Marwa Fatafta: So, my name is Marwa and I work as a volunteer/activist with a Palestinian digital rights group called 7amleh, which means “campaign” in Arabic. And we have been documenting and researching into human rights or digital rights violations that are taking place in Palestine and Israel. And one of the most recent case studies or work that we’re looking into is the use of predictive policing by Israel, which is rather a sensitive issue given that there isn’t a lot that we know about the subject.
I promise I won’t go into politics—maybe just a little bit, to explain the context in which this predictive policing system has developed and how it’s used and what are the human rights and digital rights implications of it.
So it starts in October 2015. There was a surge in violence in Israel, in the occupied Palestinian territories where there was a new trend of a rise of lone wolf attacks where young Palestinians, and mostly teenagers, would go and commit violent attacks against Israeli citizens or Israeli security officers, and that instigated another cycle of violence. These young people were not affiliated with any political group. They were not part of an armed cell. There was no sort of political leadership behind them, or organization. They just took matters in their own hands and spontaneously decided that they want to do something.
And as a result of this let’s say new trend of violence, Israel—or the government of Israel—blamed social media companies for the incitement to violence online. And so they blamed companies, and namely Facebook, for hosting and facilitating terrorism and incitement to violence.
And so in September 2016, some Israeli officials said that there’s a special agreement between Facebook and the government to monitor and try to tackle the online incitement, something which Facebook actually denied in a forum that we organized at 7amleh in Ramallah in the West Bank. They said that there is no special agreement and that the Facebook standards apply to everybody. And they deal with all kind of requests from governments the same way. So they have to take down content if there’s enough evidence that they’re actually inciteful.
However, experience speaks differently. Because right after a delegation from Facebook visited Israel and met with Israeli officials, a number of accounts belonging to Palestinian journalists and two media outlets with millions of followers were immediately suspended. So Palestinians and activists got online and started a campaign under the hashtag of #FacebookCensorsPalestine, and after that Facebook apologized and said that was just a mistake. And so the accounts were restored.
So, aside from trying to pressure social media companies, the Israeli military intelligence started sweeping and mining social media accounts to try to look for early warning signs. And so since October 2015, which is the start of the violence, we have documented around 800 cases of young Palestinians arrested simply because they posted something on Facebook. And they have been put under administrative detention, which means that you are detained without any legal process. So you don’t have a trial, there’s no charge, and in most cases these people are detained for four to six months. And when that period is over, it gets renewed again.
And some of the lawyers that follow up on these cases, they said when there was a charge, the charge was incitement to violence through social media. That some cases, the evidence was not presented because it was a state secret or it was an intelligence secret and therefore they cannot disclose that evidence. Or, some of the detainees said that they were shown screenshots of their Facebook posts, and they were interrogated and charged with incitement based on that.
I mean, some of the stories… There are so many examples. I mean there are 800 cases, right? But some of the stories, an example is the story of Tamara, which is a 14 year‐old Palestinian girl from East Jerusalem who was arrested in the middle of the night from her home because she wrote on Facebook something like, “Please forgive me.” And that could be directed at I don’t know, her boyfriend, or her family, or some teenage drama stuff. But according to the Israeli military intelligence, it kind of grabbed their attention because they interpreted that she is about to go and carry out an attack.
Another prominent case is the case of the Palestinian poet Dareen Tatour. She was… She’s still, actually, under house arrest. Same story. Her home was ransacked by security forces and she was detained from her house in the middle of the night, because as a poet she posted a video in which she expresses her anger at the murder of Palestinian children. And that was enough to get her arrested and again, accused of incitement. But obviously it’s a clear case of violation of freedom, of expression online. Now she’s banned from using the Internet, she’s under house arrest, and she’s forced to wear an electronic…I don’t know what you call these…bracelet.
So as I said there are many examples. But the reason why I mention these two cases is to highlight the fact that these arrests are not related to actual terrorism or violent attacks, or even the planning to do terrorist attacks, but they are related to censorship and the violation of freedom of expression and also privacy of Palestinians. And this is especially so if we are looking at what’s happening on the other side, right. So at 7amleh we did a study or research to see what are Israelis saying. Is this incitement to violence online only a phenomenon on the side of Palestinians?
So what we found is that in 2016, Israelis posted racist or provocative posts against Arabs and Palestinians on social media every forty‐six seconds. And almost 60,000 Israeli Internet users wrote at least one post containing either racism or hatred towards Arabs, and most of these were related or had some explicit or implicit calls for death and killings of Arabs. And none of these Internet users were charged or arrested or investigated for incitement. Which again shows that this charge of incitement is rather a very broad definition, and it could include any kind of resistance to Israeli policies.
Now, what we learned this year through investigations done by Israeli media is that behind these 800 arrests, or this mass arrest campaign, there’s a predictive policing system in place which does social media analysis and flags certain people to be arrested later, based on an attack that may or may not [be committed] in the future. In other words these 800—or let’s say the majority of these cases—were rested based on a machine hunch.
And now, how does this system exactly work? We don’t know the exact inner workings of it. It’s a work of military intelligence, after all. But what we know is that… So, when it comes to predictive policing there are two kinds of tools. One that is location‐based, so the algorithm basically flags where and when the next crime could happen so the security or the police officers would deploy people to try to prevent that crime from happening.
