Opening Pandora’s box: The right to silence in police interrogations and the Directive 2016/343/EU

This article examines the provisions of the Directive 2016/343 related to the right to remain silent with special emphasis on pre-trial proceedings and police interrogations. It focuses on the inherent contradictions and unclarities of the respective provisions, particularly when interpreted in light of the respective ECtHR case law. The article also identifies areas, relevant to regulation of suspect interrogations and the right to silence, which are not addressed in the Directive or the ECtHR jurisprudence. It concludes by critically assessing the likely effectiveness of the Directive provisions in ensuring the right to silence in criminal proceedings.


Introduction
The goal of this article is to expose the uncertainties in the interpretation of the provisions of the Directive 2016/343/EU on the strengthening of certain aspects of the presumption of innocence and of the right to be present at the trial in criminal proceedings (hereafter -Directive 2016/343/EU) 1 , as related to the right to remain silent at the investigative stage of the proceedings. It is argued that as the result of the vague and contradictory nature of the respective provisions, the effects of the Directive on the laws or practices of Member States around the right to silence are likely to be compromised. 2 In line with the theme of this Special Edition, our focus is on oral or written statements elicited from suspects in the course of the investigative stage.
Directive 2016/343/EU has been described as the most 'eccentric' 3 piece of legislation from the 'procedural rights' package. 4 It addresses an eclectic set of 'other' issues which were not among those designated a specific legislative instrument in the 2009 Roadmap setting out plans for EU action on procedural rights. 5 In addition to its peculiar adoption history, Directive 2016/343/EU differs from other procedural rights' directives in that many of its provisions concern not only (suspects') rights, but that these also lay down rules and principles of criminal procedure law (such as the burden of proof, presumption of innocence, and trials in absentia). It may therefore be viewed, theoretically at least, as a more 'far-reaching' attempt to harmonise national systems of criminal procedure, as compared to the other directives. 6 The right to remain silent and the privilege against self-incrimination are considered elements of the presumption of innocence. 7 Consequently, Directive 2016/343/EU devotes an entire article to the right(s) to remain silent and not to incriminate oneself (art 7). Other provisions relevant to the (protection of) the right to silenceas described beloware contained in art 2 (the scope), art 6 (the burden of proof) and art 10 (remedies). Recitals, which precede the text of the Directive, are important when interpreting the respective articles, although as demonstrated below they often bring less clarity than comprehension when it comes to understanding the respective provisions. The discussion of the privilege against self-incrimination (art 7(3), Directive 2016/343/EU) is relevant insofar as it applies to evidence (e.g. forensics, bodily samples or smartphone passwords) sought from the suspect in the course of an interrogation, with the view to obtaining a comment on that evidence, or a refusal to produce such evidence.
The article begins by describing the legislative drafting history of the Directive 2016/343/EU, which is important to understand why its final provisions have taken certain forms. The second part of the article critically analyses the Directive provisions, relevant to the protection of the right to remain silent at the investigative stage, with reference to the relevant ECtHR case law. It begins with the provisions on the scope (art 2), the content of the right to remain silent and the privilege against self-incrimination (art 7), the aspects of the burden on proof relevant to the content of the above-mentioned right (art 6), and remedies for breaches of the right to remain silent (art 10). The final part describes the various imprecisions and inconsistencies identified in the Directive 2016/ 343/EU provisions specifically with respect to the right to remain silent at the investigative stage. Drawing on the scholarship, which examines the link between certain properties of EU legislative texts and practical implementation of EU law, the article concludes with reflections on the likely effects of the respective directive provisions in Member States (hereafter, MS) laws and practices.

