, because he was not satisfied with what he saw in terms of military policing and prosecutorial decision making. The previous Government declined to accept that recommendation. I understand that the current Government’s position is that they want to go down the route of giving victims greater choice, and giving them the opportunity to express a view, although I understand that it will not be determinative. It is a start, but all the clause currently does is reflect the policy position that already exists, which is that service personnel are supposed to be informed that they can report serious criminal offences to the civilian justice system as much as the military justice system. I think that it risks being presented as more substantive and important than it is. Giving victims the option to express a view is obviously important. Our experiences chime with the former DSP’s evidence. Our recent experience is that people are increasingly being told about this option and being invited to express a view, but the critical question for us will be what information victims are given when deciding which route to express a preference for. That gets us into the very thorny area of rape conviction data, which I know you will be interested in. I think it is being suggested by the MOD that its rape conviction rates are on a par with, or better than, the civilian justice system’s. I respectfully disagree. To understand what the MOD is saying, you have to understand about rape flagging. Rape flagging is the system that has been employed by the CPS for many years, and it is very heavily criticised. It is now being deployed by the MOD. It means that if you report a range of offences, for example a rape, a sexual assault, a common assault and a false imprisonment, but there is a conviction only for the common assault, the conviction will still be rape-flagged. Many people have a problem with that, including the head of statistics regulation—he wrote to the CPS about it in 2023, expressing his concerns—because it inflates the figures. The CPS itself is clear that it is not to be relied on. There is helpful data published by the Ministry of Justice on this issue. UCL has done a huge amount of work to interrogate the data, so it is possible to compare outcomes for rape cases that are heard at Crown court with outcomes for rape cases that are heard before the court martial. It is possible to compare that data. When you do that, the outcomes for rape at court martial are very considerably lower than those at the Crown court. We are very concerned about this, still.

For me, the concern is about the nature of the information that is being given to victims. From what we see on the ground, there is a lot of inconsistency about that information. I understand from the evidence that was given to you by senior representatives of the service justice system recently that they were not particularly clear on what was being said to victims, at that level. I would welcome the victims code. I would welcome some really coherent, clear information being given to these victims. I do not think it is for the Ministry of Defence to advise victims of crime that, actually, it may be better for them to pursue an inquiry through the service justice system because, for example, alluding to what Emma was saying, it is quicker. I do not think that is the complete picture.

I think informed choice is crucial, but I agree with Ahmed that we need to be really clear about what information is given to victims and how it is given. That needs to be really quite robust in how we set the guidelines.

Can we pick up on that theme? We have heard evidence that victims’ preferences are taken into account. To what extent do you think that is the case? Building on that, clause 25 sets out some new guidance for victims on criminal jurisdiction. Do you have any thoughts about that clause specifically?

In our experience, victims’ wishes are taken into account, but that comes back to the point about how the discussion has panned out with the victim. We do not know how those discussions are happening. I do not know whether we are specifically recording the reasons why victims have a preference for one system or another. It would be interesting to understand what the reasoning is, from a data collection point of view. The point to make there is about recording and monitoring, but I think, broadly, that their decisions are taken into account.

Our recent experience is that women are being asked what their view is. In a couple of cases—I spoke to a couple of women about this yesterday—they said that they were told that there were the two options, but they were not given any information. My colleague Lucy spoke to another woman who was very clear that she felt very strongly steered down the service justice route. I think there is that pressure to elect to go down the service justice route, and in the absence of other information to inform the decision, inevitably that will happen.

Anecdotally, only in a very small proportion of the cases that I have dealt with have they even been told that they have a choice. I am very worried about the level and quality of information that is being provided on the choice for victims. Perhaps this Bill will be a step to remedy that.

Thanks very much for coming today. To dive into the details, clauses 5 and 6 give new powers to the service police and courts to protect individuals from sexual harm and domestic abuse. Could you speak a little about that? Does the Bill go far enough? Does it give adequate protections? Is there more that we could do there?

