SKN

Hello everyone, this is Susanne Knabe-Nicol from Police Science Dr with an exclusive interview with Professor Lawrence Sherman, who’s now the Chief Scientific Officer of the Metropolitan Police in London here in England, UK. And I will let him introduce himself and tell you a little bit about what he’s done in the past and why people should know him. Professor Sherman, welcome.

LS

Thank you very much, Dr Knabe-Nicol.

SKN

So, why don’t you tell people who don’t know you who you are, what you’ve done so far, I know it’s a long and very accomplished CV, but give us a bit of an overview if you would, please.

LS

Well, about a half century ago, I went to work in the New York City Police Department at a time when 11 police officers were murdered in one year, when the department was under investigation for widespread corruption conspiracies, sometimes involving 100 police officers or more. The place was very challenged and the reforms that were instituted by Mayor John Lindsay, who appointed Patrick Murphy as the police commissioner were substantial, far-ranging and very successful. And in my book, Scandal and Reform, which became my PhD dissertation at Yale University, I described the massive interventions that were successful in breaking up the bribery from gamblers and drug dealers and prostitution rings that had characterized the NYPD almost since its birth in 1844. So, it's a very different world that started me out for two years, actually doing service as a conscientious objector during the Vietnam War and working in the commissioner's office, sometimes doing undercover testing of procedures to tell people how to fill out complaints against the police, other kinds of interventions. I wasn't involved in criminal investigations, but I was certainly applying all the analytic tools that I had learned through my master's degree at that point, and working on neighbourhood policing as well as some of the professional standards issues. That led me to the best police researcher in the country at the time, Apple J Reese Junior, who invented the idea of proactive policing or at least the conceptual framework. And I went on to study police use of deadly force with some research that led the Supreme Court to restrict the powers of the police in the US to shoot fleeing felons, a century after the same law had been changed in the UK. I did research on issues related to domestic abuse with Minneapolis Police Chief Tony Bosa who, for nine years, having been my mentor in New York, and then my partner in various innovations, including repeat call address policing, the RECAP problem-solving approach, which we tested with a randomised trial, just like the hot spots that we discovered there, the 3% of addresses that had half of all of the calls is the first documented pattern of citywide hot spots concentration, which led us to a randomised trial with David Weisburd that showed doubling police presence in half the hot spots randomly assigned, cut crime substantially, cut robbery in half. And since then, we've had 85 replications by other scholars around the world, generally showing very positive effects of concentrating police resources where they’re most needed.

All of that was parallel to an academic career that eventually brought me to Cambridge University in 2007, and 15 years of running a master's degree program part-time for people who were already chief constables in one case, namely Sarah Thornton, or becoming chief constables or chief officers around the UK, as well as from Denmark, Sweden, Australia, New Zealand and other countries around the world. And in 2022, I was delighted to be appointed by Sir Mark Rowley, the Commissioner of the Metropolitan Police, as the first Chief Scientific Officer of the Metropolitan Police, where I have been able to work full-time with a very large police force of 48,000 officers and staff to test new scientific approaches to policing that might be able to make better results out of scarce and even declining resources at a time of increasing demand.

SKN

Well, I did not over-promise the very, very extensive CV there, so thank you very much for this overview. Obviously, one of the things you have done in the past here in England as well is, you were a founder of the Cambridge Centre of Evidence-Based Policing, which obviously provides very high-quality training to police around the world indeed, because we also have online versions of those training courses. What would you count as some of your greatest achievements in your career so far?

