1. A recent conflict of competences between the President of the Council of Ministers (Prime Minister) and the Judicial Authorities in Bologna - on an instance of State secrecy upheld during criminal proceedings for an anti-terrorist intelligence operation - has raised in practical terms the problem of the necessity, feasibility and opportunity to provide intelligence operators with operational guarantees and of their possible interference with secrecy.
I am not going to dwell upon that incident (which entailed three conflicts of competences before the Constitutional Court and the claim of unconstitutionality for some provisions on State secrecy) as I was involved institutionally, being at the time Minister of Justice.
The State Lawyer will tomorrow deal with it in detail since he is at present directly involved.
I mentioned it (whatever the final outcome may be) because it has contributed to the debate - at present developing at the political and doctrinal levels - on whether it is necessary and feasible to provide for operational guarantees for intelligence and security operators in case they violate the criminal code during their intelligence activities.
At present, operational guarantees are not provided for by law and the conflict of competences mentioned above is a demonstration of the attempts and the difficulties of giving the problem an indirect solution through a procedural guarantee with the upholding of State secrecy before the judicial authority. I wish to underline how doubtful I am about this procedural approach (as it refers only to the impossibility for the judge to gain knowledge of certain matters) to a problem that is instead substantive. The solution should be founded on the content and not only on the procedure by which a court case cannot proceed because the judge is prevented from gaining knowledge of certain matters.

2. I refer you to the detailed preliminary contribution to this Seminar for an illustration of the general legal framework.
Here I would like to touch only on the main issue (as it emerged both from practice and from theories) which underlines that of the operational guarantees: the possibility that while carrying out intelligence gathering activities an intelligence operator either directly or through a third party has to violate the criminal law, in particular privacy (i.e. telecommunications, interpersonal communications, privacy of premises and of documents). The need for such a violation is closely connected to another fundamental intelligence need, that is Secrecy.
In our legal system for both these profiles (the need to violate criminal law in order to carry out intelligence activities in an efficient and secret way; and the need for Secrecy in itself) the key issue is the relation with the judicial authorities.
In this respect I will here mention the relevant constitutional principles:
- the judge is subject only to the law;
- the prohibition to establish special courts;
- the obligation to start criminal proceedings in presence of a notitia criminis;
- the principle of the "contraddittorio" (audita altera parte);
- the right to defence in court;
- the principle of the "fair trial" which is a recent addition and will certainly enhance all the others.
All these principles with their applicative provisions (i.e. the publicity of court debates and records, also with the exceptions dictated by national security requirements) - deeply influence the issues of Secrecy and of operational guarantees.
I feel I should also mention here the present and past widespread mistrust towards Secrecy in relation to possible abuses, and towards the intelligence activity in general and its results. The inefficiency of the intelligence system has also been for some time the object of sharp criticism.
In such a context, the issue of operational guarantees has recently been widely debated, there have been doctrinal contributions and also reform proposals on the subject both from Parliament and from the Government.
It is certainly positive that the issue has finally drawn wide attention, more so if we consider the lively debate currently under way on the efficiency of the Intelligence Services and on the need for them to be reformed.
It is a fact that the problem of ensuring internal and external security is acquiring more and more relevance in relation to the new emerging threats.
Within the European Union the creation of a common area of freedom, security and justice was set as a priority by the Amsterdam treaty. It is considered an essential element for the future of the European nation, for the exercise of the fundamental rights, and both of these aspects are closely interconnected with the need to ensure European external security. This requires co-ordination and close co-operation, and an effort towards the harmonisation of legislation by the Services of the Union member States.
In Italy, a debate is currently under way on a reform project for the intelligence services aimed at increasing legality, efficiency and accountability. In this respect the 1977 law (currently in force) is due to be revised and the issue of operational guarantees, totally absent from that law, will have to be handled with the utmost transparency so that those concerned can take their responsibilities and make their political decisions within an overall reform project.