Background: The perceived need for a directive and preparatory works
The need for EU legislation on procedural rights in criminal proceedings, in addition to the already existing Council of Europe (European Convention of Human Rights (ECHR) as interpreted in the European Court of Human Rights case law) standards on fair trial rights (hereafter, ECtHR standards) was justified with reference to two main arguments. On the one hand, it was argued that the level of fair trials rights' protection in the MS was (still) so divergent, it would endanger the application of the mutual trust principle deemed indispensable for effective cross-border cooperation in criminal matters. 8 Both the legal regulation of certain procedural rights, and the practice of enforcement of these rights by national courts and other criminal justice actors differed among MS, which allegedly led to differing levels of protection. 9 On the other hand, the Council of Europe fair trials rights' enforcement system was deemed ineffective in levelling up the protection of the respective rights in MS, as violations of ECHR persisted despite the continued existence of said system. 10 Although the 2009 roadmap did not mention the presumption of innocence, suggestions for such an instrument had been raised well before 2009. 11  2006 had led to a proposalpublished in 2013for a directive on the presumption of innocence and the right to be present at the trial as examples of 'other' issues mentioned but not specified in the roadmap. 12 Notably, evidence of the need to legislate on the presumption of innocence, as presented by the Commission, was rather limited. The impact assessment itself stated that 'the general level of safeguards at national law level is in a general way acceptable' save for some specific legal problems and occasional violations of the presumption of innocence established by the ECtHR. 13 Consequently, concerns were raised by some MS regarding the added value of the Directive, given that the presumption of innocence was already protected both internationally and nationally in the MS. 14 Yet, despite the initial objections from some MS, the Directive has seen light in 2016, largely thanks to the Italian Presidency efforts facilitating the adoption of the general approach by the Council. 15 The concern about the lack of justification of the need for a directive on the presumption of innocence seems to stem from the following. Unlike with the other procedural rights' directives, the Commission did not persuasively demonstrate, in relation to the Directive 2016/343/EU, in which respect the legislative frameworks of MS were not compliant with the respective ECtHR standards or otherwise unsatisfactory (except pointing at some specific legal problems, such as the possibility to draw adverse inferences from silence). 16 This does not signify that (most) MS already have regulations in place, which are conducive to achieving a high level of protection of the presumption of innocence and the right to remain silent (as the 'country-specific' articles in this Special Issue demonstrate), but simply means that such analysis is largely missing from the Impact Assessment. 17 Indeed, the problems of MS compliance with the presumption of innocence and the right to remain silent described in the preparatory documents, namely, the 2006 Green Paper and the 2013 Impact Assessment, are mostly those related to professional cultures and practices of MS' criminal justice actors, rather than the respective legislative frameworks. With regard to the right to remain silent, the Green Paper notes, for instance, that 'coercion to co-operate with the authorities in the pre-trial process may infringe the privilege against self-incrimination and jeopardise the fairness of any subsequent hearing', and that 'the way in which the accused is 12. Commission, 'Proposal for a Directive of the European Parliament and of the Council on the strengthening of certain aspects of the presumption of innocence and of the right to be present at trial in criminal proceedings', COM (2013) 821 final, 27 November 2013. 13. The impact assessment referred to 26 ECtHR cases, where a violation of the presumption of innocence was explicitly established. The Commission argued that these cases were only the 'tip of the iceberg', as violations of other rights covered by the directive such as the right to remain silent were not included. Nonetheless, the justification provided by the Commission was criticised as insufficient. See, for more detail, Crass and Erbežnik (n 11) 26. 14. ibid. 15. ibid. 16. One Dutch commentator noted that in his view, the Commission did not provide persuasive evidence of the need for the directive, because the problems of compliance with the presumption of innocence concern the attitudes and practices of the authorities and the media, rather than gaps or deficiencies in the MS legislation, and it is unclear how these problems can be tackled by another piece of European legislation. Joost S Nan, 'Richtlijn 2016/343, betreffende de versterking van bepaalde aspecten van het vermoeden van onschuld; iets nieuws onder de zon?' 64 Delikt en Delikwent 706, 709. 17. The general nature of the provisions on presumption of innocence, and the need for more specific safeguards to enforce these provisions was repeatedly highlighted during negotiations. See e.g. Legal Experts Advisory Panel and Fair Trials International, 'Joint position paper on the proposed directive on the strengthening of certain aspects of the presumption of innocence and of the right to be present at trial in criminal proceedings' (2014) <https://www.fairtrials.org/wp-content/ uploads/Presumption-of-Innocence-Position-Paper2.pdf> accessed 1 May 2021. made aware of the right (to remain silent at the investigative stage) differs'. 18 The Impact Assessment, when describing the state of the problem, refers inter alia to abusive police interrogation practices seeking to convince suspects to cooperate and undermining their right to silence. It then states that the protection against these practices in many MS is insufficient due to the possibility of drawing adverse inferences from silence (tolerated by the ECtHR case law) and the lack of specific and effective remedies for breaches of the right to silence in the course of investigations. 19 Although improving compliance with suspects' rights in practice is certainly an underlying ambition of the EU procedural rights' directives, 20 it is not immediately obvious how adoption of EU legislation can help enhance practical compliance. The impact of supranational standards is more easily 'traceable' on the level of (national) law than enforcement practices. 