Juristica International

1. Introduction

The reform of law-enforcement law has planned Estonia’s new Order Protection Act  *1 (OPA) mostly on the basis of the dogmatics of German law-enforcement law. Although the fundamental choice to use foreign patterns has sparked some controversy  *2 , the author of the present paper does not intend to reopen the discussion on the model of law-enforcement law most appropriate for Estonia; rather, the intent is to examine the developments that the new direction has brought about in a narrower branch of law-enforcement law—namely, danger-prevention law. The fact that danger-prevention law has been perhaps the most rapidly developing part of German law‑enforcement law in recent decades, its legal-theoretical nature and practical implementation having generated a great many problems in comparison to traditional danger‑countering law and having fundamentally altered the dogmatic form of law‑enforcement law, can be considered sufficient justification for focused analysis of this area.  *3 It is clear that if we are to accept the main features of the German model of law‑enforcement law, the Estonian legislator cannot ignore the changes that occur within this model over time. Hence, it is also relevant to analyse the nature of the danger-prevention part of Estonia’s new law-enforcement law.

2. T he theoretical bases of danger-prevention law and its difference from danger-aversion law

Historically, the most characteristic feature of the Germanic legal tradition’s law-enforcement law model has been the fact that it proceeds from the concept of danger  *4 as sufficient grounds for probable occurrence of damage.  *5 Sufficient grounds means (if we simplify a little) that upon assessing the situation an objective observer becomes convinced that damage is inevitable if the causal chain runs its course unchecked. The task of countering a danger links the legal rights protected by law-enforcement law to the options of restricting the fundamental rights of the originator of danger (the disturber or the person liable for public order), thereby providing a balance between the two clusters of good. Law enforcement, therefore, consists primarily in averting danger—i.e., countering a threat that endangers public security and order by restricting the rights of the disturber. This main dogmatic scheme based on the liberal rule‑of-law ideology, in place since the late nineteenth century, has also been accepted by the Estonian OPA, as well as the Police and Border Guard Act  *6 (PBGA).

However, contemporary law-enforcement law also has another aspect, in taking it upon itself to deal with danger potentials that have not yet become sufficiently probable and to eliminate them at the onset, before the development of any real danger. In other words, while the purpose of danger-aversion is to prevent any damage to the right of protection, danger‑prevention law is an attempt to prevent even any threat to the right of protection as embodying sufficient likelihood of the occurrence of damage. In both cases, this means law-enforcement-related prevention, but that prevention proceeds from somewhat different concerns and can be expressed in different legal-dogmatic form.  *7

Here we might want to limit ourselves to bringing out three features that a law-enforcement law concerned with infringement management, as contrasted with danger-aversion law, definitely needs to have, although these are by no means the only relevant features of danger‑prevention law.

Firstly, abandoning the concept of danger when assessing the probability of the occurrence of damage means that it must be replaced with some other probabilistic thresholds (lower than that for danger) that express the likelihood of damage. Risk  *8 is the concept most commonly used in legal theory in this connection.

Secondly, the lower probabilistic threshold of prevention also means that the target of a law‑enforcement measure that restricts fundamental rights needs a different definition. The connection between possible damage and its originator is more ambiguous in the case of prevention, and one cannot speak of a person liable for public order (the disturber) in the meaning of danger-averting law. The further into pre-danger territory the threshold of intervention is drawn, the larger the circle of the potential targets of measures becomes.  *9

Thirdly, when the threshold for intervention is shifted, the specifics of the application of a rule-of-law safeguard—the principle of proportionality—which has a central role in law enforcement law, also changes. This is mainly because the more ambiguous the possibility of damage, the more difficult it becomes to weigh the right that is to be protected against the one to be restricted.  *10

3. Danger prevention in Estonian law-enforcement law

Against the theoretical background described above, it is interesting, and indeed necessary, to analyse whether and how the issues of law-enforcement-related prevention law have been resolved in Estonian law-enforcement law.

However, obtaining an overview is made more difficult by the fact that the reform of Estonian law-enforcement law is not yet entirely complete, as of this article’s writing. The PBGA based on the new grounds of the concept of danger entered into force on 1.1.2010. The OPA  *11 proceeding from the same principles was approved by Parliament on 23.2.2011 but is not subject to enforcement yet, on account of delays in the preparation of the implementing act.  *12 A number of specific laws from earlier times regulating the law-enforcement sphere are enforced concurrently in individual areas of supervision. Hence, our law-enforcement law currently consists in two rather different aspects. There are grounds for believing that the OPA’s implementing act may amend the OPA itself; among other things, the parts pertaining to prevention law might still change before enforcement of the act begins.