LS

Well, I think the Supreme Court decision that by virtue of national numbers of police officers shooting fleeing felons, which dropped almost to zero in the aftermath, anything that you can do with research that leads to saving lives, I think is the kind of achievement that one should remember. And I think that was the first big impact of research in my life. The next biggest impact, I think, was the discovery of hot spots, which everybody takes for granted as if we all knew about hot spots. And I can't tell you how many police chiefs told me in 1987 that there were no such things as hot spots, that crime could happen anywhere at any time, and you couldn't predict where it was going to occur, which massively underestimated the amount of concentration and the opportunity to do better policing by focusing priority resources where crime was most likely to occur. But the parallel to that, and I will jump way ahead now to 2016, when Peter and Eleanor Neyroud and I published the Cambridge Crime Harm Index. I think the transformation from counting crimes to weighing crimes and focusing on the severity of crime rather than the premise that all crimes are created equal, which they're certainly not. This is starting to have a major effect on the way British police talk about crime, the way they set their priorities, and indeed, may be moving the whole country away from counting crimes going up or down in the total number of crimes created equal, and moving to the severity measured by sentencing guidelines, so that if a murder is recommended for a sentence of over 5,000 days of imprisonment, and common assault is recommended for one day of imprisonment. Clearly, according to the sentencing guidelines, a murder is over 5,000 times more serious than a common assault. So, I'm not here to defend that a is right or wrong, but I'm here to say that under the parliamentary authority granted to the Sentencing Commission, comprised of judges of England and Wales, they came up with that metric. And so, under a democratic rule of law, that metric has been established, and we can use that in saying that our goal in policing is to drive down harm, not just counts. And the more we can educate journalists about that, the more we can educate police officers, as I was doing at 10:00pm last night at a roll call in the Charing Cross police station in London, where they were heading out into the highest harm from weapons area of all of London, and I was showing them the hot spot maps within this 200-metre hexagon, 6-sided, 2 metre edges, comprising about 100,000 square metres where, within that area that you can walk around very quickly, where was weapons crime concentrated most heavily? And when we say concentrated, what we mean is the amount of harm, because a murder is getting so much more than simply threatening somebody or saying that you have a knife, and all of those are counted as weapons offences. But in the past year at the Met, what we've done is to take London and cut it up into about 16,000 hexagons and rank-order them. So, eight of the top ten hexagons in London are right near that police station, and I was pleased to tell the foot patrol officers who I went out with, not for the first time, that we have this evidence, that if they can discourage people carrying knives, that there will be fewer weapons crimes, and therefore, the harm, the severity of crime, will go down in that area as it's done in ten similar experiments in the United States.

SKN

Well, I can definitely concur about the importance of the Crime Harm Index and I teach it in my classes as well, I tell students about it, and it’s completely new to them, it’s a complete mindset shift to categorise crime according to its harm rather than the count. You said that hopefully, it will change from count to weighting across the country, but actually, it’s happening all over the world. I know that other countries are building a Crime Harm Index based on what you have started. And because that’s how you taught it in some of your training videos, I also say to students ‘We don’t quantify disease based on 1 or 2; we can have 3 colds in a year, you can have 1 case of cancer, obviously the cancer is going to be much more harmful and serious than the colds. So, we don’t use count I disease, but for some reason, we use it in crime, which doesn’t make sense. Definitely a very important point to get across, and it’s such a new concept of thinking about crime that there’s a lot of retraining and educating to do. Can I just take you back to the first achievement you mentioned, you were talking about reducing police shootings, can you talk about that?

LS

Well, the United States has massive gun density. Certainly, the combination of wealth and an interpretation of the US Constitution’s Second Amendment means that everybody is at much greater risk of being shot or even shooting somebody by mistake or perhaps unnecessarily in self-defence, a massive number of deaths by accidents, especially with small children. So, it's no surprise that the United States police are way ahead, sometimes 40 or 50 times per capita ahead of the British or the German police. And the German police perhaps a better comparison because they carry guns, whereas the British police routinely do not carry guns on patrol. But even the German-to-US comparison shows the American police are far more likely to shoot in part because Americans are far more likely to be carrying guns than Germans. So that's the context.

