Wednesday, 25 November 2015

Supreme allies: Top national courts and the implementation of EU law

Daniel Sarmiento, Professor of EU Law at the University Complutense of Madrid*
In a short time-frame, two high courts of two Member States, the French Cour de Cassation and the Spanish Tribunal Constitucional, have delivered two important judgments on the implementation of EU Law by lawyers and domestic courts. The two decisions touch different subject-matters and deal with different claims, but they are equally relevant for what they represent for the correct implementation of EU Law. As I said a few weeks ago in a previous post, national high courts are becoming key players in EU Law, and the Court of Justice should cherish and look after this highly valuable ally.
Last May, the French Cour de Cassation ruled in favor of a former worker who had sued his lawyer for not making a proper defense of his client (see the judgment here). The lawyer did not invoke the Court of Justice’s case-law stated in the well-known cases of Mangold, Kücükdeveci, Petersen, etc., on discrimination on the grounds of age. As a result of it, the worker lost his case against his former employer. The Cour de Cassation stated that the claimant’s chances of success in case of having invoked the Court of Justice’s case-law were up to 80%. Therefore, the certainty of the loss suffered entitled the claimant to successfully claim damages from his lawyer.
Yesterday, the Spanish Constitutional Court, in plenary formation, ruled in favour of another worker whose claim based on EU Law was plainly ignored by the High Court of Madrid (see the judgment here). Following the Court of Justice’s case-law in the cases of Gavieiro Gavieiro, Lorenzo Martínez and others, which solved a series of cases identical to the one of the claimant, it was obvious that this case-law applied and solved the case. However, the High Court of Madrid ignored this and dismissed the claimant’s appeal.
The Spanish Constitutional Court has now stated that any jurisdiction in Spain that ignores a judgment of the Court of Justice is breaching the fundamental right to a fair trial, as provided by article 24 of the Spanish Constitution. This gives any claimant in such circumstances the chance of invoking another ground of appeal, and, above all, the use of the special procedure for the protection of fundamental rights before the Constitutional Court (recurso de amparo).
These two judgments impose considerable responsibilities on lawyers and judges. The French decision sets a high standard of professional expertise on practitioners, especially on those who are highly qualified and (as in the case of France) allowed to plead before the highest courts of the country. The Spanish judgment is a nice reminder for all courts in Spain that the case-law of the Court of Justice is binding in the strongest possible way, and therefore binding for all courts. Both cases have in common a total absence of reference to EU Law, by the lawyer in his submissions in one case, and by a court in its judgment in another.
Therefore, the sum of both decisions is not revolutionary, because it is obvious that a total lack of reference to the applicable law, whether it is national or EU Law, raises serious issues about the decision at stake. However, it is important that the highest courts of Member States are assuming the task of ensuring the correct application of EU Law. This is of course a matter for the Court of Justice, but also for its domestic counterparts too. And it is nice to see that these cases have been solved without the need to make a preliminary reference to the Court of Justice. High courts know what their role is and how it must be put into practice under national law. Now it is time for lawyers and for the remaining domestic courts to act accordingly.
Photo credit: Wikipedia.en
Barnard & Peers: chapter 6

*This post previously appeared on the Despite our Differences blog

Monday, 23 November 2015

The new Directive on immigration of students and researchers: a small step or a big leap forward?

Steve Peers

For a number of years, the EU has aimed to attract highly-skilled non-EU migrants to its territory. However, the existing legislation on this issue – the researchers’ Directive, adopted in 2005, and the students’ Directive, adopted in 2004 – have only had a modest impact on attracting more students and researchers to the EU, according to the Commission’s reports (see here and here) on the two Directives, issued in 2011.

Consequently, the Commission proposed an overhaul of this legislation in 2013. The European Parliament (EP) and the Council recently agreed on the text of this proposal (for the text of the provisional version of the future Directive, see here; the final version will be ‘tidied up’ a little legally). As you would expect, the EP and the Council compromised between their respective positions (for those positions, see here and here), which I discussed in an earlier blog post.

I’ll examine first the background and content of the new Directive, then look at how effective it is likely to be in its objective on increasing the numbers of researchers and students coming from third States.


The current students’ Directive also applies to the admission of school pupils on exchange programmes, unpaid trainees and volunteers, although Member States have an option to apply it to the latter three groups of migrants. The CJEU has ruled twice on the interpretation of this Directive. In  Sommer it ruled that Member States could not apply a labour-market preference test for students; in Ben Alaya case (discussed here), it ruled that Member States must admit students who comply with the rules on admission in the Directive. The same logically applies to the current researchers’ Directive. The UK and Denmark opted out of both Directives, while Ireland opted in to the researchers' Directive. All three countries have opted out of the new law.

The new law

The new Directive merges the students’ and researchers’ Directives, making major changes to them both. First of all, the Commission proposed that Member States would be obliged to apply the currently optional rules relating to school pupils, unpaid trainees and volunteers, as well as rules on two new groups of migrants: au pairs and paid trainees. The EP agreed with this idea, while the Council rejected it entirely. Ultimately, the two institutions compromised: the new Directive will have binding rules on (paid and unpaid) trainees and some volunteers (those participating in the EU’s European Voluntary Service), although stricter conditions will apply to the admission of trainees (more on that below). However, the rules on other volunteers and school pupils will remain optional, along with the new rules on au pairs.