3.1. Constitutional bases

The text of the Constitution of Estonia  *13 is too general in nature to allow for clear conclusions as to how a fitting proportion of danger-prevention law and danger-averting law should be understood in our legal order. The preambula, the principle of the rule of law, and the fundamental rights to protection (in particular, the general fundamental right to protection set forth in §13 (1) of the Constitution) entail the obligation of the state to defend internal security  *14 , yet much room is reserved for the legislator with respect to how exactly this obligation is to be met.

Although the Constitution mentions the concept of danger  *15 on several occasions (e.g., in §§20 (5) and 129 (1)) and theoretical literature on the Estonian Constitution too has opined that the identification of an infringement of the fundamental right to protection is related to a (specific) danger as a sufficient likelihood of damage inflicted on the right to protection  *16 , the author does not think it justified to conclude that the concept of danger is the only threshold for intervention that can be used in building of prevention-based law-enforcement law.

The concept of danger as an intervention threshold incorporates a point of balance between the goods to be restricted and those to be protected. As a universal and quite clear criterion, it serves its purpose well in so-called typical situations of law enforcement, wherein there is enough information on the causal relationships (or at least the acquisition of this information is objectively possible) and in which the goods to be protected and the rights to be restricted can be weighed against each other and are not characterised by a vast disproportion of value. However, upon accumulation of various circumstances—above all, the complexity and obscurity of causal relationships, the possibility of the shift from the pre-danger-probability phase to the damage-occurrence phase taking place within a very short time, and the potential that the goods to be protected may strongly outweigh the restriction to fundamental rights resulting from the application of measures—gathering information on the situation or even interfering with the causal chain directly in the pre-danger phase might prove to be a serious alternative to the protection of the relevant types of good.

The duty to ensure a country’s internal peace, especially within the minimal scope of protecting the fundamental rights, is not set in stone; in contrast, it requires assessment in light of the changing legal and social circumstances.  *17 Without a doubt, social relations have changed significantly since the time when classical law-enforcement law emerged. Rapid advances of technology, especially in the IT field, and the waning importance of international borders can be considered the major developments here. These new possibilities shape new behaviour patterns and create new opportunities for organised-crime organisations, terrorists, and other entities to commit serious criminal offences; make it easier for potential dangers to materialise  *18 and for people to prepare for crimes and conceal a criminal infrastructure; and complicate the causal relationships leading to the occurrence of damage. The large-scale terrorist attacks of the early 2000s are the most clearly expressed and probably the most frequently discussed aspect of the new behaviour patterns, one that has significantly hastened the development of prevention law in many states.

From this perspective, law-enforcement-related danger-prevention law is a general legal response to the development of a welfare society into a risk society  *19 , where the question of the assessment and distribution of risks becomes more important than the distribution of material values. Further, the preconditions for these developments are not absent from Estonian society.

All this considered, interference within pre-danger territory (including the restriction of fundamental rights) might turn out to be not only constitutional but, on limited occasions, even obligatory when the fundamental rights to protection are taken into account. The relevance of preventive defence is testified to also by the indisputable importance of the prevention principle as a constitutional principle of environmental law in addition to the principle of avoidance, which is equivalent to the principle of averting danger in law enforcement.  *20

Of course, the application of preventive measures (so-called preventive state logic), especially when done to a wider extent, also involves some serious dangers for a state based on the rule of law.  *21 The aim in a prevention law infringing on the fundamental rights is to interfere with the fundamental rights before the damage potential of the behaviour of the target of the measure is fully revealed; hence, the activity of the target of the measure that is carried out in pre-danger territory is still constitutionally protected  *22 and interference with it requires especially strong justification. It must be shown why danger-based law enforcement is not effective in the particular case in question. Furthermore, prevention law, at least a large part of it, requires mechanisms compensating for its rule-of-law deficiencies—above all, rules for application of measures that meet stricter requirements of legal clarity and definition, along with procedural safeguards, in such forms as permission from higher-ranking agencies or set procedures for informing the person involved. As is evident from the practice of the European Court of Human Rights (ECtHR) just as much as from that of the Estonian Supreme Court, a certain core area of rights to freedom that probably cannot be restricted by prevention-law measures also exists in our legal order.  *23

3.2. Law-enforcement-related prevention law in Estonia before the reform of law‑enforcement law

Before the reform, Estonian law-enforcement law did not use the concept of danger as a key interference threshold. Although preventive measures of administrative law were not absent per se from the range of measures employed by the police and other law-enforcement agencies, the grounds for their application had been formulated according to inconsistent legal logic, usually either very casuistically or, on the contrary, in a way that allowed for extensive interference.

