Radicalization Series – Part I: The slippery slope of ethnic profiling

The Open Society has published a report on ethnic profiling that I will discus here very briefly and give extensive quotes about the Dutch situation:
Ethnic Profiling in the European Union: Pervasive, Ineffective, and Discriminatory

Pervasive use of ethnic and religious stereotypes by law enforcement across Europe is harming efforts to combat crime and terrorism, according to this report released by the Open Society Justice Initiative.

Ethnic profiling occurs most often in police decisions about who to stop, question, search, and, at times, arrest. Yet there is no evidence that ethnic profiling actually prevents terrorism or lowers crime rates.

Throughout Europe, minorities and immigrant communities have reported discriminatory treatment by the police. From massive data mining operations to intimidating identity checks, ethnic profiling is often more of a public relations stunt than a real response to crime. The report, Ethnic Profiling in the European Union: Pervasive, Ineffective, and Discriminatory, details widespread profiling in France, Germany, Italy, The Netherlands, and other EU member states.

The report very clearly shows how and why ethnic profiling just doesn’t work and that conclusion has received much attention in the media. The report however also focuses on the way Franc, Germany, Italy and the Netherlands deal with the radicalization issue. Important in the policy theories of radicalization is the slippery slope paradigm:

Theories of Radicalization and the Slippery Slope into Extremist Violence (p. 93-94)
In practice, however, radicalization theories demonstrate a dangerous tendency to conflate an individual’s adoption of a conservative or “fundamentalist” practice of Islam with a willingness to resort to violence. Many radicalization theories rely on a “slippery slope” paradigm which posits a radicalization continuum along which individuals are believed to slide—gradually or rapidly—from increasing religious devotion, through conservative or “fundamentalist” streams of Islam, toward supporting terrorist activities and organizations until, in a limited number of cases, they end by directly participating in terrorist activities and organizations. The implication is that all conservative Muslims are potential terrorists; this constitutes a broad generalization that stigmatizes a group of persons on the basis of their religious beliefs. When such theories are the basis for police or other law enforcement operations without reliable supporting intelligence on terrorist threats, it is ethnic profiling. This “slippery slope” paradigm of radicalization is widespread in Europe and underpins counterterrorism practices of police and intelligence officials in many countries.

[…]
The “ring model” of the Dutch General Intelligence and Security Services (AIVD) provides a similar illustration of this paradigm. The model consists of four rings, one inside the other, representing (from the core moving out): terrorists, supporters, sympathizers, society. An AIVD official described the model:

The innermost ring [terrorists] represents persons who are able and willing to commit attacks. The active supporters of terrorists can be put in the second ring [supporters]. These persons can and want to support terrorists, so they are aware of the connection between their activities and terrorist attacks. The third ring [sympathizers] represents the people who feel some sympathy for the cause and who are susceptible to recruitment. In general, persons in this third ring reject the Western, Dutch society. The area outside the third ring [society] encompasses the entire Muslim society. The people in this area are in no way involved in Islamist extremism, but may fall victim to its actions. In this ring model our focus should not only be on the groups to be distinguished, but also on the interaction between the rings.

As, rightly so, argued the slippery slope paradigm by definition turns every Muslim into a potential radical, even though it is contested by many experts:

Centripetal movements can be designated as radicalization processes.
According to this approach, observant Muslims, particularly those practicing conservative forms of Islam, are potential terrorists. Behaviors that indicate that an individual is becoming increasingly devout or adopting a more conservative form of Islam thereby become tell-tale “indicators” of radicalization. Individuals so identified may then become the focus of various antiterrorism measures.

This slippery slope paradigm is contested by experts. A French counterterrorism official with over two decades of experience recognized the differences:

A Muslim who is not radical in his practice, but who incites to violence is dangerous, and therefore of interest to us. On the other hand, a Muslim who is radical in his faith, but who is above all very pious, is not of interest for us. …[A]s far as Salafists, there are deeply pious Salafists who are radical but non-violent. You can compare them to Cistercian or Benedictine monks who are very pious, but not violent.

Saimir Amghar, a French researcher investigating radicalization processes among Muslim youth notes that the term “radicalization” covers several different phenomena, and argues that there are really three types of radicalization:
(1) nonreligious political radicalization;
(2) religious radicalization involving orthodox practice of Islam but rejecting violence;
and (3) political radicalization drawing from religious doctrine that manifests primarily through violent jihadism.
While similar factors may drive individuals toward each form of radicalization, they are distinct responses. Amghar argues that the second and third forms do not represent steps on a continuum but are in fact oppositional tendencies that are highly critical of one another. Under this view, the nonviolent forms of conservative Islam are in fact a bulwark against terrorism rather than a path toward violent jihad.