Next, the Commission proposed to limit Member States’ current power to apply more favourable rules for students and researchers, confining that power to only a few provisions relating to the rights of migrants, while fully harmonising the rules on admission. The final Directive accepts the basic principle that the power to set more favourable standards should be more limited that at present, but imposes fewer such constraints than the Commission wanted. Member States will be allowed to apply more favourable rules for the persons concerned as regards the time limits on their residence permits. Many of the conditions relating to admission and withdrawal or non-renewal of the right to stay will be optional, not mandatory (as the Commission had proposed), and the Council insisted on many additional options being added. A clause in the preamble sets out the Council’s wish to provide expressly that Member States can have rules on admission of other categories of students or researchers.

Against the Commission’s wishes, the final Directive provides that the current rules on delegating decision-making to research institutions or universities will remain. Furthermore, it adds that Member States can optionally delegate such powers as regards volunteers or trainees as well.

Trainees are defined (more restrictively than the current law) as those who have recently completed a degree (within the last two years), or who are currently undertaking one. Their time on the territory is limited to six months, although this can be longer if the traineeship is longer, and the authorisation can be renewed once. But Member States retain the power to set more favourable standards as regards these time limits.  

One striking feature of the agreed Directive is a new right for students and researchers to stay after their research or study to look for work or self-employment. The EU institutions agreed on the principle of this right, but disagreed on the details. According to the Commission, the right should apply for a period of 12 months, although after 3 months Member States could check on the genuineness of this search, and after 6 months they could ask the migrant to prove that they have real prospects. The EP wanted to extend the period to 18 months, and to make Member States wait longer to check on the genuineness of the job search or likelihood of employment. On the other hand, the Council wanted several restrictions: to reduce the stay to 6 months; to allow Member States to limit students’ possibility to stay to those who have at least a Master’s degree; to check on the likelihood of employment after 3 months; and to give Member States an option to limit the job search to the areas of the migrant’s expertise. The final deal splits the difference on the period of extra stay (it will be 9 months), and accepts the various optional limits on the right which the Council wanted.

As for students’ right to work, the current Directive allows them to work for at least for 10 hours a week. The Commission proposed to let them work for 20 hours a week, and to drop the option to ban students from working during their first year of studies. The EP agreed with this, but the Council wanted to revert to the current 10-hour a week limit, and introduce a possible labour-market preference test (overturning Sommer). Again, the final deal splits the difference: 15 hours’ of work allowed per week, with no labour market preference test.

Another issue was equal treatment of those who work. Currently, the EU’s single permit Directive provides for equal treatment of most third-country nationals who are allowed to work, even if (like students) they were not admitted for employment. However, that Directive excludes au pairs from its scope, and only applies where the relationship is defined as ‘employment’ under national law; this will not always be the case for researchers. The new Directive will extend the equal treatment rules to students and researchers, even if they are not considered employees, and to au pairs whenever they are considered employees. Even non-employees will have equal treatment for goods and services (besides housing and public employment offices). But the new Directive will not waive any of the various exceptions to equal treatment that the single permit Directive currently provides for, besides a few minor exceptions for researchers.

Also, the new Directive will replace the weak rules on family reunion in the current researchers’ Directive with a fully-fledged right to family reunion. The EU’s family reunion Directive will apply to Directive will apply to researchers, and many of the restrictions in that Directive will be waived: the minimum waiting period; the need to show a reasonable prospect of permanent residence; the need to show integration requirements for family members before entry (those rules can still be applied after entry; on the CJEU’s interpretation of those rules, see here). There will also be a shorter deadline to process applications, and family members will have a longer period of authorised stay. The EP and Council compromised on the Commission’s proposal to waive the waiting period before family members could access the labour market: the Council wanted to delete this proposed rule entirely, but it agreed to it with a derogation for ‘exceptional circumstances such as particularly high levels of unemployment’. However, the EP got nowhere with its suggestion to extend these more favourable rules to the family members of students as well.

The Commission aimed to simplify the current rules on the movement (‘mobility’) of researchers and students between Member States for the purpose of their studies and research. It also proposed to extend those rules to paid trainees, while the EP wanted to extend those rules to cover unpaid trainees and volunteers as well. However, the Council prevailed on this issue, restricting the scope of these rules to researchers and students (as at present), and adding very complicated details to the proposal on this issue.

Finally, the Commission proposed to introduce a 60-day deadline to decide on applications for admission, shortened to 30 days for those benefiting from EU mobility programmes. (The current laws have no deadlines to decide on applications at all). The EP supported an even shorter period to decide on applications (30 days), while the Council wanted to raise the time limit to 90 days. Yet again, these institutions split the difference, with a 90-day general rule and a 60-day rule where institutions have been delegated the powers to decide on applicants.


The agreed Directive should be appraised in light of the Commission’s impact assessment report for the proposed Directive, which made detailed arguments for the amendments which the Commission proposed. This report provided evidence that students or researchers are attracted to a job-search period after the end of research or studies, as well as by further employment rights for students and for researchers’ family members. Certainly the new Directive addresses all of these issues to some extent.