I think it is really important. It is welcome and long overdue that we bring the orders available to the service police into line with what their Home Office counterparts have. That is absolutely the right thing. My concern would be about the practical application. I also chair a domestic abuse-related death review. As part of those reviews, the enforcement of orders is often a gap that we identify. There are opportunities to learn from how those orders have landed in the civilian policing system and to get that right. Training needs to be comprehensive, and we also need to bear in mind the unique nature of the forces. The people on general policing duties may be deployed away from those duties for a long period and then come back, in which time new legislation has come in. We need to bear that in mind. Certainly in the Home Office police, there is constant awareness of the orders, and the need for positive action in relation to breach is important. The other point I would make is that, quite often, the enforcement of those orders—whether or not they are made by the service justice system—is going to fall to the Home Office police. Given the nature of when breaches occur, who is going to be the first response? I think we need systems that talk to each other, so that we are really clear about what is in place, and it needs to be really robust. Otherwise, there is going to be a breach of an order that is not on a system and cannot be enforced. Enforcement would be my concern, but I think the clauses are really welcome.

I echo that, and I will not repeat it. The clauses are long overdue and very welcome, but I want to flag a couple of other things. For some time, we have called for the ability of commanding officers to deal with domestic abuse offences to be curtailed or removed, and it remains the case that a commanding officer can still deal with a range of offences through their powers under schedule 1 to the Armed Forces Act 2006. They really should not be dealing with offences such as assault, assault ABH, harassment offences and a whole range of service offences that could arise in a domestic setting, such as misconduct through alcohol and those kinds of things. We really think that has to be removed. The other thing I want to flag is the greater obligation on commanding officers in this Bill, which is very welcome. Any commanding officer, if they learn of a service offence, has to refer it to the service police—that is very welcome—but some offences that are not in schedule 2 to the original Act really should be, such as voyeurism, coercive control in an intimate relationship and the domestic abuse offences I just referred to. We think those also need to be made clear on the face of the Bill.

That is really helpful.

I would echo all of that. Domestic violence is not in my wheelhouse or expertise, so I would defer to my colleagues. I just point you to clause 17, which I think is a very important aspect of the Bill. It puts an obligation on commanding officers to report serious crimes of this nature, and I wonder whether that could be expanded.

We are coming to that.

Finally, do you have any brief thoughts on clause 9, on the extension of MAPPAs, or multi-agency public protection arrangements? Is it enough? Does it need to do more, or is it in the right place?

Again, I think it is really welcome, and my comment would follow similar lines, as it is about making sure that the systems speak to each other.

Yes, it is about co-ordination.

I would agree with that. In our written submission, we made the point that there may be a gap: if a commanding officer is still able to deal with an assault ABH in a summary hearing, that may not count as a service offence within the meaning of the MAPPA arrangements, so it might slip through. In circumstances where you want to beef up the service police’s powers to deal with domestic abuse offences, they need to have sight of everything. That is something that could probably do with a bit more scrutiny.

Clause 17 will expand the duty on commanding officers to report a serious offence to the service police. How much difference do you think that new duty will make in practice?

I think it will make a difference, but I would love to see the clause be given teeth. It should be coupled with an insistence that commanding officers who do not report such serious crime may face administrative action. While that is not criminal prosecution, they will at least know that their record may be at stake. If you could give this clause teeth, it would be something I would recommend wholeheartedly. I think it ties into the recommendations made and evidence submitted following the inquiry into servicewomen’s lives. A large proportion of that evidence pointed to the fact that there were stories of commanding officers sweeping these sorts of issues under the carpet. Once we get the word out, all commanding officers—not just the commanding officers with responsibility for the victim’s chain of command—will have to speak up when these things happen, and I would encourage that.

I will not repeat myself, as I think I answered this question in my previous reply. It is undoubtedly a very good thing, and we have recently been involved in cases where this has been an issue. I just want to re-emphasise the point that schedules 1 and 2 to the Armed Forces Act 2006 need looking at again, because there are other aspects of commanding officers’ powers, in relation to their ability to deal with domestic abuse offences, that also need to be addressed.

I would agree with everything that has been said. I think it is very welcome.

