If a judge is confronted with illegally obtained evidence, the question arises whether such evidence is admissible in court. Before answering this question, we will clarify under which circumstances evidence is considered to be ‘obtained illegally’. The Court of Cassation clearly defined ‘illegally obtained evidence’ in 2004. Evidence is considered illegal if it has been gathered (i) by committing an offence, (ii) in violation of criminal procedural law, (iii) in violation of the right to privacy (art. 8 ECHR), (iv) in violation of the rights of the defence (art. 6 ECHR) or (v) in violation of the right to human dignity. The grounds of illegality are often interchangeable. For instance, most offences will infringe upon criminal procedural legislation, violations of human dignity will often be covered by a criminal offence and most infringements upon criminal procedural law will also entail a violation of the right to privacy since the interference with private law will not be in accordance with the law as required by article 8 ECHR

The category of violation of criminal procedure legislation is furthermore not limited to specific formal rules of procedure. The prosecutor and investigating judge have a statutory duty to safeguard the loyalty or integrity of the evidence gathering process. Evidence gathered in a disloyal manner must be considered as ‘illegally obtained’, e.g. when a formal procedure is deliberately averted from its purpose to render illegally obtained evidence legal (‘evidence laundering’, a pretended ‘accidental’ discovery of evidence which was actually ‘staged’) or when the entire criminal investigation is based on the assumption that someone is guilty of an offence. The latter situation will of course also run afoul of the presumption of innocence and will thus be in breach of the right to a fair trial.

Electronic evidence can be altered easily and is very volatile by nature. Yet Belgian law lacks specific technical rules on how it should be gathered and treated. Rules do exist on the circumstances in which law enforcement agencies can access electronic evidence, but there is no detailed legal regime for technical requirements regarding the execution of a search for digital evidence, nor clear requirements regarding the tools which may be used to perform such searches. Criminal procedural law only refers to ‘suitable technical means’, which should be used to gather and preserve digital evidence to ensure its integrity and confidentiality. Guidelines seem to exist at the level of the public prosecutors’ office and police forces but they are not accessible to the public. The statutory requirement to use ‘suitable technical means’ entails that if a judge deems the specific technical tools used be the investigators ‘unsuitable’, e.g. when electronic data are altered while executing a search (such as when the configuration of the computer system under investigation is changed), the evidence will have been ‘illegally obtained’. This will inevitably entail difficult factual and technical discussions.

Evidence is not only considered to be have been obtained illegally when the criminal prosecution authorities engaged in the illegal action. Illegal gathering of evidence by a private person may also affect its use in a criminal procedure. The Court of Cassation, however, only considers such evidence as ‘illegally obtained’ if two conditions are met: (i) the person who committed the illegality is the one who reported the offence to the law enforcement authorities and (ii) he or she engaged in the illegal action to obtain goods or information with the intention of using it as evidence in a criminal case.

If the person who reported the offence to the authorities came across the data or goods by accident - albeit in an illegal manner, e.g. by burglary - said data or goods will not be considered ‘illegally obtained’ and can therefore simply be used as evidence in a criminal case. The same is true for the data or goods that were handed over to law enforcement authorities by a person who was not involved in the illegal action.

The Court also distinguishes between the actual illegal gathering of evidence and the mere illegal discovery of an offence. In the latter case, the person (including foreign authorities) who discovered the offence can report it to Belgian law enforcement authorities. This will not affect the legality of the evidence that is subsequently gathered by the Belgian law enforcement authorities.

In a broader sense Belgian courts distinguish illegally obtained evidence from ‘police information’ which is only used to start or direct the criminal investigation and is not used by the trial judge to establish his deep-down conviction on the existence of the offence or the criminal liability of the defendants.