21 The legislative history of Directive 2016/343/EU, as of the other procedural rights' directives, is that of a rather painstaking negotiation process 22 between the three parties involvedthe Commission, Council (representing MS interests) and European Parliament. The proposed texts were closely scrutinised and debated, and compromises were inevitably made. As a resultand quite predictablythe provisions from the initial Commission draft were 'watered down' in the General Approach and in the final version of the Directive. In particular with respect to the right to remain silent, the Commission had initially proposed to render the right and the privilege against self-incrimination absolute by outlawing any negative consequences for the suspect of remaining silent or refusing to cooperate. Another suggestion was to include an explicit provision on the inadmissibility of evidence obtained in breach of the above-mentioned rights. 23 Both proposals were rejected in the course of negotiations and replaced by less 'radical' provisions. This was notwithstanding the participation of the European Parliament in the Directive drafting process, whose position typically reflects the desire to ensure a higher level of protection of rights. Provisions relevant to the right to silence in interrogations Scope and moment of application of the Directive including the right to silence (art 2) Article 2 of the Directive 2016/343/EU states that 'it applies at all stages of the criminal proceedings, from the moment when a person is suspected or accused of having committed a criminal offence, or an alleged criminal offence, until the decision on the final determination of whether that person has committed the criminal offence concerned has become definitive'. 25 This means that, unlike the first procedural rights' directives, the Directive 2016/343/EU takes effect from the moment a person is under suspicion and 'even before that person is made aware by the competent authorities of a MS, by official notification or otherwise'. 26 However, no specific mention is made, unlike, for instance, in the Directive 2013/48/EU on the right of access to a lawyer, that Directive 2016/343/EU applies to individuals who are being interrogated by police authorities, or to those who are called to the police station to testify as witnesses or 'concerned persons', but who become suspects in the course of the interrogation. 27 This might create uncertainty as to whether Directive 2016/343/EU applies to these categoriesand as yet, no clarification on this has been givenbut note that the ECtHR case law clearly extends protection to such persons. 28 Even more problematic are the references, in art 2, to 'all stages of the criminal proceedings' and in recital 11, to the exclusion of 'civil proceedings or administrative proceedings, where the latter can lead to sanctions', giving a range of examples of such proceedings (e.g. for road traffic or tax offences). This formulation suggests, as one possible way of interpretation, that the Directive would rely on domestic classifications of such proceedings, although it contradicts the ECtHR approach, which applies an autonomous definition of 'criminal charge' based on so-called Engel criteria. 29 The latter interpretation is rendered even more plausible, if the travaux preparatoires are being considered, as the European Parliament had advanced a proposal to align the respective text with the Engel criteria, which was ultimately rejected. 30 Although it is in no way certain that the CJEU would prefer to rely on national definitions of 'criminal proceedings', as it has engaged with autonomous interpretation of EU (criminal) law in the past, 31 the Directive provisions cited above arguably create greater confusion, rather than clarity, regarding the scope of application of the right to remain silent. 32 It is also unclear what would be the status, under the Directive, of information obtained in 25. The possibility to extend the scope of the Directive 2016/343/EU to legal persons and not only to natural persons (which was ultimately rejected) was an important point of debate around art 2, which is not covered in this article. See for more detail on this issue, Stijn Lamberigts, 'The Directive on the Presumption of Innocence: A Missed opportunity for Legal Persons' Eucrim 36. 26. Recital 12, Directive 2016/343/EU. The first three Directives (on the right to interpretation and translation, on the right to information and on the right of access to a lawyer) defined the moment of application as 'from the moment that the person has been made aware' of being suspected or accused of a criminal offence, which might signify a later moment than when the suspicion arises (see art 2 of the respective directives). 27. See recitals 20 and 21, directive on the right of access to a lawyer (n 4). See  punitive administrative proceedings which should have been considered criminal according to the Engel criteria, in subsequent criminal proceedings commenced at a later time. 33 General provisions on the right to remain silent and accompanying safeguards (art 7, para 1) The provisions of Directive 2016/343/EU on the right to remain silent, like many others, have a declaratory character. Article 7(1) states that 'Member States shall ensure that suspects and accused persons have the right to remain silent in relation to the criminal offence that they are suspected or accused of having committed'. Recital 26 further states: 'the right to remain silent and the right not to incriminate oneself should apply to questions relating to the criminal offence that a person is suspected or accused of having committed and not, for example, to questions relating to the identification of a suspect or accused person'. The initial proposal by the Commission encompassed more detail concerning how this right should be effectuated, including how suspects should be informed about it (in particular, that they should be informed of the consequences of remaining silent). 34 The Parliament report added even more detail, however the Council objected to including a paragraph on informing suspects about the right to remain silent altogether, considering it superfluous as it is already mentioned in the Directive on the right to information. 35 As a result, the proposed text was deleted, although importantly, the above-mentioned Directive does not mention the obligation to inform suspects about the consequences of remaining silent.
