Home Affairs Committee
Oral evidence: Hate crime and its violent consequences, HC 683
Tuesday 6 February 2018
Ordered by the House of Commons to be published on 6 February 2018.
Members present: Yvette Cooper (Chair); Rehman Chishti; Sir Christopher Chope; Sarah Jones; Tim Loughton; Stuart C. McDonald; Naz Shah; John Woodcock.
Questions 171-258
Witnesses
I: Professor Neil Chakraborti, Director of the Centre for Hate Studies, University of Leicester; Sam Smethers, Chief Executive, The Fawcett Society; Dr Loretta Trickett, Centre for the Study and Reduction of Hate Crimes, Bias and Prejudice, Nottingham Trent University; and Professor Mark Walters, Co-investigator of the Sussex Hate Crime Project, University of Sussex.
II: Dr Jelena Dzakula, Fellow, and Professor Damian Tambini, Research Director, Department of Media and Communications, London School of Economics and Political Science; Dr Martyn Frampton, Reader in Modern History at Queen Mary University of London and Co-Head of the Security and Extremism Unit at Policy Exchange; and Dr Martin Moore, Director of the Centre for the Study of Media, Communication and Power, King’s College London
Examination of witnesses
Professor Neil Chakraborti, Director of the Centre for Hate Studies, University of Leicester; Sam Smethers, Chief Executive, The Fawcett Society; Dr Loretta Trickett, Centre for the Study and Reduction of Hate Crimes, Bias and Prejudice, Nottingham Trent University; and Professor Mark Walters, Co-investigator of the Sussex Hate Crime Project, University of Sussex.
Q171 Chair: I welcome everybody to this session of the Home Affairs Select Committee inquiry on hate crime this afternoon, and I welcome our witnesses today. The focus of today’s evidence sessions is on what the response should be, and what further policy measures we should be looking for. Can I ask each of our panel to introduce themselves and say which organisations you are from? In the spirit of Vote 100 today, I think I should start with The Fawcett Society.
Sam Smethers: Good afternoon, everyone. I am Sam Smethers, Chief Executive of The Fawcett Society.
Professor Chakraborti: Good afternoon. I am Neil Chakraborti, Professor of Criminology at the University of Leicester and Director at the Centre for Hate Studies, also at the University of Leicester.
Dr Trickett: Hello. I am Dr Loretta Trickett from Nottingham Trent University. We have a centre in the criminology department that deals with hate.
Professor Walters: I am Mark Walters. I am a Professor of Criminal Law and Criminology at the University of Sussex.
Q172 Chair: Staying on the Vote 100 theme to start with, what are your views on how abuse and hatred in public life has changed? This is in the context of a postcard sent to Emmeline Pankhurst that is currently being circulated on BuzzFeed today, which reads, “You set of sickening fools. If you have no homes, no husbands, no children, no relations, why don’t you drown yourselves and get out of the way?” Abuse is not new. We have had abuse around gender, race, sexuality and different aspects over very many years. What do you see as being the main features of hate crime or abuse in public life that you are most concerned about today?
Sam Smethers: There are two things. One is the sheer volume of it. I think people in public life today are dealing with an unprecedented volume of hate against them on a regular basis. You cannot participate online without encountering it, so that is something that we did not see 20 years ago. It is relevant for me to say that I used to work for Bernie Grant, who was one of the first black MPs, and he certainly used to get hate mail. There is no question about that; we used to see quite a lot of hate mail but it was posted to him. It was still vicious and violent, the worst kind of hate mail, but he was not getting it bombarded at him all the time. If everything is coming through your phone and that is what you need to communicate with the outside world, then you can’t stop it because everything is coming at you through that phone and effectively you are compromised in how you can function as a candidate or a representative. That is the first thing.
Secondly, we seem to have a no-holds barred kind of approach to the thresholds for this stuff now. Some things that people feel free to tweet, they would not say to someone’s face, they would not say in a letter, and they would not probably even put in an e-mail. There is something about the nature of the medium that brings out the worst in us all. It is both of those factors combined that make it quite a potent thing. We have now lots of evidence that women are being particularly targeted, certainly black and ethnic minority women are being very heavily targeted. We have to be alive to the fact that this is very real and the risk of crossover between the online world and the offline world is also very real.
Dr Trickett: I think we are living in concerning times. We have seen over here what is happening globally: hate speech directed at politicians, a lot of female politicians in particular, one of our MPs being murdered on the street. That is a wake-up call for all of us. It is a problem for our young people in particular because, as Sam was saying, with new forms of communication and children with smartphones, it is about those children retweeting messages and knowing what is acceptable and not acceptable.
We have seen, certainly in the last 10 years since I have been working in hate crime, that we have gone backwards in some ways, because hate crime is on the rise. We are seeing it very close to home, across Europe and elsewhere globally. As Sam said, it is about the capacity for identifying what we should tolerate and what we should not tolerate. I think these are pressing times for hate crime. We are trying to do a lot with legislation, but sometimes that legislation is not effective to tackle all the variants that we see.
Some people are more willing to express these views on public transport. I think at the moment people feel that they will get away with it, so although we have legislation and the public voice, a lot of people feel that there is going to be no penalty and no response to it, so they can air these views.
Professor Chakraborti: I would go along with those points. What concerns me most about the escalating level of hate crime we are seeing is that to a large extent it is experienced by people who often find themselves in very marginal positions within society. These are people who often have very little faith in authority, people who find themselves in positions where they have very little economic mobility or political clout, and people who are very unlikely to report their experiences.
I have heard for more than 20 years now people talking about how we can take official figures with a pinch of salt because they give us only a partial picture. We know that and we have known that for years, yet we have not done enough to overcome those barriers to justice felt by so many victims of hate crime across all strands.
Professor Walters: We are at a crossroads. Some commentators are saying that we have entered an era of hate. That is combined with a really toxic political rhetoric, especially after the EU referendum, and we have huge numbers of hate crimes being reported and recorded—an increase of 29% last year, over 80,000. Some recent statistics from the Crime Survey for England and Wales show that there are still over 200,000 hate crimes estimated to be committed each year in this country alone. That is probably an underestimation. There is an issue of not just offline hate crime, but of huge amounts of online hate crime being committed every day. It is not going away, despite the reforms that have been made to the law and to other criminal justice practices. There is a lot more work for us to do. I think that, if we get a chance to come back to this, law reform in particular is needed now.
Q173 Chair: We want to explore what different policy responses might be, but just on the overview, if you were to pick one thing, what would be your top measure or your top response that you would take or urge us to be pushing for?
Professor Chakraborti: Can I seek some clarification? Is that criminal justice policy that exists or—
Chair: To be honest, it could be absolutely anything—you have just described the problem as you see it but perhaps you could pick your top one thing. We will ask you about all of them, but I am interested in what you see as being the greatest priority or the most important thing for us to definitely not lose sight of in the things that need to be done.
Professor Chakraborti: I would like to see the police and other partner agencies developing reporting processes through meaningful consultation with diverse communities, rather than imposing third-party reporting processes or police reporting processes that do not really work—they are not accessible for victims or for witnesses—and consulting with diverse communities and saying, “What would work for you? What would make reporting more accessible?” I would like to see that across the board.
Sam Smethers: I would like to see misogyny recorded and recognised on a par with other hate crimes. It may be that you could add it to the current framework or you might need to design a framework that is unique to misogyny. There are different views on that and there are issues you would have to think about in how you implemented it, because of domestic violence, FGM and crimes that are clearly not stranger crimes. It is difficult to present it as hate if someone purports to love you but is still committing a crime of violence against you. You would have to think about how you implemented it, but it is absolutely time we did it.
Dr Trickett: Sam has taken some words out of my mouth. We are both doing some of the same things, so I would echo everything she said on that. I would like to see real improvements in how we train police officers. I have done some research on this and at the moment we are not fully investing in that training. How we train the police in England and Wales is very disjointed and quite a lot of it is online training or PowerPoints.
It is particularly difficult with hate crime. As Neil was saying earlier, we are talking about sometimes very marginalised communities, many of whom the police have not much exposure to, because the police have much less exposure to hate crimes. I know we are saying there is a lot of it but they have a lot less exposure to hate crime than more traditional forms of offending with other types of street-based offending, such as, I suppose the run-of-the-mill crimes that the police are dealing with. There are some issues around that. One is the police understanding how hate crimes manifest and how they impact on not only the individual but also on wider communities. It is very difficult if they are not having that sort of outreach contact.
The other thing is it that it builds into how we protect people if they have reported to the police, because we have no standardised risk assessment for hate crime in this country. We are working on that at the moment at Nottingham Trent with the Nottinghamshire police. What that means is sometimes the police, with the training they get, are very ill-equipped to deal with hate crime victims and put in place those protective measures. They are also ill-equipped, I think, for recognising what hostility looks like in these different sorts of situations.
A prime example of that would be with disabled people. We have a woeful record on prosecuting disability hate crime, for example. I believe that is because the training is not fit for purpose. That links into you having to have good investigation and good evidence gathering and safeguarding. I would like to see prosecutions going up. I think that is the whole point of doing this. Hate crime prosecutions across the board for all of the strands as they are recognised at the moment are still extremely low. For some groups, it is even lower than others. As Neil was saying, if the reporting, as we know, is the tip of the iceberg and then we look at those, we cannot be saying we take hate crime seriously if we are not prosecuting at quite a high level.
Professor Walters: It is difficult to choose one. If I may, I will come back to restorative justice later on, but I will opt for law reform of the way in which hate crime is prosecuted. A lot of the criminal justice responses and policing of hate crime pivots on the legal framework we have, and at the moment there is a disparity in protection of different types of hate crime within the legal framework. We have specific criminal offences, such as racially and religiously-aggravated assault, for example, which are substantive offences in the criminal law. Then we have other types of hate crimes, such as sexual orientation, transgender or disability-aggravated crimes, which are only dealt with at sentencing. Then you have other crimes that can be aggravated by gender identity or age, which can be dealt with under sentencing guidelines as a form of aggravation.
You have a hierarchy of hate and a disparate way in which the law treats different types of hate crime. That has resulted in different levels of response to different types of hate crime. Until we get parity and a single legal framework through which we can deal with hate crime, we are going to have differences in the way in which different types of hate crime are treated. Even if we have parity in protection, that is not necessarily going to equate to parity in enforcement of the law.
One of the problems, as Loretta pointed out, is disability hate crime. The law requires that there be some form of hostility for there to be an actual criminal offence that can be aggravated by, for example, racial hostility or sexual orientation hostility or disability hostility. That is very difficult to prove in court for most disability hate crimes, where a lot of disabled people are targeted because of who they are, because they have a perceived weakness or because they are perceived to be vulnerable, but it is difficult to say that that is about identity-based hostility.
We were doing some research at the University of Sussex last year and one of the CPS lawyers said to me, “It is like hammering a square peg into a round hole”. The legislation that we currently have is not fit for purpose for certain types of hate crime. I would like to see parity in protection but also an amendment to the wording of part of the legislation to change the way in which we prosecute and convict hate crimes, so that we can do it more successfully and have more prosecutions and more convictions. From there, we can think about smarter punishments and different ways to address the causes and consequences of hate.
Q174 Naz Shah: Talking about legislation, it has been suggested that aggravated offences should be extended to include other forms of hate crime beyond those based on race and religion. Do you believe this is a good idea and what impact is it likely to have? Perhaps we can start with you, Professor Walters, as you were talking about it.
Professor Walters: That is exactly what I am referring to. The big difference is that if you have criminal law offences, that has a real impact on our criminal justice system and the way in which those types of offences are treated. If you have racially-aggravated offences, for example, the criminal justice system is more likely to treat those more seriously because the police have to gather evidence so the CPS can prove in court that a particular offence has been committed. The hostility makes up a substantive element of the crime.
If you do not include, for example, sexual orientation or disability within the Crime and Disorder Act 1998, you are prosecuting only a basic offence, which then becomes aggravated at sentencing. It means that a lot of the hostility is filtered out of the system, because the police are not necessarily required to gather the evidence to prove it in court. The criminal law has a powerful role to play in challenging hate crime. It determines what type of offence is going to be prosecuted. The label you have, the type of penalty you have, will be flagged on the system and that will then filter through to post-conviction, probation services, community rehabilitation companies and so on. I believe that all hate crimes should be treated the same and they should all be prescribed within the Crime and Disorder Act, which creates specific criminal offences.
Q175 Naz Shah: Do you think it is a good idea that aggravated offences should include others?
Professor Walters: Absolutely, because it will mean that all types of hate crime are treated seriously. It also has symbolic messages. If you say that one hate crime should be dealt with in the criminal law, and another at sentencing, one has a higher penalty range than the other and you are sending out a message that the hate crimes that fall under sentencing provisions only are less serious—less important. What kind of message does that send to the communities that we are trying to protect from hate crime? I think it is very important for practical reasons and symbolic reasons that we cover all those types of hate crimes in the criminal law.