Evidence must always be subjected to judicial scrutiny and adversarial proceedings. By contrast, police information, i.e. information from police sources which the enforcement authorities do not (wish to) disclose and the exact origin of which is therefore typically unknown (informer, anonymous source, independent foreign investigation, …), will not be subjected to full judicial control on whether the procedural rules applicable to the gathering of such information have been respected. Police information can therefore only serve to launch the criminal investigation and steer its course. It can for instance help selecting places or persons as a target for surveillance or searches, whereas the actual evidence will only be gathered in the course of the surveillance or the search. In the same sense, the Court of Cassation has considered that “information gathered by the police that may give rise to further investigation is only ordinary information that must be verified by a preliminary or judicial investigation”.

Police information can only be challenged by defendants who can make it plausible that it has been obtained in an irregular manner. Mere ignorance of the origin of the information does not prevent it from being used to start or direct the investigation: that is not in itself sufficient to establish an irregularity. The Court of Cassation indeed held repeatedly that art. 6 ECHR does not preclude information of unknown origin from being taken into account as ‘steering information’, a basis for subsequent gathering of ‘real’ evidence, as long as it does not appear that the information has been gathered irregularly. This has been criticised as offering too much scope for ‘evidence laundering’.

When a defendant makes it plausible that the police information was gathered in an illegal way, the law enforcement authorities will have to clarify how the information was obtained. If they do not succeed in establishing the legal origin of the information, the judge will assume it has been obtained in an irregular way. The rules on the use of illegally obtained evidence (threefold test, infra) do not apply to illegally obtained police information. The judge only verifies whether the right to a fair trial would be violated if the police information is used.

Traditionally (from the 1920s onwards) the courts and doctrine stressed that illegally obtained evidence could not be used in court against the suspect. From 2003 onwards, the Court of Cassation, however, drastically curtailed the exclusionary rule (developing the so-called Antigoon-theory), and actually turned the tables: it made exclusion of illegally obtained evidence the exception rather than the rule. On several occasions, the Constitutional Court has condoned this new approach.

In 2004, the new approach was codified for the first time, but only in relation to evidence obtained abroad. In 2013, and once more without much legislative debate, a new article 32 was introduced in the Preliminary Title of the Code of Criminal Procedure (hereinafter: PT CCP), which codified the Antigoon doctrine.

Under art. 32 PT CCP the judge may declare illegally obtained evidence inadmissible only if:

statute law explicitly mentions nullity as the sanction for neglecting the formality;

the irregularity affects the reliability of the evidence; or

the use of the illegally obtained evidence would violate the right to a fair trial.

Said three grounds for exclusion of evidence have to be considered as exclusive: only in the given three scenarios the evidence can and has to be excluded. As Maes points out, neither the Court of Cassation, nor the drafters of article 32 CCP have given theoretical reasons or invoked an explicit underlying vision on the function of the exclusionary rule to justify why it should be limited to these three grounds. The ratio, both behind the Court of Cassation’s U-turn and Parliament’s decision to codify it, seems to be an increasing irritation with suspects and defendants invoking procedural errors and a perception - not really backed up by empirical examples - that many guilty persons would profit from the strict rules on procedural errors while their interests have not really been harmed. Neither the judiciary, nor the legislative body ever really addressed the issue that poorly drafted and incoherent legislation on procedural rules may be a cause of law enforcement authorities’ errors. Commentators pointed out that this remarkable leniency towards sloppy law enforcement and the judiciary stands in contrast to the increasingly strict application of procedural rules with regards to suspects and defendants, e.g. with regards to procedural deadlines and formalities.

