Over 100 Roma migrants evicted by police in Sweden. Demotix/Rikard Stadler. All rights reserved.
The year 2013 was significant with respect to surveillance in Sweden, maybe even more so than in the rest of the world. In December 2013 to January 2014 a major debate erupted about the profiling practices by the Swedish police while carrying out internal border control in the search for illegal migrants (the REVA programme).
In June, a new regulation on video surveillance was passed that significantly loosened up the earlier regulation. This however drowned in the media flow as Edward Snowden and the Guardian journalists began releasing the data on NSA mass surveillance. Finally, in September the daily Dagens Nyheter broke the news that the Swedish police had created a registry of all Swedish Roma – over 4700 men, women, children, and even some deceased.
These ‘events’ can and should be viewed as instances where surveillance practices play a dominant role. It is thereby interesting to note how few connections were made between them in public discourse, despite the rich soil for addressing issues such as generalized suspicion, the ontology of threat in Sweden and the west more generally, and discriminatory surveillance. Below we discuss the results of a study of the news media coverage on two surveillance ‘events’ in Sweden: the NSA mass surveillance and the Roma database.
While the debate that followed the NSA leaks showed strong similarities with the debate on the agency’s Swedish counterpart in 2008 – the National Defence Radio Establishment (in Sweden referred to as the FRA) – the debate on the REVA programme and the Roma surveillance was carried out on entirely different premises.
Below we discuss these differences. We pose the question: what really is the problem with surveillance in these different contexts and who is it who is considered to be victimized by surveillance practices? The resulting analysis tells us a lot about the legitimation of surveillance and why discriminatory practices by the Swedish state can continue unchecked.
Research on surveillance in Canada and the United Kingdom shows that surveillance is considered more problematic when it is directed toward the general population rather than a particular sub-group. Targeting ‘ordinary citizens’ to curb organized crime provokes harder reactions than similar measures targeting a national minority, according to this thesis. We studied the total coverage of NSA mass surveillance and the Roma database ‘events’ in the four main national Swedish newspapers, primarily with the aim of finding out more about the media discourse on surveillance in Sweden. More particularly, though, we were interested in representations of the Swedish Self and its constructed Others in the context of surveillance.
In both cases the surveillance is described as problematic and misguided. But we see a clear difference between the events as regards the cause behind the problem. In the coverage of the NSA ,invasive state surveillance was depicted as a clear threat to the Swedish population (e.g. Expressen, 2013-06-08). However, when the newspapers covered the event of the illegal registering of Roma and their relatives, the surveillance aspect completely vanished from the picture.
The minute the distinction between Us and the Other becomes very obvious, the act of surveillance as such becomes unproblematic. Instead the racist motif of the police was the object of critique. The problem at hand was thereby transferred away from the debate on surveillance and privacy to the area of racism and xenophobia. This is a place where it becomes less threatening to the general public, because it does not concern them directly. Instead, this opens up the possibility to (rightly!) criticize racist law enforcement agencies, whilst ignoring the surveillance infrastructure that potentially makes it possible to register everyone in this way.
The government’s official position with respect to the two ‘events’ were fairly easy to predict, given the nature of the problems and the involvement of the Swedish state in both instances. Thus, the strategy with respect to the Roma database has largely been to repair the Swedish image, and in the case of NSA surveillance, to pretend it did not really happen. Both strategies have proved impossible, of course.
The repair strategy was deployed first and foremost by politicians and in editorials. Maintaining an image of Sweden as a country where this must be an isolated event, carried out by deviant individuals and independent of any racist social structures was central to most actors. A prime example is Erik Ullenhag, the Minister of Integration, who stated for example that, “I thought that it was completely unthinkable that something like this would happen in Sweden. Registering people based on ethnicity is illegal, it’s unworthy.” (Expressen 2013-09-24). In a somewhat naïve fashion the mere fact that the surveillance is illegal convinces the minister that it won’t happen, (after all, this is Sweden!).
With regard to reporting on the NSA surveillance, when questioned by journalists the Swedish government has attempted to disregard, downplay, or refer to applicable legislation. A much cited example of downplaying was made by Carl Bildt, Minister of Foreign Affairs, with respect to the surveillance of Angela Merkel (that was possibly carried out by the Swedish FRA at least in part).
