When discussions about interventions into the fundamental rights of citizens occur, critics are often calmed by the argument that these can only occur under strict requirements and only with a “judicial reservation”. Referring to the judicial reservation is a common argument: the citizens’ trust in the judiciary is much greater than their trust in the government, according to surveys. At the moment, arguments are again being made using the judicial reservation; this time it’s about the planned reintroduction of data retention.
The control instrument of the judicial reservation has been accused for many years of being less effective in practice. For example, two well-invested studies published by Bielefeld University and the Max Planck Institute for Foreign and International Criminal Law came to this conclusion in 2003. Both studies documented multiple deficiencies in the process at the time. The Max Planck Institute, for example, came to the conclusion that a surveillance measure would fail to be granted only in absolutely exceptional cases.
The Bielefeld University study stated at the time that only a quarter of surveillance measures were arranged according to the process regulations. Moreover, the surveillance measures would mostly consist of orders, which allows the assumption that judges do not reach their decision independently.
A public prosecutor who was surveyed by Max Planck Institut researchers at the time stated the following on the topic of email surveillance, for the record:
[translation] In the area of email surveillance, an update and clarification is required. Chaos rules. There are the most crazy legal concepts and regardless of which application I submit, the judge allows it in these cases.
Source: Studie des Max-Planck-Institutes für ausländisches und internationales Strafrecht, p226
We have also grappled for some time with the question of how the control instrument of the judicial reservation intended by the legislator for surveillance measures has developed in Germany – and how its effectiveness is controlled or evaluated. The occasion for this question was (among other things) telecommunication surveillance (TKÜ) that was ordered for which both we and our lawyers found the offence stated to be insufficient. Incidentally, if you think that a surveillance measure (TKÜ) couldn’t affect you because you haven’t committed any crimes, you are incorrect. In practice, people within the sphere of a suspect also have their communication surveilled or seized, even if there is absolutely no suspicion of a crime committed by that person.
If a suspect is ascertained and police officers with the public prosecutor instigate the seizing or surveillance of an email account, legal protection for the affected party is severely restricted by the secrecy of the measure. It can not be heard before the decision of the determining judge responsible. The judge should compensate for this deficit: the judge checks the case and if convinced that the telecommunication of the suspect should indeed be surveilled or seized, allows the public prosecutor’s application. Information as to how often a judge rejects an application for a surveillance measure is therefore an important indicator of how effective the control instrument of the judicial reservation really is. If, for example, all applications for surveillance were to be approved in a particular state, this would be a strong indication that the state is on the way to becoming a surveillance state.
How often a judge declines a surveillance measure mostly can not, however, be ascertained in Germany. In the Federal Office of Justice’s yearly report, only the number of rulings passed is specified, in which measures under § 100a Abs. 1 StPO were arranged, as well as the number of surveillance measures undertaken (cf. § 100b Abs. 5, 6 StPO). The German federal states have to supply these numbers to the Federal Office of Justice. Numbers such as how often an application for a surveillance measure is not satisfying to a judge are not included in the statistics, however. The judicial reservation is therefore a control instrument whose efficacy it is actually largely unknown.
We wanted to know if the corresponding numbers were perhaps available in the German federal states. At the start of the year, we therefore asked the justice ministries for information in writing.
Initially, the responses were disappointing. We received the same responses time and again in which we were told that no statistics were kept as to how often applications for a surveillance measure were denied. The number of cases in which applications for surveillance (TKÜ) were denied was supposedly unknown. The fact that the number of refused applications for surveillance was not collected was supposedly because there was no necessity for reporting in the law.
The Bavarian state justice ministry (among others) explained to us:
[translation] The necessity to report under § 100b Abs. 5, Abs. 6 StPo does not stipulate any requirement to compile denied applications, which is why no statistics on this exist.
The state justice ministry of Hesse told us that this would require an effort of manual analysis
[translation] that seems disproportionate to me and criminal investigative authorities can not be overburdened.
We then received the information that we sought: from Berlin, we received a reply that the senate in Berlin had collected the number of denied surveillance measures since 2006.
And we were shocked.
Since 2007, not a single application for telecommunication surveillance has been denied in Berlin. (See the respective yearly reports from the senate on the practice of telephone surveillance under §§ 100 a, 100 b StPO)
In total, between 2008 and 2014 in Berlin, 14,621 applications for surveillance were made – and approved. The number of surveillance measures arranged increased markedly over these years.