Can I turn our attention to the additional complexities of the service justice processes for personnel serving overseas? What are your observations of the challenges for anybody serving abroad? Do you have any thoughts on how the measures in the Bill will affect those personnel who are serving abroad?

The services available are broadly the same. For example, our service works internationally, and the Victim Witness Care Unit is available. Where you perhaps get some variety is in the initial response and the wider safeguarding work that happens around victims and perpetrators. Again, it is down to individual commanding officers and their response. We have certainly had a lot of really positive results, particularly with the Royal Navy recently in how it has managed removing perpetrators from ships. That is a bold move, but a really welcome one. I think there is variety in the response, depending on whether the setting we are dealing with is a ship, a base or somewhere else. If there is an opportunity to offer some guidelines or to standardise the codes of practice in any way, that would be useful. We can get good results, but it is determined by individuals at the moment.

I would agree with all of that. We had a case, for example, where there were no standing arrangements in place to deal with an allegation of sexual assault overseas. In that situation, there was inevitably a delay while arrangements were made for medical assessments, sexual assault referral clinics or their overseas equivalent—there was nothing in place. And inevitably that delay was, from the victim’s perspective, very harmful to her. We have other cases where the person assigned to, for example, a ship to deal with disciplinary matters or problems—allegations of assault—is, in the victims’ view, part of the problem. I do not underestimate the practical challenges of ensuring that access to justice is meaningful in those rather remote locations. It is difficult.

It is a credit to the MOD that its policies are very expressly stated to apply everywhere across the globe. This Bill also expressly says, in its jurisdictional clauses, that it will apply to sovereign base areas and overseas territories, for example. But you have—I think it is a very good question—the practical challenge that when you are deployed abroad, you can often be incredibly isolated as a victim. There needs to be an emphasis in the guidance to commanding officers that when you are abroad, when you are in a foreign jurisdiction, whatever it may be, you need to remember that the same rules apply. There are also the logistical challenges that Zoë and Emma mentioned, such as being able to take somebody away from those circumstances. I have acted in cases in which women have complained about very serious incidents on board ship and have asked to be removed, but that has not been practically possible. There will be things that are possible and things that are not possible, but I think that having that discussion, and having the educational requirements for commanding officers so that they know that it is within their power and that they may be expected to give those things some thought, is going to be very important.

It is even more complex if you are serving operationally on a submarine. I don’t know whether you have come across cases like that.

I have, yes.

Thank you very much for your evidence. It is super-useful and always well thought through and balanced. And thank you for your work in collaboration with the MOD to make the system better. This is a broader and more holistic question. Do you agree that there is a need for a service justice system, and if so, why?

Historically, it was created to keep people in line when they were fighting wars abroad. That has changed, and the service justice system can and must change with the needs of society. I see many reasons why the service justice system is needed, but I do not see that it has been capable of dealing with serious sexual crime, particularly in respect of servicewomen. I think that is what we are really concerned about, and it is where our work and advocacy come to the fore. Yes, it is an incredibly important system. It is a system that is trying to improve itself, and we have to give credit for that, but I do not see it as fit, currently, for dealing with the complaints and issues that servicewomen raise.

I would agree with that. Yes, absolutely, there is a need for a service justice system. As long as we have to deploy people overseas, we have to have a system of legal justice that travels with them. So yes, absolutely. I agree with everything Ahmed said. The service justice system has been on a journey. When I think about how it was 10 years ago, it is undoubtedly much improved, but see my earlier answer: I still think there are very serious challenges in relation to victims of rape being able to expect good outcomes at court martial. As long as the differences remain as stark as I think they are, there are very serious concerns about that. Yes, of course, there needs to be a service justice system.

I agree with everything that has been said. The evolution needs to happen at pace. The work that has already been started needs to continue. But yes.

Zoë, Emma and Ahmed, thank you very much for being with us. To pick up on Emma’s point about the concept of service justice travelling with soldiers on overseas deployments, have there been any recent cases of budgetary constraints, whereby people have not been able to go out to investigate in a timely way because the budget has not been there?