The CCP hardly ever provides nullity as a sanction for failing to meet a procedural formality (infra). There always has been much case law on the ‘exclusionary/admissibility rule’ and the legislation concerning procedures for intrusive investigative measures has become ever more detailed and extensive. That makes it all the more striking how little legislative attention has been paid to the appropriate sanction for illegally obtained evidence. To a certain extent this can be explained by the fact that, originally, the code’s drafters assumed that including procedural formalities in statute law was sufficient to guarantee the respect thereof. It appeared self-evident that what happened outside of the bounds of the law, would be inadmissible in subsequent criminal proceedings. This is why the CCP did not contain the initial ‘exclusionary rule’ and furthermore seldom explicitly provides nullity as a sanction for failing to meet a procedural formality. Nullity sanctions were included to stress that a certain requirement was important even though it could at first sight appear to be rather trivial. Since the U-turn in the case law in 2003, Parliament has included only few nullity sanctions in the CCP (infra), because nullity leaves absolutely no margin for appreciation for the judge. Lawmakers seem unwilling to take a stance and prefer to leave it to the judges to balance the interests at stake. For the same reason, one of the few nullity sanctions which was provided by law (in relation to the wiretap, which was introduced in 1993 with quite some reluctance and distrust towards law enforcement at the time) has been abolished in 2016. Parliament indeed considered that the consequences of the nullity sanction were too dramatic and disproportionate in the serious cases for which wiretaps are used.

It should be noted, however, that case law did develop a category of so-called “substantial formalities”. Although not accompanied by a statutory nullity sanction, judges considered those formalities nevertheless to be essential. Would a violation of said formalities lead to exclusion of evidence? Following a Court of Cassation decision of 16 November 2004, it did not seem to be relevant whether a formality was considered substantial or not. The consequences of a violation of said formality had to be assessed in the same manner. This therefore seemed to make the distinction between substantial and non-substantial formalities practically irrelevant. However, in 2013 the Court of Cassation did accept that a breach of a specific category of substantial formalities should lead to the exclusion of evidence: those relating to the organisation of courts and tribunals. The case concerned a home search which was mandated by a judge who did not have the power to authorise those searches. Though it seemed like an important correction, this case law was explicitly abandoned when Parliament introduced art. 32 PT CCP. Later case law confirmed that even in case of a breach of substantial formalities relating to the organisation of courts and tribunals, no exception applies.

Whether judges consider a procedural safeguard to be substantial or not, has therefore become irrelevant. Only the violation of legal requirements explicitly accompanied by a statutory nullity sanction will automatically lead to exclusion of the evidence obtained. The sanction of nullity is currently only linked to some specific requirements for the seizure of real estate (art. 35bis CCP), testimony by a witness granted full anonymity (art. 86bis and 86ter CCP) and when performing a polygraph test as of 1 January 2021 (new version of art. 112duodecies CCP which should take effect on said date).

Furthermore, statute law provides for an explicit sanction when statements are made during questioning which infringes the rules on questioning in violation of the right to legal assistance. Art. 47bis in fine CCP stipulates that no person can be convicted based on statements he made in breach of the right to and rules on prior confidential consultations and the assistance of a lawyer during questioning. This is, however, not strict nullity: such statements can for instance be used to convict third persons.

As a second rule, judges must exclude illegally obtained evidence when the irregularity taints its reliability. Of course, judges always take reliability into account, also when assessing the probative value of legally obtained evidence. However, when the judge assesses the probative value of legally obtained evidence, unreliable evidence will not be ‘excluded’ in the strict sense of the word, i.e. it will not be declared inadmissible. Rather, the judge will simply decide not to rely on said evidence in deciding upon the defendant’s guilt.

A prime example of evidence that may be unreliable as a result of the way in which it was gathered, is evidence that is dependent on the free will of the person providing it, notably witness statements and confessions. The reliability could clearly be tainted by illicit pressure or influence from the police. However, also other types of evidence can be unreliable due to irregularities. DNA tests for instance are regulated in detail. A violation of these rather technical regulations may just as well affect the reliability of the test results.