The Dutch agencies have been very active in operationalize slippery slope paradigm by using an approach that uses indicators to identify persons who could radicalize:

The Netherlands: Operationalizing the Theory of Radicalization (p.94-96)
The objective is to enable these actors to identify persons or organizations of potential interest to the police and intelligence services. In 2006, an official from the office of the National Coordinator on Counterterrorism, the institution responsible for developing policy and coordinating anti-terrorism measures in the Netherlands, explained their approach to profiling:

We are also working on some initiatives in Amsterdam and Rotterdam to develop indicators—some kind of criteria of what to look for—that can also be used by people that are not specialists in using profiles in a critical way so that they can understand what kind of behavior should or would be potentially of our interest…We are refining them and refining them—trying to put in an administrative system.

The city of Rotterdam took the lead with a program called “Join in or get left behind,” initiated in February 2005. According to this program, indicators of radicalization include particular behavior patterns, such as frequent travel or hosting gatherings at one’s home, and changes in behavior, such as a man of Arabic origin who suddenly acquires more traditionally religious Muslim approaches to hair style, dress, mosque attendance, or physical contact with women in public. Dutch officials have taken pains to avoid the inclusion of ethnicity or nationality as suspicious criteria, but the indicators developed nonetheless draw attention to individuals who are becoming more orthodox in their practice of Islam. Essentially, a Muslim who shows outward signs of more conservative practice would become suspicious. Likewise a non-Muslim who outwardly shows signs of Muslim practice, indicating that he is a convert, would also become suspicious.

Those trained to watch for these indicators are reportedly told to watch not just for one change in behavior, but several. When they believe someone is radicalizing, they are asked to report the individual to the information “switch-point,” which verifies the situation and determines the most appropriate follow-up action. According to an evaluation report by the information switch-point, the Rotterdam program alerted police to 17 cases during 2005. Although the numbers to date appear relatively few and the consequences of identification benign, the indicators of radicalization clearly target Muslims and are likely to stigmatize a far larger number of Muslims than those actually identified as at risk of radicalization.

The indicators of radicalization used in the Netherlands continue to conflate orthodox religious practice with a tendency to use violence. A government guide for companies on detecting radicalization among their employees gives this advice: In determining whether there are radicalized personnel in your employment, a combination of factors must be taken into consideration. The following list provides a number of indicators which might signal the presence of radicalized personnel:

• Possession of extremist literature, pamphlets, or sound and data recording equipment,
or the perusal of extremist literature by means of the internet. This can be difficult for
companies to assess, as such activities are often carried out in another language, such as
Arabic.
• Seeming approval of terrorist attacks.
• Travel to regions or countries in which a terrorist conflict is taking place or in which there
are terrorist training camps, such as Chechnya, Kashmir, Iraq, and Pakistan.
• A sudden aversion to “Western customs” such as mixed activities (male/female), or drinking
alcohol, and requesting specific Islamic meals.
• Wearing specific clothing and symbols, or a sudden change of clothing style.

The Open Society report is a bit to fast sometimes with its assessment of the Dutch approach but fortunately they also acknowledge a gradual change in the counterradicalization practices:

On a more positive note, some Dutch authorities are also distinguishing between  “extremism” involving support for violence and orthodox religious practice. The Amsterdam “switch-point on radicalization” has moved away from indicators focused on orthodox religious practice, and emphasizes the need to separate religious practice from political views—particularly whether an individual supports the use of violence.
Rotterdam is also moving away from the use of indicators of radicalization, although it remains in a process of flux. Across the Netherlands, there is a shift away from an approach that stigmatizes individuals and groups, and toward policies that address root causes of radicalization, such as discrimination, exclusion, and social polarization.