In that context, we were able to make a dent in the number of people shot to death by police in the late 1980s after a 1985 Supreme Court decision, which was unfortunately nullified, or shall we say, mitigated by another Supreme Court decision in 1989, that sort of undid a big cut in police shootings. And until that 1989 decision is overturned, what we are doomed to see is a steady increase, as we have seen since 2020, when we had about 1,000 people a year being shot to death in the United States by police officers, and it's now up to about 1,400. So, that 40% growth in the number of police officers killing people with their guns can be compared to less than ten police shootings that kill people a year in a country of 60 million, which is about one sixth of the US population. So, here we have the wonderful benefit of a gun-free society and a gun-free police force in the UK. But in the US, the need to try to find ways to save lives is huge. And I've spent a fair amount of my life on efforts to have police, for example, give first aid to the people they've shot. I even had the Trinidad Police put these special bandages, haemostatic bandages that stop the blood flow and increase the odds that people will be kept alive, and sadly, the police in Trinidad refused to use them, because they said they had no interest in having people who might shoot them get out of this shooting incident alive. It is a terrible thing to hear, but this is part of the huge diversity across the world in the cultures of policing, not just within police forces, but also by the public and what they expect. So, within that context, we didn't, I think, make as much difference as we could have done, but I think that there is hope for other ways to save lives in the US, which could still happen with enough culture change and political change over time, especially with things like the medical treatment of people who have just been shot. That alone could possibly cut the deaths in half. And nobody's really embraced that idea politically. But it's still possible, and we have the evidence out from some of the conferences I've convened and articles published by medical scholars, public health scholars who show how we can still save lives in the United States. Just because we know how to do it, doesn't make it happen, but it also means that that knowledge will be there 50 years from now, if somebody wants to use it.

SKN

And you mentioned the first aid provision by police to people who have just been shot, potentially by police. Was that the main change that was applied and sanctioned by the court or was there something else?

LS

No, the main change said that you can't shoot somebody who's running away from you and is not armed and is not threatening to kill anybody, even though that was the law of the land in England in the 17th century. But the Blackstone's Law Treatise in the United States, which was brought over from England, was never changed in the US, so most of the states had this common law rule that if somebody is running away from the police who have a reason to arrest him, if the person doesn't stop, you can shoot him. And kill them. And the Supreme Court, based in part on my research, which they footnoted in the decision, said ‘No, you can't do that under the US Constitution’. It violates other terms of the Constitution. And just because it was in the common law doesn't mean that it's porting to our Constitution. So again, hundreds of lives were saved for at least the four years after that decision was made, and not undermined by the following decision by the Supreme Court with frankly Republican appointees differing from Democratic appointees.

SKN

Well, even those hundreds of lives in that relatively short time span count. Thank you for clarifying that. Can you tell us a little bit about your current role? You are the first Chief Scientific Officer of the Metropolitan Police. Why was that role put in place? What does it entail?

LS

The role was put in place because Sir Mark Rowley wanted to take a much more precise approach to the use of police resources. And as a maths undergraduate, he appreciated the importance of precision. He appreciated the relevance of evidence of outcomes, so that if you try to do things in a slightly different way or more precise way, such as where you assign police to do certain activities, as opposed to this general area, but going precisely here, you get better results. He's also very interested in using precision to set priorities. And so, the most important thing that I've done, I think, since coming to the Metropolitan Police, was to identify all of the men who had been accused of at least two incidents of domestic abuse or crimes of violence against women and girls, which is called VAWG in the United the United Kingdom. What we did with that over 100,000 people being accused of crimes against women and girls over the age of ten, and the offenders have to be over the age of 18, there's over 100,000 in one year in London with those criteria being met. And if you rank-ordered them from the amount of harm they've done, you will go from zero under the sentencing guidelines for non-crime incidents, all the way up to about 15,000 days of recommended imprisonment, the equivalent of three murders in one year, which you can achieve by being accused of three murders or by having enough rapes, which are also very high in the sentencing guidelines to add up to that. If you take the top 100, out of these 35,000 for which the median number of days of punishment is only 11, and you're looking at 15,000 [days] at the top of the top 100 [offenders] group. You see these huge differences. It's not just a question of who's a high-risk offender. It's a question of who's on the thousandth floor of a very tall building overlooking a city in which most houses are maybe five or ten stories tall. And these super dangerous people deserve a lot more police attention than they've been getting, especially because if you can get them a life sentence, and we've just done that this week with one of these offenders in the top 100; if you get them a life sentence, then they're not going to be threatening death or rape to all these women and girls in London who are at risk until you can put them into sadly, a situation that's necessary because they're not the kind of people who are going to be responsive to rehabilitation programs, at least not in the short run. And the risk that they'll kill or rape many other females is too great, I think, in the view of the courts that have decided to give them life imprisonment.