Hence, a clear distinction between use of the probabilistic levels of danger and danger prevention as the state’s two preventive activities could not have existed in such law‑enforcement law either.  *24 Offence-prevention (crime prevention)—which lacked a legal definition stemming from the legislation—was understood in a broad sense; it could mean the prevention of offences in the longer term but also the prevention (blocking) of an already existing but not yet punishable offence. More than any specific police measures, however, prevention was taken to include, in the author’s opinion, various social and educational measures, as well as those intended to deal with the consequences of offences—i.e., strategic work on the causes of crime.  *25

As an area wherein the duty to prevent dangers (in the meaning of the OPA) was most likely accomplished at least in part through employing of measures that restricted fundamental rights even in Estonia’s earlier law-enforcement law, surveillance of a law-enforcement-related nature can first be pointed out. The Surveillance Act  *26 provided for an option of applying various data-collection measures that restricted fundamental rights in order to prevent and hinder crime. Although some have expressed the opinion that this formulation should be understood more narrowly than the prevention of offences in its broader (criminal preventive) sense  *27 , there is no reason to claim that prevention in the sense of surveillance measures only applied to crime threat in the sense of a specific danger. Another area of action that can be classified as danger-prevention activity has been inspection-type state supervision  *28 , or (mostly random) surveillance in various potentially dangerous spheres of social life that is independent of a suspicion of danger (see Subsection 3.3).

No single tendency that would point to increasing relevance of a prevention right and the extension of prevention measures to lower-probability areas can be pointed to in Estonia’s earlier law-enforcement law. Rather, the dubiousness—from the perspective of the principle of legal definition—of law-enforcement measures that have been formulated in a disproportionately broad fashion and thus enable interference also in pre-danger territory has become understood over time.

3.3. Danger-prevention law and the reform of law-enforcement law

The reform of law-enforcement law raised the question of the relationship between repression and prevention in state activity in general with regard to Estonia’s legal order. Preventive administration began to be discussed as the ideological basis for the new law-enforcement law.  *29 As a side effect of the introduction of the danger criterion, a clear distinction between danger-aversion and the pre-danger territory became evident. Judging the earlier law by the new criteria, one finds that a good proportion of the law-enforcement powers hitherto exercised had set the interference threshold lower than sufficient probability of the occurrence of damage for an objective observer.

Both the PBGA and the OPA distinguish clearly between danger-prevention and averting danger already on the level of describing the tasks of law-enforcement agencies.  *30

Pursuant to §2 (1) of the OPA, law enforcement consists in the prevention of a danger to the public order, the identification of danger in a case of suspicion of danger, countering of danger, and elimination of disturbance to public order. A (specific) danger, in the sense of §5 (2) of the OPA is a situation wherein, from an objective assessment of the circumstances that have arisen, it can be considered sufficiently probable that a breach of public order is about to take place, whereas danger prevention, pursuant to §5 (7), is defined as the collection, exchange, and analysis of information and the planning and implementation of activities to counter any future dangers to public order, including the prevention of offences. The fact that suspicion of a danger (§5 (6)) has been defined separately lends itself to the assumption that the prevention of a danger was something clearly prior to the suspicion of a particular danger for the authors of the draft OPA.

In the prevention of a danger, law-enforcement agencies are bound by the same principles (proportionality, effectiveness, protection of rights, and co-operation among agencies) as they are in the danger-aversion phase, pursuant to the OPA.

The importance of the prevention of offences as an area of danger prevention is demonstrated in the OPA by the fact that the act contains a separate chapter on offence prevention (Chapter 2, from §17 onward). At the same time, it should be noted that the definition of the prevention of offences as given in §17 does not directly link prevention to the pre‑danger territory; rather, it speaks of the prevention of offences and of ensuring public order in a broader sense. This raises the question of whether it refers to pre-danger activity only or to the prevention of offences at any stage before the occurrence of the action. In general, it must be said that the chapter on the prevention of offences does not add anything substantial to the working logic of the OPA and, in fact, seems alien to the act. For example, the division of preventive measures into social ones, measures directed at circumstances, and measures designed to eliminate the consequences (§18) seems more appropriate for legal theory and entirely useless with regard to regulation of legal relationships, at least in this act. In particular, it adds nothing to the state supervision measures described in the later chapters. If anything, it creates confusion by extending the concept of a law-enforcement measure considerably in comparison to how it has been used elsewhere in the act. Other sections too (on the duties of the state in prevention of offences, offence-prevention council, financing of prevention of offences, etc.) regulate the administrative planning process of prevention without addressing any specific activities of a direct offence-preventing effect.