Or it could be people‐based, where there are certain profiles developed. So you develop certain profiles of likely attackers, and again the algorithm monitors people’s activities online and if there’s something that’s suspicious, it flags that person as a likely attacker. And that system is what Israel uses. So they have built a number of profiles of likely Palestinian attackers based on certain data points, some of it related to age, location, and the psychological buildup of these people. And according to media reports there were psychologists sitting [in] on interrogations to basically drag information and see how do these young people think and what could be seen as some warning signs.
But also they look for certain trigger words which are often used in the Palestinian discourse. Things like the word the “shahid,” which means martyr, and this is a term that we use for anybody who dies from the conflict; words like “Al‐Quds,” which means Jerusalem; “al‐Aqsa,” which is the holy Muslim site; any Quranic verses from the holy book; poetry; if you put a picture of a person who just was killed or arrested by the Israeli forces. And they particularly look at the activity of friends and families of people who just got arrested, to assess whether these people would retaliate.
So for example, a 23 year‐old Palestinian kid with Down syndrome was shot dead and then two months later they arrested his brother. And during interrogation, the interrogators showed him screenshots of his Facebook and that he posted a picture of his dead brother. And that was enough for them to arrest him and to ask him straightforward, “Are you planning a retaliation attack against Israel, because you had the picture of your brother up online?”
So what does this all mean, this predictive policing business for Palestinians? It means that if you’re a Palestinian living in Palestine, it means that you are a suspect by default and you could be arrested for whatever thing you posted online. And that this is for an imaginary crime that you may or may not commit in the future.
And this is one of the things that some ex‐veterans of… The elite unit of the Israeli army is called Unit 8200. It’s an elite signal intelligence unit that is part of the IDF and is often compared to the NSA in terms of its surveillance capabilities. So, some veterans or ex‐veterans at the time, they wrote a protest letter to the Israeli Prime Minister Benjamin Netanyahu to protest against the unethical, unlimited surveillance that is being committed against Palestinians. And he says, and I quote, “There is no distinction between Palestinians who are, and are not, involved in violence. Information that is collected and stored harms innocent people.” So again this is to stress that predictive policing reinforces this governmentality that everyone is a suspect and everyone therefore should be surveilled and monitored.
But for us here sitting in the room, we have to look at this predictive policing in Israel from a different perspective, or a wider perspective. Because this unique combination of having this algorithm as a policing system, and also the ability to detain people and put them under administrative detention, which I assume in Western democracies that could not be the case—you can’t just simply…at least in theory, detain someone and not press charges and not go through the proper legal process.
So that combination is somehow unique to Israel and maybe cannot be replicated elsewhere. But that of course does not mean that such technologies are not being employed and growing in other police and security departments around the world.
In fact, the EU Counter‐Terrorism Coordinator (which I can never pronounce his name) said in Israel in meetings with the Israeli government and in some conference that the EU is interested in adopting this predictive policing system used in Israel, specifically to try and stop lone wolf for attacks in the EU. And in other EU states like Denmark, the Danish police have bought a similar predictive policing system from one of the companies that allegedly is involved in the development of the system that we now know in Israel. And only recently, like two months ago, the UK…there were some reports in The Independent that there is a delegation going from the UK government to Israel to exactly cooperate on predictive policing systems after the recent attack on London Bridge. So again, they want to look at how to stop lone wolf attacks by monitoring social media and doing social media analysis.
So, of course that is not a surprise, given that Israel is a leader in exporting cyber arms, cybersecurity, surveillance technologies around the world. And I think the reason behind this prominence as the revolving door between its army and the technology sector. So, some of the leading Israeli companies that export surveillance technologies, or cyber arms, cybersecurity products, they were founded by ex‐veterans of Unit 8200. And there is no legal barrier for these people who while working in this unit to take ideas developed there related to military intelligence and then later sell them as private sector companies. Which means that, in the words of an Israeli scholar, the Palestinian territory’s become like a certain lab to test certain technologies and fine‐tune them, and then sell them later to countries around the world.
And then of course the Israeli government and other governments justify the use of predictive policing by saying that it drops crime rates, or it’s effective in stopping terrorism. But I think that these claims, just exactly like the systems that they support, they lack statistical evidence. So a lot of the research that was done on the effectiveness of predictive policing shows no backing whatsoever that they are indeed effective. So one of the studies, or actually two, were done by the Rand Corporation in the US. They tried to look at the use of predictive policing systems in certain US communities, and that they found that they neither reduced crime nor increased public safety.
And there was another research done by the Human Rights Data Analysis Group to look at how predictive policing recycles existing biases, something that Maya just discussed. They used algorithms and tested test in Oakland, US on drug crimes data, and they found out that the algorithm was telling them to go exclusively to areas where there are only black residents and low‐income residents.
So I guess, and I’m ending here now, that the argument whether predictive policing is effective or not is not what we need to think about. I think predictive policing changes the traditional logic behind law enforcement, which means that if you committed a crime or you’re planning to commit a crime and there is hard enough substantial evidence to arrest you, then you’re prosecuted. But I think we really need to stop and and think very hard about systems that change this logic and you might get arrested and prosecuted based on a crime that you may or may not commit in the future. Thank you very much.