At a press conference in October he stated that this was merely business as usual. “I know what world we live in”, he stated, and continued to say that it was not the American surveillance he was worried about, implicitly pointing towards Russia (Aftonbladet 2013-10-27, Dagens Nyheter 2013-10-25). Both repair and disregard are quite problematic strategies because they strengthen the tendency that we have already identified and prevent connections being made between the different surveillance ‘events’. They do this by isolating the substance of the surveillance to a specific context and making any discussion applicable only to that specific case.
However, when looking at the terminology used to describe the surveillance ‘events’, on the other hand, there are great similarities. On a conceptual level both debates explicitly focus on privacy, the Swedish translation of which (“integritet”) many before us have noted to be highly ambiguous. It signifies privacy, autonomy as well as physical integrity in the sense used in English. The violation of Roma peoples’ “integritet” occurred through registration in the coverage – not surveillance. The NSA, however, is clearly discussed as a case of surveillance. While both practices conceptually are “surveillance”, as understood in social science, the Swedish discourse makes a clear separation between law enforcement and secret intelligence agencies. This may seem a petty detail, but it also prevents Swedish writers from drawing connections between the two instances: In our data set of ca. 1650 texts only two (!) made any substantial connection between the two ‘events’.
The two surveillance events also have their significant differences when it comes to who expresses the problematic nature of surveillance. During the FRA debate and the debate following Edward Snowden’s revelations the journalists themselves, oftentimes in the editorials, express their concern about the surveillance.
With regard to the Roma registry the problem is often formulated during interviews with Roma representatives who have been subjected to surveillance themselves. On the one hand, giving the concerned party the right to express their views publicly may be something positive. On the other, the newspaper editors place the responsibility for convincing the readership that this was indeed something of great concern onto the victims of surveillance themselves, while at the same time lamenting about the fact that something like this could have happened in Sweden.
Given the Swedish history of institutional surveillance and racism it is actually not that surprising, but rather a sign of continuity of a long tradition of oppression of, among others, Roma people, as the more recent coverage of the Roma registry has shown.
Doing unto others
The effect here is that surveillance becomes limited to a particular level. It never elevates to some level of generality, as in the case of the FRA or NSA surveillance, where news coverage instead focuses on the threat to privacy of all people.
This ‘particularity’ de-politicises the matter and actively removes the debate from the general level that acknowledges each and every person’s right to privacy, down to the level of a particular police district’s lack of knowledge about the long tradition of Swedish oppression of the Roma.
The fact that we find almost nothing written about the Roma registry that indicated that the surveillance in itself was problematic shows the accuracy of earlier research in Canada and the UK. This is also the Swedish case: when surveillance is applied to the Others it is no longer problematic.
Notwithstanding the essential fact that the Roma registry is a modern expression of age-old antiziganism, we believe that the coverage by Swedish mainstream media clearly displays (and reproduces!) the view that Roma people, in 2013, still do not qualify for the status of Swedish citizens.
In Swedish coverage the right to privacy and integrity, guaranteed for all citizens, are not considered essential democratic rights.
While in fact intimately tied to the very rights designed to protect every citizen against discriminatory practices, the institutional racism of the police is viewed as completely separate from the problematic of surveillance.
This disregard for the police’s modus operandi, which on a more general level is a threat to the rights of every Swedish citizen, may result in a failure to address the mechanisms by which the Roma registry was created and maintained in the first place, i.e. the vast surveillance infrastructure in the hands of Swedish authorities.
We wonder how it can be that the newspaper editors’ and politicians’ We are only victims of state mass surveillance, while the Others are targets for a racist ‘anomaly’ by the police?
Read more from our 'Joining the dots on state surveillance' series here.
 Aftonbladet, Dagens Nyheter, Expressen, and Svenska Dagbbladet. NSA mass surveillance N=1156; Roma database N=459. The study covers the first six months of each ‘event’.
 Even though the Chancellor of Justice, the highest representative of the citizens vis-à-vis the state in Sweden, has judged the Roma database to be an infringement (May 7, 2014), the settlement sum of roughly €550 actually does more damage than repair trust in Swedish law enforcement.
 The first article by Ewa Stenberg (Dagens Nyheter 2013-11-16) actually addresses both the issue of institutional racism and the surveillance culture in Swedish public administration, whereas the other is an opinon piece by legal expert Mark Klamberg about institutional oversight (Expressen 2013-11-07).