I am not aware of anything where the budget is a problem. If there is a concern about an absence of standing arrangements, that probably comes back, ultimately, to money. I am thinking about a case I did a few years ago where exactly that happened—there were no standing arrangements—but no, that is not something that has been specifically said to me.

Not budgetary constraints, but operational constraints in the sense of being deployed abroad and somebody raising a complaint but not being separated from the person she has accused of harassing her, for example. Often, the answer is, “Look, we’re deployed in a small space, and there are only so many places we can put them.” There are those kinds of considerations, but not budget.

Not that I am aware of, no.

To go back to clause 17, do you believe that appropriate processes are in place for service personnel to report offences, and are they comfortable using those processes?

I do not think they are comfortable. I do not think the statistics currently reflect a system where there is a lot of faith in it. One has to think about the service complaints process in tandem with the service justice system, because from what I see, a lot of complaints that are raised about inappropriate behaviours go through that process, too. As I said before, I welcome the change in the sense of putting the onus on commanding officers. Certainly, since the inquiry into servicewomen’s lives, I have seen an uptick in the number of inquiries, and I think it is because servicewomen, in particular, are feeling more able. They have a voice and they can step forward, and that is important. Currently, there is not much faith in the system, and I hope that it is building.

Giving victims multiple options for how they report is important. Some people are not going to feel comfortable going to their commanding officer. A large number of our referrals are self-referrals, which is reflective of some of the challenges that have been mentioned.

To reiterate, in terms of wider unacceptable behaviours, we know that the vast majority—I think it is 90%—do not report them at all, which suggests there is a problem. Historically, women would repeatedly say they did not even know that they could go to the civilian police. That is less and less common now. It will be critical to give women enough information and lots of options to report, and the independent legal advice option that will be made available may also assist in that.

Are there any more questions? Do you have anything you want to add that we have not covered in our questioning? Okay. Thank you very much for giving evidence this afternoon. We are really grateful. Examination of witness Witness: Mary Cowe.

This is today’s second panel of the Select Committee on the Armed Forces Bill. You are very welcome, and thank you for taking the time to come and give evidence to us. Could you introduce yourself, and then we will get straight on with the questions?

My name is Mary, and I am the Director of Service Prosecutions.

Thank you for joining us, Mary, and thank you for everything you do. To begin with, can you talk us through the role that the Service Prosecuting Authority plays in providing support for victims of service offences?

The most fundamental thing we do for victims is to aim to charge the right cases and to prosecute them properly. The Committee has heard from other witnesses about specific measures for victims—the Victim Witness Care Unit, and our liaison with the police to try to make sure victims are kept informed and appropriately supported at court. But fundamentally, our core business—that which we can give victims—is making good decisions. In answer to the question, a few things specific to the service justice system—in contrast to my background in the criminal justice system—potentially mean it has the edge. I will go through a couple of those things, in terms of how we deal with cases. One thing I have noticed is that cases in the service justice system routinely benefit from early advice. Police officers contact prosecutors in the service justice system on a very regular basis to ask for assistance at an early stage. They are mandated to do that in any case of rape or sexual assault. That does not happen in the criminal justice system; it is not that joined up, and there aren’t the resources. The benefit that gives victims is that you can anticipate legal hurdles and try to overcome them. You can make sure that the investigation is as complete as possible, as early as possible. That is one advantage. Something else I have noticed is that all cases in the service justice system are looked at by two lawyers before charge. In my experience of the criminal justice system, one CPS lawyer would make a decision. It might then be briefed out to counsel, who would look at it, but a direction of travel will have been set. In the service justice system, two reviewing lawyers, two prosecutors, have to sign off on every decision, in effect, and it seems to me that that level of input and thought before charge is a real benefit. Recently, I have introduced having a third lawyer check where a view has been taken that we would not direct charge in a rape case, because I am interested in monitoring trends around rape. That is something I can institute overnight because I have a small workforce. I think it would be impossible for the DPP to roll that out overnight over all the areas for which he has responsibility. There are lots of other little things that we can do to help victims, including being involved in the prevention agenda or going along to commanding officers designate courses to talk to them about how they ought to run summary hearings. There are lots of other things, but in terms of our core business, the fact that we are relatively small and have good working relationships with the police actively helps victims.