The third ground for exclusion is the most important one in practice. The Court of Cassation has set out a number of criteria to determine whether evidence obtained in an irregular manner must be excluded to ensure the right to a fair trial. It confirmed these criteria after the introduction of art. 32 PT CCP. In the assessment, the judge may in particular take into account:

the intentional or inexcusable nature of the rule-breaking by the investigators,

the seriousness of the irregularity in comparison to the seriousness of the offence under investigation,

the fact that the illegally obtained evidence only relates to a material element of the existence of the offence, and not to the involvement of a specific person in committing the offence,

that fact that the irregularity is of a purely formal nature,

the impact of the irregularity on the freedom or right that the violated norm is intended to protect.

This list is not exhaustive. The Court of Cassation also refers to the respect police forces had for the loyalty principle enshrined in the CCP as a factor that judges can take into consideration in the exercise of their discretion. On several occasions, it has also taken into account circumstances which cast doubt on the reliability of the evidence obtained because of concerns as to the impartiality of the person who conducted the investigation. This consideration differs from the second ground for exclusion because it need not be established that the reliability of the evidence has indeed been tainted. The fact that circumstances cast serious doubt on the reliability of the evidence suffices to conclude that its use would lead to a breach of the right to a fair trial.

The criteria are indicative, they are not binding, nor is any single one of them decisive. For instance, the fact that law enforcement agencies committed the irregularity intentionally does not necessarily lead to exclusion of the resulting evidence. The judge should take into account all the elements of the case to decide whether the use of the illegally obtained evidence would be contrary to the right to a fair trial. As a general rule, the judge is, however, not obliged to include all of the abovementioned criteria in the assessment and is free to use other criteria. In a 2018 decision, the Court of Cassation seemed to imply that in the assessment the judge must include the criteria that the parties to the proceedings have raised. In the same sense, the European Court of Human Rights has ruled that the right to a fair trial presupposes effective judicial control. The suspect must therefore be given the opportunity to challenge the admissibility and authenticity of the evidence. However, this possibility should not be purely theoretical and illusory. The judge must take a concrete position on the arguments raised and provide an appropriate response to the crucial points raised.

Whereas the broad, non-decisive and non-exhaustive criteria seem to leave a wide judicial discretion when assessing whether the right to a fair trial is breached, in practice it is mainly used to deny there is such a breach. Evidence is almost never excluded in court nowadays.

The Court of Cassation has always stood by the “fruit of the poisonous tree doctrine”: if the conditions for exclusion are fulfilled, this entails not only that the original illegally obtained evidence but also its fruits must be excluded. If the original illegal action is covered by any of the three grounds for compulsory exclusion, judges cannot take into consideration “neither directly nor indirectly” any evidence found as a consequence of the original error. Nonetheless, both national judges and the European Court of Human Rights have accepted that ‘other evidence’ was unrelated to the illegally gathered evidence, that metaphorically it was ‘fruit of a different tree’ and that the unrelated evidence was sufficient to found a conviction. The decisions on pre-trial preventive detention also seem to be ‘a different tree’: the Court of Cassation stresses that illegally gathered information can and often even should be used by judges deciding on the arrest or the continued detention on remand of suspects.

Under the existing CCP, which provides for two types of criminal investigation (preliminary investigation and judicial inquiry), the moment in the procedure at which the evidence will be excluded may differ. At the end of the judicial inquiry, the so-called pre-trial chambers (like the Chamber of Indictment of the Court of Appeals) should eliminate the illegally gathered evidence (art. 235bis CCP). The idea was to avoid that the trial judge would still be psychologically influenced by the evidence he is supposed to ignore. However, if the issue has not been raised in the pre-trial stage, it will still be possible to raise it at the actual trial. Furthermore, the Court of Cassation has suggested that exclusion of illegally obtained evidence by the pre-trial chambers is only possible in case of nullity or violation of the right to a fair trial. The second ground, unreliability caused by the irregularity of the gathering, cannot be judged by the pre-trial chambers, as it requires an assessment of the probative value of the evidence and this is not their task. It should be left to the trial judge. Regardless, the overwhelming majority of investigations are preliminary investigations conducted by the public prosecutor, which are not subject to the supervision of the pre-trial chambers, so any decision on exclusion or admission of illegally gathered evidence will be taken by the trial court. There is an important exception, though: if so-called special investigative methods have been used (undercover agents, secret visual surveillance, but not wiretap or covert search in IT-systems) the pre-trial Indictment Chamber will perform a control of legality even after a preliminary investigation (art. 235ter CCP). The pre-trial proceedings are not open to the public and the procedure is organised in such a way as to allow judicial control without revealing too much about the methods and technical means used or the identity of the people involved in the covert police operations.