But they only get to make that decision if the police focus on those people, those very dangerous people who are getting repeated statements by victims that they're suffering at the hands of those offenders. And case by case, it may not be provable that they did that, that you can convict them of the accusations and give them an incapacitating prison sentence. But what we can say is that if we see that pattern, we can put more resources for investigation that are more likely to support a prosecution, conviction and incarceration of these very tiny percentage of all offenders against women. But they have been ignored, frankly, with the massive investment in the very low level, in terms of sentencing guidelines, high volume offenses, which reflects so many people. And it's understandable why we want to help so many people. The problem is that if we do nothing, 75% of them won't have any future harm anyway. So, in a way, it's spending a lot of police time unnecessarily with people who don't need it nearly as much as the victims of these monsters, if I may say, who are repeatedly committing rapes, much like John Worboys, who was finally put in prison after allegedly committing 104 rapes as a Black Cab driver in London. And to be able to take that high end seriously, at the same time that we help every victim, that we respond to every victim's concerns, that we record the crime, that we look for these patterns. That's a kind of balanced approach. But the reason that Mark Rowley wants to be more precise is that it's only when you ask these questions and you ask, not just who's in a higher risk category. When that higher risk category is as tall as the Shard building, the tallest building in London, it's more important to know if they're on the top floor or midway down or at the bottom. And that's a method, much like the crime harm index itself that had never really been developed prior to the appointment of Sir Mark Rowley as Commissioner of the police in London.

SKN

Again, that is the necessity of rethinking: Where’s the harm? Where's the damage? What do we need to do? Because in the 80s you were told by police officers, ‘It's all random. There are no patterns. There are no such things as hot spots.’ Whereas now, we're seeing there are hot spots geographically. There are harm spots in terms of who is doing most of the damage. It's just so important to quantify properly and look at where is the most need. Because I can't have a scratch on one arm and a pending amputation on the other arm and say, well, it's one and one, they're equal. That’s exactly the mind shift we need to have in policing.

LS

It applies to victims as well as offenders, as well as places. And I would say that at this point, we've done a lot about places, we're beginning to do things about stacking offenders, much like we're doing with terrorism and trying to figure out who are the ones who are most likely to try to kill people the soonest. But the thing we haven't really stacked – with a few exceptions like the Dorset Police, which was the first place to do this study - if you rank-order all the victims as they did in 2016, what you find is less than 4% of the victims have 85% of the harm. And that is the most concentrated pattern of harm that I've ever seen anywhere. And it's the one that's least likely to be used in any police force in the world because they see themselves as offender-focused and serving victims. But they’re not seeing this next step to arranging in rank order of priority, which victims require the most urgent and substantial attention, because they might be victimised by multiple offenders. And that's where I think the opportunity to learn from rank-ordering the victims as well as the offenders and the places is sort of the next step in the process of maximising the harm prevention with scarce police resources by using greater precision.

SKN

And that's the scientific bit there, isn't it? You've talked about rank-ordering the offenders who perpetrate violence against girls and women. It’s probably been just less than two years, that you've been in post? What changes have you managed to make since you've been in post?