Conversely, would-be migrants are deterred by the great variety of national rules and the rules on mobility between Member States.  On this point, the new Directive will only reduce the variety of national rules modestly, and will install mobility rules more complex than those applying at present.

Presumably, it is also a deterrent for would-be students and researchers who are already legally present to leave the country to make their applications. To address this, the EP wanted to oblige Member States to consider in-country applications for researchers, but ultimately it could not convince the Council (or the Commission) to change the existing rules, which give Member States only an option to allow this.

As for the additional scope of the Directive, it is striking that the new binding rules on admission only apply to trainees who are undertaking or who have completed higher education, and to volunteers in the EU’s own programme. The latter change in the law is necessary in order to ensure the effectiveness of that programme, but the former change in the law is another example of the EU focussing its migration policy upon highly qualified employees. (Remember that according to the preamble to the new Directive, the admission of trainees who have not entered higher education is left entirely to national discretion). It’s unfortunate that at least the rules on equal treatment aren’t binding for all volunteers, school pupils and au pairs, to ensure that these migrants are not exploited and that domestic labour standards are not undercut.

Many of the changes in the Directive intending to attract qualified migrants would make even more sense if they were part of a ‘joined up’ policy – for instance, allowing trainees to make an in-country application for studies or research, or waiving some of the conditions in the EU’s ‘Blue Card’ Directive for highly-skilled migrants (reducing the income threshold, for instance) for graduate trainees, researchers, and students looking for work under this new Directive. Fortunately, there will be a chance to address this issue in the near future, as the Commission will soon be proposing an amendment to the Blue Card Directive (on the reform of that Directive, see here).

Overall, then, the new Directive has gone some distance towards accomplishing its intended objectives, but its effect could be further augmented in the near future by a broader reform of EU law on highly-skilled immigration in general.

Barnard & Peers: chapter 26

Photo credit:

Monday, 16 November 2015

Crime victims: does EU law now confer extra rights?

Steve Peers*

*This analysis is adapted from the forthcoming 4th edition of EU Justice and Home Affairs Law

Most crimes have victims. Those victims are often not only devastated by the impact of the crime, but also frustrated by the insensitivity of the criminal justice system towards their concerns. To address this, back in 2012 the EU adopted a Directive on crime victims’ rights, which Member States must comply with by today’s date. This law replaces a previous EU law on the subject, a Framework Decision dating back to 2001. What is different about the new rules? How much impact could they have on victims’ rights in practice?

Previous law: the Framework Decision

The Framework Decision had to be applied in phases between March 2002 and March 2006. It defined a ‘victim’ broadly, as meaning ‘a natural person who has suffered harm, including physical or mental injury, emotional suffering or economic loss, directly caused by acts or omissions that are in violation of the criminal law of a Member State’. According to the CJEU rulings in Dell’Orto and Eredics, in light of this definition, the Framework Decision did not apply to legal persons as victims. Nor did it require Member States to make legal persons criminally liable for their acts (Giovanardi).

It applied to ‘criminal proceedings’ defined in accordance with national law, and the Court of Justice confirmed in Katz that this included private prosecutions. But the CJEU also ruled (in Gueye and Salmeron Sanchez) that it did not harmonize substantive criminal law, such as domestic violence legislation.

As to the substance, the Framework Decision provided first of all generally for ‘respect and recognition’ for crime victims, requiring that each Member State ensure that victims have a ‘real and appropriate role in its criminal legal system’, that they ‘are treated with due respect for the dignity of the individual during proceedings’ and that they ‘recognise the rights and legitimate interests of victims’. In the case of ‘particularly vulnerable victims’, there was an obligation to provide ‘specific treatment best suited to their circumstances’. Member States also had to make provision for victims to supply evidence, but to refrain from questioning them any more than necessary. Bringing these points together, the ‘most vulnerable’ victims had to able to testify in a manner which protected them from the effects of giving evidence in open court, by means compatible with national legal principles.

In the well-known Pupino judgment, which concerned very young children who were allegedly abused in a nursery by their teacher, the Court of Justice unsurprisingly ruled that such victims must be considered ‘vulnerable’ pursuant to the Framework Decision – leaving aside the bigger question of whether all minors must be considered ‘vulnerable’. So these victims were entitled to the protection of a special procedure in which they did not have to give their testimony in court, as long as this was consistent with the right to a fair trial. But in the later judgment in X, concerning alleged sexual abuse of a child, the CJEU ruled that this did not entail an obligation to use those special procedures where in effect the victim was asking for their use as a means to overrule the discretion to bring proceedings which national law gave to prosecutors.

Similarly, in the Katz case, the Court ruled that a person bringing a private prosecution did not have the right to demand, in light of the generality of the Framework Decision, that he have the status of a witness; but nevertheless the Framework Decision required that he must be able to submit evidence in the proceedings in some form. The Court later clarified (in Gueye and Salmeron Sanchez) that this right to be heard in the proceedings entailed the possibility for the victim to describe what happened and to express an opinion, but not to insist on any particular penalty.

Next, victims had the right to receive information on a number of issues, inter alia on the conduct of the criminal proceedings following their complaint and on the release of the accused or convicted person, at least where there might be a danger to the victim. If victims were parties or witnesses, Member States had to take necessary steps to reduce any communication difficulties they face (presumably by providing for translation and interpretation; this fell short of the later EU Directive on translation and interpretation for suspects). Member States also had to ensure legal and non-legal aid was provided to victims who are parties, and that victims who were witnesses or parties might receive reimbursement of their expenses. (Compare to the proposal on legal aid for suspects, discussed here).