In one situation illegally gathered evidence is always admissible: if the evidence is exculpatory, suspects can always rely on it. It does not matter whether suspects or defendants were personally involved in the illegal actions or not. For instance, when a witness statement containing some exculpatory information is illegal (e.g. violation of the public nature of the court sessions), the suspect can use the information in his defence all the same. As it happens, evidence which is exculpatory for one of the suspects can indeed be quite damaging for another suspect. Judges will have to consider it in favour of the former, but ignore it when dealing with the latter. That is easier said than done.

Privileging material truth-finding and punishment of perpetrators over the procedural protection of individual rights, Belgium has made it very difficult for judges to exclude illegally gathered evidence. That this system seems compatible with European (human rights) law, does not mean there is no room for improvement.

We do not advocate a return to the old strict exclusionary rule. There are good reasons to mitigate it, without the overcompensation of current Belgian law. The underlying rationale of the exclusion of illegally obtained evidence has always been and remains that this sanction will have a dissuasive effect on overzealous police officers, prosecutors and perhaps even investigating judges. In real life this type of sanction may not reduce the unwanted behaviour as much as it would stimulate creativity in searching ways to avoid the sanction, e.g. ’evidence laundering’ by third parties. Moreover, the distance - both in terms of time and psychologically - between the illegal action and the potential exclusion of the evidence is considerable. Under Belgian criminal procedural law, the people behaving improperly during the investigation are rarely called to answer in person at the trial for their procedural misbehaviour. For unintentional, inadvertent breaking of the rules by investigators, one should not expect too much of a deterrent effect for exclusion as a sanction. The biggest policy problem is that a consistent application of the exclusionary rule produces results which both practitioners and public opinion find hard to swallow. The exclusionary rule is mostly - or only - to the advantage of ’guilty’ suspects. When a case does not entail criminal proceedings, innocent persons whose rights were violated, will not benefit from this sanction. In conclusion, the exclusion of reliable, but illegally gathered evidence does not always sanction or reward the right persons. Sometimes admitting the evidence does make sense.

The current Belgian system lacks coherence. On the one hand it is legalistic to the extreme, in the sense that it links obligatory exclusion to the violation of a formality for which statutes explicitly provide nullity. This is rather peculiar, as everyone knows that in Belgian statutory law such provisions are very rare. One could say the Court of Cassation passed the buck to Parliament, which should provide nullity for every formality which it deems essential. So far, the suggestion has fallen on deaf ears. On the other hand the current rules turn on the criterion of the right to a fair trial as a kind of panacea. The Belgian standards - which used to be relatively high - are reduced to the absolute minimum level of Strasbourg case law. As such, violations of the right to privacy do not entail an infringement on the right to a fair trial. Belgium took Strasbourg case law, which should be an absolute minimum states should never fall below, and turned it into the only standard and consequently into the maximum level of protection.

One can understand that the Belgian authorities are concerned that ever more detailed procedural rules in combination with more adversarial style defence tactics will lead to trials in which procedural discussions overshadow the merits and the substantive issues of criminal law. Is this a nostalgic longing for trials with debates centred more on the defendants and less on the investigators? Much as we appreciate this concern, we would like to warn against phobic traits in the judicial dislike of defence arguments related to procedural errors. Any pharmacologist knows that small doses of poison can sometimes be beneficial for an organism. Still, the Belgian criminal justice system’s appetite for poisoned fruit can be dangerous, if not lethal, in the long run. It is unduly lenient for sloppy performance and even for mischievous practices by investigators.