Dr Trickett: I think some of the categorisations have become almost exclusive. One of the panel members was talking earlier about the hierarchy of hate, and Professor Walters was just talking about that. We have to recognise what we are trying to do with the legislation. What you are talking about for me is a demonstration of hostility or prejudice against a person on the basis of an aspect of their identity, a characteristic. I believe, like Sam does, that we should look at gendered hostility and violence against women. That is one of the things we are trying to do in Nottingham with misogyny as a hate crime—a policy that is being introduced there. But there is room to think about other victims within that who are being targeted for an aspect of their identity.
We need to think about the historical context in which hate crime has evolved. It has come about from the victims’ movement, largely from the States and Canada. We were influenced by that and we started to look at victims in the criminal justice system and try to be more inclusive of that. But what that meant was we started off with race and religion and we have added these other categories afterwards, almost like bolt-ons, that are not given the same status.
For me, one of the issues is that although we are predominantly looking at minority groups, it is whether or not we are saying, “We can only afford people protection for hate crime on the basis of some sort of minority status”. That has been one of the arguments put forward for not including women, for example, because they are not a minority group, or are we looking at tackling prejudice and the impact of that prejudice on that individual and that wider community? These acts of misogynistic street harassment, for example, impact on all women. Mark has written an article on this in the past, but you can look at rape, for example, and the impact of rape on all women and women’s freedom of movement and fear of crime and sexualised street harassment.
I guess it is looking at what are the components of the law and what we are looking for it to do. For me, the key thing is the prejudice and the impact on a person and on the wider group from which that person comes. I think we need to introduce new categories.
Sam Smethers: I agree with much of that. The question with misogyny, as we have said, is whether you just add it to the current system or you design something bespoke for it, because it is not a minority category, it is a much bigger thing. Genuinely, I am not clear what I think the right answer is to that. I am quite pragmatic: sometimes if you think you can just add it to what is already there, that is probably what you are going to do, but is that really the right way to deal with misogyny? That is one of the things we need to get to grips with. At the moment we have seen conflicting evidence on that.
There is some quite interesting work being done by Engender in Scotland, feeding into the Scottish review of hate crimes. I would point you to that, because that is a really interesting paper. They have done quite a lot of thinking about it. They have come up with a recommendation that there should be a standalone piece of legislation on misogynistic hate crime and it should be a much bigger concept as it does not add it to the existing strands.
When we did our law review, we came up with the opposite recommendation—that we would just add it to the existing strands. It does need a bit more careful thought about how it fits in with the existing criminal law framework to deal with existing crimes against women. Fundamentally, do we think we ought to be recognising misogyny in our society and taking it seriously and dealing with it? Absolutely, yes. It is just how we do it.
Professor Chakraborti: I agree that parity across all of the monitored strands of legal provisions is absolutely appropriate. There are practical benefits, and there are symbolic, declaratory effects, as Mark has alluded to. It is less so in deterrent value, but I think we would all agree on the value.
The only caveat I would say is that this is just one part of the solution for delivering justice to victims. We know now there is mounting evidence to suggest that victims of hate crime do not always want the conventional criminal justice response. There is not overwhelming demand on the part of victims to get caught up in a slow, intimidating, incomprehensible criminal justice system. Longer sentences can be entirely appropriate in some contexts, but it is not always the right solution. Evidence suggests that restorative interventions, more of a focus on education interventions, and programmes of rehabilitation inside and outside of prison, can have really powerful effects, not just on offending behaviour but also in what victims want from the justice system. It is really important when we are thinking about criminal justice outcomes to think about alternatives as well.
Q176 Naz Shah: Dr Trickett, you are doing some work with Nottinghamshire on misogyny. If misogyny is categorised as a hate crime, in your opinion should it be extended to other groups, for example gypsy travellers, travelling communities or Goths?
Dr Trickett: Yes, I absolutely think that. At the moment, the legislation that we have is too restrictive and we need to look at why hate crimes were introduced. You could look at it and say that potentially anybody could be the victim of a hate crime. We are obviously not always going to be able to prove that, but I do not think it can just be that you would be recognised as a hate crime victim because you also happen to come within a minority identity. I accept that we are talking about power differentials with hate crime offending, but within that moment the power imbalance is such that the person who is committing the hate offence is demonstrating power over a person.
We need to look at extending the categories. That is going to involve revisiting the legislation as it is currently constituted and looking at redefining the law, but we need to include other groups. I do not think, as a woman saying that I want misogyny to be included, that I can sit here and say that I would not want that. What we are doing with the research at Nottingham Trent at the moment is using focus groups and surveys, and asking men what they think about the introduction of misogyny as a hate crime policy. That is because we are aware of the fact that only a minority of men—a significant minority but only a minority of men—commit those sorts of offences, and for many men, their wives or girlfriends or daughters are on the receiving end. I think that men have a duty in society to think about how they tackle that. We have a lot of support for that campaign from some of the men who are feeding back for us, but it is also about how you challenge people’s attitudes.
If I can draw a parallel there between misogyny and gypsies and travellers, for example, there is a lot of prejudice against gypsies and travellers, which comes from all walks of society. They are one of the most stigmatised groups that we have. To go back to parity of dealing with hate crime and victims, although at the moment you have some examples of good practice across the country on different hate crime steering groups and different police initiatives, the impact upon them as a community is really detrimental. I do not think the support networks are there.
In order to have support networks and parity of prosecution or even restorative justice, if you are developing those approaches they need to be taking account of different types of hate crimes against different types of communities so you can get that knowledge: what is it for those groups? You might find some groups are particularly open to restorative justice, for example. That is something to consider, but I do not think we can continue to remain with the protected groups that we have.
Q177 Chair: Can you give us an example of something, if misogyny was made a hate crime, that would be covered or that you think should be covered that is not being at the moment?
Sam Smethers: Going back to the Nottinghamshire example, it is quite interesting that it is not so much new offences. It is rebadging existing offences and women having the confidence to come forward about those offences, whether it be an assault—and there were one or two cases of assault, but also street harassment—where women have felt able to report them and are encouraged to report them because they have had more confidence in the system supporting them and recognising that.
The women’s organisation in Nottingham has played an active role in training the police, so it has been quite a positive relationship between the women’s organisation supporting women and improving the way the police are policing those sorts of offences against women. It is a mixture of criminal offences and hate incidents, which are not necessarily criminal offences but are the kind of thing that women want to stop and that inhibit their daily lives, whether it is how they walk down the street or the route they take home or whatever it might be. It has to encompass whatever the woman feels fundamentally is the offence against her. We are not drawing new offences particularly; it is about improving the way the existing offences are policed.
Q178 John Woodcock: Dr Trickett, you mentioned challenging hate crime. Can I ask you and the rest of the panel about counter-messaging campaigns and, very broadly, what you think is effective and what ought to be the role of Government and the public sphere in supporting or enabling them?
Dr Trickett: In Nottingham we have had a lot of initial backlash to the misogyny policy, as you perhaps would expect because it tackles—certainly when you are talking about sexualised street harassment of women in public places—an activity that to a certain extent has been normalised. As Sam was saying, we have these existing offences, for example public order and assault, a variety of offences that we could use, but women have put up with and tolerated it because it has been normalised in society.
One of the messages is that the general public is still quite ignorant about what hate crime is. I think that is the message we need to get out there, because they think it is about extreme hatred. When you talking about targeting women, for example, it is often dressed up as complimentary, “It was a compliment about your appearance”. The police are filtering some of the things we are getting through at Nottingham, because there will be a continuum for lower-level offences and quite high-level offences, some of which are sexual assaults. But I think it is about getting across the message of what hate crime is. Quite a lot of members of the general public still think it is about extreme hatred.
I think there needs to be an understanding about what the hate crime legislation is built on. It is about hostility and prejudice and what sorts of behaviours might fall under that, because there is still real misunderstanding. It has not been helped by reporting in the media. By doing that, you can then start to challenge the behaviour, because you are getting people to think about their behaviour in a new light.
One of the things about the misogyny policy is trying to get people to think about sexualised street harassment of women in a different way, using a different message to challenge what has become acceptable behaviour for some people. We need a drive on educating the public about what hate crime is. We have lots of posters about hate crime, but people are not really sure what the basis of it is and they do not know it is in addition to other forms of offending either. We need a little bit more education on that.
Q179 John Woodcock: Is the primary message of your misogyny campaign in Nottingham, “Don’t do this because it is not nice for the victim” or is it, “Don’t do this because there could be consequences for you, the perpetrator”? Did you make a conscious decision about that?
Dr Trickett: One of the difficulties we are going to face at some point, and I would suggest in the not too distant future, is that we have had a substantial amount of reporting but that is not necessarily matched by prosecutions for hate crime. That is the case across the board. There is a lot of media interest at the moment, asking us, “How many prosecutions have you had for hate crime?” But I think our conscious decision is that we want to change the status quo in Nottingham about this sort of behaviour. We believe it is fuelled by hostility against women and so we think that hate crime is a good way of trying to tackle that.
The longer-term consequences are that we want to change the behaviour of men who engage in it. What you hope is that they start to realise that, “This is unacceptable behaviour in Nottingham and there might be some sort of consequences, potentially prosecution but the police taking some sort of action against me”. It is a bit of both, but at the moment we started off by being more concerned about changing the perception and encouraging women, saying “You do not have to put up with this. What is happening to you is a crime, anyway, and you can go to the police about it”. It is mostly about women to start with, filtering through to warning men that, “If you continue this sort of behaviour, you could get a criminal record”.
Professor Chakraborti: I agree absolutely with what Loretta has just said. Within our analysis of campaigning and what seems to work on hate crime and what does not, three themes broadly emerged as good news stories. There was some campaigning that did not work, but there were three themes that came out.
Chair: We have a Division; we have a vote. It is an occupational hazard, I am afraid, for us. We will adjourn for 10 minutes for the vote. Thank you for your patience.
Sitting suspended for a Division in the House.
On resuming—
Chair: We will continue our evidence session. John Woodcock.
Q180 John Woodcock: You were just getting into your stride, Professor Chakraborti, but you were cruelly cut off. Can you remember what I was asking?
Professor Chakraborti: If I can remember what I was going to say. We will help each other along, but I think it was three themes that we have noted as working particularly well in hate crime campaigns in different contexts. One is campaigns that focused on bystander intervention, so campaigns that took the focus away from the individual victim per se exclusively and acknowledged that this is not just a problem for victims and not just a problem that the police need to find a solution to, but it is a wider societal problem that we all need to respond to collectively. I think post the spike after the referendum debates, we have seen the need for collective endeavour in responding to hate crime. Bystander interventions and advising people on how to step in and create safe spaces has worked very well.
We have also seen campaigns work quite well when they reach within and across different communities, so rather than relying quite tokenistically on engagement with so-called community leaders, extending beyond that and engaging with the real diversity that makes up different parts of the country. That has been successful. I think Loretta acknowledged earlier that sometimes it is the tip of the iceberg when we are engaging. We know that victims come from so many different communities, so many different walks of life, but often most victims do not even recognise that term “hate crime”, even if they are encountering hate crime on a very regular basis. Campaigns that have meaning to different communities from different walks of life is so important. That would be the second thing.
The third one would be success story campaigning. Often campaigns that have just an alarmist message do not work. What seems to work quite well among some victims and some witnesses is success stories, so a positive outcome from reporting a hate incident or a positive outcome from a prosecution. Quite often we do not see those success stories. There is not a large number of them, but those success stories are not widely shared and instead we see the negative headlines and the problems.
Those are three of the themes that I think can be quite relevant if we are going to try to produce effective counter-narratives within our campaigns.
Q181 John Woodcock: That is really helpful. Sam Smethers, do you have anything to add?
Sam Smethers: Our first campaign is to get misogyny as a hate crime at all. One of the attendees at our law review launch stood up and said she would like misogyny to be recorded as a hate crime because it would really help her, as a teacher, with the education of her teenagers at school. She was saying this would help with sex and relationships education and misogyny to effectively work together. If it was recognised outside the classroom as a hate crime, that would strengthen her hand in the classroom in dealing with the behaviour and conduct of young people in the school. I thought that was an interesting message. We know relationships and sex education is going to be introduced. It needs to be gendered, good quality and inclusive. We don’t know yet how good that is going to be, but we want to make it as good as it can possibly be. I think it would be incredibly powerful if at the same time we had the message coming from that side of the classroom too.
Q182 John Woodcock: Professor Walters, I think your research questions the effectiveness of infinite resources for education campaigns if there is no evidence that they have been successful. Is that the right characteristic? How would you suggest improving that?
Professor Walters: It is not quite the right characterisation—
John Woodcock: Forgive me.
Professor Walters: —but it is almost. I think you are referring to a report that we published last year called “Preventing Hate Crime”. What we have argued was that education measures or counter-narrative initiatives, whatever it might be, should be evaluated as they are implemented, so before you invest in, “Okay, we are going to go with this type of measure”—
Q183 John Woodcock: It is not that you think they do not work; it is just we do not know if they work because they have not been properly evaluated?