This move away from using stereotypical indicators is a good thing, and not only because the method itself is ineffective and reprehensible, but also, and perhaps most important, it creates an atmosphere of distrust that I have come across my fieldwork as well. This is something that the writers of the report also show very clearly:

Netherlands (p. 103-105)
According to some representatives of Dutch Muslims, Dutch intelligence services have used religion as a basis for monitoring, focusing their attention on Muslim organizations. Thus, it is alleged, monitoring has sometimes been conducted based on generalizations about the type of Islam that groups practice rather than specific information about activities in a particular mosque or organization. Such monitoring, some Muslims complain, has targeted even nonviolent streams of Islam, on the theory that they may be hotbeds of terrorist radicalization. Dutch authorities assert that surveillance has become narrowly targeted. There is wide gap between the perceptions of Dutch Muslims and Dutch law enforcement authorities of the scope and impact of surveillance practices.

Dutch Muslim organizations believe that Dutch intelligence services monitor many of the country’s mosques. Dutch intelligence services have argued in public reports that Salafist and other extreme streams of Islam are very active and in some cases are trying to influence or even take over less extreme mosques in the Netherlands. A 2005 Dutch General Intelligence Service (AIVD) report on the links among Saudi Arabia, Salafism, radicalization processes, and terrorism in the Netherlands, was based on information gathered from monitoring mosques considered radical. In a 2007 interview, Dutch counterterrorism authorities stated that very few mosques are under surveillance as “hotbeds of radicalization.” One official explained it this way:

We are not condemning the general thinking of a group… [T]his concerns less than one percent of mosques. It is really a very small number. And all of the mosques surveilled have Hofstadt group connections [and] imams who use violent rhetoric…”

In a June 2007 presentation, Deputy National Coordinator for Counterterrorism Lidewijde Ongering testified before a U.S. Senate committee:

A small number of locations in the Netherlands, such as a few Salafist centers and mosques, have been identified as potential gateways to radical milieus.…Experience has shown that for some young people, non-violent Salafism is a first step towards further radicalization. The Dutch authorities keep a close watch on the imams and governing bodies of these institutions and remind them of their social responsibilities. Our message is clear: we will not allow them to cross the line and publicly preach intolerance. We also expect them to exclude jihadist recruiters and stop young people from opting for violence. If people in or around these centers prove to be promoting radicalization or spreading hatred, we do not hesitate to prosecute them or deport them as a threat to national security.

Surveillance is largely conducted through direct contact with mosque authorities and individuals who attend mosques, and through established informants. However, in an unknown number of cases, intelligence services also tape record sermons, especially Friday prayers. A Muslim community leader from Rotterdam, Brahim Bursic, called attention to the taping:

We know that all Friday prayers in mosques are taped. I told the imams not to be afraid; we are a democracy and they are not doing anything against the law. I also publicly suggested that if the intelligence officials are interested in what is being said in the mosques, the Friday prayers could be broadcast on TV.

A senior Dutch police officer said that such recording only occurs in “very limited, specific cases.” In addition to mosques, Dutch intelligence officials monitor Muslim organizations they believe to be spreading or supporting radical Islam. In a 2004 report, the AIVD defined “radical Islam” as “the politico-religious pursuit of establishing—if necessary by extreme means—a society which reflects the perceived values from the original sources of Islam as purely as possible.” The report commented on the different views within “radical Islam”:

Radical Islam consists of many movements and groups that, although related (in particular concerning faith and anti-Western sentiments), may harbor very different views on aims and means. This means that various kinds of threats can emanate from radical Islam, one of which is terrorism. In addition to radical Islamic organizations and networks which concentrate on the jihad (in the sense of armed combat) against the West, there are other groups, which principally focus on “Dawa” (the propagation of the radical-Islamic ideology), while some groups and networks combine both.

Both organizations classified as “jihad-focused” and those classified as “Dawafocused” fall within what Dutch intelligence sources described as the AIVD’s “professional interest,” and some are kept under surveillance. While the AIVD recognizes that only “jihad-focused” groups pose an immediate threat of violence, it believes that the “Dawa-focused” groups pose a longer-term threat by feeding processes of radicalization. The AIVD’s 2006 guide for local authorities explains the Dutch approach of targeting “hotbeds of radicalization”:

A hotbed of radicalism is an organization, group or place that serves as a breeding ground for activities and views that are instrumental in radicalizing individuals and can ultimately result in terrorist activities. … Hotbeds of radicalism can also serve as an ideological breeding ground for extremists. They can function as a first step on a path that may lead to violence. This danger exists in particular in the case of organizations that advocate extreme, intolerant isolationism or promote an intolerant “us vs. them” mentality. …