3. The issue of operational guarantees - as I said - is conceptually distinguished but closely interconnected with that of secrecy, in particular when an intelligence project having as its objective the protection of national security may require acts violating privacy or similar. Such acts could entail violation of the criminal law or an infringement of individual rights, most of which are strongly protected by the Constitution (exclusive jurisdiction on matters referring to domicile, mail and communications).
The problem is to select the right alternative. Either the problem is "ignored" and the intelligence activity is carried out in compliance with the rules protecting privacy and, in particular, respecting the provisions for the involvement of the judicial authorities. This alternative is essentially a theoretical one, as it would entail the end of the intelligence activity, unless specific preventive measures are enforced by the judicial authority. Still this option appears to be in contrast with the Judiciary's constitutional characteristics as mentioned in para. 2, and also with the Constitutional Court's indication (par. 4) defining national security as a primary constitutional value.
The second alternative finds in the nature of the intelligence activity a justification for possible violations of rules and rights according to an evaluation of precedence or at least of equivalence of the interests granted by the Constitution.
This second option entails firstly a precise identification of the interests involved and then a careful control on the warranty procedures to be adopted and finally the identification of the body who should enforce those procedures.
It is also important to identify which values may and which ones may not be infringed upon during intelligence activity (with a specific indication for the first ones or with the explicit exclusion of the second ones). Finally the parameters of necessity, scope and extent for such acts should be fixed with utmost rigour. All this should be aimed at ensuring the legitimacy of the intelligence activity and its adherence to the institutional purposes, at preventing and severely punishing abuses and at ensuring careful scrutiny.
4. In the Constitution a clear indication in this sense can not be found (with the exception of art. 126 which mentions national security as a possible cause for dissolution of a regional council; the protection of privacy in a wide sense is instead well provided for within the Constitution). The first step therefore should be to verify if in the Constitution unequivocable indications for a possible infringement on the right to privacy, justified on national security grounds, exist. Future legislative solutions should be founded only on such constitutional indications.
The starting point should undoubtedly be the Constitutional Court judgment n. 86/77 on secrecy. This judgment refers to, and further develops, some indications already present in judgment n. 82/1976.
In this judgment the starting point is the interest of the State as community to preserve its territorial integrity and independence, in other words its very existence. This is considered a "supreme constitutional value". To this value the Court adds national security plus the defence of the Motherland and links them to other constitutional principles: national independence, national unity and the indivisibility of the national territory, and the protection of the democratic system.
The Court was of the opinion that these institutional interests of the State as community are "supreme and fundamental": therefore they legitimate secrecy on security grounds, in a reasonable means-end relation. Their enforcement is an institutional duty that the highest State authorities responsible for the survival and the integrity of the State and for democracy have to fulfil. The Court in even clearer terms states that national security requirements must have top priority and must come before all other public functions. Further, national security is "a primary and essential interest of the community, it takes precedence over any other as it refers to the very existence of the State."
Finally, on the relations with the Judicial Authorities as to secrecy, the Court confirms that the evaluation of the appropriate and necessary means to guarantee national security is of an exclusively political nature and does not pertain to the judge. All this taking for granted the unacceptability of uncontrolled or uncontrollable and irresponsible powers and the general and institutional accountability of the Government before Parliament.

5. The path indicated by the Constitutional Court in relation to secrecy and its constitutional legitimacy, to the relations between the political and judicial powers and to the priority of the constitutional value of national security should certainly be followed as an authoritative, clear and unequivocable guideline.
The issue of operational guarantees has been up till now only dealt with indirectly, that is through the upholding of State secrecy within court proceedings by the political power, thus preventing the Judicial Authority from gaining knowledge of certain matters with the ensuing impossibility to continue a criminal action where the knowledge of what is covered by secrecy is essential to the case.