LS

Well, the use of retrospective ranking based on past accusations against individuals has been a good teachable moment for the government, for the journalists who cover police activity, for the voters and consumers of police services. I think it's been very important for people to understand the differences between these concepts and the ways in which policing had been done for the last two centuries. But what we're still hoping to do is to not be looking backwards as much as forwards. And that's where we get into this very complex world of predictive analytics, of saying that ‘This person is five times as likely to commit a high-harm offence in the next two years as that person’. And we've been doing that with potential murderers in Philadelphia, when I was at the University of Pennsylvania and working with Richard Burke. We were able to predict that of 50,000 people on probation, which would be the 500 who would try to kill somebody or succeed in killing somebody in the coming two years? And by using new statistical approaches that don't assume the normal distribution, but which use massive supercomputers to look at every possible prediction with a combination of their criminal history, how young they were at the first offence and many other factors, all of that allowed us to predict murder more accurately at the individual level than ever before, Dr Geoffrey Barnes who worked on that is currently principal criminologist at the Metropolitan Police in London, he took that system to Durham Police and developed a custody suite ranking of inmates as high, medium or low risk of committing a very serious crime. And he demonstrated that, if you use professional judgment and experience [of the group of officers in the study], that you get the centre pretty much in agreement with the algorithm. But the quantitative formula identified three times as many people who are high risk as the experienced custody officers did, and the custody officers were also much less likely to force people out of low risk and up into the medium, where you might be keeping them in prison or in jail longer. You might not be looking at alternatives that are appropriate for low-risk people. So, this is a way of trying to look into the future and to ask, how often do you get it right? How often do you get it wrong?

The problem is people don't trust algorithms because they don't like ‘The computers says no’. It's taking a while for people to have confidence in what is ultimately going to be a more accurate. What is ultimately going to be a more accurate approach is to do prospective forecasting based on entire populations as opposed to retrospective analysis of who's been reported to the police by victims. The second one sounds much more legitimate, much more plausible as a reason to be doing intensive investigations. The problem is, they may be getting it wrong. That method may focus us on people who are actually less likely to kill somebody or rape women and try to get away with it, than if we were using the prospective forecasting. So, there's an education issue with prospective forecasting. There's also a technical issue, because the populations are so large we have to use big data sets, and that runs us into the complexity of the Information Commissioner's Office regulations that make it hard to pull together all of the relevant records to make the forecasts. And that's where I think we will have much greater precision to the extent we can overcome those hurdles and move into a predictive phase of analysis that can be compared to the retrospective. So, the prospective can be seen to be more accurate than the retrospective, I think that's what we'll find. But it's not until we can show the data to the public and to the parliament and to the journalists to say, this is why we should be using the prospective tools, and that's what we should be funding. So, I think there's a great opportunity for policing to be improved by further collaborations with data science firms to come up with the accuracy as the standard. How many people do we miss who actually killed somebody? How many people did we successfully predict? In an early comparison with probation, judgment that actually predicted 5% of the serious crime by probationers in London, we developed a prospective algorithm that identified 40%. So that's six times more accurate in identifying serious offenders. That's why we should be going forward to the prospective modelling of who is most likely to cause the highest harm.

SKN

That leads into my next question which is, how would you describe the current state of science in policing? I'll leave it up to you if you want to have a global look on that or UK-based or American-based.