Victims’ privacy and safety had to be protected, inter alia from reprisals from the offender. This could entail special methods of giving testimony, ensuring lack of contact with the offender in court proceedings, and limiting photography of victims in courtrooms. The CJEU clarified in Gueye and Salmeron Sanchez that these rules aimed ‘to ensure that the ability of victims adequately to take part in the criminal proceedings is not jeopardised by the possibility that their safety and privacy is placed at risk’. But victims’ right to a private life didn’t mean that they could influence the penalties which courts could impose upon offenders, such as a mandatory injunction in domestic violence cases, since these provisions in the Framework Decision did not aim to regulate any indirect consequences to the victims’ private life stemming from the imposition of criminal penalties upon offenders.

Member States had to ensure that it was possible for the victim to receive a decision on compensation from the offender in criminal proceedings, unless in certain cases compensation is provided in another manner; and Member States had to return victims’ property that was not needed for the purpose of criminal proceedings. The Advocate-General’s opinion in Dell’Orto argued that the former right had to include compensation for pecuniary losses, and that any exception from the possibility to obtain a decision on compensation had to be limited to certain cases only and take place usually within the framework of the same proceedings which resulted in a conviction of the offender. As for the return of property, the opinion argued that the obligation to return it only applied where the ownership of the property was undisputed or had been established in criminal proceedings; otherwise the issue is a matter for civil law.

The Framework Decision also required Member States to ‘seek to promote penal mediation’ between victim and offender ‘for offences which it considers appropriate’. The CJEU clarified this obligation in Eredics: Member States had discretion to decide which offences are covered by such proceedings. While their discretion might be affected by a need to use objective criteria to decide on which cases to cover, it was not a breach of the Framework Decision to confine penal mediation to cases involving offences against the person, transport safety, or offences against property. Member States could equally exclude domestic violence cases from penal mediation (Gueye and Salmeron Sanchez).

There were specific provisions for victims who are resident in another Member State, and for cooperation between Member States. Finally, Member States also had to promote victim support organizations, train personnel in contact with victims (particularly police officers and legal practitioners), and ensure that intimidation of victims cannot occur in venues such as courts and police stations.

The Commission’s first report on the national transposition of most provisions of the Framework Decision was quite critical regarding the lack of reported national measures which fully met the specific requirements of the Framework Decision. Its second report concluded that implementation of the Framework Decision was still ‘not satisfactory’, due to the continued variations and omissions in national law and the decision of some Member States to implement the Framework Decision by non-binding means.

The 2012 Directive

From today’s date, the Directive has fully replaced the previous Framework Decision (except in Denmark, which had an opt-out). One difference with the previous rules is the legal effect of the law: unlike the Framework Decision, the Directive can confer directly effective rights on victims, rather than indirect effect only (as confirmed by the CJEU in Pupino).

Furthermore, there are a number of substantive changes to the rules, which overall increase the standard of protection for victims’ rights. First of all, the Directive contains a new provision on its objectives, including a general requirement of decent treatment, including non-discrimination; there is also a specific general rule on child victims (Art 1(2)). The definition of ‘victim’ now expressly includes family members in the event of a victim’s death (Art 2(1)(a); ‘family members’ are defined in Art 2(1)(b)). There are wholly new rules on the victim’s ‘right to understand and to be understood’ (Art 3), followed by greatly expanded rules on the victim’s right to receive information (Arts 4 to 6; compare to the ‘letter of rights’ Directive for criminal suspects).

Victims have a ‘right to interpretation and translation’, which is much stronger than the rules on ‘communication safeguards’ in the previous Framework Decision. In fact, these rights are essentially a short version of suspects’ rights to information and translation, set out in the EU legislation referred to above. They also have a ‘right to access victim support services’, which again is much stronger than the rules on ‘specialist services and victim support organisations’ in the Framework Decision. On the other hand, the right to be heard for victims has not changed significantly.

A potentially important new right for victims is the right to review a decision not to prosecute (Art 11), although this does not go so far as to require all Member States to ensure a prosecution following every complaint by a victim (which some Member States provide for already in principle). While the ‘procedural rules’ for such reviews are determined by national law, Member States do not have any discretion as regards the underlying obligation to provide for such reviews, or to limit the substantive grounds which might be pleaded in such challenges. For instance, it should always be possible to argue that a decision not to prosecute was discriminatory, in light of the obligation to deal with victims and respond to victims’ complaints in a non-discriminatory manner (Art 1(1)). The preamble (recital 44) suggests that this right also applies ‘where a prosecutor decides to withdraw charges or discontinue proceedings’.

However, the Directive includes some special rules on this right. Where (under national law) the role of the victim is established only after a decision not to prosecute has been taken, only the victims of serious crime have such a right of review (Art 11(2); on the definition of ‘serious’ crime, see recitals 8 and 18 in the preamble). Also, the right of review does not apply to decisions taken by courts (recital 43 in the preamble), so victims have no right to review of a sentence, or to early release, although they have the right to information about such developments (Art 6).  The right to review does not concern special procedures, such as proceedings against members of parliament or government, in relation to the exercise of their official position (Art 11(5)). Member States can also override the right to review in cases where a prosecutor decides not to prosecute following an out-of-court settlement (Art 11(3)).