Other proposals to include more detailed and enforceable guarantees of the right to remain silent such as the safeguards against unintentional 'waivers' of the right, or provisions on audio-visual recordings of suspect interrogations, 36 likewise, have not been taken on board in the negotiations process. In contrast, Article 9 of the Directive on the right of access to a lawyer envisages that a waiver of this right is valid if the suspect has been provided, orally or in writing, with clear and sufficient information in simple and understandable language about the content of the right concerned and the possible consequences of waiving it, and the waiver is given voluntarily and unequivocally.
Remarkably, although both the 2006 Green Paper and the 2013 Impact Assessment express a concern about the use of improper compulsion and abuse of suspects' rights during police interrogations, Directive 2016/343/EU does not include any concrete provisions targeting these problems (other than reiterating in recitals that suspects should not be forced or compelled to provide information). 37 One could expect, for instance, the inclusion of specific provisions to further clarify what 'improper compulsion' means in the context of suspect interrogations. A proposal (partially) to this effect was made by the European Parliament in the course of negotiating the Directive provisions concerning the use of compulsion to obtain information or evidence. Namely, the Parliament report stated that 'the Directive must clearly state that the use of physical or psychological violence or threats against suspects or accused persons is banned, on the grounds that it constitutes a violation of the right to human dignity and the right to a fair trial'. 38 The Council, however, suggested a 'milder' approach (accepted by all parties), namely, including a reference to ECtHR case law when determining whether the right to remain silent or the right not to incriminate oneself has been violated. 39 This approach is especially unhelpful in clarifying the limits of 'improper compulsion' in the context of police interrogations, because the existing ECtHR case law, which mostly concerns the most extreme forms of compulsion, such as physical or psychological illtreatment and torture, does not allow one to draw much guidance on this matter. 40 Adverse inferences from silence (art 7, para 5) The question of whether or not Directive 2016/343/EU should authorise the drawing of adverse inferences from suspects' silence was one of the most debated in the negotiation process. The ECtHR case law tolerates adverse inferences from silence, provided that safeguards are met such as the existence of a prima facie case against the accused, and where the evidence against the suspect clearly calls for an explanation. 41 The Commission, having expressed an intention to follow the ECtHR case law on the matter in its 2006 Green Paper, has reconsidered its position following consultations with the relevant stakeholders, including the European Criminal Bar Association. 42 As a result, the 2013 Proposal suggested that any negative consequences attached to suspects' silence should be unequivocally banned. 43 The Commission proposed the following text: 'Exercise of the right to remain silent shall not be used against a suspect or accused person at a later stage in the In this article, 'adverse inferences from silence' are understood as 'conclusions, negative for the suspect, drawn from their silence or failure to provide certain information in the course of criminal proceedings, which can impact upon the determination of criminal responsibility or the assessment of evidence'. This may include the use of inference as 'evidence' but also for other purposes such as undermining the credibility of a defence; as corroborating or supporting evidence; to assist the interpretation of other evidence; or as validation of the prosecution case in the absence of a counternarrative from the defence. See  proceedings and shall not be considered as a corroboration of facts'. 44 The Parliament went even further, suggesting that in addition to prohibiting the use of suspect's silence to 'corroborate' facts advanced by the prosecution, there should also be an explicit prohibition on the use of suspect's silence for the purpose of ascertaining criminal responsibility, or as a reason in itself to order pre-trial detention. 45 The Council, whilst seemingly agreeing to the ban of adverse inferences from silence in principle, rejected the Parliament's suggestions complementing the Commission's proposal. It has also suggested replacing the phrase 'corroboration of facts' (to denote the purpose for which inferences from silence cannot be used) with 'evidence that the person concerned has committed an offence…' in the said proposal. The ban on the use of silence as 'evidence' appears narrower than on its use as 'corroboration of facts'. First, the former might be interpreted as the use of silence in a more formal manner, namely, as a prohibition against (formally) mentioning the suspect's silence as a piece of 'evidence' in the reasoning part of a judgement. Secondly, the ban on the use of adverse inference as 'evidence' suggests that it implies a prohibition to use suspect's silence to fill in obvious gaps in the prosecution evidence, but not for other, more indirect purposes, for instance, to undermine the credibility of the accused's defence. Even more importantly, the Council added recital 28 to the Directive (which was accepted) stating that the prohibition on the use of silence in evidence 'should be without prejudice to national rules or systems which allow a court or a judge to take account of the silence of the suspect or accused person as an element of corroboration of evidence obtained by other means, provided the rights of the defence are respected'. The Directive provisions on the possible consequences of suspects' silence, particularly viewed in light of the travaux preparatoires described above, are extremely ambiguous. The operative text of the Directive seemingly establishes an unequivocal ban on the use of negative inferences from suspects' silence, which creates an impression that Directive 2016/343/EU expands on the respective ECtHR standard and renders the right to remain silent (almost) absolute, or in any case very 'strong'. 46 It is, however, unclear whether and to what extent the Directive standard on adverse inferences from silence differs, in reality, from the ECtHR standard, which authorises limited use of such inferences. The current ECtHR standard prohibits the use of inferences as 'sole or main evidence' of suspects' guilt, and requires the existence of a prima facie case established by the prosecutor before such inferences may be drawn. The ECtHR, furthermore, ruled that drawing an inference would be permissible where evidence adduced by the prosecution is so strong, the only commonsense inference drawn from suspect's silence should be that he has no case to answer. 47 Directive 2016/343/EU prohibits the use of adverse inferences from silence as 'evidence (of guilt)', but it might tolerate, on one possible reading of article 7(5) in conjunction with recital 28, their use as an element of 'corroboration of evidence'. Recital 28 thus muddles the interpretation of art 7(5), especially given that recitals, having no independent legal value, are used to interpret the respective operative provisions. 48 As the text of recital 28 is not obviously in contradiction with art 7(5), several interpretations are possible, which adds to the ambiguity of the respective directive provisions.