You mentioned review by two lawyers and the possible introduction of a third lawyer if a case is not taken on. How do you think that works alongside the victim’s right to review?

It does not in any way remove the victim’s right to review. At the end of the case, if a decision is made not to direct any charges at all, the victim has the same right that they would have in the civilian system: another independent lawyer looks at it, and quite often that would be me or my deputy, or it would be sent out to independent counsel for a completely fresh look. My overall feeling is that most cases get more attention than in the civilian system.

Has the right to review overturned a decision in any cases?

Yes. I think it is in about 15% of cases, so a small number of cases every year. I think victims exercise their right to review in perhaps 20 or 25 cases—it fluctuates, obviously, from year to year—and the decision being overturned in four or five cases would not be unusual.

Clause 10 will introduce a new victims code of practice. What should the code look like to support victims to navigate the service system, in your opinion?

As you heard, there is already an armed forces code of practice. Having spoken to others in the SPA who have lived in barracks or been on a ship—to talk through what bespoke things should perhaps be in the armed forces code that are not in the code for civilians—it seems to me that one of the major challenges is that, if you are reporting a crime in the service justice world, in the services, you are almost definitely making a report about someone who is known to you. You probably have the same employer; you might be in the same housing; and you will have the same welfare services. Something I have wondered about is whether a positive duty could perhaps be put on commanding officers or units to look at that issue of safeguarding. They can control safeguarding much more than in the civilian justice system. Obviously, there is a potential conflict of interest where a commanding officer is being asked, “Take this person off duty,” when at the same time they are trying to deliver a service or are being deployed. I wonder whether there is space for someone external to the chain of command to perform some kind of risk assessment, so that a victim has the benefit of knowing that when someone is deciding, “Can you safely stay in the same barracks or on the same ship?”, the person looking at that decision is not also the person who has to deal with a manoeuvre or an operation. That is one thought I had.

As we heard from the last panel, there are operational as well as behavioural issues. If they are both in the same service, trying to remove one person for safeguarding raises quite significant concerns. Is there a logistical component as well?

I am sure there is, but if we are going to say that victims in the service justice system have the unique part of their service recognised, which is that it will be harder for them to report, the right to at least have another pair of eyes that is not primarily focused on operational matters to look at it and say, “This is safe; this is not safe,” might be a good idea.

Can I briefly explore the statistics on the number of reported offences that come to you, in respect of serious sexual offences and rape, and the number of charging decisions that you make? Do you know roughly what the figure is?

I cannot give you the precise number, but I know that it is roughly half, and I know that it is roughly what happens in the civilian world. The fact that it is roughly what happens in the civilian world does not mean it is right, which is why I have introduced a third lawyer check in rape cases: one person, who is our rape and serious sexual offences lead—a civilian lawyer from CPS—looks at all the cases in which we are not going to charge rape, to identify any trends, ensure we are trauma-informed and just keep note. I think that is roughly the answer.

Can I ask one more thing? It has been said by others in evidence to us that the chain of command tries to influence things at times. Can you confirm to the Committee that you are absolutely independent in your decision making, and that the MOD and the chain of command cannot influence you?

Absolutely. I am a civilian. I have never been in the military, and I do not report to a Minister of Defence. I operate under the superintendence of the Attorney General. If anyone in the chain of command contacted me to say, “Prosecute,” or “Don’t prosecute,” I would let the Government’s chief officer know immediately. It would be entirely improper.

As the DSP, could you explain the earlier evidence that we heard about the flagging system? What are your thoughts on how we are recording that data?