LS

Well, science is global knowledge, and Professor Mazzerolle at University of Queensland has a global police database on research studies that is probably not used by many police forces in the world or even academics who don't know about it, haven't been able to access it. But the science itself is growing. We had one randomised controlled trial in police decision-making in 1980, and now we have about 750 that are completed or near completed. There's millions in medical trials, but of course, doctors have been using medical trials since the early 1950s, and I think the science itself is very robust. There's much growth, much innovation and new revelations from scientific research all the time. The problem is to build that bridge between science and policing. And to be able to put the science into the policing, I think is going to require a new generation of tools that are not just about software programs, but a combination of data engineering and training, so that the systems for decision-making about how police should spend their time can be joined up with the data analytics that help estimate with precision, where is the most productive choice in the range of choices that we could make about what these officers in any group at any level of analysis in a police agency, what are they doing today compared to what they could be doing? Can we get that much more productivity out of their time if we move them into the things that science has shown will yield less harm in the community, if the police do them, than if they're trying to do something else that will reduce harm, but it's not going to reduce as much harm? And so, to get sophisticated about that, I think we need a new generation of police leadership education of the kind that we've been trying to develop at Cambridge since 2007. The focus on evidence-based policing is certainly attracting thousands of police from around the world. That's a very good sign. And so, I think we're revving up the engines and over the next several decades, I think we should see tremendous advances as long as we keep the growth of interest. In not just scientific discoveries, but the use of science, the application of science to everyday decision-making. And we see a lot of it already with people taking for granted some of these concepts of concentration. What we have to do next, I think, is to build that into the strategic level, allocation of police resources and not just as an independent police decision, but as a dialogue with communities to make it legitimate to have them understand why the police aren't doing bike theft, because they are doing prevention of rape and murder. And we've got to make a convincing case, because if you've just had your bike stolen, you don't care about other people getting murdered. You want to get your bike back. And if you see it on YouTube or eBay, then you have a reason to say, well, why don't the police get that bike for me? And that's where we've got to have more dialogue about well, do we want to ignore all these serious crimes to be dealing with the huge volume of property crime in particular, even though that affects people's safety, and we can't say that it's not something we can afford. That's for the government to say. But that's why I think if the police don't make it clear that we have to make choices, and I must say, in democracy, nobody wants to hear ‘We can't afford it’. We want a promise to do everything, but we know where that gets us. If we don't make the right choices, if we don't even fund enough police officers from 2010-2024, as many people say, we didn't, then the consequences of not even checking the background of the police we're hiring and the crisis in vetting, which in 2019, was a massive backlog of not being able to hire police because we didn't have enough people to do the background investigations, and to the extent that they might have been rushed and some things might have been missing, then we have horror stories of police officers who do terrible things including murder. Obviously, that undermines the trust in the police. But if you follow me on this idea, we got to make tough choices, one thing that we can take away from the last ten years in British policing is that you can never let the checking of background of police officers be put to the side because we're too busy handling public order or other issues that may come along. They might be more urgent than doing background investigations, but doing those background investigations thoroughly and then actually looking at them with the science of prediction, which we're very close to being able to do. And in the United States, there's a company that's tracking over 100,000 police officers across over 100 police forces, and they're combining those data to predict and indeed prevent police misconduct, where you see a very high risk of it developing by using the statistical scientific approach. And so, with greater science, I think we can have even greater police forces and democracies who are likely to have officers who are at risk not get hired in the first place, and if they become at risk while they're serving over ten or 20 years, that we can support them with mental health or whatever other steps are needed to keep them from doing something criminal, something that will bring great disgrace on the police service. And that's where I think science has the greatest potential to help advance trust and confidence in the police, which is fundamental.

SKN: Can you give us a utopian scenario of a future where policing is led by silence? What would a typical day look like? What kind of decisions and actions would be determined by that?