In simple terms, the judicial authority can not proceed with a case on the grounds of its impossibility to gain knowledge. The only alternative for the judicial authority is to gather other elements of evidence on the notitia criminis, that is on the violation of the criminal code committed during intelligence activities aliunde that is outside the cover of State secrecy. This is a consequence of the lack of legal provisions granting substantive immunity in cases of violations of the criminal code required by the secret nature of intelligence activities. The Bologna events and the various conflicts of competences confirm the inadequacy of this procedure, apart from the doubts over the choice of approaching a problem that is substantive through procedural measures.
The Constitutional Court in relation to secrecy has provided significant guidelines for the drafting of a specific extenuating circumstance:
- the democratic order and its defence are absolute priorities;
- so is national security to this end;
- the pursuing of this objective and all the connected functions take precedence over all other State functions;
- political discretionary power, responsibility and scrutiny over the enforcement are entrusted to the highest State authorities;
- the need for parliamentary oversight;
- constitutional legitimacy of the impossibility for the Judicial Authorities to gain knowledge of certain matters and ensuing impossibility to continue the criminal proceedings.
Using these indications as reference it is possible to discipline the conflict between constitutionally significant interests, not only as it has happened between the intelligence activity and the possibility for the judicial Authorities to know; but also when such conflict emerges between the intelligence activity's modus operandi aimed at preserving secrecy and other constitutionally significant interests.
Further confirmation comes from the limitations to fundamental rights on the grounds of national security provided for in the UN 1976 international agreement on the political and civil rights and, before that, in the 1950 European Convention on human rights.
Sec. 8 of the European Convention provides for the protection of private and family life, of the domicile and the mail and still it poses national security in a democratic society as an extenuating circumstance to be provided for in the law. This is particularly significant because the fundamental rights provided for in national Constitutions and in the Convention are the general principles at the basis of the European Union's legislation; and also because this EU legislation has as its main objective the creation of a common European area of freedom, security and justice.

6. The first aspect to consider in order to define an operational guarantee is its possible framing within an extenuating circumstance; then a source able to evaluate its applicability will have to be identified (Parliament, Government or Judiciary); thirdly its content and application procedures will have to be established. Considering the general lines of our criminal system, the choice of a justificatory cause appears preferable when compared to the option of granting immunity on the grounds of lack of the mens rea element (still there have been cases of acquittals due to the lack of mens rea; it is nevertheless difficult to assume that the operator was not aware of the nature of his conduct).
A legal excuse (general or special) applies to actions which are normally punishable, but that become lawful because authorised or even imposed by a legal provision. The ensuing judgment should be referred to the whole legal framework so that the action, object of the legal excuse, can not be subject to any penalty within the criminal sphere nor outside it.
The law can not contradict itself, allowing an act and prohibiting it at the same time. The values and interests underlying the provisions allowing or prohibiting a conduct should be carefully evaluated and compared, priorities must be identified before excluding the illegality of a conduct (which would otherwise be considered a crime).
In general terms the doctrine contemplates "non legally provided" justificatory causes (besides those specifically provided for in the criminal code or in special legislation) usually stemming from the analogic interpretation of certain rules. In the topic under examination here, nevertheless, the interests at stake (i.e. privacy) are so significant (so much so that they deserve a specific constitutional mention) that it appears advisable and justified to respect the principle of legality with the utmost rigour.

7. Logic and experience indicate that during intelligence gathering activities in order to maintain them secret it may be necessary - in absence of other possible ways - to intercept communications or to install listening devices or to view documents in ways which imply a violation of the law, even of the criminal provisions protecting privacy and property. In this respect the constitutional principles on security, directly drawn from the interpretation of the Constitution given by the Constitutional Court with reference to secrecy (see para.4) should be referred to as they are valid also for the intelligence activity aimed at the protection of national security.
In simple terms there is an extremely close connection between secrecy and intelligence. Secrecy obviously implies that information has been obtained; but this acquisition can only have taken place in secrecy.