To look at the area of rape conviction rates at all, it is important that the Committee understands the number of cases we are talking about, because the different statistical systems do not make sense unless you have that understanding. In the civilian world, the CPS prosecutes about 1,500 to 2,500 cases of rape a year: a big, four-digit number. We prosecute about 20 to 30 cases a year. The reason I start with that is that it makes comparisons inherently volatile. If a prosecutor makes the appropriate but bold decision to say, “I think that there is a realistic prospect—just—of conviction in this case, and I’m going to put it forward and let a board decide,” and the board says no, those decisions in one or two cases are going to account for a 10% or 15% increase or drop. That is the first thing to understand. I agree with Emma Norton that rape flagging is not a good system for the service justice system. Rape flagging means that a case has been referred to the CPS by the police, the police have said that it might be a rape, and at the end, if there is a conviction for any offence, it effectively counts as a rape conviction. I agree that that does not work, particularly for a small pool of cases. What was happening was that the rape flag data, which treats any conviction for any offence as though it were a rape conviction, was being compared with data from the Military Court Service that looked only at rape convictions; it discounted even an attempted rape or sexual assault conviction. You were comparing two completely different pools of information: in one pool, with the CPS, an ABH is counted as a rape, and in the other pool, with the Military Court Service, an attempted rape does not count at all. For that reason, I think, the MOD released information trying to apply CPS statistical methods to our data. In fairness to it, all over the transparency release it says, “We aren’t particularly happy doing this, and we won’t do it again, but this is effectively to remove the idea that there is some kind of smoking gun.” When it performed that exercise, it came to roughly the same percentage of cases as the CPS. Whether or not that gives you any information is unclear, but it certainly does not bolster a case that there is something fundamentally wrong. It seems to me that we have such a small number of cases every year that the best thing to do is to aggregate data from several years. What we have done internally is look at data from 2018 onwards—the eight years since the Military Court Service introduced official statistics. If you do that, the percentage comes out at about 42% convictions for rape in rape trials. I disagree with the previous evidence you have heard that you can compare that to the CPS figure. You cannot, as it does not count its data like that. All I can tell you is that when you look at our data on its own terms—rape convictions and rape trials—it is about 42%, not the much lower numbers that have been put out there. Reference was made earlier to Professor Cheryl Thomas’s work in the civilian system, trying to unpick CPS methodology. The thing to understand about that is that the military court data is about convictions per defendant; her work is about convictions per charge. Can I explain why that matters? In the civilian system, in particular, you might get an indictment with 25 charges, and in the way she has counted them, if you get a guilty verdict in one trial, it is 25 guilties and it counts 25 times. If you lose 24 trials in which there is a single count of rape, they effectively balance each other out. In a big population of 2,000 cases, no doubt that all evens out, but if you compare her per-charge data to our per-defendant data, we will always look worse, because we do not do that many multi-charge counts. The best thing I can give you in terms of evidence—I have just gone through the numbers—is actual hard numbers rather than statistics. I wonder if that would help. In 2023, we prosecuted 15 defendants for rape; four defendants were found guilty. In 2024, we prosecuted 15 defendants for rape; 10 were found guilty. In 2025, we prosecuted 13 defendants for rape; six were found guilty. I do not think that means our competence is varying by margins of a third or a half every year. I think the calendar year into which difficult decisions fall will be very volatile. Where I end up is that the statistical evidence, it seems to me, does not leave you with a smoking gun, but what it does say is that when you have really small numbers, you have to have lots of other assurance processes. You cannot just look at numbers, because numbers can fluctuate. You have to do your second lawyer checks. You have to have scrutiny panels. You have to make the right decision in the right case. I would be concerned about the conviction rate. Is it a method of diagnosing whether there is a problem becoming a target? I want my prosecutors to have regard to the law and the evidence, not to be thinking, “If I charge this case, what will it do to the statistics?” That would be wrong. That is where I end up. I do not think there is a smoking gun; there may be an evidence gap. What I would like to do, personally, is publish on the SPA’s website the information that I have given you: “In this year, we prosecuted x number; this is how many rape convictions there were; this is how many sexual assault convictions there were; this is how many acquittals there were.” That would be fairer, more honest and faster, frankly, than trying to unpick all the statistical information.

Thank you very much for your evidence, Mary. It gives me great confidence when I hear someone speak with such conviction and with all the information and experience behind it, not only from the service justice system but from the civilian system. While we are discussing concurrent jurisdiction, can you talk us through the rationale for not taking forward the Lyons review recommendation on reserving certain offences to the civilian jurisdiction?