Article 7(4) of Directive 2016/343/EU states, furthermore, that it is up to MS to 'allow their judicial authorities to take into account, when sentencing, cooperative behaviour of suspects and accused persons'. However, the meaning of 'cooperative behaviour' is not further explained, which might be a possible source of confusion, and may be understood as authorising 'perverse incentives' to plead guilty, thus compromising the suspect's right to remain silent. 49 Compulsion to provide information other than oral statements concerning the suspected offence (art 7, para 3) The right not to incriminate oneself potentially applies not only to oral statements of the suspect with regard to the alleged offence, but also to other kinds and sources of (incriminating) information, the obtainment of which might require the suspect's cooperation, for example, written documents, forensic samples, smartphone passwords, or even such 'futuristic' sources as brain imaging. 50 Such information usually falls under a different legal regime and is subject to lower protections than oral statements concerning the incriminated offence, which fall squarely under the scope of the right to silence. 51 At the same time, distinguishing between these two types of information, namely, (oral) statements concerning the incriminated offence, and other information or material obtained from the suspect can, in practice, be problematic. Suspects may be required to produce samples, documents or passwords in the course of the interrogation. Investigating authorities also routinely ask suspects to comment on the information or material obtained from them, as well as on the reasons for (eventual) refusal to provide such information. Such refusal may lead to drawing an adverse inference in certain legal systems. 52 It is, however, unclear to what extent the privilege against self-incrimination and the right to remain silent apply, under the ECtHR regime, to information or material obtained from the suspect, other than to oral statements concerning the incriminated offence(s). 53  protection of the privilege. A major source used to discern such distinctions is the enigmatic side passage from Saunders v United Kingdom, a leading ECtHR case on the matter: 'The right not to incriminate oneself […] does not extend to the use in criminal proceedings of material which may be obtained from the accused through the use of compulsory powers but which has an existence independent of the will of the suspect such as, inter alia, documents acquired pursuant to a warrant, breath, blood and urine samples and bodily tissue for the purpose of DNA testing '. 54 Reading this passage together with the other relevant ECtHR case law (which also seem to partially contradict Saunders 55 ), commentators have suggested several possible rationales for deciding which material obtained from the suspect falls under the privilege against self-incrimination. These rationales include, for instance: whether active cooperation of the suspect is required to produce the material ('means-based' criterion); 56 the nature of the material obtained (e.g. real evidence versus testimonial evidence); 57 the use to which the material is put in the criminal proceedings; 58 or the extent to which the obligation to provide information interferes with the mental freedom of the suspect ('mental privacy'), as opposed to their physical freedom. 59 None of these rationales, however, seems to be consistently applied across the entire body of the respective ECtHR case law.