The aim of the approach is to make clear through joint, coordinated government action to those in charge of the hotbed of radicalism and to its visitors that activities of a radical nature will not be tolerated and that the authorities are monitoring activities closely. It is not clear how often monitoring by intelligence services is founded on intelligence-based evidence and how often generalizations about ethnicity or religion are the determining factor. Nor is it clear how often more intrusive monitoring techniques are utilized. Certain Muslim places of worship and organizations are clearly viewed as suspicious, even without specific evidence indicating involvement in any terrorist activities or incitement to violence. They are instead held to be potentially dangerous due to generalizations about the stream of Islam that they practice, albeit peacefully, and the theory that such practices represent a first step in the process of radicalization. In the absence of information about support for terrorism, covert surveillance is inappropriate and law enforcement efforts should instead focus on outreach to Muslim communities and voluntary information sharing. There is a wide gap between the way Dutch intelligence officials describe their monitoring and the way it is perceived by Muslim organizations and individuals. Dutch authorities claim their practices have become more narrowly targeted over time, but this cannot be independently verified. To this day, many Dutch Muslims believe that discriminatory and profiling-based surveillance is widespread and this perception, accurate or not, has negative consequences for policing in the Netherlands.

(p. 110-111)
When individuals are stopped on the street for identity checks, when police surround a mosque, when a business is raided or an individual arrested, the general public naturally assumes that law enforcement officials are acting because there is a reason to do so—that these persons present a real threat. The lack of any significant counterterrorism outcomes—such as detection, charges, or convictions—as a result of ethnic profiling does not serve to mitigate the damage done; the bare fact of being singled out in the context of counterterrorism measures is sufficient to create the stigma. A representative of the Dutch Association of Moroccans and Tunisians described the change in perception:

Everyone thinks that when there is smoke, there is fire. Since 9/11 and the murder of Theo van Gogh, relations between Moroccans and the broader society have deteriorated. People look at each other with suspicion. When a Moroccan man walks in the street with a beard people look at him differently than before. When the police react in that way, it creates a bigger problem. It affects people’s perceptions. For instance, when a train was stopped recently and two men in Arabic dress were handcuffed and taken off the train by police, this image has a strong effect on those watching. And it makes big news in the media. Afterwards when it turns out that these men were not planning anything at all, they were just practicing Muslims, but it’s too late, the damage has been done.

The distrust among Muslims is very clear and many of them are very afraid of the negative consequences as was recently shown in the case of the ‘Ikea-threat‘ in Amsterdam which proved to be a false alarm. The information that lead to the arrests of the people was public information and could have been gathered by anyone. The fact that these seven people were Muslim, one of them being a relative of a person involved in the Madrid attacks was sufficient for the arrests. The hoax actually used ethnic profiling to make itself credible and probably the person making the fake call relied on his stereotype of the Dutch police using stereotypes. The Open Society institute refers to several reports that make clear that ethnic profiling doesn’t work because it is too static, too inclusive and too exclusive:

(P114-115)
[…]there is no single pathway to violent radicalization and the nature of Muslim practice is not a consistent or reliable factor in radicalization. These studies highlight a fundamental problem of using ethnic profiles: they are both overinclusive and underinclusive. They are overinclusive in that the vast majority of the people who fall into the category are entirely innocent; and they are underinclusive in that there are other terrorists and other criminals who do not fit the profile and who would escape attention if the profile were strictly applied. While overinclusion imposes
an unnecessary burden on “false positives” (persons who are innocent but match the profile), underinclusion may divert police attention from actual threats that lie beyond the prescribed profile. Thus, it was reported that, prior to the July 2005 attacks on the London public transport system, the leader of the bombers “had come to the attention of the intelligence services as an associate of other men who were suspected of involvement in a terrorist bomb plot. But he was not pursued because he did not tick enough of the boxes in the pre-July profile of the terror suspect.” The most authoritative report to date on the 7/7 London bombings concludes that “there is not a consistent profile to help identify who may be vulnerable to radicalization.” Another fundamental problem of profiling is its failure to account for the dynamism of its target: the subjects of profiling evolve in response to policing and law enforcement tactics. When a terrorist profile is known, terrorists can adapt to it through strategies of evasion and substitution. They may evade detection by recruiting individuals who do not fit the profile.

I think in general, besides being a little to black and white sometimes, the report is very good and gives a fair account of (at least) the Dutch situation. I do want to raise two issues: one about ethnic profiling and crime and the other one about radicalization.