Procedurally, victims must be given sufficient information about their right to review ‘without unnecessary delay’ (Art 11(3)). Normally the review must be carried out by a body independent of the body which decided not to prosecute (recital 43 of the preamble), but where the decision not to prosecute was taken by the highest prosecution authority and no review of that decision is possible under national law, the decision must be reviewed by the same authority (Art 11(4)). Implicitly, it is not necessary for a court to carry out the review, but that interpretation is questionable in light of the right of access to court in Article 47 of the Charter.

The Directive is silent on what happens if the review is successful. However, logically the principle of effectiveness of EU law requires that in this case, at the very least the prosecutors must reconsider their decision not to prosecute to the extent that it was flawed, and produce a fresh decision following that reconsideration.

Next, the Directive provides for safeguards in restorative justice services, in place of the prior rules on ‘penal mediation’ (Art 12; it follows that the case law on penal mediation is no longer relevant). But a series of rules (Arts 13-19) have not been fundamentally altered: the right to legal aid; the right to reimbursement of expenses; the right to the return of property; the right to a decision on compensation from the offender; the rights of victims resident in another Member State; the general right to protection; and the right to avoid contact with the offender.

Finally, there are a number of changes to other important rules: the rules on protection of victims during criminal investigations (interviews, legal assistance, medical examinations) have been expanded (Art 20); the right to privacy of victims has been elaborated further (Art 21); the provisions on victims with ‘specific protection needs’ have been hugely expanded (Arts 22-24); there are expanded provisions on the training of practitioners (Art 25); the rules on cooperation between Member States’ authorities have been expanded (Art 26(1)); and there are new provisions requiring Member States to make victims more aware of their rights (Art 26(2)).

Compared to the previous legislation, the Directive not only has stronger legal effect, but also has increased substantive rights for victims as regards: non-discrimination; the ‘right to understand and to be understood’; the right to receive information; the ‘right to interpretation and translation’; the ‘right to access victim support services’; the right to review a decision not to prosecute; safeguards in restorative justice services; protection of victims during criminal investigations; the right to privacy of victims; and victims with ‘specific protection needs’, including victims of hate crimes. The Directive is therefore likely to make a significant contribution to the protection of crime victims' rights in the EU - assuming, as always, that it is fully and correctly implemented.

See also: the Commission’s detailed guidance document concerning implementation of the Directive. 

Barnard & Peers: chapter 25

Photo credit:

Sunday, 15 November 2015

The new Directive on the presumption of innocence: protecting the ‘golden thread’

Debbie Sayers, Legal Research Consultant,

"Throughout the web of the […] criminal law one golden thread is always to be seen that it is the duty of the prosecution to prove the prisoner's guilt”

Woolmington v DPP [1935] UKHL 1

Around nine million people are the subject of criminal proceedings every year in the EU.[1] All of them are entitled, by law, to a fair trial irrespective of the charge faced. The presumption of innocence (set out in Article 6 (2) ECHR and Article 48 (1) EU Charter) is the cornerstone of the right to a fair trial. It is rooted in the need to protect the individual against the improper use of coercive state power. The principle is derived from, and entrenched within, the constitutional traditions of all EU Member States. In essence, it relates to both the procedural burden and standard of proof (the prosecution must prove the case against the defendant beyond reasonable doubt) while also constituting a privilege in its own right.
In reality, the principle is persistently under attack within EU Member States because of concern about crime and security, because of rampant managerialism in a climate of austerity[2] and because of penal populism. Too often, a more ‘effective’ criminal justice system is portrayed solely as a system which convicts more people. Consequently, any discussion on how best to protect the presumption in contemporary criminal justice systems is to be welcomed. In the EU context, the effective standard promotion and enforcement of human rights is to be particularly encouraged because EU criminal cooperation, via mechanisms such as mutual recognition (e.g. the European Arrest Warrant - EAW), has laid bare the variability within criminal justice systems. Although fair trial standards are set out in the ECHR, and reflected in the EU Charter, their basic level of protection is neither uniformly transposed nor consistently enforced. This has undermined a model of criminal cooperation built on ‘mutual trust’ between Member States. The EU’s response has been the creation of EU-specific standards in the form of Directives issued in pursuance of the Roadmap for strengthening procedural rights of suspected or accused persons in criminal proceedings based on Article 82 of the TFEU. Thus far the following instruments have been agreed:

·         Directive on the right to information in criminal proceedings establishes that Member States must inform individuals of their rights, including the right of access to a lawyer and the right to remain silent;
·         Directive on the right to interpretation and translation in criminal proceedings;
·       Directive on the right of access to a lawyer in criminal proceedings and in European arrest warrant proceedings (the UK and Ireland have opted out);
·        Commission Recommendation on the right to legal aid for suspects; and
·    Commission Recommendation on procedural safeguards for vulnerable persons suspected or accused in criminal proceedings.