Whist the interpretation based on the first rationalenamely, whether active cooperation of the suspect is required to produce informationappears most accepted 60 (although it is derived mostly from other cases than Saunders, namely, Funke v France 61 and J.B. v Switzerland), 62 it is questionable whether Saunders supports this interpretation. 63 Namely, active cooperation of the suspect may be required in the particular examples mentioned in Saunders, and particularly the production of documents. The reference to the 'material ... which has an existence independent of the will of the suspect' in Saunders is likewise unclear, particularly because in Funke and J.B. the Court extends the scope of the privilege to pre-existing documents (which applicants were required to produce under compulsion), i.e. material that appears to exist 'outside of the accused's will'. 64 The Directive 2016/343/EU does not provide much more clarity on the issue of what material obtained from the suspect, other than oral statements with regard the incriminated offence(s), falls under the scope of its protection. Article 7(3) states that 'the exercise of the right not to incriminate oneself shall not prevent the competent authorities from gathering evidence which may be lawfully obtained through the use of legal powers of compulsion and which has an existence independent of the will of the suspects or accused persons'. Thus, the Directive simply repeats the ECtHR language in Saunders, which as demonstrated above, is unfortunately rather confusing. In the recitals, the Directive refers back to the respective ECtHR case law (recital 27) and lists the kinds of material to which the privilege against self-incrimination does not apply (recital 29), namely: 'such as material acquired pursuant to a warrant, material in respect of which there is a legal obligation of retention and production upon request, breath, blood or urine samples and bodily tissue for the purpose of DNA testing'.

cSaunders v United Kingdom
Whilst listing the types of material not covered by the privilege against self-incrimination might be the only sensible way of dealing with the matter, it is not clear whether and under which conditions the Directive allows the exclusion of other types of information or material not mentioned in recital 29 (e.g. smartphone passwords). One possible reading of the recital is that it only provides (some) examples, and that the scope of art 7(3) can be extended to other types of information which 'exists independently from the will of the accused'. Thus, the Parliament's suggestion that the list of exceptions should be made exhaustive (it argued that the 'non-extension of the principle of presumption of innocence to other potentially self-incriminating elements should apply only in clearly identified cases') 65 was not retained in the final text. In the absence of clear guidance on the criteria or rationales for excluding certain kinds of materials obtained from the suspect from the scope of the privilege, there is a risk that MS may interpret the respective directive's provisions as allowing for a wide range of exceptions. 66

Burden of proof (art 6)
Article 6 of Directive 2016/343/EU focuses on the burden of proof stating that 'Member States shall ensure that the burden of proof for establishing the guilt of suspects and accused persons is on the prosecution'. This provision primarily aims to ensure protection of the presumption of innocence during trial and it seems to have little relevance to the right to remain silent at the investigative stage. However, the significance of the burden of proof provisions for the protection of the right to silence is two-fold. Firstly, it might function to reinforce the various elements of the right to silence, such as the prohibition on the use of confessions obtained in breach of the right, 67 or the limitations on drawing negative inferences from silence, or on the use of suspect's silence to the benefit of the prosecution. Although nothing in the text of the Directive or in the travaux preparatoires suggests such a link, it may be established, for instance, by the CJEU when interpreting the respective directive provisions. Secondly, the burden of proof provisions are indirectly relevant to the right to remain silent at the investigative stage via the rules concerning the presumptions de facto placing the burden of proving certain circumstances on the defence. Presumptions are commonly used in traffic offences, 68 environmental crime, money laundering or drug-related crimes and 'work in the way that a fact is considered proven by a reasoning that infers the existence of an unknown fact from a known fact'. 69 Clearly, the existence of such presumptions creates an incentive for suspects to provide an explanation and waive their right to remain silent, as in the absence of such an explanation certain incriminatory facts will be presumed. Whilst it remains questionable whether and to what extent such presumptions (partially) shift the burden of proof onto the defence (e.g. through creating probative shortcuts in favour of the prosecution), 70 they are clearly authorised by the ECtHR case law, albeit appropriate limits. 71 Accordingly, the Commission proposal included a provision in the-then art 5(2) stating that 'MS should ensure that any presumption, which shifts the burden of proof to the suspects or accused persons, is of sufficient importance to justify overriding that principle and is rebuttable'. The Council proposed to replace the words concerning shifting the burden of proof with 'presumptions of fact or law', arguing that such presumptions create modifications to the burden of proof, rather than shifting it entirely on the defence. However, the European Parliament was strongly against the 'reversal' of the burden of proof allegedly authorised by the proposed provisions. 72 Eventually, the Parliament ensured the deletion of the respective provision from art 6 of Directive 2016/343/EU in exchange for a number of important concessions, including on the wording of the right to remain silent. Yet, recital 22 does mention that the provisions on the burden of proof should be 'without prejudice' to the use of presumptions of fact or law concerning the criminal liability of a suspect or accused person. As a result, the Directive creates somewhat of a 'grey area' around the use of such presumptions.