Professor Walters: Exactly. There will be aspects of those measures that work really well and aspects that might be counter-productive. We need to know what parts of those measures are the most effective. You have to build research and evaluation so that when you are putting all of that investment in you get it right. That is the point.
Chair: Apologies for having to pause you again. We will suspend this sitting for a further 10 minutes.
Sitting suspended for a Division in the House.
On resuming—
Chair: I reopen the Committee’s hearing. Sarah Jones.
Q184 Sarah Jones: Dr Trickett, you were talking earlier about training for the police in dealing with hate crimes. Can you talk a bit more about what that training should be and whether or not you think there should be specialist hate crime officers or that everybody should be having the same level of training?
Dr Trickett: There are two issues. When you have specialist officers, you train officers and then you have that specialism in-house and that is ongoing, so you can supervise people. In Nottingham they had a hate crime manager who left and they found that had a negative impact. I am not saying that there is no need for specialism, but I think all officers need to be trained on hate crime. One of the issues that we face at the moment is that some of the more workshop-based training has been largely sacrificed because of austerity measures, because you have to take officers off duty to do that sort of training.
It strikes me—I have just finished writing an article on this—that we train the police a little bit in isolation from other professions. In the training that the police get there is not active learning going on—it is very passive. They are getting PowerPoint presentations or they are getting often NCALT training, but I think it is about engaging the police in tackling hate crimes within teamwork in a classroom environment. Of course, in addition to that you do need impact for the communities that are affected by hate crime.
At Trent we have looked at how you might improve computer-based training as a sort of back-up, as an ongoing resource—because obviously if you have a workshop the police will do that and then they will go off—and at how you might be more interactive, like those scenarios, and also to complement workshops in computer training, drawing on some other areas of expertise elsewhere in the university to look at how you might stimulate situations and get police officers to do some of that online. We have looked at training that is provided for medics, for example, so at how you might get police officers to think about risk assessment, which is something that I think we do need to include in training, but we must then look at what range of interventions you might put in—“What are the issues for this particular victim?” and then getting officers to work together as a team to get through that.
This is the time to be thinking about it. At the moment we are shaping that policing and thinking, “Do police officers need degrees? What might future training look like?” However, I do think there is quite a way to go on training. At the moment it is not standardised across the country. There are some pockets of good practice, but I think we need a major rethink about that.
In terms of specialist officers though, it is very difficult for police forces to be able to continue good practice if they do not have somebody in post as a hate crime manager. The reason for that is that often hate crime gets lost, whether you are looking at training or at particular examples of when hate crimes occur within the realm of policing in a particular city. I think we need effective training for that to change, but I would vote that we would have some sort of specialism within forces, and that at least you have one hate crime management post to embed that in the longer term. I do not think you can just have training and then not have some sort of supervision of how that is overseen.
Professor Chakraborti: I do not disagree at all, but I would say that the training should not be confined solely to police officers or to specialist hate crime officers. I agree that all forces need hate crime officers, but investigating officers for individual cases often are not hate crime officers; they are police officers who lack any kind of specialist training on hate crime. If we want those investigative officers to show the compassion, empathy and sensitivity required in hate crime cases, we need to think about hate crime training rolled out in a more systematic, widespread manner than it is currently.
We also need to go beyond the police and recognise that victims of hate crime are often labelled as vulnerable or as an antisocial behaviour problem and not necessarily as a hate crime victim. Therefore practitioners from a whole range of agencies—local authorities, housing, education—all need an in-depth understanding of hate crime. At the moment that seems to be lacking across the piece.
Finally, I think we need to evaluate the training. Our analysis at the University of Leicester has showed time and time again that training is rolled out in a piecemeal fashion: it is outdated, it is not updated and it is not evaluated, so we do not know whether it is having any kind of positive impact or not.
Q185 Sarah Jones: Whose responsibility do you think it should be within the police for making sure training takes place?
Dr Trickett: There needs to be leadership. Just to pick up on what Neil said there, we need to draw on insights from other agencies and make sure that other agencies are trained, because we are talking about a multiagency response to hate crime. Police cannot do everything. One of the things I found in the research that I did was that frontline officers felt very much that the training they had was a tick-box exercise so the force could say they had been trained, but you need leadership at the top of the police that filters down.
What you often see—I know Jon Garland has written about this, for example—is a distance between frontline officers and more senior officers, and I am sure Leicester will find this as well. That means that you need the same training to be delivered to different rank and file, because otherwise you cannot come up with the commitment that hate crime is being taken seriously by the force. The other thing that is important is that we need to address aspects of police culture. It came up time and time again in my research that there is a different culture for frontline officers. We have the idea that we have an homogenous policing culture, but quite a lot of police culture can work against the training that you are trying to deliver. Police often work with very stereotypical views of some minority groups.
I also think it should be linked to promotional schemes for policing. Hate crime for me is part of a bigger picture of community policing and how we think about communities and minority groups within society. It needs to be thought out at a wider level as well.
Professor Walters: Let us deal with what Dr Trickett and Professor Chakraborti have said about having a specialist officer or what we sometimes refer to as a hate crime lead within the force and how important that is—that that person becomes such a hub of knowledge. It is their leadership that often galvanises support from other officers who have more training and specialist knowledge on hate crime. As we know, hate crime has a real ripple effect. It has a big impact on communities that are targeted.
We did a large survey of about 3,000 LGBT and Muslim people across the country and we asked them about their confidence levels in the police and the criminal justice system. There was such high support for police services to have a specialist procedure and a specialist police officer who deals with hate crime. That is really important. I know one police force, for example, that has just abolished its hate crime force lead. That is worrying because it is important for community confidence in the police to know that there is specialist knowledge and that people are being trained in that force.
Q186 Rehman Chishti: On that point, Professor Walters, about a lead in hate crime, out of all the constabularies in the country, which police force would you say is doing well in tackling hate crime and leading with the right criteria and structure?
Professor Walters: I don’t think I can answer that question. I think they are all doing slightly different things. The Metropolitan Police Service probably has the most in terms of steering groups and strategies. They have an online hate crime hub, which no other force has at the moment. I think they are leading the way in many aspects. However, they are one of the only forces that does not really use restorative justice for hate crime, so they are lacking in other respects. I wonder if this is more of a question for Neil or Loretta, who have done more research in this area.
Professor Chakraborti: I do not disagree, but I think it is very difficult to specify one individual force over another. There is an inspection led by HMICFRS that is going on currently. The idea is to share good practice across all of our 43 police forces and to identify what is working.
Q187 Rehman Chishti: Is one of the strands of the HMIC inspections that take place across the board looking at how police forces deal with hate crime?
Professor Chakraborti: Yes. That is currently under way and Professor Walters and I are part of that.
Q188 Rehman Chishti: That will then give us the answer in relation to which police forces are dealing with these matters adequately or not; is that right?
Professor Chakraborti: Yes. I think that we have to exercise caution when we say what is good practice and what is not. Sometimes police forces are wary of saying that they are doing good things, because if you are doing good things you will have high numbers of hate crime occurring in your local police force area. That is a success story, because you are encouraging victims to come forward and share that, but the way in which it is interpreted and reported by some sections of the media might be construed as if you are failing because you are not tackling hate crime in your local area. How we judge success and positive police practice is something that needs a bit of work, but there is an inspection going on at the moment that will pull together the strands of good practice across the country.
Q189 Rehman Chishti: On the reporting and recording, what practical measures can be taken to improve reporting of hate crime? I know organisations like Tell MAMA deal with the issues of Islamophobia and hate crime in that. How successful has that been and what more can be done to address the issue of reporting?
Professor Chakraborti: I think Tell MAMA is an example of really good practice within the context of Islamophobic hate crime and encouraging greater capacity within communities, but it is confined to one strand of hate crime. There is poor practice in relation to most strands of hate crime up and down the country, primarily because very little is done locally to engage in dialogue with local communities about how they would like to report. We know that many victims are unlikely to come forward and go to a police station, but what we do not perhaps do is engage with them enough and find out, “What would make you report?”
For instance, third-party reporting structures are commonly placed in venues that victims do not go to, like a Citizen’s Advice Bureau or a library, rather than hubs within local communities. There needs to be much more thought and much more dialogue across all of the five monitored strands and beyond those strands in order to identify how people will report. There is no one-size-fits-all solution to all strands.
Q190 Rehman Chishti: I get that. You said Tell MAMA is working very well. Are you saying that kind of model should be extended elsewhere to tackle all forms of hatred?
Professor Chakraborti: That model works well because it is well known, it is well publicised and it works on one particular strand. That is an advantage for Tell MAMA. I think the disadvantage of coming up with a generic reporting structure is that what works well for certain victims might not work as well for others, so I think reporting structures and processes need to be tailored for particular communities. That is certainly what has worked well for Tell MAMA and the Community Security Trust as well. That works well, so they are pockets of good practice.
Q191 Rehman Chishti: How effective is the system in making people aware of how and where to report hate crime?
Professor Chakraborti: Not very good. I can go on, but it is not very good at all. We have talked at length about recorded figures being the tip of the iceberg, and we have continually said that for many years as a country, but we have not done enough to try to address the gap between the vast numbers of incidents taking place at ground level and the tip of that iceberg. That suggests that much work needs to be done, not just in the context of criminal justice and prosecution, but at ground level in capacity building, confidence building and trust in authority.
Q192 Rehman Chishti: If we have not done that well, are there other comparable links that you can make with elsewhere, whether in Europe or elsewhere internationally, where they have done it well and we can learn from them?
Professor Chakraborti: Worryingly, I think practice within this country is probably—and I will be intrigued to see what the panel think—streets ahead of what is going on elsewhere across Europe and in the United States and beyond. The policy domain, the framework, the legislation, within this country is quite impressive. It is not without its flaws, as we have seen today, but it is progressive and sends out that symbolic declaratory message that, “Hey, it isn’t just unpalatable, it is criminal”. It is important to acknowledge the positives within the framework. The problem is making that policy framework come to life for the communities that are affected by hate crime. That is where the disconnect exists, not necessarily in policy terms but in bringing it to light for the people most affected.
Q193 Rehman Chishti: Should all crimes where the victim is in one of the five strands be automatically flagged up as a potential hate crime?
Professor Chakraborti: If the victim feels that the incident has been motivated by hatred or if a witness feels that, yes, it should be automatically flagged up.
Professor Walters: Where there is evidence that someone who is disabled has been targeted because they are disabled, not necessarily because there has been a form of hostility or prejudice evinced towards them, that should automatically be flagged as a hate crime. A supervising officer can then remove that flag if there is not enough evidence to show that there is prejudice or hostility. That links back into my issue earlier about the law and the word “hostility” not quite fitting the actuality of a lot of disability hate crime. We need to reform the law further, but for now I think that where people are targeted because they are disabled should be flagged automatically as a hate crime.
Q194 Rehman Chishti: On two final clarification points on the law, we heard from Dr Trickett that not enough prosecutions are taking place for disability hate crime. Are you saying that the CPS guidance on the realistic prospect of conviction is not being applied properly?
Professor Walters: No.
Q195 Rehman Chishti: Is it the case that cases are not being brought with sufficient evidence in the first place, or are you saying that the officers that we have collecting evidence are not in a position to, or do not collect evidence properly, in the first place?
Professor Walters: Yes, there is that. The problem is at both ends of the criminal process. It is the policing part in the investigation, gathering the evidence that you will need to prove in court that there is a disability hate crime, and it is at the other end, where judges are refusing to accept evidence of disability hostility and instead saying, “This is about vulnerability, this is not about hostility”. Research we did last year showed that there an estimated 34,000 disability hate crimes are reported to the police each year. That resulted in 84 convictions, with a declared uplift, using hate crime legislation. That is 0.2%, so 99.8% of disability hate crime cases are falling out of the system.
I think the CPS is doing a pretty good job. There has been disability hate crime training. I conducted interviews with 20 CPS lawyers and they had a really high-level understanding of the relationship between perceived vulnerability and hostility. They were taking hundreds and hundreds of cases to court. It gets to the judge and the judge just says, “No, I will aggravate this on vulnerability, not on hostility”. Of course there are thousands of cases that do not make it to prosecution because the police are not asking the right questions. When we asked practitioners, “Why aren’t they asking the right questions?” they said, “There is a fear of getting it wrong. If I ask someone a question about their disability, I might offend them and I might be accused of being prejudiced myself” so they are just not asking the questions. If you do not say anything, you cannot be accused of being prejudiced or saying something wrong.
It is about evidence gathering and about judges making a decision here. That is why I strongly believe that the law needs to be changed, not just so that we have parity of protection for disabled victims of hate crime but also so that the wording of the legislation is changed. I would advocate that that wording is changed to remove “motivation of hostility” and replaced with what is called a “by reason” test.
Q196 Rehman Chishti: Looking at disability hate crime and also at the prosecution of religiously-aggravated or racially-aggravated crime, you have given us the figures and the percentage of successful prosecutions for disability hate crime. To see how the criteria apply in relation to religiously-aggravated hate crime and racially-aggravated crime, what are the percentage figures on that?