It is in this context that the EU has recently agreed a compromise text on the Proposal for a Directive on the strengthening of certain aspects of the presumption of innocence and of the right to be present at trial in criminal proceedings. The stated purpose of the Directive is “to enhance the right to a fair trial in criminal proceedings by laying down minimum rules concerning certain aspects of the presumption of innocence and the right to be present at the trial” (Recital 4a). This should: “strengthen the trust of Member States in the criminal justice systems of other Member States and […] thus help to facilitate mutual recognition of decisions in criminal matters. Such common minimum rules should also remove obstacles to the free movement of citizens throughout the territories of the Member States.” (Recital 5)
Before looking at some of the key points of the Directive, there are two points to note: first, although compliance with existing law (ECHR) is noted to be a problem, EU standard setting has not been based on any empirical legal study. The precise scope and substance of the presumption of innocence can be harder to define within comparative contexts and this required further consideration. Second, Article 52(3) of the Charter confirms that the EU may raise standards beyond those of the ECHR but it cannot permit States to fall below them (note also the non-regression clause in Article 12 of the Directive). The ECHR is thus the core baseline for any assessment of the efficacy of new standards. Consequently, if the agreed instrument is not, at the very least, consistent with the ECHR, it is difficult to conclude it will have any positive benefit for the accused or address the problems it seeks to resolve.

The Directive

Article 1 confirms that the Directive is intended to lay down minimum rules on “certain aspects” of the right to the presumption of innocence in criminal proceeding” and the right to be present at the trial in criminal proceedings. The Directive is not intended, therefore, to be an exhaustive study of the principle and the ECHR will still be the main guide to those aspects which are not included in the text.

Article 2: Scope

Article 2 confirms that the Directive will apply at “all stages from the moment when a person is suspected or accused of having committed a criminal offence, or an alleged criminal offence, until the final determination of the question whether the person has committed the offence concerned and that decision has become definitive”. Recital 6 confirms that it applies “only to criminal proceedings, as interpreted in the case-law of the Court of Justice of the European Union (Court of Justice), without prejudice to the case-law of the European Court of Human Rights. Administrative proceedings, including administrative proceedings that can lead to sanctions, such as proceedings relating to competition, trade, financial services, traffic offences, or tax, including tax surcharge, and investigations by administrative authorities in relation to such proceedings, as well as civil proceedings, should not be covered by this Directive.”

There is no recognition of the possible consequences that admissions made in administrative proceedings could have on subsequent criminal proceedings. Further, despite the meaning of the term criminal proceedings being “without prejudice” to the definition established by the ECtHR, the Directive seems to contradict this. Article 6 (2) ECHR refers to a “criminal offence” but this has been interpreted to encompass types of cases beyond the classically “criminal”, for example, professional disciplinary proceedings or certain administrative offences which may fall within the ambit of the criminal head of Article 6 (e.g. Lutz v Germany, No. 9912/82, 25/08/1987; Bendenoun v. France 12547/86, 24/02/1994).

Further, the Directive applies only to natural persons and therefore excludes legal persons (e.g. companies). In a Joint Position Paper in 2014, Fair Trials International, noted that this leaves “their protection to existing safeguards, while acknowledging that the case law of the ECtHR has not clearly recognised the right of silence for legal persons” (para 12).[3] Given that legal persons can clearly be affected by the mutual recognition agenda, this is an opportunity missed in terms of clarifying and enhancing protection.

Additionally, as FTI has noted, unlike the Directive on access to a lawyer (Article 2(3)), this Directive does not extend protection explicitly to those “persons other than suspects or accused persons who, in the course of questioning, become suspects or accused persons”. There appears to be no justification for this inconsistency.

Article 3: Presumption of innocence

Article 3 is simply a restatement of the principle. It sets out “Member States shall ensure that suspects and accused persons are presumed innocent until proven guilty according to law”.  There is no attempt to articulate the nature of the provision further or set out the core aspects of the presumption for the purposes of the Directive.

For example, the ECtHR has confirmed that, in practice, for the presumption to be meaningful, certain procedural safeguards must be in place. For example, the prosecution will need to produce evidence of guilt in the trial (Barberá, Messegué and Jabardo v. Spain, No. 10590/83, 6.12.98) and the defendant must be given the right to be heard in his or her own defence (Minelli v. Switzerland No. 8660/79, 25.3.83). It will also be unlawful to base a conviction solely on the silence of an accused (Murray v. UK, No. 18731/91, 8.2.96). Consequently, the presumption is closely tied to the ability of the suspect to defend him/herself by receiving information about the charge so that s/he may prepare and present her/his defence accordingly (Barberá, Messegué and Jabardo v. Spain).  A document aimed at ensuring consistency should have articulated these issues more clearly.

Article 4: Public references to guilt before proven guilty

Article 4 puts Member States under an obligation to “take the necessary measures” to ensure that there are no public statements made by “public authorities”, as well as judicial decisions (save for verdicts), before suspects have been proven guilty according to law (see also Recitals 13 and 13a). This does not include prosecution attempts to prove the case or the public dissemination of information on the proceedings when it is “strictly necessary” for reasons relating to the criminal investigation or for the public interest. Appropriate remedies must be made available in the event of a breach (Article 10).