First of all, although the report is not about crime, but the same questions can be asked. In the Netherlands professor Bovenkerk showed new statistics yesterday with an ‘alarming’ overrepresentation of ethnic groups.

Fact or stigma? Police registration of ethnic backgrounds in Rotterdam is highly disputed | Radio Netherlands Worldwide

Crime figures are normally compiled on a yearly basis. However, Professor Bovenkerk has looked at criminal behaviour spanning a number of years, between the ages of 18 and 24. His approach has resulted in an entirely different outcome. The professor linked information based on ethnicity to crime figures.

This is not permitted in the Netherlands, but the police department, youth care organisations and advisory groups in Rotterdam have been allowed to make the link, thanks to a special legal construction introduced in 2002.

Shock
Despite the multicultural projects, and the efforts of community leaders and youth services to promote integration, 55 percent of Moroccan youth between the ages of 18 and 24 are picked up by police as crime suspects.

The same goes for 40 percent of Antillean and Surinamese youth and 36 percent of Turkish youth. These statistics are based on the city of Rotterdam, but Professor Bovenkerk does not see any reason why the statistics would be differ for any other major Dutch city.

After being very reluctant to calculate the ‘ethnic dimension’ of crime in the 1980s and the beginning of the 1990s, since then it has become some sort of Dutch hobby to present new statistics every now and then which are in all instances very alarming. So we have the ‘fact’ here that we have an overrepresentation of migrant youth in crime statistics, in particular Moroccan-Dutch youth and than still ethnic profiling would not be useful? Is research like this not a particular modality of ethnic profiling? It indeed is and as professor Henk Effers says it can be useful to know exactly ‘where it hurts’ and subsequently base our policies on these findings. But just as the slippery slope paradigm turns every Muslim into a potential radical, ethnic profiling with regard to crime does something similar. It assumes or produces the image that the problem is in fact ethnic or cultural. And that is highly questionable as Effers (and Bovenkerk for that matter) argues:
Fact or stigma? Police registration of ethnic backgrounds in Rotterdam is highly disputed | Radio Netherlands Worldwide

Above all, he thinks it important that Professor Bovenkerk clearly points out the ‘ethnic explanation’. Saying that they are criminal because they’ve been brought up that way doesn’t make sense:

“On the one hand, there is the socio-economic question. Many Moroccans, but also other ethnic groups, are deprived socio-economically speaking. On the other hand, the social control within these groups is much lower than within the broader Dutch population.”

These two things are not unrelated, Professor Effers explains. When children are sent outside to play because there is not enough room inside, there is little social control on the street. The professor also remarks that his colleague’s report has brought to light the fact that almost 20 percent of native Dutch youth have had a brush with the law.

Scoring
Professor Bovenkerk’s report touches on a delicate subject within political circles. Politicians like anti-immigration proponent Geert Wilders could use these figures to score political points. Mr Bovenkerk would be greatly disappointed if his report were indeed to be used in this way. He has clear ideas on what a good method of tackling the problem would be:

“In any case, I think the job market plays an important role. There’s a lot possible but the youth need to be brought in. And then it’s really a question of stimulating more social control. Within the ethnic groups but also within the families.”

And I would like to add another thing to that. How long are we going to talk about the children of migrants as being ‘Moroccan’, ‘Turkish’ or in the case of people living in the kingdom of the Netherlands (but outside the Dutch state) as Antillians?

The second point I want to raise very briefly here is the issue of radicalization itself and the way it is treated in the report. The Open Society report takes radicalization among Muslims as a matter of fact. It does not really question the whole issue nor does it provide with a critical alternative for the radicalization theories prevalent in policy circles and counterterrorism agencies. It criticize the slippery slope paradigm but it does not ask the question what radicalization is according to them, what the actual developments are and how troublesome that may be (or not). Furthermore it only focuses on counterradicalization policies focused on Muslims. A different form of ethnic profiling however also occurs in the case of youth of the radical right, in particular in the past debates about ‘Lonsdale’ youth that designates white youth belonging to a particular youth culture as potential radicals. Furthermore by only focusing on Muslims it again appears that radicalization is only an issue when it comes to Muslims, thereby ignoring the violence and disturbance caused by radical left and radical right in the past 20 years.

I will come back on the issue of (defining) radicalization in a series of entries the coming weeks thereby (in the end) presenting my views and a possible working definition of radicalization. So, if you want to read that, stay updated and have not subscribed yet, you can do that HERE.