Professor Walters: I just happen to have them with me.
Rehman Chishti: Very good.
Professor Walters: Race is 4% and religion is 1%. I do not have the percentages, but I know, for example, that there were an estimated 13,000 reported cases of religious-based hate crime, resulting in 185 cases with the declared uplift—so still really bad, still really low, but just a little bit better. Disability is by far the worst.
Q197 Rehman Chishti: It is not just simply about disability, it is across the board?
Professor Walters: Disability is the worst, but it is across the board.
Dr Trickett: Can I add something to what Mark was saying on disability? In research that I did with police officers, I focused quite heavily on disability, and on officers were saying to me that they feel that people are targeted on the basis of vulnerability, but in their eyes that is really reprehensible. One of the issues in that evidence gathering is also being able to prove that the relationship that you are talking about is based on hostility, which is what Mark is talking about. Quite a lot of them are similar to grooming cases where you have a person who is targeted. I think Neil has written in the past about these perceptions of people as vulnerable: not that the person is vulnerable, but they are in a situation of vulnerability if somebody is willing to exploit them.
To go back to what Mark was saying, if you change the definition of hate crime—because for the police finding the evidence is one of the reasons. It is not that they are not open to it. Training needs to address it better, but it is just “Why go to all of that trouble if when you get the case there is not going to be a prosecution?” You go for the underlying offence or, in the worst-case scenario, disabled people are moved out of their homes away from people who are targeting them because it is easier to address it through housing policy and moving them on than it is taking it somewhere else. I would second everything that Mark has said on this, that at the moment we are not getting to grips with disability hate crime for those reasons.
Q198 Tim Loughton: Following on from that, and this issue of vulnerability as a motivation of hostility, so why should it be any worse a crime if it is a vulnerable person, however you define a vulnerable person? If the motivation is there to do damage to somebody, which is what hate crime is all about, it does not matter how damageable that person is, does it, in the seriousness of what you are trying to do?
Dr Trickett: I think some of the explanations of hostility that we have are quite poor. For me, if you have singled out a person on the basis that they are disabled because you think they are more vulnerable, they are an easier target and there is less likely to be any comeback for you—you are selecting on the basis of prejudice simply because of their disability, which takes us back to the characteristic.
Neil has written about this before and it clearly says that in some of the academic literature, but that is not the understanding of hostility that the police and judges have. If somebody is targeted because they are perceived to be vulnerable on the basis of disability, you are still making a conscious decision to target that person because they are disabled over a non-disabled person. That for me is a hate crime, but that is not coming through in how we are training and prosecuting.
Q199 Tim Loughton: If one takes anti-Semitism and somebody is vehemently anti-Semitic, we do not say that because a person is Jewish and not disabled they are not more vulnerable. That characteristic is not a vulnerability, is it? They are trying to inflict harm on somebody on account of them being a type of person—be it race, religion, gender, disability—that they have a problem with. What I am trying to get at is that clearly there is a problem of prosecuting these cases in the courts and partly it is down to training and, frankly, probably the police being swamped with so much of this stuff happening online as well as rationalising how you deal with it.
Do we need to be smarter with the punishments? Do we need to have some form of particularly using online dissemination of naming and shaming the wrongdoers, or do we just treat it as any other crime and have to reach the threshold of proof that we would with anything else, if it were physical harm done to somebody face to face rather than hate online?
Dr Trickett: There is a problem in saying that we would tailor it to specific circumstances. One of the things that we have to decide is whether or not we are going to treat that sort of offending as a hate crime. With the sort of offending you are talking about, if somebody is demonstrating some sort of religious hostility or racial hostility, you will usually have some evidence of language there, whether it be online or offline. There will be something that you can refer to as hostility. The hostility that underlies most disability hate crime is the person in question who is committing it considering the person whom they are committing the crime against as being a lesser human being than they are, either because they are not able bodied or they have some sort of learning disability. It is about how we explain that.
What we need to look at is what the baseline offence is and then apply for an uplift tariff. The uplift tariff says that if you have targeted somebody on the basis of prejudice, your motivation in targeting that person needs to be taken more seriously and the impact on that victim might be harsher. I would be reluctant to say we are looking at tailoring a punishment here, if you are talking about anti-Semitism and tailoring that to disability. I do not think that is doable because then we will not have parity of punishment. I think what we need to do is to follow up with robust prosecutions and disability has to be rethought. We either say hostility against disabled people usually manifests like this—it is still hostility, but that is how it manifests, and then we have the uplift tariff.
Q200 Tim Loughton: I think we are slightly talking at cross-purposes. The aim that we all want to achieve is that we want more people guilty of this stuff brought to justice and to pay a penalty, and for there to be seen to be justice done so that other people are deterred and do not think it is an easy, soft thing that you can get away with. Several of the panel talked about restorative justice earlier. I wondered what form that takes. How do we have smarter punishments for people doing this sort of crime that would make it easier to throw the book at them?
Professor Chakraborti: I will let Mark talk in depth about restorative justice because it is his pet area in the context of hate crime. As a broad set of principles, I absolutely think we need to get better at thinking about smarter forms of justice, not necessarily harsher forms of justice. What victims want is a sense of justice, which is not happening at the moment. What society wants is a reduction in offending behaviour. Finding forms of justice that facilitate those needs is key. There is a way and there is evidence now to support the different strategies for delivering smarter justice. On that note, I will hand over to Mark so he can say his piece about restorative justice.
Professor Walters: I have to say my piece on it. I did a PhD on restorative justice and hate crime, so I have to say something.
Tim Loughton: Okay, make it short then.
Professor Walters: I think that restorative justice can be used as an alternative to prosecution where perhaps the victim does not want the case to go forward, will not come to court, or there is not enough evidence for a case to go to court. It can be used as an alternative.
It can also be used in addition to punishment and it could be used, for example, as part of an uplift whereby the victim and the offender and other community members will be prepared through indirect meetings. They will meet to discuss what has happened and during that meeting quite often what happens is that the victim will talk about the harm that has happened to them, but also talk about what it is like to be different in the community. It is that dialogical process that enables the offender, the perpetrator, to start to hear about what it is like for them to be different and about the kinds of harms that are caused by being anti-Semitic or targeting someone because they are disabled. If you are talking about smarter justice interventions, you can tailor the form of reparation to what has happened.
I was involved in an anti-Semitic case, with a victim and an offender. The offender was convicted of racially and religiously-aggravated harassment. They went through a restorative justice process as part of their penalty. The family did not want the offender to pick up litter. They did not want him to go out and paint the school or something. They wanted him to do a study on the effects that the Holocaust had on the Jewish people. He was supervised by the offending manager for two weeks, and he had to come back and present that to the family. He had to provide a reflection in his document. He went through the chronology and then the reflection was, “I had actually no idea that being anti-Semitic had this kind of impact. I did not know that all of these people died during the Second World War”. Who knows whether that had a genuine impact on his life moving forward, but it is probably going to have a better chance at doing that than sending someone to prison for an extra two weeks or, quite often if it is a fine for a public order offence, giving them an extra £20 on top of the £50 fine.
If you are talking about smarter penalties or smarter interventions, restorative justice has a lot of potential. There is a real policy resistance to it at the moment for hate crime because of the focus on increasing penalties. I have done another study recently, a survey of 3,000 people, and 60% of those individuals said they would prefer a restorative justice process to an enhanced jail term. There is a lot more support for these alternative practices and process than we think there is.
Q201 Tim Loughton: That sounds interesting and practical, and hopefully has an effect on prevention. For those who would not go down that route or if it would not be appropriate, where you need something to name and shame those people who use the internet and social media for hate crime in the privacy of their own sitting room with their terminal or whatever, when they are brought to book should we be using things like a cyber offenders website where you can see some of the hateful things that some of these people have been putting in full public glare?
Professor Walters: I am not a fan of naming and shaming. I do not think it has any positive impact on the victim, the community or the offender in their future behaviour. It is antithetical to restorative justice theory and practice. I do believe that you have to have robust prosecutions with a symbolic, declaratory announcement of a conviction and a penalty. That is really important, but if you want communities to come together and you want people to have better relationships, you need dialogue that is focused on understanding, focused on harm, and on how you repair that harm. It is not necessarily about stigmatising people. There is very little evidence to show that stigmatising people has any benefits to our society in terms of reoffending or people being rehabilitated. It seems like a good idea, a visceral response, “Yes, do that, that sounds like what we should be doing”, but the evidence does not bear it out as being a productive means.
Dr Trickett: I think we could do more with community sentencing than we currently do. In order to have restorative justice—and Mark has done a PhD on it and is more of an expert than me—obviously you have to have all the parties in agreement. I do think we could be more intelligent about a community aspect and look at more intelligent ways of how offenders might make some recompense to the community that has been affected than we do, rather than simply using fines and imprisonment.
I am not in favour in naming and shaming either because I think what happens in those situations is that you increase the justification of anger in the person that has been named and shamed. The effect of that might be that they go and redirect their anger or hatred elsewhere. You might end up just displacing that crime off the internet. I think it can end up antagonising the situation rather than calming it down.
Q202 John Woodcock: Is there not potentially a deterrent effect, though, if someone is seen to have been shown in the public eye?
Professor Walters: The evidence does not bear that out.
Professor Chakraborti: There is no evidence to suggest that at all, nor is there any demand. Over the past five years, I have worked with more than 2,000 victims of hate crime from different backgrounds, walks of life and across all the monitored strands and beyond, and not a single one has expressed a demand for that kind of justice, for naming and shaming. There does not seem to be that demand among victims, nor does there seem to be any evidence whatsoever to show that that will be effective. For every one that you might deter, there will be another person coming up. Deterrence does not really work in that context.
Q203 Stuart C. McDonald: A number of times in evidence sessions, including today, in response to questions we have had answers that suggest that what needs to happen is that there has to be greater co-operation and collaboration between policing and justice systems and external agencies such as community groups and organisations. For example, today we have had answers like that about police training, offender rehabilitation and even reporting mechanisms. Is that happening? Are there barriers to police and justice systems going out and speaking to these community organisations? If so, how do we overcome those barriers and who should be responsible for it?
Professor Chakraborti: There are barriers, but many of those barriers are perceived barriers as opposed to real structural barriers. I think the common thread unifying a lot of what we have said today is effective dialogue and effective communication. When it comes to that reporting part of the piece, there is nothing to stop police forces and partners, local authorities, housing, whoever it may be, engaging with diverse communities and talking, getting the message across, not only in condemning hate crime but also in how we can provide support.
That does not seem to be happening. When we drill into why it is not happening and what those barriers are, there is a variety of reasons but primarily there seem to be shortcuts and tokenistic responses, “I have done my community consultation because I held a workshop on hate crime, five people came to it, so that is that box ticked”. That seems to be happening time and time again across the board in different contexts.
I could give a very long answer, but I think the barriers that are there are sometimes imagined and once you drill into it and identify some of the solutions you realise that it is not just about resource. There are meaningful things we can do.
Q204 Stuart C. McDonald: I suppose that to an extent brings us back to questions of leadership in, for example, police forces and so on.
Dr Trickett: I think there is a real issue. One of the problems in policing is that we are living in tough times for policing in terms of finance and budget. One of the things for me is that many of the police officers that I spoke to—frontline officers, in particular beat managers and PCSOs—are very committed to trying to deal with hate crime but it is quite resource intensive. You have to get sometimes within a neighbourhood context what is going on and that sort of pattern. They saw it as being part of a community policing role, and some of their frustration was that they are often being taken off and moved around to go and cover other roles and often could not get things finished. They would build up a rapport with victims and get moved on.
I also think we need to be a bit more intelligent about how we share the information and where it is kept. For example, if you are just looking at academia and the engagement of academia with politicians and also with police and organisations, we could do with a database, some sort of repository where you can find out what work is being done in the area. What tends to happen is the Home Office will put out a call for some work in an area, or a politician will raise something as being of concern, and there will often be a lot of research that is being done on these areas that could be useful to politicians and Government but they are unaware of it. It has been published sometimes in academic journals. Universities are doing more to try to make that available because of the REF criteria, but it would be really useful to have some sort of repository.
We are part of an international network on hate scholars, for example, and we share some of our reports globally. Those are good pockets of practice and expertise because what tends to happen as well is if you are talking about multiagency working, for example, and the police may be drawing on intelligence from other agencies, some of that work is replicated—too many agencies are doing it because nobody knows who is responsible for that particular role. We have a lot of repetition of the same things and money may be lost.
We need to think more intelligently about how we share information and maybe have somewhere where people can get hold of information. If an MP is interested, for example, in what work is being done on hate crime, they can go and find it and think, “There has been a small study done there in Portsmouth, but that generated some really interesting results. I might ring that person up”. The dialogues are actually quite distanced from each other.
Q205 Stuart C. McDonald: That brings me on to my final question, which is about how Government and police engage with academic and research institutions. We heard in the context of another inquiry—I think it was far-right extremism—that there was not really any sort of Government interaction with academics and experts in the field. Is it similar here or is it just not structured enough?