It appears therefore feasible to provide for an operational guarantee through a special extenuating circumstance following the priority scale of the constitutional interests connected with national security. The recent government reform bill on the Intelligence and Security services and the document drafted by the Committee established ad hoc by the previous government, both follow this path.
Security is a supreme national interest. The State must identify acts, facts and information needed to safeguard security, that is national unity and integrity, the defence of the democratic order, in a close ratio between means and ends. Also a proportion (equivalence if not prevalence) must exist between the interests involved: privacy on the one hand, and national security on the other.
This is the reason why the stress should not be placed on the single illegal operation needing authorisation; rather it should fall on the overall intelligence project to be implemented in order to achieve the institutional purpose of protecting the supreme constitutional value of national security. The means necessary to achieve such purpose must be also focused upon; among them the possible commission of unlawful acts which must be unavoidable and adequately proportionate to the interests pursued (the gathering of information; the violation of privacy in a wide sense).
The strict necessity, the functionality and the adequate proportion between the illegal conduct and the gathering of information exclusively for security purposes, should provide reasonable protection against possible abuses or distortions. Illegal conducts not strictly related to the information gathering activity within the limits indicated above would obviously be excluded from the extenuating circumstance. The same holds good for violations of interests and values other than the ones expressly indicated within the extenuating provision.

8. In order to determine the content of the extenuating provision for the operational guarantees, the proper and exclusive purposes of the intelligence activity should be clearly identified – and this also to comply with the legal requirements of exactness and rigour of definition.
All we have said about comparing the interests at stake should primarily found on a strict compliance with the intelligence activity's institutional purposes. This is one of the objectives of the Government reform bill and of the Committee's work.
The intelligence activity should therefore be carefully defined in its scope and limits distinguishing it from the investigation activity carried out by preventive and judicial police forces. These are differently regulated and entertain different relations with the Judicial Authority. It is obvious that the intelligence activity could never intermingle with police investigation activities; the two nevertheless need to be linked in a transparent and documented way, whenever information held by the intelligence bodies may refer to the competence domain of the Police forces.
Further, it is necessary to establish carefully the scope, the purposes, the procedures and the responsibility for each of the intelligence projects which might require the application of the extenuating circumstance; also identifying the political and the administrative-operational authorities to be entrusted with the responsibility for the relative authorisation. This authorisation entails a violation of the criminal law, and should therefore pertain to the highest political authorities.
The extenuating circumstance should be defined considering both the values and interests at stake, and should be founded on:
- the impossibility of reaching the goal in any other way but the one requiring the violation of the privacy;
- the absolute necessity of the unlawful conduct ;
- the adjustment of the conduct to the institutional purpose which is to be achieved.
In order to prevent abuses effective and serious penalties should be provided for (in relation to the planning of the project, to its enforcement and the use of its results); also the documents reporting the activity should be kept so as to guarantee secrecy and to allow later checks.
Finally, and above all, the illegal conduct which might be authorised should be precisely identified and described, together with the values that can not in any case be violated (life, personal freedom, physical integrity, public safety and well being, according to a list drawn from the Government reform bill and from the document of the ad hoc Committee).

9. Various reform bills contain proposals which raise doubts as to their feasibility from the procedural point of view. According to them the judicial authorities (that is specifically charged judicial bodies i.e. the General Prosecutor at the Cassazione) would be responsible for a preventive evaluation on the applicability of the extenuanting circumstance. In other terms the Judiciary would grant a sort of authorisation to the intelligence project.
This entails the risk of mixing the authorising stage with the scrutiny phase, and of confusing the judicial function with the political-administrative one. This would contrast with the unequivocable indications supplied by the Constitutional Court (see para. 4). We are not talking here of a warrant tout court to commit crimes, rather of an authorisation for an intelligence project which might entail the possibility-necessity of illegal behaviour.