Yes. I should say first of all that I am conscious that I am a civil servant and that this is a policy matter. If I can answer it in this way, I hope that—without trespassing into anything I shouldn’t—I can perhaps identify a few factors that I think the Committee could consider when deciding whether it is the right decision. One factor is the issue of victims being able to express a preference. To be frank, when I arrived in the Service Prosecuting Authority and was told that victims all wanted to stay in the service justice system, I was a little sceptical. What I have understood since then is that cases move between the service and civilian systems much earlier than I had anticipated—long before they get to us—and that victims who say to police officers, “I don’t want to be in your system,” will sometimes have their case passed on long before it gets to us. Our numbers are probably masking the fact that some victims are opting out earlier on, before they get to us. Something that has impressed me is that our victim attrition rate for rape offences—victims withdrawing from the process—in the last two years was zero. One possible inference from that is that victims are probably not being railroaded into making decisions to stay in the service justice system, because that would probably emerge at some point in our engagement with them. I am concerned to hear the suggestion that victims are effectively being told, “In some way, you’re supporting the service justice system by electing to stay with us.” I agree that all of that could be tightened up. But I think, considering everything that is happening in the criminal justice system right now, that there is some evidence that victims are making a real choice. One factor for the Committee, potentially, is whether we want to continue with a system in which victims can have that preference considered and it remains an option for them, or whether we effectively decide for them. That is one issue. The second—I’ve only got three, I promise—is about expertise. The Lyons report did not say that all cases should be taken away from the service justice system. It acknowledged that some rape cases would have to stay in the service justice system because of offending overseas. It also said that perhaps for some other cases it would be better and you would have to seek AG consent. Whatever happens, there will always be a core of cases that have to be dealt with by the service justice system. I suggest that one thing to consider is the context in which you want those cases dealt with. Do you want them dealt with by prosecutors who otherwise never touch a rape case, who have perhaps been deskilled, or do you want them dealt with by prosecutors for whom it has kind of become part of our core business? People go on RASSO training every year, exactly the same as in the CPS. We are having vulnerable witness training to make sure that our vulnerable adults and children are spoken to correctly at court. There has been a big build-up in our ability to deal with those cases, so what happens if it is withdrawn? It is about acknowledging that someone will have to be around to deal with the couple of rape cases that happen overseas every year, and about the culture you want them to be operating in. Lastly, I think it was said earlier that Lyons’s view came from what he had seen and heard, and his confidence in the system. Re-reading his report in preparation for today, he really said it as a point of principle, as I read it, that rape and serious offences should just be tried in the civilian system and that that was a right for defendants to have, because they are serious cases. In the Henriques report, the judge made, I think, the sensible point: if you were just saying, “It’s a question of competence,” why focus on rape? Why let them do anything? My point about that is that it is very difficult for the Service Prosecuting Authority to engage and say, “This is how we’ll change,” when it is not clear—there is no clear evidence that I can point to that says, “This is why a victim should feel that the CPS will deal with their case better than the SPA.” It seems to be an ideological thing in some places, perhaps, and expressed by Lyons as an axiom, that rape belongs in the civilian system. I will finish by saying that whatever the difficulties are in reporting service offending, it will have to be dealt with in some way, either by the CPS or the SPA. I invite the Committee to bear in mind the changes that have happened, for good or ill, in both the service justice system and the criminal justice system since 2021.

Can I draw your attention to clause 5, “Sexual harm prevention orders and sexual risk orders”, clause 6, “Protection from domestic abuse and stalking”, and clause 7, “Service restraining orders”? In your written evidence, you identified concerns about a legislative gap. Could you explain to the Committee what the gap is and how it might be addressed?