Professor Walters: The aim of the International Network for Hate Studies is to bring policymakers, practitioners and academics together. Paul Giannasi, who is the hate crime lead for the Government, is on the board and we have been working closely with him. I sometimes think that the Home Office, the Ministry of Justice and police can do a little bit more to engage with us. We try our best to make our research as accessible as possible and to come to meetings like this, but it would be great if those agencies spent a little bit more energy and time investing in the research to make sure that the policies are evidence based.
Dr Trickett: An example of that is sometimes if you are at university and you are trying to address a problem—for example, I want some money at the moment to study disability hate crime—you have to apply to funding bodies here. Everyone is competing for stuff. Then what happens is the Home Office will go and do a piece of research that actually replicates something that someone has done five years ago or 10 years ago and we keep trying to reinvent the wheel. I think there is a need to find where the work is being done in this area. Do we need something done? We are talking about evaluation of police training, for example. There has been virtually no evaluation done of police training, apart from the College of Policing doing a learning needs analysis, which Paul Giannasi instructed and was engaged in at some point.
There needs to be much more open dialogue between politicians and Government because you are dealing with the things in Westminster every day, the things that are changing all the time. Politics is constantly evolving by the second and sometimes you could come to us and say, “We have this problem. What do you think about it?” I have been consulted by the press sometimes more times than I have been by MPs because the press are interested in something and they will ring you up for an opinion. I think there is definitely a need for better information sharing between Government and universities.
Chair: Thank you very much for your evidence today. We really appreciate your time. Thank you very much for your patience as well, when the votes happened.
Dr Jelena Dzakula, Fellow, and Professor Damian Tambini, Research Director, Department of Media and Communications, London School of Economics and Political Science; Dr Martyn Frampton, Reader in Modern History at Queen Mary University of London and Co-Head of the Security and Extremism Unit at Policy Exchange; and Dr Martin Moore, Director of the Centre for the Study of Media, Communication and Power, King’s College London
Q206 Chair: I welcome you before the Select Committee. Thank you for your patience and for waiting. I realise we are starting later than we said we would. Given the constraints of time—and apologies again for that—I am going to suggest that as we ask you a whole series of different questions, do not all feel the need to reply to every single question and come in on the ones that you feel that you have most to add to. Can I ask you each to introduce yourselves and say which organisation you are from?
Dr Dzakula: Good afternoon. Thank you for inviting us again. Jelena Dzakula, and I am at LSE.
Professor Tambini: Damian Tambini, London School of Economics, where I teach media policy.
Dr Frampton: Martyn Frampton. I am a historian at Queen Mary, but I am also here with my Policy Exchange hat on, where I work with them on extremism and security.
Dr Moore: Martin Moore. I am from King’s College London and I run a centre called the Centre for the Study of Media, Communication and Power.
Q207 Chair: We have taken a lot of evidence on the nature of online hate crime and abuse, and we have taken evidence from social media companies about some of the things that they are now doing in response to that. What one thing would be top of your list that needs still to be done, either by social media companies or by other organisations or by Government, to address online hate crime?
Dr Dzakula: For me, the first thing is increasing transparency of how artificial intelligence is used. There is a lot of discourse about it being a technical solution, which it really is not. There is a lot of human labour and human judgment involved in deciding how to use artificial intelligence. For me, that is top of the list, especially because it is used to a large extent and will probably be used even more. There are some really good solutions and there are some really good uses, but we do not know a lot about how it is used and what decisions are being made on how it will be employed.
Q208 Chair: Can you give me an example?
Dr Dzakula: For example, DeepText. If you want to use an algorithm for something, you need to train it or record a piece of text. We also do not know to what extent artificial intelligence is used as a supplement, or is it the final decision-maker; what happens when a decision is made? That is one example.
Q209 Chair: Okay. Professor Tambini?
Professor Tambini: I saw your evidence and I sympathise with you because you were talking about specific instances where content was not taken down. That was clearly illegal and after repeated notification from you it still was not taken down. It does seem that there are certain categories of content that, even with direct notification, is not being taken down.
I think we can see that within the general scope of European regulation in this area different content types seem to be subject to different removal regimes. Intellectual property notification seems to lead to things being taken down relatively quickly; child abuse content seems to be taken down relatively quickly; it seems that hate speech is not being taken down so quickly.
I think that the Committee should tread very carefully when thinking about recommendations. Having been watching this policy area for a long time, I have noticed that, to put it in a simple way, the way we regulate these things in this country seems to be becoming more like the way they regulate them in China in regulation of social media, and not the other way round. We need to be careful about the free speech implications of anything we do.
I would second the idea that we can improve transparency, and an important part of the solution is helping consumers to empower themselves and make the market work better—for example, people could switch between social platforms and put some competitive pressure on. Part of that is about transparency, but I do think we need to think about co-regulatory solutions. I think the Committee has noted the new German NetzDG law. I can say a little bit more about how the Germans have introduced a new law and also how—
Chair: We will come on to that.
Professor Tambini: —there is a new European code of conduct, which I think is already having an effect. There is also a message of “let’s see how things develop because they appear to be moving in the right direction.”
Q210 Chair: Can you give me an example of something where you think that the British approach is becoming more like that of China?
Professor Tambini: If we think about approaches to social media liability more generally—I take your point and I am not speaking about specific changes in the law; I am speaking about a change in the way these things are discussed and the “something must be done” atmosphere. The point I am making is that we should not take a kneejerk-type response. We should not rush to legislation. I think the German law as it is currently drafted and being implemented has been quite robustly criticised by the UN Special Rapporteur on Freedom of Expression and by Reporters Without Frontiers, and we need to be careful about the elements of that approach that we adopt. I would not pick any specific area but the general—
Q211 Chair: You do not have any examples of anything that Britain has done that are remotely like China.
Professor Tambini: I will think about that, but I take your point. There is a general—
Chair: I am just suggesting you might want to withdraw that sentence or come up with an example of something that is a way in which—
Professor Tambini: I will withdraw it for now and I will try to come up with an example. Thank you.
Q212 Chair: Dr Frampton?
Dr Frampton: One of the most interesting parts of the big report that we did at Policy Exchange, “The New Netwar”, is that it showed the disconnect in how the public look at this issue and where we are in the legal and regulatory framework. I think it is clear there is a crisis of confidence out there, and high polling showing that the public are unhappy with the situation in the way the social media companies operate and the sense that things are out of control. I think I am probably going to speak slightly against the last point.
In that context, there is a moment now with Brexit and thinking about the legal regulatory framework that the social media companies face. At the moment, it is framed very much by the ecommerce directive of the EU from 2000. Even the EU, if you look at the stuff it has been saying over the last four or five years, feels that that directive is out of date with the evolution of social media companies. We have been stuck in this binary of whether we should look at the social media companies as platforms or as publishers. The key is to try to find new language that captures what they are doing and perhaps—in a way I think I am picking up on something that the Committee on Standards in Public Life talked about in their report quite recently—shifts the balance of where liability lies.
At the moment the framework under the EU directive is that the presumption is that these companies are not liable for the content that appears on their platforms providing they do XYZ. One of the options would be to shift that around to the presumption that they are liable unless they have fairly robust practices in place for notification, removal, making sure that content does not reappear, and transparency as was mentioned. I think we are at an interesting moment and there is a real opportunity to perhaps reframe the debate and reframe the law on all of this.
Dr Moore: If there was one thing I would urge, it would be caution in the short term and ambition in the long term. It is funny, I came straight from teaching at King’s before I came here. I was teaching a class and we were talking about folk devils and moral panics. It seems very strange because as I have been studying this over the last few years we seem to have moved from a position of slightly doe-eyed infatuation with some of these tech giants across to almost—I am not saying that we are yet—veering towards a direction where we have a slight moral panic about them, sometimes justified. I think sometimes there are serious problems, particularly political and social, but there is a significant risk that we overreact in the short term and underreact in the long term.
If I was to come back to what Jelena said at the beginning, from my own perspective I think we have a fundamental problem around transparency, not necessarily just about artificial intelligence but more generally. For example, in order to try to figure out how we respond to things like hate crime, it is very difficult to do that when we do not know exactly who online is committing it and what their motivations are. We do not know, for example, if we introduce particular design changes within the platforms how substantial an effect that will have, and whether that will reduce it significantly or not. For Facebook particularly, we do not know the effects of some of the algorithmic changes that Facebook makes. News organisations themselves are always throwing their hands up about the fact that one day they can be getting 600,000 views to their articles and the next day they can be getting 100,000, and they have no idea why. It is almost directly as a consequence of Facebook changing its algorithm.
At the moment we have hugely opaque systems, which make it much more difficult for us to establish who is committing the offences, how the offences are being committed, and what effect the interventions that they are already doing are having.
Q213 Chair: Sticking on this algorithms issue for a second, do you think you could get that level of transparency without legislation, or do you think in the end it will need legislation to get additional transparency?
Dr Moore: There is something that I am slightly astonished that these companies have not done yet, which seems to me very bizarre. They can see worldwide that there is increasing anxiety about their algorithms, not just in terms of hate crime but discrimination, politics, and lots of different areas, yet they have not set up governance systems internally to give greater public confidence that they are adhering to certain principles and to essentially have dialogue with government or regulators to show them that they are not. Their defence is always, “If we make the algorithms themselves more transparent, they will be gamed more than they are already”, which is already quite extensively.
I think it would be a step forward if they were to have, for example, just as a comparison, a governance system that was more like, let’s say, the Scott Trust and The Guardian or a corporate organisation that has governance in place, whose primary responsibility is to look at the way in which these algorithms are being both created and used and the effects that they are having, so that they can give greater confidence to the public and to regulators and to the Government that they are not being misused.
Dr Dzakula: I am not a big fan of very open public transparency. I do not see the value of it being public because those things are confidential and there are issues of competition as well. I think it is either transparency to an independent body or a multi-stakeholder rather than publishing very publicly on their websites what exactly is happening.
Q214 Chair: You think it needs some sort of regulatory framework to establish the transparency?
Dr Dzakula: Or it can be voluntary, yes. I am not quite convinced that it has to be legislated for us to get there.
Q215 Chair: How much do you think that in the end this is a problem that is built into the business model that they have? For example, we probed them on the issue of online extremism where if you are interested in, for example, far-right extremism on YouTube or on Twitter, you get more of it suggested to you. Therefore, there is an automatic process of radicalisation, but it is built into their business model because they want you to stay on the site. They know from all of their algorithms what is most likely to keep you clicking. Is there a fundamental problem here with the business model and the advertising nature that they need?
Dr Moore: This is what I meant about being cautious in the short term and ambitious in the long term. It seems to me that the problem is quite fundamental and the problem is across our digital sphere. It is particularly with companies that have chosen a particular type of business model, which is based around advertising.
To extend your example, something that was exposed partly by a Times investigation last February and has come up repeatedly since then and will come up repeatedly beyond that, is the degree to which advertisers are inadvertently funding extremist or hate-filled content. Part of the reason for that is because the way in which a lot of internet advertising now works programmatically is such that you have these real-time auctions whereby the advertisers themselves have absolutely no idea where their adverts will be placed because all they are doing is choosing to direct it at particular individuals. They do not care where those individuals are. Those individuals might have gone from The Telegraph or The Guardian or wherever on to a YouTube video that has extremist or hate-filled content. What the advertiser is doing is buying those eyeballs. They are not buying a space on The Telegraph or The Guardian. As a consequence, and because they are essentially blinded from seeing where their content is going, they are inadvertently supporting a lot of this stuff online.
It is exactly as you say. I think we have a much deeper problem across the board in terms of the business model and, of course, of the dominance, which I think is central.
Dr Dzakula: Transparency does have to have some sort of a purpose. One of the purposes is that it might change the behaviour of the tech giants; another is that it might empower the users a bit more, but it is also about accountability. If we do have the information, who can do what and where do we go?
Q216 Tim Loughton: I am surprised at the defence of free speech against clamping down on some of this stuff. There is a world of difference, surely, between somebody who expresses anti-Semitic views personally against an individual or anti-Muslim views, as happened in the case of the terrorist who was jailed last week for mowing down Muslims coming out of a mosque after prayer—there were tweets or Facebook posts saying, “Shame that he didn’t take out more of them” or whatever. That is very clear violence and incitement to violence against an individual or a group of individuals. At the same time there is people’s right to express comments about, “The Government are unduly influenced by the State of Israel or are doing far too much to accommodate Muslims in this country” or whatever. They may be distasteful views but those people have a right to express them and be challenged. That is not a hate crime. It becomes a hate crime at certain points of it.
I am afraid this was the defence that some of the American social media companies in front of us said, “Well, basically, everyone has a right to offend anybody else, so that is why we are not doing so much about it”. That was really where they wanted to be. That cannot, surely, be a defence.
Professor Tambini: I will clarify a bit. I should confirm that I withdraw the earlier comment about China. What I should have said is that we are in danger of becoming more like China.