The fact that between 2008 and 2014, not a single one of the 14,621 applications for surveillance in Berlin was denied, certainly clarifies in our perception that doubts regarding the effectiveness of the control instruments of the judicial reservation are not only justified, but also that there is a need for clarification. How can it be possible that judges grant every single application for surveillance of a citizen over many years? What do these numbers say about the state of our constitutional state? The numbers from Berlin provide a wide overview of a large time period. In our view, they clearly prove that the instrument of the intended controls has actually not been of sufficient quality for a long time and a debate is necessary.
Over the years, the situation has got worse: the Max Planck Institute study of 2003 came to the conclusion that only 0.4% of applications for surveillance measures were not approved, and the rate in Berlin over the last seven years in a row is 0.00%. (Source: Max Planck Institute study, p177, PDF p197 and yearly reports from Berlin.) That all orders occurred conforming to the process regulations is doubtful: in any case, the Bielefeld University study of 2003 came to the conclusion that 75% of all surveillance reviewed was not ordered in accordance with the process requirements.
Without the yearly reports from Berlin, which the federal state of Berlin has voluntarily given out since 2006, there would be absolutely no numbers available on the effectiveness of the judicial reservation in Germany. For us, this is incomprehensible for reasons of democratic controls alone. The fact that every request for surveillance is approved according to the numbers available to us is, due to the lack of reporting requirements, not only unknown to the wider public, but the legislator can not evaluate the effect of its own control instrument. In our view, the legislator absolutely must compile nationwide statistics for the purposes of evaluation and control on how often applications for surveillance measures are actually granted, and how often judges decline surveillance (TKÜ). Only when appropriate statistics are available is control possible. Alarming developments can then be recognised early and debated.
We therefore recommend that adapting the reporting requirements for control and evaluation purposes under § 100b Abs. 5, Abs. 6 StPO to that effect, so that not only the number of surveillance (TKÜ) measures arranged can be statistically recorded, but also the number of denied requests for surveillance (TKÜ) in order to check the effectiveness of the judicial reservation.
According to studies, a lack of time and personnel in the courts has for years also contributed to the situation. We can see an important starting point to strengthen the control of surveillance processes here. The Max Planck Institute study from 2003 already explains, for example, that the judge for an investigation, with evidence of a heavy workload, only has ten to a maximum of 30 minutes to check a decision for surveillance (TKÜ). Another judge stated then that he was forced to put his “checking priorities” into more serious cases, like bodily attacks or arrest warrants. The study determined, moreover, that police initiation of surveillance (TKÜ) was regularly taken over by federal prosecution and the judge in the investigation. The reasons for the order for surveillance (TKÜ) were, [translation] “according to the records and after self-assessment of criminal investigators surveyed, almost exclusively written by the police”, not by the judges themselves.
With regard to the judges’ workloads there appears to have been no improvement in the last few years:
From a current study, the Roland Rechtsreport 2014, it emerges that nine out of ten judges and federal prosecutors surveyed think it is necessary for additional judges and federal prosecutors to be employed. 85% of those surveyed said they have too little time for their legal cases. A vast majority (72%) of judges and federal prosecutors were of the view that the framework of conditions for jurisdiction in Germany are currently deteriorating. The main reason for this was that there are too few personnel.
We think it is alarming that such conditions have clearly existed for many years, and that there have clearly been no efforts made since the study from 2003 that would have led to an actual improvement in the controls of surveillance processes. This clearly leads in practice to statistics like those from Berlin, which in our judgement are no longer fair in a constitutional state. If the possibilities for surveillance in Germany continue to develop while these deficiencies linger, this is a development that can not be beneficial to democracy.
As the government is currently planning to reintroduce data retention and will authorise these public positions to make attacks on fundamental rights that should be subject to controls by the instrument of the judicial reservation, we summon the German Minister for Justice, Heiko Maas, to stop the draft legislation. If the possibilities for surveillance in Germany continue to be expanded while the deficiencies outlined in our transparency report still exist and clearly every application for surveillance is approved, this is a development that can not be beneficial to democracy. Data retention would allow public positions to make attacks on fundamental rights that are supposed to be subject to control by the instrument of the judicial reservation. According to the numbers we have documented, the instrument has not fairly performed its intended control tasks for many years. Controls of the information process are also deficient. Often there is no requirement to keep statistics or reports. In the practice of requests for user information under § 113 TKG, chaotic circumstances rule: almost all requests that reach us are illegal. We fear that the introduction of the law would lead to a further increase in illegal requests.