Remedies (art 10)
The absence of effective remedies for pre-trial violations of the right to remain silent in many MS was described as part of the problem (presumably) to be tackled by the Directive in the Impact Assessment. The provisions of Directive 2016/343/EU on remedies was another 'hot potato' in the course of the negotiations. The limits of EU action on remedies for procedural rights for procedural rights' violations were debated, as (full) harmonisation seems to imply the need to restructure 68. InTelfner v Austria, for example, it was the defence who had to prove that the accused was not the driver of a car registered in his mother's name and causing the traffic accident. national criminal procedure systems, which goes beyond the EU mandate. 73 The debate spun around the insertion of an exclusionary rule of evidence obtained in breach of the right to remain silent, proposed by the Commission and the European Parliament. The Commission's initial proposal read that 'any evidence obtained in breach of this Article shall not be admissible, unless the use of such evidence would not prejudice the overall fairness of the proceedings'. 74 The LIBE Committee of the European Parliament proposed an unequivocal exclusionary rule by deleting the reference to the 'overall procedural fairness', which, especially in light of the recent ECtHR case law, would potentially allow most evidence obtained in breach of art 6 rights at the investigative stage to meet the admissibility requirements, as long as the so-called 'counterbalancing safeguards' of procedural fairness are met. 75 Instead, the Rapporteur of the European Parliament suggested that the respective provision should state that evidence obtained in breach in breach of the right to remain silent shall not be admissible at any stage of the proceedings, and shall be removed from the case file.
The Council, however, objected strongly to the texts proposed by the Commission and the Parliament, arguing that a categorical exclusionary rule would run contrary to legal traditions of MS, and particularly those which operate a system of free evaluation of evidence. Reference was made to art 82(2) TFEU, which states that minimum rules adopted by the EU should take into account the differences between the legal systems and traditions of the MS. 76 The compromise text on remedies for breaches of the right to remain silent included in para 2 to the Directive 2016/343/EU mirrors the analogous provision in the Directive on the right of access to a lawyer: 'without prejudice to national rules and systems on the admissibility of evidence, Member States shall ensure that, in the assessment of statements made by suspects or accused persons or of evidence obtained in breach of the right to remain silent or the right not to incriminate oneself, the rights of the defence and the fairness of the proceedings are respected'. It should be noted that the respective provision of the Directive on the right of access to a lawyer (art 12, para 2), was likewise the result of a political compromise, following similar objections of the MS to the inclusion of a categorical exclusionary rule with regard to breaches of the right of access to a lawyer. 77 The final provision is therefore rather vague, and interpreting it in conjunction with the respective ECtHR case law blurs its meaning even further. 78 The ECtHR has radically changed its course on the matter first by adopting an unequivocal exclusionary rule for evidence obtained in breach of the right of access to a lawyer in Salduz v Turkey, 79 and then relativising it in Ibrahim and others v UK. It is therefore unclear which of these two directions, if any, should be followed when interpreting art 12(2) of the respective directive. A similar dilemma is likely to arise with respect to interpreting art 10(2) of Directive 2016/343/EU.

Critical analysis
This section aims to assess the likely 'effectiveness' of the respective directive provisions, namely: whether they are likely to cause change in the Member States' laws and/or legal practices, leading to enhanced protection of the right to remain silent. 80 Effectiveness of EU legislation rests on a number of conditions, or dimensions. Synthesising the relevant literature on practical implementation of EU legislative acts, Bondarouk and Mastenbrouk distinguish three such dimensions: substance, scope and effort. 81 Whilst 'scope' relates to the breadth of application of the respective legislation, and 'effort' concerns the intensity of practical action put into implementation, 'substance' is closely linked to the content of the given legislative instrument. It is defined as consisting of two elements: the definitional details, and the (clarity of) objectives of the legislative action. 82 Although empirical knowledge on EU compliance, and factors facilitating or hindering EU action compliance remain fragmented, 83 the 'substance' dimension seems to be most extensively addressed in the respective literature. 84 Several studies have documented a link between more detailed and less ambiguous definitions used in EU legislation, on the one hand, and clarity and transparency of the underlying objectives of EU laws, on the other hand, and their effective implementation in practice. 85 The high number of recitals can adversely impact on the speed of transposition of directives. 86 Studies into the implementation of EU legislation codifying individual rights within EU asylum and migration policies, 87 another substantive theme within the Area of Freedom, Security and Justice, have documented common failings in the design of such instruments hindering their effective implementation. A recent empirical study into the implementation of the Asylum Procedures Directive, for instance, found that it 'is compatible with different, sometimes even contrasting properties of domestic asylum policies', and that 'the risk of failing common standards and the principles of international refugee law is… not only a matter of implementation, but is already inherent in the Directive's design'. 88 The shortcomings identified were the great deal of flexibility afforded to MS in interpreting and implementing the Directive and the ambiguity of its language, including inherent contradictions of its underlying objectives (e.g. claiming to strengthen procedural rights of asylum-seekers but introducing compulsory accelerated asylum procedures with limited procedural guarantees). 