LS

You'll forgive me, I'd rather not go with utopia, in which I have dabbled. I've written about police forces that would be started from scratch, especially in the United States, with much higher education requirements, much higher salaries, and a whole new look for the police institution. It's not going to happen. Yes, it's utopian, but I think it’s not as useful to talk about as what is going on in the world of business. And what you see there is the combination of science with data analytic technology and programmable decisions that are supported and monitored by expert humans. But if we have the data engineering, that will actually do the calculations that would, for example, give guidance to a constable on the beat that I was on last night, instead of just showing him a map, which is what I did at the briefing, I’d say, here's 43 hot spots within Leicester Square, and you can go to this bigger one or you can go to this other one, but, you should be paying attention to the map. It's actually pretty hard, because you're walking around looking at people and seeing who's threatening or who's suspicious. If instead, if what we had was a kind of Google Maps voice, saying, ‘Turn left here, turn right there’, or ‘This place had five knife robberies in the last three months’. And that kind of information is programmed to be delivered to the decision-maker, who is the police officer who is out encountering potential offenders. We're never going to get away from that. Police are not going to be losing their jobs with artificial intelligence. But what they can be is far more effective if we have that kind of precision in giving guidance or at least suggestions to police officers on how to spend their time, where to go, what to do. Much like we're used to when we drive through an unfamiliar city with the voice of Google Map, saying, ‘Go through this set of lights, and then at the next one turn right’. That's doable. We know how to do that today. We've got the hot spots mapping. We've got the technical tools. And to my knowledge, there's not a police force in the world that has put the science and the technology together with the operations to deliver that kind of maximised deterrent effect of a limited number of police.

SKN:

What would you say is currently the biggest challenge that is faced by policing in your opinion?

LS:

I think the biggest challenge that police face in democracies is the pressure to take sides, the pressure against nonpartisan, neutral policing. And that is where Britain I think leads the world because it has embedded, since 1361, this principle of operational independence of the local decision-maker, and then in 1361, it was magistrates, and the metropolitan police commissioners were magistrates, and therefore, all the constables were in effect magistrates’ agents. The idea was that the King couldn't possibly make all the decisions that needed to be made, and they should be made by people with local expertise. And that's exactly what each police officer is and can be. So, if what we say is necessary is that the police be not two-tiered, which is what they've been accused of in the wake of the tragic murders of three girls in a Liverpool-area village, Southport. The anti-immigrant protests that came up from groups that self-describe as far right, said the police were biased in favour of the immigrants, in favour of ethnic minorities, and one of them, who's been arrested by the police for throwing bricks at the police, said ’Hold on. I'm British. Let me go. Obviously, I'm not the problem. It's those other people who are the problem’. So, if this country, the mother of democracy, the birthplace of the idea that actually constructed in the 19th century that we have one law for all people, which we didn't much before 1833, but we certainly do now, and to have this neutral position of the police under a rule of law is the way in which the police make the world safe for democracy, but they are under constant pressure to be the kind of democracy in which 51% gets to torture the 49% is simply because they're greater in numbers. Now, that's some people's vision of democracy. The British vision of democracy, which we emulate in the United States and many other countries is that it's a democratic governance of a rule of law, but the rule of law stands independent of the party of the day. And the Prime Minister who asked the Commissioner of police in London to stop a pro-Palestinian demonstration on 11^th November 2023, and the police commissioner said ‘No’, that's operational independence. I don't know of another country in the world in which, we have centralised policing in the way where the Prime Minister could make a direct request to a single police executive. But I certainly don't know of another country in which the Police Chief would say ‘No’ to the head of government. And that's something many people could debate. But if you go back to this, what's the biggest threat, the biggest threat is that political parties will take over the police. And what's just happened in Bangladesh, where the police were serving a prime minister who was unpopular, and in the course of protest, killed 100 people to defend the prime minister they were working for. Now the prime minister's gone, and ‘the police have fled’. We can't have police taking sides with a prime minister who might be driven out of the country, because it leaves the police vulnerable to having the army kill them or the population kill them. And it's all because the police in Bangladesh were not neutral. They were not seen to be neutral. They didn't act in a neutral way with respect to defence of life principles. And I say this having had the occasion to lecture to the Bangladesh National Police Leadership Academy about violence against women, as it happens. But if you had asked the people in that room, are they at risk of having to flee for their lives because they're doing what the Prime Minister asked them to, they would say, ‘No, of course not. Everybody does what the Prime Minister asked them to’. Well, not in Britain. And that's a good thing, and we should be proud of this operational independence in Britain, and we should never underestimate how hard it is to hold onto it, because that's why the British police are being battered with the accusation of two-tiered policing. And which is, I think, foremost in their minds, every time there's a demonstration, every time there's an issue in which there is a question of not using too much force, which can then later be characterised as favouring one group over another. When really, what they're trying to do is to make all decisions proportionate in a neutral way. That's our challenge that I think our leadership understands fully, doing a great job without science. So, the threat is really not about how good is the science. The threat is how well and how articulate the police leadership can be managing this threat. And I must say that on this date in August of 2024, things have gone a lot better because of the very strong foundations of British policing going right back to Robert Peel. They've gone a lot better than they might have done. But also, I think we have to give credit to the current leadership who have applied those principles and kept us on an even keel in a very stormy time.