The case law of the ECtHR is quite substantial in this area (e.g. Allenet de Ribemont v. France, No. 15175/89, 10.02.1995) and more detailed guidance may help to drive up standards. Additionally, reference could have been made to the Council of Europe’s standards (CoE Recommendation Rec (2003)13 on the Provision of Information through the Media) particularly Principles 1 and 2 to assist in clarifying the scope and content of this Article.
Article 4a - Presentation of suspects and accused persons

This provision obliges Member States to “take appropriate measures” to ensure that suspects are not “presented as being guilty, in court or in public, through the use of measures of physical restraint” (see also Recital 13c). An exception is made under Article 4a (2) if the measures are required for security purposes or to prevent suspects absconding or having contact with third persons.

The ECtHR has established that an accused should not be treated in a way which undermines the presumption of innocence, e.g. by being in a caged dock without justification or being required to wear prison uniforms (Ramishvili and Kokhreidze v. Georgia, No. 1704/06, 27.1.09; Jiga v. Romania, App. No. 14352/04, 16.3.10). The Directive arguably offers less protection by focusing solely on physical restraint. The only concession to this lies in the rather weak assertion in Recital 13d which requests that Member States “refrain from presenting suspects in prison clothes where “practically possible”.

Article 5: Burden of proof

Article 5 deals with the burden of proof. It requires Member States to “ensure that the burden of proof in establishing the guilt of suspects and accused persons is on the prosecution”. This is an important issue. The burden of proof refers to the fact that the prosecution who must prove the case against the accused. The initial draft of Article 5 initially contained an article permitting the burden of proof to be shifted to the defence. The European Parliament’s Civil Liberties Committee successfully proposed an amendment deleting this burden of proof shifting provision.

However, Recital 14 permits presumptions of fact and law “confined within reasonable limits, taking into account the importance of what is at stake and maintaining the rights of the defence, and the means employed have to be reasonably proportionate to the legitimate aim sought to be achieved. The presumptions should be rebuttable; in any case, they may only be used provided the rights of the defence are respected”. Presumptions of fact and law create reverse burdens of proof which can clearly undermine the presumption of innocence. The ECtHR has permitted the evidential burden to be shifted to the defence but the importance of what is at stake and the safeguards which exist to protect the rights of the defence must be considered when determining whether a reverse burden is acceptable (Salabiaku v. France, Nos. 10519/83, 10519/83, 7.10.88).  No reference is made in the Directive for the need for specific safeguards to protect the rights of the defence in cases of presumptions of fact or law.

Article 6: Right to remain silent and right not to incriminate oneself

Article 6 (formerly Articles 6 and 7) requires that the suspect has the right to remain silent “in relation to the offence that they are suspected or accused of having committed”. This should surely have been extended to the right to silence in relation to the commission of any offence.

The Directive also confirms that “suspects and accused persons have the right not to incriminate themselves” (Saunders v. the United Kingdom, No. 19187/91, 17.12.96). Under Article 6 (1) ECHR, the right not to incriminate oneself presupposes that the prosecution must prove their case against the accused without recourse to evidence obtained “through methods of coercion or oppression in against the will of the accused”. The importance of informing a suspect of the right to remain silent is crucial yet the Directive makes no comment on this and makes no direct link in the operational text (as opposed to the Recitals) between this right and the Directive on the Right to Information or the Directive on the Right of Access to a Lawyer. Access to a lawyer is part of the procedural safeguards to which the ECtHR will have regard when examining whether any procedure has undermined the privilege against self-incrimination. If an accused has no lawyer, s/he has less chance of being informed of his/her rights and there is less chance that they will be respected (Pishchalnikov v. Russia, No. 7025/04, 24.9.09).

The ECtHR has noted that even where a person willingly agrees to give statements to the police after being informed that his/her words may be used in evidence against him/her, this cannot be regarded as a fully informed choice if s/he has not been expressly notified of his right to remain silent and if his/her decision has been taken without the assistance of counsel (Navone and Others v. Monaco, No. 62880/11, 24.10.13; Stojkovic v. France and Belgium, No. 25303/08, 27.10.11).

Yet, the Directive contains no reference to waiver of rights. The ECtHR has said, consistently, that a person can waive any fair trial guarantees of their own free will, either expressly or tacitly, but that a waiver requires safeguards for it to be effective, namely it must: (i) be established in an unequivocal manner; (ii) be attended by minimum safeguards commensurate to its importance; (iii) be voluntary; (iv) constitute a knowing and intelligent relinquishment of a right; and (v) if implicit from the accused’s conduct, it must be shown that s/he could reasonably have foreseen what the consequences of his/her conduct would be. Further, reasonable steps should be taken to ensure the accused has a level of understanding commensurate to their personal situation (Panovits v. Cyprus, No. 4268/04, 11.12.08). States will need to take additional steps to protect the rights of vulnerable suspects such as persons with disabilities and children, for example by arranging for third parties to support the individual.

It is of note that, in contrast, Article 9 of the Directive on the Right of Access to a Lawyer creates three conditions for a valid waiver: (i) the suspect must be 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; (ii) the waiver must be given voluntarily and unequivocally; and (iii) it must be recorded in accordance with the law of the EU Member State. However, it should be noted that the draft Directive on procedural safeguards for children suspected or accused in criminal proceedings confirms that children may not waive their right to a lawyer. Further, an EU Recommendation on procedural safeguards for vulnerable persons recommends that it should not be possible for vulnerable persons to waive their right to a lawyer (Recommendation of 27 November 2013 on procedural safeguards for vulnerable persons). Waiver is not addressed in the Directive and it is not clear why similar protections to the right to silence have not been established. Although Recitals 25c and 25d make reference to vulnerable suspects, they make no specific direction on the substance of the protection they require.