Secondly, an a priori authorisation by the judicial authorities might interfere with the principles establishing the independence of the judge and the obligation to start criminal proceedings in presence of a notitia criminis. Local judicial authorities with jurisdiction over a specific area, may well decide to ignore this a priori "judicial authorisation".
Entrusting another State authority with the responsibility of such an authorisation might raise constitutionality problems with reference to sections 14 and 15 of the Constitution, which define the principles of the exclusive jurisdiction and of the exclusive regulating power of the law in specific domains. Still, in this particular instance it is not just a question of devising a legal provision allowing the political powers to authorise an infringement on privacy without informing the judicial authorities; but rather of finding a justificatory cause - referred to the supreme constitutional value of national security - for a conduct that is illegal also because it contrasts with the exclusivity of the jurisdictional function.
A last remark touches upon the evaluation on the subsistence of the extenuating circumstance: should it be up to the highest security authority to propose it and confirm it, upon application by the judicial authority and evaluation by a Committee of Garanti (as in the Committee's draft document). Or should the judicial authority give its approval anyway, leaving open the option of upholding State secrecy in case of disagreement (as in the Government reform bill). In consideration both of the recent disputes between the judicial and political authorities and of the possible confusion between the procedural phase of secrecy and the substantive one of the operational guarantee, the first option appears preferable. It binds the judicial authority to take note of the extenuating circumstances when this is set forth by the Service's technical authority and then confirmed by the highest political authority. In case of a disagreement the judicial authority may raise a conflict of competence case before the Constitutional Court.
10. To sum up we can state that the extremely delicate issue of operational guarantees for intelligence operators can be disciplined through brave legislative initiatives. They should be devised within an overall intelligence reform of the tasks, the regulations and the necessary overseeing system.
Going from the procedural perspective (where the protection of the Services' activity is achieved imposing restrictions on the acquisition of evidence) to the substantive one (in which the correspondence between the conduct and the interests to protect is the origin for the exclusion of illegality) should not be thought of as creating a zona franca where on grounds of national security legislation renounces its function of fixing the borders between legal and illegal acts or, even worse, delegates one of the parties to do it.
The difficulty to define with absolute precision the concept of "national security" and the traditional inclusion of the Services' activities within the arcana imperii raise in many excessive diffidence. This however at the same time shows that operational guarantees must be set in a clear framework, the responsiblities and the oversight system must be clearly defined so as to prevent abuses.
There should be no confusion between national security requirements and the concept of the raison d'Etat. In history many arbitrary acts aimed at protecting the State in abstract terms and at preserving the status quo have been justified with the raison d'Etat. In a democratic order is not the State in need of protection, but rather the institutions entrusted with the defence of the constitutional values.
In this respect, the Government reform bill in listing the functions of the two intelligence agencies refers to "the protection of national independence and integrity against any external danger, threat or aggression" for one; and to "the protection of the Republic and its institutions against any danger, threat and aggression and form of subversion coming from within the national boundaries "for the other. It is worth mentioning that in both there is a reference to the State in its territorial expression. Further, it is worth mentioning how in the definition of the agencies' functions the bill is much more detailed that the 1977 law. The acceptance of the need for operational guarantees therefore should not be viewed as the rule of law giving way to the raison d'état.
It is instead possible and feasible to fasten in a dynamic and positive way the Intelligence Services' functions with the protection of the constitutional values; but at the same time the goals should be clearly framed within the defence of the institutions against specific threats.
The object of the intelligence activity should also be defined with as much precision as possible. In this respect the evolution from the "secret services" to the "intelligence and security services" and to the "intelligence system for the security" marks a significant change of perspective. The services should carry out any activity needed to counter threats against security.
The general framework for the Services' work appears therefore to have changed completely. In this perspective the intelligence operators should not oppose the drafting of provisions establishing strict penalties for "deviant" activities. These behaviours will in future be easily identified thanks to the clear and detailed definition of the institutional tasks. So they should not trouble those loyally committed to their duty.