89 Another study into the transposition of the EU Family Reunification Directive into national law of MS found that although the Directive mostly resulted in liberalising reunification regimes in most MS, the chosen 'minimum standards' approach, the wide margin of appreciation for MS, and the unresolved tensions with the respective ECtHR standards, have weakened its potential effects on national legislatures. As a result, the study concluded that none of the examined MS transposed the Directive fully and correctly into their national law in the first year of its implementation. 90 More specifically, with respect to EU legislation on suspects' rights in criminal proceedings, an argument was raised 91 that in order for this legislation to have added value, it should contain more detailed and enforceable safeguards of fair trial rights than those already formulated by the ECtHR. 92 Indeed, save for a few exceptions (such as the right of early access to a lawyer), at the moment of drafting the EU procedural rights' legislation, the respective procedural rights of suspects were already generally protected in most MS. 93 Although empirical information on the implementation of EU procedural rights directives is (still) very scarce, a recent study on the transposition of art 12(2) of the Directive on the right of access to a lawyer (remedies) has indeed shown that very few MS amended their national legislation to transpose the respective provisions. This was due to the vague and general nature of the respective article and the great margin of appreciation afforded to MS, which led the respective departments responsible for transposition in MS to believe that no adjustments were necessary. This was despite the fact that, as the study's author concluded, most MS did not provide for an 'effective remedy' for the breaches of the right to a lawyer in their laws, as interpreted based on the object and purpose behind the respective directive provisions. 94 Assessed against the findings described above, the effectiveness of the provisions of Directive 2016/343/EU on the right to remain silent (at the investigative stage) is likely to be compromised. 95 The language of the respective directive provisions, as follows from the paragraphs above, is very general and creates more confusion than clarity, particularly when attempting to interpret them in conjunction with the respective ECtHR case law. The lack of detail of the provisions of Directive 2016/343/EU is remarkable, especially when compared with the provisions of the other procedural rights' directives, for instance, the Directive on the right of access to a lawyer, which clarifies among others, the timing of access, conditions for valid waiver of the right, and the obligations of state authorities to inform suspects about the right and to facilitate access. 96 On the other hand, the number of recitals is equally quite high: 51 recitals and 10 'substantive' articles. 97 Many relevant and arguably indispensable elements concerning the right to remain silent at the investigative stage of the proceedings are omitted from the Directive 2016/343/EU. These include, for instance, informing suspects about the right (providing the right to silence 'caution'), conditions for waiver of the right, or the clarification of improper compulsion in the context of police interrogations. The margin of appreciation afforded to MS in implementing the analysed provisions of the Directive 2016/343/EU is very large. This follows not only from their vague and general language allowing for differences in interpretation, but also from numerous references in the recitals and operative text stating that they apply 'without prejudice to national rules', which seems to suggest that the relevant national legislation would prevail over the EU law provisions. In our view, the negotiations history of the Directive reveals a number of 'lost chances' to clarify the accompanying safeguards and limitations of the right to remain silent instead, as has happened, of engaging in principled arguments on the scope of the respective rights. It is commonly accepted in international human rights law that the rights to remain silent and not to incriminate oneself are not absolute. Therefore, the clarity and enforceability of safeguards, and the careful and restrictive wording of limitations of the right become key in ensuring their practical effectiveness. Rather, Directive 2016/343/EU seems to 'camouflage' the existence of (widely recognised) exceptions in the recitals, for instance, the use of adverse inferences from silence or the partial reversal of the burden of proof; at the same time, it uses vague language which allows MS to interpret these exceptions rather broadly. One aspect which is assessed favourably by commentators is that the Directive adopts a (supposedly) stronger standard than the ECtHR with regard to the safeguards against drawing adverse inferences from silence. However, this view is not supported by our analysis above.
The lack of coherence of the objectives, or clarity on the 'object and purpose', of the respective provisions of Directive 2016/343/EU and of procedural rights' directives in general, is rather remarkable. For example, since the very first legislative proposals on suspects' procedural rights, it has been unclear whether their goal was to improve compliance with the existing ECtHR standards or to provide for higher standards by ensuring the highest possible level of protection of the respective rights. 98 Therefore, there is constant confusion about whether or not the respective directive's provisions should be interpreted in line with the ECtHR case law, or more expansively than the ECtHR pronouncements. More specifically with regard to Directive 2016/343/EU, it is not very clear from the Impact Assessment nor the other preparatory documents which practical or legal problems it aims to address, and there is a lack of thematic coherence between the problems described in these documents and the finally adopted provisions (for instance, they do not address the issue of improper compulsion in interrogations).