SKN

And over the time that you have been in England and in the UK, what changes have you seen in policing?

LS

I think the biggest change in policing has been this growth of curiosity and commitment to innovation in the police leadership cadre. And of course, the diversity in the last quarter century with more women, more minorities, and others, making the police force look more like the population of Britain and of London. Where we have improved in our perspective and understanding of how best to deal with diverse societies. I think it has been through the skill derived from lots of hard lessons, lots of mistakes, lots of commissions of inquiry and reports, but if you spend time with British police officers, they talk about those reports. They go right back to a report written in the 1980s after the Brixton riots and then reports on the Stephen Lawrence tragedy and other things that the general reader of the newspapers might forget about, but it becomes a dominating theme in British policing that we can't make those mistakes again. We may make new mistakes because it's a new world. But I think what's really good is the fact that the well educated police leader will be very conversant with all of those issues and have a very nuanced grasp of how to manage them in a neighbourhood meeting, at a roll call with patrol officers about to go out to work, in front of a parliamentary committee, or in countless interviews with news journalists and explaining things to the public. I think the police are much more articulate, not just about what the police are doing, but why they're doing it and what the various considerations are that they must be balanced, to try to pull together the conflicts of a society that needs to have a consensus and to be able to work together despite differences of opinion. That's what I admire so much about the British police leadership.

SKN

And as my last question for you, Professor Sherman, if you had a magic wand and you could make one major change in policing, what would that be?

LS

The one magic wand change to make in policing would be, in my view – so many important things to choose from, but I think would be the starting salary for police, which once was much higher relative to other occupations. And a lot of the police leaders I've worked with were hired under much higher starting salaries. If I could wave a magic wand, I'd maybe increase by 50 to 100% the starting pay of police officers, that would give us far more applicants and allow us to take much more rigorous steps in picking the best of the best of people who want to commit their lives to police service. But I think the chances of that happening are very small, but you did say a magic wand. So, if you believe in magic, that's the way to go.

SKN

Indeed, I did say magic wand and this is a very interesting point you bring up because it's not just a matter of finances, is it? It's about like you say, good people who make the decision that policing is a better option for them than something else that they might go into, and it makes policing more robust. Somebody was telling me recently that Canadian police have a relatively higher salary compared to many other countries. And they are more interested in their own development as a police officer, and they might invest in their education and their skill set. You don't have that in many other places because they can't afford it, and everywhere you have underpaid police, you get corruption and you get problems, so it's not just about the finances, it's actually much wider reaching, isn't it?

LS

Very wide reaching, and if you increase the pay and the education that they can get after joining the police and to find the best way to do that, so they're not trying to both go to the classroom and go out to the streets on the same day, which has turned out to be difficult for many officers. So, we can get good at not only recruiting but developing people who have both substantial levels of moral commitment to the values that are reflected in British policing, but also that they have the cognitive ability to make the best decisions very quickly with all of the support of the technology that is going to be possible.

SKN

Professor Sherman, thank you very much for your time.

LS

Thank you, Susanne.