The free speech issue is not something that we can just simply sweep aside, is it? Today of all days I agree that we have to say that this kind of online abuse is very often about silencing people, particularly silencing women and minorities. The free speech arguments cut both ways and I am not for a second taking the position that this is about a principle of free speech that must at all costs be protected by protecting the interests of the tech giants.
The issue behind this, which was mentioned by Martyn Frampton, is this question of liability and the shield of liability that particular platforms have been granted by the ecommerce directive and equivalent legislation in the US. What it effectively does is establish what US legal scholars acknowledge now is a subsidy to these platforms because it gives them a holiday from liability. They are not liable until they are told they are carrying illegal content. Demonstrable hate speech is illegal content and this is something where the necessary balances in any judgment about free speech are clearly in favour of restricting that speech and taking it down as quickly as possible. What all this settlement for internet content does is establish a general category of illegal content but, as I was briefly mentioning before, for the different subcategories within that category—intellectual property infringement, child abuse content, defamation and hate speech—there are different mechanisms at play, which is why some gets taken down and some does not get taken down.
Q217 Tim Loughton: But an awful lot of it does get taken down and that is the point, isn’t it?
Professor Tambini: The reason that hate speech is difficult and the reason that the things that this Committee has been wrestling with—I think rightly—are profoundly difficult, is because they not only involve political speech, where there is generally a high level of protection, but they also involve public figures, where there is a high level of protection of speech. These are difficult and expensive judgments for private platforms to make.
Q218 Tim Loughton: I understand, but the, “It is all a bit too difficult” excuse is not a defence. It was very difficult for Google, for example, but they did develop IT that would detect preventatively child sexual images, certain sexual terms used by sexual predators, so that it is very difficult now to access a lot of that stuff. They put a lot of effort and a lot of resource into doing it and they have done it. A lot of resource has gone into preventing IP infringement. Nowhere near the amount of resource or priority, apparently, has been put into designing or adapting that technology to filter out preventatively various forms of hate speech, notwithstanding that you should not prevent legitimate attacks on the policies of politicians or various other special interest groups. They have not done it.
Dr Moore’s point was interesting. You now have companies like JP Morgan, which is almost the largest bank in the world with major advertising expenditure, who have now had to employ their own people in order to make sure that their adverts that they have paid for are not appearing alongside some of the hate sites. It can be done, but it is done by the customer rather than the purveyor of those services. That is surely the wrong way to do it. If it can be done, the liability should be on the person who is benefiting from it financially and selling a product, but who appears to be unprepared to take proper precautions to make sure that that product is not wittingly or unwittingly committing a crime or being used as a platform to commit a crime.
Dr Moore: The key there is the “what” they do in the sense that there are a whole number of different things that they could do pre-legislation that could have beneficial effects. The Internet Safety Strategy Green Paper talks about some of the actions that Tumblr has taken to alert people when they search for certain terms. Whether they are to do with extremism or anorexia or various other terms, they are alerted to some of the issues surrounding those terms.
One can design in particular elements to different publishing platforms to make people more aware of the potential liability of what they are doing. One could imagine, for example, if you were to fill in particular terms prepublication and then an algorithm was alerted to the fact that they appeared to be in breach of the law, one could be prompted before having pressed “publish” to say, “Are you sure you wish to publish this because this could be liable under XYZ?” That does not stop them from publishing it, but it certainly alerts the user.
There are certain ways in which one can design platforms so that it becomes built into the design that you discourage that sort of behaviour and so that you also alert people to some of the potential liability that they are exposing themselves to.
Q219 Tim Loughton: Would you agree that that is not happening at the moment?
Dr Moore: It is certainly not happening on some of the platforms. It is clearly not happening on Twitter and elsewhere.
The other thing I suppose I am slightly confused by, to come back to Jelena’s point about the use of algorithms, is that it seems to me as though there are significant opportunities—and again this comes to prelegislation—for these companies to have algorithmic flags, a little bit like you were just talking about. There are specific types of behaviour, for example—I am thinking particularly about co-ordinated attacks, mobbing and lynching and so on—which follow a familiar pattern and which one could entirely imagine would be flagged up automatically so that the organisation themselves can at least be aware of it and can choose to intervene or not intervene, depending obviously on what the activity is.
Dr Frampton: I agree with the thrust of your question that the political will, if you want to use that phrase, on the part of the companies is key. That is where I think the legislative or regulatory framework is so important. The initial laws that were framed around 2000 were premised on the principle of protecting these companies from the power of the state, and that was very much the ethos. No one could have foreseen at that point the scale of companies like Facebook or Google, which arguably have influence far exceeding that of most states. Again, I think that is why we need a change in the framework to compel or drive them to readjust their priorities—and to answer a point that was made earlier, I am not advocating moral panic. I do not view these companies as inherently iniquitous or particularly worse than any other major corporation. I just think that it is about forcing them to readjust their own priorities to, as you say, invest more resource in dealing with this.
Q220 Tim Loughton: How do you force them? That is the question. It was interesting when we had the first hearing with Facebook, Google and Twitter in front of us, that the publicity that generated in showing up some of the sites that they were prepared to host, even though they had been warned against them, led to a number of corporates voting with their feet and taking away their advertising. The Government should be doing that as well because the Government are an advertiser there. That I think is what forced them into saying they have to take this more seriously. Whether they are taking it seriously enough still and fast enough and employing enough people to do something about it is a moot point. Do we have to go down the German route—and it is early days for the German route—where effectively it is very clear from the start that if you do not show willing speedily enough, preventatively or in a rapid takedown, then you will be committing a crime for which you will be fined? You are going to force the pace financially either from the client or from the state. Which is the better or more effective way of going if one has to go that way?
Dr Frampton: My own view is that I think one has to be prepared to regulate. I see no reason why a subdivision of Ofcom or a body like Ofcom could not play that kind of role in monitoring the processes that the companies have hinted that they are going to produce—this word transparency again—and being more transparent about how they operate their notification processes. When they receive complaints about material how many of those are acted upon and what is the process for that? They have hinted that they are going to provide more of this, but why should that not be overseen by an independent regulator?
Q221 Tim Loughton: Will that be a stage too far, Professor Tambini, given your earlier comments? Would it be censorship?
Professor Tambini: Not necessarily; I do not think so. There are elements of the German scheme that I think are very good, notably setting out very clearly what kind of transparency you need. There has to be a clear set of criteria for what you are taking down and why. There have to be clear objectives for different categories of content. They separate out different categories of content, stuff which is manifestly illegal that has to be down in 24 hours; illegal, down after a week; and then another category of content, which is too difficult, which gets referred to an independently audited self-regulatory body. There are elements of that scheme that I think can work very well.
Part of the problem and part of the challenge to the scheme as it is in Germany is that the regulated self-regulatory body reports directly to the Ministry. Given the discussions we had about the royal charter and the press, I do not think that that would go with the grain of media policy in this country, for example.
There are elements of transparency. Any regulatory oversight has to be much more independent from the state. My personal position is that one of the things we have to be aware of is not only the state’s role in relation to freedom of speech but the platform’s role in relation to freedom of speech. These are monopolies. They have some of the characteristics of monopolies. It is also very difficult for consumers to switch between them. One of the reasons that kind of harassment, bullying, hate speech and all of these difficult categories of content are so difficult is because it is difficult for consumers to leave those platforms, without immense cost to them personally.
If we are shifting all of that decision-making authority and all of that discretion to the platforms to decide what is okay, what is manifestly illegal and what is to be taken down, saying, “You deal with it”, we need some kind of corresponding transparency about exactly what they are taking down and why they are taking it down. The German law has it for the procedure and criteria. However, if that is going to be automated we need it for the algorithm too for this immensely powerful organisation that is having a big impact on the circulation of ideas in society. If the algorithm says, “This has the characteristics of something like bullying” we need to be very careful that that bot, that piece of software, has not simply recognised a legitimate protest, for example, or a legitimate form of speech. These are difficult areas. Transparency for the procedures, yes, and also transparency for the algorithms.
Q222 Rehman Chishti: In getting clarity from social media about what they have done and how effective they are being, between all of them—Twitter, Google, YouTube and Facebook—which would you say has been more effective at taking down material that ought to be taken down?
Dr Moore: This is where we come back to the transparency question in the sense that it is extremely difficult to evaluate these companies.
The one that is easier to evaluate is Twitter because it is far more open. Facebook is remarkably opaque, to the extent that at the moment we are relying—it is very early, I know—essentially on anecdotal evidence to understand what effect the German law is having on people’s particular newsfeed. I have a student from Germany who said he saw a distinct difference between his newsfeed while he was in Germany and his newsfeed while he was in the UK. That is, as I say, simply anecdotal evidence. At the moment we have no ability to test what is being removed, what is being kept and the difference between what is happening in Germany and what is happening here. That, of course, would have a huge benefit to understanding whether it is of value or not.
To reply to your question, the answer is that Facebook probably has much greater ability to intervene, partly because it is a closed platform, for the very same reason that it is even more difficult than the other platforms to evaluate.
Professor Tambini: I mentioned before a general note of caution in the sense that things are changing. We have had some specific and really shocking examples of individual pieces of content that this Committee has notified to the platforms—things that are clearly illegal, clearly offensive and of the worst kind—and that have not been taken down. The EU Hate Speech Code of Conduct, which was published in 2016 and has the buy-in of all the major platforms across the European Union, has been reviewed, I think three times. During that period there has been a shift. I think the second report said 50% of clearly illegal content was removed within a week and the latest report, from January, said 70% of pieces of content. They are headline figures and it is going to be necessary to look at what is behind that. That is not enough transparency and it may be necessary to legislate for transparency. It is the combination of that kind of transparency and also information for consumers.
An area where the Committee might want to make recommendations is in social networks, for example, which could be considered to be safe for children. That is an area where parents and schools, for example, are constantly asking for information and guidance. Media literacy is a key part of the solution.
In general there is a shift going on. Smart incentives to try to make the platforms deal with these issues is the right instinct. Given that we are giving them so much power in doing that, we need also to have transparency about how that is used.
Q223 Rehman Chishti: Being completely frank, on the points some make about erring on the side of caution when compelling sites to do things, when one sees the comments last week by Mrs Justice Cheema-Grubb in relation to the evil acts of Darren Osborne—looking at the short space of time from him seeing the material, to being radicalised, to carrying out evil and terrorist acts—the first duty of the state is to protect its citizens. Therefore, if you have social media companies out there that repeatedly get asked to take action but do not do it, on that basis the Government have a duty, linked to the first duty to protect their citizens, to compel them to take that action. Linked to that point, and where there may be defects in the German legislation, is that the German legislation says, “You have to take it down within 24 hours or face sanctions”. Some reports—correct me if I am wrong—say that when material is put online it is shared within two hours up to 75% of the time. Even if the German legislation comes into place, unless you address the early onset of getting that material taken down or prevent it, you still have problems. Am I right in that or not?
Dr Dzakula: Can I say something? You presume that a lot of those cases are very straightforward; they are not. Sometimes 24 hours is not enough to decide whether that content—
Rehman Chishti: I get that. I am saying this is what the German legislation has put forward, 24 hours. Some reports that we have say that once material is loaded on the system it is shared by individuals within two hours. Therefore, even having 24 hours is not enough because the material is shared more quickly than that. The onset from seeing it, to being radicalised and taking action is now a very short space of time in some circumstances. How do you address that?
Dr Dzakula: Maybe with preapproval, because there is a database that is done with images of child abuse. It is instead of waiting until that time to decide on whether the material is hate speech or not.
Q224 Sarah Jones: We have already covered a bit of this and Tim did a lot of this. However, is there anything further you have to add to what the Prime Minister said in Davos about how investors should put pressure on the large companies to do more, and also encourage large platforms to help smaller platforms do more? We have talked about where the Government’s responsibility is. Do you have any other thoughts about who should be taking responsibility? Are there other any other measures the Government could do, or do you think Theresa May’s approach is the right one?
Dr Frampton: The unaddressed problem is small platforms such as Telegram, which has become a major focal point for online extremism and ISIS-related material. They are using diverse platforms in different ways. I do not think, for example that they are now using particularly Twitter or Facebook in the way they were perhaps three or four years ago. I agree to some extent with Professor Tambini that there has been a shift. There is still some there but there is less extremist distribution on Twitter—or Twitter is less the first port of call. The first stage happens on, for example, Telegram. Telegram is then used to co-ordinate interventions into more mainstream space.
The reason why those mainstream companies are still so important is because they set the tone. There were some remarks recently from Max Hill about whether one should be wary of forcing things underground, and whether that makes it more difficult to pursue and to know what is going on. I disagree with that. It is crucial for the state to shape social norms in terms of what happens online. That is where these mainstream platforms are so crucial. The state has a role to play in saying, “This kind of material is completely unacceptable on these platforms. We want to work with you to remove it”. We recognise there is a problem with more rogue platforms like Telegram, or certainly platforms that until now have shown very little inclination to play a more positive role. However, there is still a key role in stigmatising their failures and helping the mainstream companies to live up to their responsibilities.