Certainly. These provisions are very welcome because they close a gap at one end of the system, whereby if someone is still serving at the time of a court martial but leaves straightaway, the order by the court preventing them from, for example, having contact with someone can be enforced. The issue I have identified is this: what about a situation in which someone is serving at the time of the behaviour? Let us say that they were even serving at the point at which my prosecutors are talking to the CPS about the right jurisdiction. They get charged and are in the service justice system, and then they leave the services. As things are drafted at the moment, a service court could still punish that person in other ways, but it could not make a protective order. I cannot transfer it back to the civilian system once we have charged it. My tentative suggestion—I have mentioned this to people working in the policy department, and better minds than mine are on it—is that there should be a way to change the wording so that service courts can make orders in respect of people who were, for example, serving at the time they were charged, and that those orders should immediately convert to civilian orders so that they would be enforceable. My understanding is that work is going on, but there is an acknowledged gap.

That has been pointed out to the Committee in previous evidence, hasn’t it, Chair? Thank you very much.

Mary, thank you for your very in-depth evidence this afternoon. Clauses 20 to 25 will make several changes to service courts. From your perspective, how effective will those measures be in improving the efficiency of the service justice system?

The main thing about those provisions of the Bill is that, overall, they regularise things between the service justice system and the civilian system, for example by allowing hospital orders to be made. All those things are useful: for example, that is one factor that a victim does not have to consider when expressing a preference. If there is equality of disposal at the end, other factors that might be more important to them, such as timeliness, independence or trust, can come to the fore. All the measures that create or attempt to create parity of disposal between service courts and criminal courts are to be welcomed. I do not think that I have seen anything in particular to which I would want to draw your attention.

What role do you think the clause 25 guidance on criminal jurisdiction will play in helping victims to navigate concurrent jurisdiction?

I think it will help. At the moment, from everything I have seen and read internally, victims are being asked about preferences, particularly in more serious cases; it is in cases involving fighting and so forth that we are patchier. However, the point has already been made to the Committee this afternoon that there is not very clear guidance, or guardrails, for the police on having that discussion. I think the guidance could usefully cover content, including being neutral. It is not anybody’s job in the police to advise a victim, and neutrality is important, but we should be open about information. On timing, generally speaking, we want victims’ views early, because we do not want an investigation to progress for a long time and a victim then to feel that they have to stay with the service police, for example. We should also make sure that there are proper records, so that we can have some confidence that it is a genuine and informed view. If there were some capability—for example, through the independent legal advocacy pilot that has been discussed—to allow victims to speak to another person who is not a serving police officer, that would be good. They would have another sounding board. In general terms, giving guidance so that we know that victims understand the benefits and risks of each system would be very useful.

Is it fair to say that the majority of courts martial are the other ranks and non-commissioned officers, rather than commissioned officers?

I do not have that data. I am sorry.

This may or may not be relevant to the question that I am going to ask, but we took some evidence last week regarding the lowest rank that can sit on a court martial board. It is OR-7, I seem to recall, which is equivalent to a staff sergeant in the Army, whereas in the criminal justice system you could be a 25-year-old grocer from Tesco sitting on a murder trial. In the service justice system, that is not the case. What is your view as the Director of Service Prosecutions? Should the lowest rank potentially be reduced?

I think it requires thought and research. I agree: when I first heard about the system, I was concerned that it sounded a little antiquated, with officers judging officers and that kind of thing. My understanding is that it is thought to be a guarantee of independence for members of the board not to be in the position of making a judgment call on someone who might, in whatever way, one day end up being their boss, so that they do not feel intimidated. Whether that is the only way to resolve the issue, or whether you could have a board with mixed ranks, I do not know, but that would require thought and policy. I have noticed, from reading into it, that the minimum rank appears to decrease every five years, so there is an acceptance that boards have to be representative, generally speaking. The fewer apparently artificial restrictions you have, the more representative they will be. It is certainly not something I would be opposed to; it is just about the balance. I agree that it feels alien to say, “You can sit on a jury, but you can’t sit on a board.” I am just conscious that if you are sitting on a board, you do not want to be in the invidious position of judging someone who, in some nebulous way, is or might be your boss. I suspect that there is a balance to be struck.

As there are no other questions, that completes our session today. Thank you very much for taking the time to give evidence; we are really grateful.