In the last two decades a lot has been said about the relationship between liberty and security, liberty and secrecy, and of course, about the role of the ‘deep state’ in contributing to security by means of secrecy among other. The intelligence community has been both praised and bashed for data collection successes and failures and has found itself in the spotlight of the public debate on legitimate means and boundaries when tending to its business.

A number of developments have followed those clashes between the public and one of the oldest trades. As a response to systematic foreign surveillance measures, for instance, the EU enacted a comprehensive data protection legislation. At the same time, however, many of those EU jurisdictions implemented comprehensive intelligence reforms and codified some of those notoriously complained about and previously unregulated intelligence practices. Thus, little to nothing changed in the good old liberty vs. security or liberty vs. secrecy paradigm.

The tension between what intelligence agencies do and how the public reacts to it is, however, built in from the very beginning and can be also explained by looking into the underlying logics of secrecy that govern those counterparts. Secrecy in intelligence is endemic, institutionally embedded and part of the raison d’etre of the organisations. For the public, however, secrecy is bad news – if governmental institutions keep secrets, it is understood that this is mandated by something that cannot be justified in public and would thus lack legitimacy. Evidently, the here underlying secrecy rationales are also quite different – the secrecy of the intelligence agencies is functional; the one the public is dealing with is relational.

These differences are well explained by Horn’s approach in tracing the historical evolution of secrecy. She designates state secrets mysterium, arcanum and secretum according to their chronological but also political placement. Arcanum – originating from arca, meaning treasury or chest –, is that hidden or locked away piece of information. Framed in this manner, secrets do not have to be something unknown per se. An arcanum is already something that is simply left unsaid or locked under the lid of silence. It is ultimately a code of conduct, a government technique of concealment that exercises secrecy to sustain and expand power, and thus a particular form of political management.

Secretum in its turn comes from secernere – to exclude, to secret – and unlike arcanum is not that much about concealing information, but about the effects of that concealment. As it creates a relation between those who know and those who do not, a secretum’s essence lies in its capacity to influence social inclusion and exclusion. Secrecy understood in this way is also close to Simmel’s sociological approach, according to which it is relational and external. Consequently, a secret’s existence is rather presumed. It excludes outsiders, allowing the insiders to create hidden communities alongside apparent ones, and makes for a correspondingly strong feeling of possession – those who know a secret enjoy a position of exception over those who do not. A secret thus becomes an adornment, a secret jewel, empowering its owner both through possession and potential disclosure.

Thus, when it comes to the liberty vs. security debate the two secrecy logics exist in parallel but also occasionally clash and collide. The intelligence trade is very much guided by an arcanum attitude, according to which secrecy is simply good housekeeping. Everyone outside of the inner circle that lacks knowledge on intelligence goals, practices and outcomes, however, falls prey to the power of secretum – they are in conscious lack of possession of that information, while simultaneously knowing who is in control of it and speculating about its contents. Accordingly, the public perceives the entire realm of the secret services as a necessary evil. An evil that at times seems needed (especially considering the fact that ‘national security’ as a justification for intelligence behavior has a habit of spreading), yet highly explosive and quickly condemned when those who determine what is and should remain secret misjudge the liberty-security balance. Intelligence practices have thus become the antonym of law and social contract, and there is oftentimes a question mark hanging above data gathering tools, excluding them from the space of legality and morality.

Further, although secrecy does prevail in intelligence establishments, contemporary examples teach us that the ‘deep state’ is facing unique difficulties to keep its secrets in the information age. As revelations and leaks become the norm and cyber exploitation tools become public, intelligence departments and practices are visible more than ever before. However, while each revelation adds to understanding the intelligence conundrum a little bit better, it has the opposite effect on secrecy – it extends the public’s awareness of secrecy at work, but does not make it more comprehendible. In this context, and much like the intelligence agencies that keep pushing both technological and legal boundaries, secrecy is something constantly in need of legitimization yet never really legitimate. No wonder the liberty vs. security debate feels like playing with Rubik’s cube.