Dr Moore: Going back to where I started, with caution in the near term and ambition in the long term, there are two big fundamental problems here, one of which is associated with the dominance of these companies and how big they are. Directly connected to that is the extent to which their business model drives an awful lot of this stuff. It seems to me that is where we ought to be paying a lot of our attention, not least, for example, thinking about ways to introduce friction to some of the business models around data and personal data. At the moment that is not accounted for in lots of legislation. When it comes to competition law and so on, there are serious deficiencies with regard to taking into account the use of personal data to expand in other fields.
In the short term, in terms of intervention we need to be careful about the law of unintended consequences. If we look at the example of “right to be forgotten” introduced in 2014, I think it was in 2016 that the chief counsel of Google said, “We have created a whole new jurisprudence around privacy and freedom of expression”. That is an entirely private jurisprudence. Jurisprudence has been developed within Google as a consequence of legislating—I use that word—or ruling on all of these different cases themselves, developing their own privatised version of what should and should not be private and what should and should not be freedom of expression.
My concern at the moment is there are so many issues that we are, rightly or wrongly, blaming on these social media companies--everything from fake news, to filter bubbles, to echo chambers, to hate speech, to other social and political problems. If we simply try to legislate to resolve one issue it will have an awful lot of knock-on effects on others, and those knock-on effects could be seriously detrimental.
Q225 Sarah Jones: When you said “increasing friction”, how could you do that? What is that?
Dr Moore: One of the fundamental issues with these companies is that they collect, gather, organise, reuse and resell data. They do that with a tiny amount of friction. There is very little cost to them in collecting more personal data, so they might as well. In fact, they are now into an arms race where effectively they are selling to advertisers on the basis that they have more and more data and are tracking further and further. Making it clear to them that there are disincentives—whether those are tax-based incentives or whatever else—to the continued collection, reuse and resale of data would introduce a lot of sand into their cogs, and make them start to think, “Hang on a minute, do we really need to collect all this particular data? Do we really need to track people on this basis?” Only that way will they start to rethink the way in which they do business.
Professor Tambini: To underline that, the timing is good in the sense of how the general data protection regulation is implemented this year and the judgment calls that will be made on things like consent, and data only being able to be used in a way that has been consented for. That may give some opportunities—in ways I have not fully thought through, I have to admit—to make some recommendations on how that is implemented and things this Committee might like to flag up.
Q226 Sarah Jones: I chair the all-party parliamentary group on knife crime. We have had companies in to talk about some of the violence online and how that can be stopped. It is the same bag of, “These things are difficult because when is it freedom of expression” and so on. However, there are lots of videos of people with guns that should be taken down from YouTube that are still there. It is a similar kind of problem. How do you think we—not just this Committee but Parliament—should be looking at this issue? It seems you have a group of us talking about hate crime, a group of us talking about knife crime, and a group of us talking about child exploitation, and we are all talking to the same companies. Are we going about it in entirely the wrong way? Should we be flipping it around and trying to ensure their commercial drivers stop them from doing all of these things? I wonder if we are getting it slightly wrong.
Professor Tambini: If I can mention a couple of things I was struck when you complained about individual pieces of content and said that they were not taken down. The current scheme is that from the moment of notification the platforms are liable. If they do not take the content down the obvious next step is that somebody should sue them because that is how the system works. In the case of intellectual property you have industrialised pressure on the platforms to take stuff down. In the case of hate speech, harassment and other kinds of content you do not have the same incentives for them. Thinking about where they are standing in terms of the cost benefit, where they invest the difficult judgments, in making decent decisions, automating them and so on will depend ultimately on those economic decisions. I would suggest trying to think from their point of view about where the incentives are.
Regarding the general terrorism material that was mentioned before, the difficult thing to contemplate is the fact that the tech fix—the simple “this must be taken down”—might not work. It may be the fact that there are many more people out there and taking things down may have the effect of getting many more people to repost it in some cases. The platforms have that difficulty. The fine detail of asking for something to be taken down is that it may have been taken down by somebody but the same piece of content may have been reposted by somebody else.
Q227 Chair: They seem to cope with that with Katy Perry videos. Why can they not cope with that if it is illegal extremism?
Professor Tambini: They can, but I do not think the incentives are there because I do not think the threat to sue is there.
Chair: On those, of course, they might be criminally liable but there is not a financial issue. There is not anybody to sue them.
Professor Tambini: The DPP has guidance on social media crime. There are different categories and they have priorities for what will be publicly prosecuted. There are a limited number of cases here and we are talking about huge potential volumes. It is a practical task of dealing with the volume. You need some sort of co-regulatory automated process and some sort of transparency and oversight.
The German model has some good elements. I have been cast as pro-free speech and, “let the platforms off the hook”. That is not at all where I am coming from. However, I do think it is complicated.
Sarah Jones: From that point of view there is what they do in China. They do not believe in free speech for one minute. They just say they do because it suits their argument. We let them have that because we believe in free speech but, fundamentally, they do not.
Q228 Chair: To follow that up, if you were going to do an amended version of the German model, what would it be?
Professor Tambini: It would be more of the first part of the German model, which is about transparency, and clearly setting out the procedures. In relation to the press we have this model that has been held in abeyance. Maybe—heaven forbid!—we should have a royal charter for the platforms. Then we could hold the second part, which is a recognised self-regulatory body pretty much like the press system we introduced with the Press Recognition Panel, which is held in abeyance if the platforms do not deliver on the first part.
On the point about parliamentary relationship that was raised, as I mentioned before, we must think about how they are accountable to Parliament, how to formalise that and how to get better quality of reporting with the procedures that are in place—the codes and so forth—and the complaints would be useful.
Dr Frampton: I will add one other thing to the original question. It does make sense to try to view these problems in the round rather than siloing them. They do overlap so much—hate crime into extremism and anti-Semitism. A lot of these things do come round. One of the things I found surprising was that the “Internet Safety Strategy” consultation paper at the end of last year explicitly said it was not going to deal with extremism, terrorism and child sexual exploitation. That seemed slightly odd. I think it makes sense to put this all together and ask the bigger questions about how the companies operate and—your key word, Professor Tambini—incentivising them to change the way they operate.
Q229 Stuart C. McDonald: Dr Frampton, in one of your reports you talk about a sliding scale of responsive regulations that begins with tightening up codes and so on, moving through financial penalties to using legislation to prosecute repeat offenders. How far along that scale do you think we need to get before it has a significant effect on the behaviour of social media companies and to the extent that we are strung along by them? Every time we say, “You do not have enough people taking down stuff” the next time they come along and, lo and behold, there are a couple of hundred extra.
Dr Frampton: Certainly we have probably passed step one, which was more robust community guidelines for conditions. If you look at what the community guidelines say about conditions for things like Facebook and YouTube, on paper they should be quite robust. However, clearly they are not being implemented.
The next stage is to seriously consider the idea of a hypothecated levy on the companies that are used to support things like the Counter Terrorism Internet Referral Unit. It begins, again, to shift the balance of that relationship. At the moment—this is why it relates to the broader question of the framework in which they operate—it falls to the police to inform the companies that there is material they do not like. The police often complain that, in a different context, they are not informed if something is taken down and that it is potentially an investigation that needs to be followed. In an ideal world that relationship should run the other way, which is that the companies remove material that is problematic automatically. If they are able to identify people who are constantly trying to disseminate this kind of material, they inform the police—flip that relationship on its head. That is probably the next stage. I am not advocating criminal prosecutions of Facebook or whoever but, despite them constantly talking about tightening up their code of conduct and all the rest, there has not been that sea change in their approach.
Q230 Stuart C. McDonald: Does anyone else want to express a view about the idea that there is an extent to which we, as MPs, have been strung along for as long as social media companies can get away with doing the bare minimum to keep us off their backs a little bit? I think you said, Professor Tambini, that change is happening but you were referring really to legislative change. Do any of you detect a clear indication that social media companies get it and that they are going to make significant changes to how they operate, or is it just going to carry on like this until, four or five years down the line, we finally get to the stage where we say, “Right, we are going to do a German-plus model” or something else?
Dr Moore: Yes and no. They clearly are extremely conscious, particularly because of what happened in the US, of their greater responsibilities in the civic sphere. They have been reminded of that and are asked to sit in front of Senate House committees and various other things. At the same time they are also very conscious that when you have 2.2 billion users worldwide and your whole model is based on scale, automation and keeping human costs to a minimum, trying to instigate some of these processes—particularly the ones that are more interventionist—threatens their whole business model. If one was to flip liability, these companies would go under overnight and probably flip liability. They simply do not have the ability to police the amount of content that 2.2 billion people, in Facebook’s case, and 320 million people, in Twitter’s case, are posting online. In some sense they are very concerned. Equally, they are almost certainly limiting the number of steps they take because otherwise they potentially put themselves in a very difficult position from a commercial perspective.
Q231 Stuart C. McDonald: Following up on that, in your book, Dr Moore, you refer to progressive responses and that a laissez-faire shrug of shoulders is not attractive. Equally, some of the regulation is not attractive as well. When you are talking about progressive responses, are they competition law or data protection laws? What is it you mean?
Dr Moore: No. I would say that was more legislative amendment and changes. By progressive interventions I mean that we would not like to be starting from here. We have gone an awful long way down the road already, particularly in creating a privatised public sphere. It seems to me that in the context of thinking about where we go next we have to think about how we want people to participate online.
A good proportion of our lives is being spent in the digital sphere. However, most of those lives are being spent in private companies—private companies that make their own rules and some police them better than others. What we seem to have very limited amounts of is more positive interventions, whether that is news, discussion, engagement, enabling people to question their authority, or enabling people to do the sorts of jobs we used to associate with the fourth estate but that the fourth estate can no longer do. It seems to me that there is a huge potential for making those sorts of interventions at this stage. As far as I can see, that is not being thought about very much at all.
Q232 Chair: Give me an example.
Dr Moore: There is a wonderful civil society organisation called My Society, I do not know if you have come across it. It is 10-plus years ago. It creates websites that help people find out what their MPs are doing in their name. You might have come across it. “They Work For You” is one of the sites they built. They built a site that helps people with Freedom of Information requests. They built a site called “Fix My Street”. Essentially they build civic technology to help and empower people when it comes to engagement with authority and their representatives. Now they have funding to do it around the world so they are taking most of their tools elsewhere.
If you imagine, for example, an institution that was producing and helping to develop civic technology that enabled people to engage much more in the digital realm in the same way that My Society did—but My Society times 100—then you would have potentially a huge change in the whole way in which people engage with politics.
Q233 Chair: I still do not understand what it is you are proposing. I thought what you were saying was there were things you could do to better democratise or open up those private spaces, the Facebooks or the social media platforms.
Dr Moore: No, I am talking about parallel spaces. A comparable, although very different, intervention was 100 years ago with the BBC. It was a very particular type of intervention to address what they saw at the time as a particular problem. It seems to me like we have to have similar—although obviously very different, recognising that it is 100 years on—types of positive intervention to enable people to have the types of space where they talk in their real names or as constituents, where they talk about members of a particular geographic area, and where they know they are talking to a real person—to a councillor. We need to have ways in which people can properly look at the way their representatives are acting in their name. At the moment we have legislation around open data, for example, that is, frankly, useless from the perspective of citizens because they can download in seven-point font a pdf on what their local councillor is spending. There are an awful lot of things that one can do that help empower people and help give people the opportunity to engage online in a much more representative way than simply tweeting insults at their politician or whatever else.
Chair: It sounds like the aspirations behind Matt Hancock’s app but I am not entirely sure it quite worked out like that.
Dr Dzakula: There are already a few alternative social media networks being built on Blockchain that have a completely different type of governance, where users have a vote on how the network will govern. There are a few alternatives.
Q234 Chair: Did you name one then, sorry? You can let us have more in written evidence.
Dr Dzakula: There are a few of them and they have a different business model so, yes, it does work differently.
Q235 Stuart C. McDonald: In your scheme, Dr Frampton, you mention a regulator for social media. Generally for the panel, is that a realistic option? What exactly would a regulator be tasked with doing? Who should the regulator be, if we need one?
Professor Tambini: You have one in the German law. You have a self-regulatory body that has a very minimal role. It is more like an appeal court. If the object is to say these platforms are not dealing with takedowns in a way that we understand and they are not doing it quickly enough, they have incentives to deal with manifestly illegal content within 24 hours. If it is really too difficult—if it is a complex issue of defamation, if it raises difficulties in balancing free speech, if there are things like political satire that are sometimes very difficult to deal with—that can go to the regulatory body.
The German model is for an independent self-regulatory body. I think it is publicly funded. It has to meet requirements of being independent from the state and from the industry broadly. This deals with those issues that the platforms find too difficult. As long as they refer difficult cases to this body within the right amount of time, which is seven days, and as long as they accept that they will be bound by the decision of that body, they have met their obligations and they are not then liable to be fined.
Part of it would be that procedurally-focused check—do the procedures work, are they transparent, are they working within the rules broadly?—on the work that is done by the platforms and the algorithmic transparency that we were referring to. A US legal scholar called Frank Pasquale has theorised this as an independent regulatory authority to whom details about principles behind the software code and the objectives and even the code itself can be disclosed to an expert body that is capable of seeing where the decision-making is occurring, how it is occurring and if it is meeting the principles that the platform claims to be implementing.