11. Going back to the topic of the operational guarantees, we can now say that within such a neatly-defined domain, exceptional actions calling for the application of the extenuating circumstance should be framed within clear authorisation and overviewing procedures. The scrutiny activity should be performed so as to satisfy the confidentiality requirements. At least as long as the activity is being carried out or until its disclosure does not endanger any longer the interests it legitimately was supposed to protect.
The oversight system should be thought of separately and not necessarily in connection with the upholding during proceedings of a justificatory cause.
The special extenuating circumstance as operational guarantee refers to the intrinsic illegal nature of the conduct, but within court proceedings it should be called upon only in the exceptional case of the very conduct becoming the object of the judicial inquiry. This would mean that during the intelligence activity the operator has had "an accident".
The legitimacy of the behaviour must be ensured independently on any judicial inspection.
In para. 7 the close connection between secrecy and intelligence has been underlined. From this close connection stems the incompatibility between intelligence and judicial means as an ordinary channel for control over the intelligence activity. The judicial activity by its very nature entails in fact the respect of the opposite principle, that is "publicness" and accessibility.
For this reason in the same paragraph the importance is to define the activity "to be exonerated from criminal liability" within a previously-approved intelligence project. If an infringement of a constitutional individual right is to take place, it becomes legitimate only if aimed at protecting another constitutional interest. This should be established through an advance evaluation. Otherwise the inevitable risk would be to justify on the grounds of national security, maybe even a posteriori, all activities interfering with the individual sphere, even if not clearly illegitimate, introducing into the system an unacceptable "immunity on grounds of duty".
The evaluation of the proportion between interests should take place in advance and should be founded on the following principles:
- it should be indispensable;
- adequate;
- the same result could not be achieved otherwise.
These principles should refer to the case concerned, and they should be verifiable and evaluable. Otherwise the constitutional foundation of the extenuating provision could encounter insuperable objections (for the reasons mentioned above it is not excluded from the exclusive jurisdiction on matters referring to searches and interceptions - para. 7). In view of the importance of this concept, it would be advisable to preserve the wording adopted by the Government ad hoc Committee.
The highest political authority should be entrusted with the evaluation of proportionality in view of its importance. It not simply a question of balancing the sacrifice of an individual's right with the efficiency of a Service (its highest administrative authority could do that); but instead individual situations should be compared and evaluated with respect to the primary institutional interest. This evaluation should be done by an authority institutionally and politically responsible for the protection of this interest, this authority can not be but the highest Government authority.
For the same reason it is absolute necessary that within court proceedings the mechanism to confirm the applicability of the extenuating circumstance should be entrusted to the President of the Council of Ministers. This appears to be easily accepted and in some way resembles the well-rooted parallel procedure of the upholding of State secrecy.

12. Last but not least, another cornerstone for a valid system of operational guarantees is the need to keep records of and for storing documents relating to the authorisation procedures and to the intelligence activities concerned.
In this respect too, one should not fall victim of the common belief that secrecy and transparency are incompatible, in the same way as are the respect of the rules and the raison d'Etat.
The documents referring to the authorisation procedures and to the intelligence activities concerned represent above all a protection for all the intelligence operators. They should be the first to wish these documents to be as thorough as possible and to be properly stored. At the same time this would allow the necessary checks which do not contrast with the confidentiality requirements. There could be scrutiny within the intelligence body, in ways to be determined; by the parliamentary oversight committee with reference at least to the number and general details of the operations approved; then a control on the respect of the institutional limits might be needed in case a conflict of competence is raised before the Constitutional Court (such a conflict is the natural outcome of a disagreement between the judicial authorities and the Presidence of the Council of Ministers over the existence of a justificatory cause); finally there would be a control, indirect but extremely effective, stemming from the application of a time limit to secrecy (still to be enforced with due care in order to prevent possible damages).