Rather than the (individual or general) deterrence of investigators, it should be the overall image of the criminal justice as trustworthy and the exemplary ‘proper, lawful’ behaviour of those representing it, that should guide the legal regime for illegally gathered evidence. Like other authors, we have suggested that Parliament should make exclusion the rule again, but allow judges to use the evidence in a number of exceptional situations. Unlike now, investigators who infringe the rules should not be sure from the outset that their illegal actions will not affect the evidence and judges should explain why they do use evidence in spite of the fact that it was illegally gathered.

Now that Belgium is finally getting rid of the Napoleonic CCP and considering the adoption of a new one, Parliament should grasp the opportunity to redress matters. Whether the proposal for a new CCP will eventually be adopted, remains to be seen. In any event, its drafters are intent on changing the rules on illegal evidence. Normality would be restored in the sense that legal rules must be followed and if they are not, as happens with illegally gathered evidence, the principle is that it should not be used à charge. The Drafting Committee is right in stressing that under the Rule of Law, a state cannot take as a norm that illegally gathered evidence should be used. Information gathered in violation of the rights of the defence, the right to privacy, the right to a private life, the integrity of persons, in violation of rules on the competence of judges and courts or in violation of the secrecy of journalistic sources, should be excluded, as well as information derived from it (‘fruit of the poisonous tree’ will be poisoned).

However, the drafters do not envisage a return to the old strict exclusionary rule: judges will be allowed to make exceptions. They should consider the seriousness of the violation, the legal interest at stake and other circumstances of the specific case to allow the use when

1° it is not the result of an intentional and unjustifiable violation of the protected legal interest,

2° the seriousness of the legal rule or interest that was affected in the specific case does not outweigh the interest of society in prosecuting the evidence and punishing the perpetrator and

3° the use of the evidence is not detrimental to the integrity of the criminal justice system.

In addition to this broad discretion to judge the appropriateness of excluding or using illegally gathered evidence, there are two instances in which judges have an absolute obligation to exclude. This would be more strict than the case law of the European Court of Human Rights: evidence which results from inhuman and degrading treatment must always be excluded. The other will prove even more controversial (and might be amended by Parliament): if the law required authorisation (a warrant) by a judge for an investigative action and such authorisation has not been obtained. Less serious violations could be the basis for exclusion if the judge feels they are detrimental to the integrity of the criminal justice authorities.

Since the integrity of the criminal justice system is at stake, not just the rights of the specific individual, exclusion is due regardless of whether the person requesting the exclusion was personally affected by the illegal action. If only the rights of someone who is not being prosecuted have been violated, for instance by an illegal search, the evidence can be excluded and the defendants could ‘profit’ from this.

The drafters reject the distinction made between ‘police information’ on the one and ‘evidence’ on the other hand, since they believe the two are often inextricably linked. It does distinguish, as the Court of Cassation has always done, between the knowledge of the existence of offence (which is not affected by the illegality) and the actual use of illegal evidence. Only if the initial knowledge has its origin in torture, inhuman or degrading treatment, any prosecution would become impossible.

The new rules would have as a consequence that debates on how far the authorities can go will once again take place in open court, but not that serious criminals would be acquitted because of minor bureaucratic errors. Intentional wrongdoing will have more consequences than unintentional mistakes. The trial judge, which will have heard all the interested parties, is best placed to balance the conflicting interests and decide what is best for society. Unlike now, the investigators will not know from the outset that whatever happens, the evidence will always be used. Such uncertainty can only be a stimulus to respect the law. That is what officers have sworn to do, anyway.

Corresponding author: Frank Verbruggen, Charlotte Conings, KU Leuven, Belgium. Email: frank.verbruggen@kuleuven.be, charlotte.conings@stibbe.com.