The danger-prevention part of the new law-enforcement law is much more complex and multi-layered with regard to law-enforcement powers (that is, law-enforcement measures).

A distinction must be drawn between preventive measures that restrict fundamental rights and those that do not. The application of non-restrictive preventive measures may be, as other law‑enforcement measures can be, directly based on the prevention task. A large amount of the preventive activities that consist in the collection of data from publicly available sources can be accomplished by means of non-infringing measures also in our legal order. However, the fact that the grouping of a measure with those infringing or not infringing on the fundamental rights can prove controversial may cause problems in practice.

The concept of a danger-prevention measure that infringes on the fundamental rights is ambiguous, as it can be given several meanings.  *31 Preventive measures probably should not be taken to include only the law-enforcement measures explicitly defined as measures for the prevention of dangers; rather, the content and purpose of any given measure should be taken into account. From the standpoint of logic, all measures whose application is not based on suspicion of a specific danger, a specific danger, or damage that has already occurred are applied to the ‘preliminary territory’ of danger. However, at least one other criterion must be added to this—namely, the dogmatics of the German law-enforcement law take the pre‑danger measures to include inevitably also those that lack a clearly definable range of targets (i.e., measures whose target is not clearly a person liable for preservation of public order). This holds even in cases wherein the presence of a specific danger is deemed necessary grounds for the application of the measure.  *32

On the basis of such a material definition of the preventive measures, at least three groups of preventive measures that admit the infringement of fundamental rights can be distinguished within the structure of the OPA.

Firstly, §24 of the OPA includes, expressis verbis, measures to be applied by a law‑enforcement agency for prevention of danger.

This is an area that is referred to in the dogmatics of Estonian law-enforcement law  *33 as inspection-type state supervision and is typically not addressed by German general police laws, in which its regulation is instead left to the specific laws of the law-enforcement field.  *34 Inspection-type supervision is exerted over areas of social life that involve potential danger (and that therefore usually require prior permission or at least notification) and hence require preventive control in addition to averting of danger. Classic examples include supervision of economic activity of various kinds, traffic supervision, health protection, supervision of migration, supervision of adherence to the Weapons Act, etc.

The act enumerates certain law-enforcement measures directed at preventive gathering of information (entering premises, carrying out questioning, requesting documents, etc.), allowing the law-enforcement agencies to apply these ‘for prevention of danger’. As a further condition, the right to employ the measure must also be provided for by a specific law regulating supervision of the respective field (e.g., the Weapons Act, the Technical Supervision Act, or the Traffic Act).

One differentiating feature of such preventive measures, as compared to measures for countering a danger, lies in the fact that their application does not allow for the possibility of applying direct coercion of the target of the measure (see §24 (3) of the OPA).

The second type of measures—no longer explicitly called preventive but essentially of a preventive nature if one judges by the criteria mentioned above—is rooted in §25 of the OPA. Pursuant to §25 (1), certain law-enforcement measures may also be applied in cases of ‘a danger or for the ascertainment of a serious danger’ with regard to a person there is no reason to deem a person liable for public order. Although it is not clear from the text of the act or the explanatory memorandum, the legislator probably had in mind the scenario of conducting a police operation motivated by a specific danger or suspicion of a danger while the originator of the threat remains unknown or has not been captured. The application of measures is possible only with the permission of the relevant minister (or, in urgent cases, with that of the head of a law-enforcement agency) and within specified territorial and time limits.

Thirdly, upon closer analysis of the rest of the OPA measures, they can be seen to include those that may by their nature be applied with regard to persons whom there is no reason to deem liable for public order, or whose grounds for application do not include the criterion of a specific danger, suspicion of a danger, or any corresponding level of probability. These are, above all, the measures listed in §23 (2) of the OPA—questioning, so-called video surveillance, and prohibition on stay, along with some cases of security-check application, such as the application of a security check to everybody entering the premises or territory of a public authority, regardless of a suspicion of a danger. Subsection 23 (2) of the OPA explicitly allows the application of such measures with respect to persons not liable for public order. However, the list given in the article is not exhaustive—for example, the checking of a movable for purposes of ‘ensuring the safety of a safeguarded person or object’ provided for by §49 (1) 6, which is not included in the list, must in its essence be classified under the same category.