Q236 Stuart C. McDonald: Anyone against a regulator? Why shouldn’t there be a regulator?
Dr Moore: One of the things that is not particularly distinguished in the German legislation, and which I think we can learn from the discussions and recommendations from the Leveson inquiry, is the clear distinction between content and process, and the extent to which any regulation looks at the adequacy of processes that are put in place to deal with things, as opposed to looking at specific content that is posted up. That distinction is not really made in the German legislation. Are the processes by which these companies are dealing with this stuff adequate? Can they show they are dealing with them in a fast enough way? Can they show that they are providing adequate responsiveness and adequate redress as opposed to policing the content itself as a regulator?
Q237 Stuart C. McDonald: How can you make an assessment of the adequacy of the processes unless you are making a judgment call on what the end result of those processes is? If you are not making a decision on whether or not certain content is good or bad, how can you say whether or not the processes are adequate? Is it easy to separate those two things out?
Dr Moore: The obligation needs to be on the companies themselves to show that the processes are working adequately. Damian mentioned previously that the transparency report that is being recommended seems to be going in that direction, but it is not very clear exactly what the criteria are by which we judge, by which the process will be judged, nor is it clear who is going to be regulating that process and who is going to be overseeing it. Those seem to me to be two central questions as to exactly what criteria these companies are being judged on and how they are proving and demonstrating that they are doing an adequate job and who is judging it.
Q238 John Woodcock: Could you describe to us what you think a world where social media companies were categorised as publishers would look like? How would they change?
Dr Frampton: I am not sure I would go as far as saying that they are publishers. I think you need to come up with a new way of thinking about it.
Q239 John Woodcock: I am not asking, first of all, whether you think it is a good idea or not, but clearly it is seen as a step too far for many people. Say it did happen; what do you think would be the effects?
Professor Tambini: They would have to moderate user-generated content and a lot of the free flow of ideas on these platforms, which have so many beneficial consequences, would end.
Dr Moore: Certain platforms would almost certainly close. Twitter has people questioning its future anyway. I think certain platforms would be unwilling or unable to do that, so they would not necessarily close worldwide but they would certainly remove certain services from the UK. We need to bear in mind that for many of these companies the UK is a priority market, but it is a relatively small market compared to some of their others.
Professor Tambini: In a sense, the question, if you will forgive me, is brilliantly provocative but simplistic, isn’t it? Isn’t it possible that there is a third category and that merely by presenting them as publishers we are in a sense driving along looking in the rear-view mirror?
Q240 John Woodcock: Yes, possibly, but the reason I asked the question is because for simplistic people like us it seems to be common sense. It is helpful for us to be able to visualise a bit more how fundamental a change it would be for both the existence of platforms and user experience. Would you envisage, if Facebook became a publisher, each post would have to be moderated before it would go on? What could you envisage would be the safeguards that those platforms would take? Clearly, none of you are fans of it.
Professor Tambini: I think there are different areas of law in which a definition of a publisher would change different things about defamation versus some of the other categories of harm that we might consider. It would be complicated, but broadly I think the business model would change entirely, particularly for user-generated content. I do not think it is an attractive solution. I do not really think it is a solution because in a sense it is oversimplified. We need to think a bit more creatively about changing the sense.
Dr Frampton: I have wrestled with that issue a lot and tried to come up with what the ways of conceptualising it are. Could you have a notion of almost media curatorship, which more in a way captures what they are doing, that they are constructing the flow of content produced by others?
Q241 John Woodcock: Could you explain that a bit more?
Dr Frampton: What they do has elements of being publishers and that is why they are certainly not, as the original law conceives of them in the European context, neutral, passive disseminators of information. That is why there is little in the way of liability that attaches to them at the present time. I go back to where I started. It is how you transform that equation, and I do not have the answers, really.
Q242 John Woodcock: That is what my next question was going to be. Potentially, this is where we are going. I do not know, Professor Tambini, if this goes to your original China point, but we have a committee recommending a step towards that. Can any of you envisage what that might be in a constructive, workable-type way?
Dr Frampton: I would repeat what I said earlier, which is that I think the presumption of law should be that they are liable unless—
John Woodcock: You change the presumption?
Dr Frampton: Yes. I would change the stated presumption because I think that gives you more powers to be more robust and more assertive about what you require of them.
Q243 John Woodcock: Do others think that is workable? Is that a direction that others would be comfortable with on the platforms?
Professor Tambini: Liability is one thing and that means that somebody with the power and resources to sue them will have their rights protected. That is not everybody, so some sort of a self-regulatory, co-regulatory system, which the platforms are involved in, would deal with a much greater volume of material that would protect more people. What we have now is not simply that they have no responsibility at all. They are responsible, they are liable, when they are notified.
One of the things that is very often not clear is when a notification has occurred. Making clear in some way that notifications can be any complaint to a platform, working on things that are familiar in terms of complaints buttons, and designing in more effective complaints procedures to the platform itself could be very good. In a sense, it is developing that notification and takedown procedure rather than saying by a sweep of a gavel that the platforms are now publishers. I think it is more realistic. The backlash when people find that they are not able to communicate on platforms they love in ways that they are very used to communicating with would be too difficult to deal with.
Q244 Chair: In theory, if someone is libelled on Twitter, for example, and they inform Twitter of that and Twitter does nothing, could they then sue Twitter?
Professor Tambini: Yes.
Q245 Chair: Is there any example of anybody having done that?
Professor Tambini: I cannot cite you a case, but yes, this is—
Q246 Chair: Suing Twitter itself as opposed to suing the person who also libelled?
Professor Tambini: Yes, because when they are notified that they are hosting this content, that is the moment their liability commences and that is how the system works.
Q247 John Woodcock: With your complaints button, one option for a complaints button could be to get a level of automation or certainly build in greater ease in informing the platform that it is libelling it?
Professor Tambini: Yes.
Q248 Chair: Suppose it is an anti-Semitic hate crime. In those circumstances is there any civil action that could be taken or are you just reliant on the CPS deciding that it is in the public interest to prosecute if it has not been taken down in a week?
Professor Tambini: I will preface this by saying I am not a lawyer so I would suggest that you hear evidence from some lawyers if you want to develop this area, but I would presume so. I will not go any further than that.
Q249 Chair: Part of the reason I am pushing this is about whether or not it operates as a remedy. If in the end this simply relies on the incentive being whether or not there is going to be a criminal prosecution, the reality is, with the pressures on the police and the CPS and so on, if something is taken down in a month rather than in 24 hours, it is still as long as it is taken down unlikely to cross the threshold of whether or not it is worth the CPS and the police going through an entire criminal prosecution. However, the harm done by something being up for a month rather than 24 hours is still very substantial. It does not feel like the criminal liability in that respect is going to bite as much as an incentive.
Professor Tambini: That is why there is no credible threat as there would be in relation, for example, to intellectual property because the platforms and the hosts of the content do not see huge risks that would make it rational for them to invest a huge amount in effective review and takedown procedures.
Dr Frampton: The way that the current law has operated, the way it is phrased actually, is the idea that once they have actual knowledge of this material, as Professor Tambini says, their liability commences and then they must remove it expeditiously. Of course, there has been all kinds of wriggle room as to what that means. That is where I think the German model is interesting because it, for the first time, puts that timeframe there.
Professor Tambini: I should mention something briefly as well. Article 15 of the ecommerce directive also prohibits member states imposing an obligation on platforms to monitor. Going back 20 years, the objective of this framework was to foster free speech, innovation and economic growth in this wonderful new thing, the internet. They wanted to clearly underline that governments should respect this shield for liability. In countries with a less robust democratic tradition, you can see why it might be necessary to say, “Be careful. Do not oblige the platforms to do your dirty censorship work for you. You, the state, can’t oblige them to monitor on your behalf”. That is the system we have. This Committee asking platforms to monitor would actually contravene the directive.
Q250 Naz Shah: That final point is an interesting one. I have two very brief questions. The first one is to Dr Moore and Professor Tambini. You will be aware that we have had the social media giants in as part of our inquiry. Do you think we should be extending that to print media, given your expertise? Should we have print media in before us to talk about hate crime?
Dr Moore: Have you not?
Naz Shah: No. It is just a yes or no, really.
Dr Moore: Yes.
Professor Tambini: Certainly.
Q251 Naz Shah: Thank you. Dr Frampton, given your Policy Exchange hat, the Policy Exchange has a rather poor history among many British Muslims, whether due to the alleged fabrication of evidence according to “Newsnight” and its report on your report “The Hijacking of British Islam”, to recently being found to have misrepresented a poll of British Muslims. Given that, do you feel that you are in a good position—or any position—to be effective in your positioning on hate crime against British Muslims in relation to anti-Muslim hate crime?
Dr Frampton: I will speak to the poll because that is the report that I authored. First, I do not think we misrepresented anything. I think there was a complaint that the newspaper article had misrepresented our report. We did not misrepresent anything.
If you look at the content of that report, the key messages were that British Muslims are no different from the rest of the population on most issues. Their interests are secular and very much in line, perhaps more in line, with “British values” than is often said to be the case. We merely raised a set of issues where we thought there were particular wrinkles of that around things like conspiracy theories, about a certain manifestation of conspiracy theories, and certain social attitudes. What we found there was no different from various other reports that have been done in the past. I think it is a slightly odd question and not entirely relevant to what we have been talking about.
Naz Shah: I feel that it is rather relevant given that we are talking about hate crime, and hate crime impacts on British Muslim communities as it does on other communities. I would put it to you that while I accept and I agree with what you have just talked about, in the past when we have had evidence particularly from the MCB, the Policy Exchange did not exactly give the same kind of answers that you have given. Your answers have definitely been more balanced, which I welcome, but in the past that has not been the case. Thank you very much.
Q252 Chair: I will ask a quick couple of final questions. On the different policy options and responses, I think you all said there ought to be more transparency and you would include in that legislating for transparency. Is that fair?
Dr Moore: Potentially, yes.
Q253 Chair: Secondly, you were all interested in the possibility of some form of regulation, but there would need to be some care attached to the nature of the regulation and that a version of the German model might be effective. Again, is that fair or do some of you disagree with that?
Professor Tambini: I think we should come up with our own model, which may be inspired in part by the German model.
Dr Moore: I also think there is an awful lot we could do prior to setting up a regulatory body in both progressive interventions and making these companies do more.
Q254 Chair: Using the existing law or additional legal measures?
Dr Moore: Yes, using the existing law.
Q255 Chair: The top one on that list would be which?
Dr Moore: I think design changes by the companies themselves in the way in which they alert their users and make clear to their users when they are potentially liable for some of the content they are publishing.
Q256 Chair: I think you were all uncomfortable with the idea that the platforms should be fully liable, but thought there might be more scope to approach it in a different way or to use the existing law in a different way. The Prime Minister’s speech this evening is talking about there being a new social media code of practice. What would you put in a social media code of practice for the UK that would be different from the EU code of practice already in existence?
Professor Tambini: Nothing. We have a lot of codes. The challenges are in the procedures and the level of transparency and when decisions are being made ensuring that the difficult decisions are referred and there is a sufficient level of due process in those decisions. Some of those things can be written into the code and I think some of them are written into the EU code of practice.
Q257 Chair: Do you see any issues with jurisdiction? Germany has found one way to address jurisdiction. Do you think there will be jurisdictional issues given that a lot of these companies in the end are US based?
Dr Frampton: I think there is a jurisdictional problem but the companies are clearly sensitive, nonetheless, to the way in which they operate in European and British jurisdictions. There is a whole set of bigger questions about the provenance of the material more generally, but there are mechanisms that you can bring to bear on them. It is clear from their reactions to the reports of this Committee and what the Prime Minister is saying that this is something that they are worried about.
Q258 Chair: I am conscious of the time. Unfortunately, we have another meeting and we have obviously taken your time up hugely. We are really grateful for your time. Your evidence has been immensely interesting. If we have any further questions, can we write to you and ask for any further thoughts you have? If you have any further things that you think that we should be considering, please do give us any further written evidence as well.
I am conscious, Dr Moore, of your advice to us to not be too much in haste in the short term but to be ambitious in the long term. The challenge for us, not just in Parliament but in parliaments across the world on this, is that it probably took decades to bring in the kind of antitrust legislation to deal with the initial industrial monopolies, just as it took decades to bring in the seatbelt legislation to deal with cars and traffic accidents. We cannot wait decades to deal with these sorts of concerns. Therefore, anything more you have that either we can pressurise the social media companies to do themselves or we should be recommending to Government that they do would be very welcome to us. Thank you very much for your time and evidence.
Dr Moore: Bear in mind that the original antitrust legislation that was brought in in the US, the Sherman Antitrust Act, was only discussed in 1887 and was brought in in 1889. It was a two-year timeframe that, given it was in an analogue, predigital age, was really rather fast.