The present overview would be incomplete if we failed to mention that, in addition to these three groups of preventive measures included in the OPA, some essentially preventive law enforcement measures will be regulated outside the OPA also in the future. One important subset of these measures includes law-enforcement-directed secretive data-collection measures or so-called law-enforcement-related surveillance, which is currently regulated in connection with criminal-procedure-related surveillance pursuant to a legal-political decision by the legislator  *35 , although the legal theory situates these within the purview of law‑enforcement law.  *36

If we are to draw conclusions from the grouping of preventive measures that is sketched out above, we must first acknowledge the heterogeneity of the preventive measures, like that of the preventive tasks, just as much as the lack of consistent logic of legal regulation. Some of the preventive measures could be called ‘horizontal’, as their danger-preventive nature is evident from the formulation of the measure itself and they are mixed up with measures for countering a danger, whereas other preventive measures might be called ‘vertical’—i.e., their preventive quality is due to the supplementation of threat-countering measures with the additional conditions set forth in §§24 and 25 of the general part of the OPA. Vertical preventive measures reduce the OPA’s clarity of arrangement. The group of preventive measures provided for by §25 is, in the opinion of this author, especially unwieldy with regard to both formulation and regulating logic. The possibility of conducting so-called police operations against an undefined range of persons is certainly necessary in some situations, but the same purpose could be achieved by means of other legal-technical solutions (comparison with the police checkpoint regulation used in the German police law might be of help, for instance). However, the fact that the legislator has sensed a need for a further administrative check in cases of police operations, as a mechanism balancing the infringement of fundamental rights occurring on probabilistically uncertain grounds, can nevertheless be considered positive. Furthermore, an attempt has been made to specify the circumstances of acceptable infringement with great accuracy.

With §25 of the OPA too, it must be acknowledged that preventive inspection-type supervision in some areas is constitutional in principle and also definitely necessary; however, problems arise from the large number and variety of areas of application of inspection-type surveillance. The aspiration with the OPA is to constitute a general part for all of them, but this leads to a high level of abstraction of the regulation and to the result that several acts (the OPA and the specific law applying to the particular area of inspection-type supervision in question) must be read in combination if one is to understand the regulation fully.

The Ministry of Justice, who prepared the OPA, has acknowledged that defining the measures of inspection-type supervision only by reference to the general goal of danger prevention may not conform to the principle of legal determination and, therefore, has proposed its amendment such that inspection-type supervision would be carried out on the basis of periodically compiled danger projections that specify the conditions for conducting of the supervision.  *37

The view taken in the act with regard to inspection-type supervision—that the forced execution of such measures in the form of direct coercion is ruled out—is also not without its problems. This solution is based on the legislator’s assessment that probabilistic grounds that have not yet condensed into a specific danger cannot outweigh the concern about infringement of fundamental rights resulting from direct coercion as a means of coercion that very strongly restricts the fundamental rights. On the other hand, however, this is cast into doubt by concerns over the effectiveness of law enforcement  *38 and also by the question of why the legislator has not followed the same lines in cases of other groups of preventive  measures.

4. Conclusions

In view of the foregoing considerations, it would be an exaggeration to say that a strong tendency to emphasise prevention over the averting of a danger can be observed in Estonian law-enforcement law, which is still only adapting to the concept of danger as an intervention threshold. The bulk of the weight of the law-enforcement task will rest with danger-averting law also in the future. However, danger-prevention law that conforms to the characteristics brought out above (see Section 2) also has an undisputed role (mainly as law‑enforcement‑related information law) in our law-enforcement law. At present, law‑enforcement-related prevention law has been only weakly worked through in Estonia, both theoretically and with regard to legislative practice.  *39 However, the developments of the globalising world and the fact that we belong to Europe’s unified security area are likely to force our legal scientists to pay more attention to the issues of prevention law. In the process of adopting the traditions of Germanic-family law-enforcement law as danger‑averting law, the prevention law being developed in Germany could also provide a source of comparison and ideas for the further shaping of Estonian law-enforcement law.