Under Article 6 (2), in line with the ECtHR jurisprudence, “the exercise of the right not to incriminate oneself shall not prevent gathering evidence which may be lawfully obtained through the use of legal compulsory powers and which has an existence independent of the will of the suspects or accused persons”. The ECtHR has noted this includes documents acquired pursuant to a warrant, breath, blood and urine samples, and bodily tissue for the purpose of DNA testing (Ortiz and Martin v. Spain, No. 43486/98, 15.6.99).

Article 6 (2b) of the Directive permits Member States’ judicial authorities to take into account “the cooperative behaviour of suspects and accused persons when sentencing”. No explanation is given for what “cooperative behaviour” means and certainly an “admission of guilt” is not excluded. This appears contradictory, confused and potentially undermining. Discounts for ‘cooperative behaviour’ are common in many criminal justice systems but they may create perverse incentives to plead guilty. All incentives to guilty pleas may compromise the right of defendants to be presumed innocent as they relieve the prosecution of the burden of proving guilt, and place pressure on suspects to admit an offence. Further thought should have been given to the implications of this Article on practice.

Article 6 (3) notes that “the exercise of the right to remain silent and of the right not to incriminate oneself shall not be used against a suspect or accused person and shall not be considered as evidence that the person concerned has committed the offence which he or she is suspected or accused of having committed”. This is welcome and appears to go further than the ECtHR which has found that an accused’s decision to remain silent throughout criminal proceedings may carry consequences, such as ‘adverse inferences’ being draw from the silence. (Condron v. United Kingdom, No. 35718/97, 2.5.00; Murray v. UK, No. 18731/91, 8.2.96).
Under Article 6 (5) Member States are not precluded from deciding that “in minor offences, the conduct of proceedings, or certain stages thereof, may take place in writing and/or without questioning of the suspect or accused person by the police or other law enforcement or judicial authorities in relation to the offence concerned, provided this is in conformity with the right to a fair trial”. The Commission has made a specific statement in relation to this provision stating that it should not be used to allow derogations from the right or to allow Member States to draw negative consequences from the exercise of the suspects' right to remain silent.
A proposal by FTI for the audio-visual recording of police interviews to prevent violations was not taken on board. Thus, the circumstances in which an accused agrees to talk remain extremely important. This is particularly important as there is no specific provision reinforcing the prevention of subterfuge to undermine the right to silence by eliciting confessions (e.g. by the use of informants) and no specific prohibition on the use of such evidence at trial. The ECtHR has clearly held that the privilege against self-incrimination includes the right not to incriminate oneself through coercion or oppression, in defiance of the will of the accused: Allan v. the United Kingdom, No. 48539/99, 12.11.02. This should have been clearly restated.
Article 8: Trials in absentia

Article 8 (and Recitals 21 and 22) re-affirm the individual’s right to be present at their trial. The ECtHR has confirmed that this is implicit in the right to a fair trial by way of a public hearing (Jacobsson v. Sweden, No. 16970/90, 19.2.98) and that it is difficult to see how anyone can exercise their defence rights without being present at their own trial (Colozza v. Italy, No. 9024/80, 12.2.85).

However, this principle is not absolute and this is reflected in the provisions of the Directive. Articles 8 (2) and (2a) permit Member States to hold trials in someone’s absence (and to enforce the decision of that trial) only if: (a) the suspect or accused person has been informed in due time of the trial and of the consequences of a non-appearance; or (b) the suspect or accused person, having been informed of the trial, is represented by a mandated lawyer, who was appointed either by the suspect or accused person, or by the State. This is narrower than the position set out by the ECtHR which has indicated that only “certain established facts might provide an unequivocal indication that the accused is aware of the existence of the criminal proceedings against him and of the nature and the cause of the accusation and does not intend to take part in the trial or wishes to escape prosecution” (Sejdovic v Italy, No. 56581/00, 1.3.06, at [99]). For example, “where the accused states publicly or in writing that he does not intend to respond to summonses of which he has become aware through sources other than the authorities, or succeeds in evading an attempted arrest …. or when materials are brought to the attention of the authorities which unequivocally show that he is aware of the proceedings pending against him and of the charges he faces” ([99]).

Under Article 8 (3), if Member States cannot comply with Article 8 (2) because the suspect or accused person cannot be located despite reasonable efforts having been made, “Member States may provide that a decision can nevertheless be taken, and that such a decision can be enforced”. However, in this situation, Member States “shall ensure that when suspects or accused persons are informed of the decision, in particular when they are apprehended, they shall also be informed of the possibility to contest the decision and of the right to a new trial, or another legal remedy, in accordance with Article 9”. The ECHR makes no such distinction in relation to when an accused has a right to a retrial. Indeed, in the case of Sejdovic v Italy, (No. 56581/00, 1.3.06), the ECtHR confirmed that “a denial of justice … undoubtedly occurs where a person convicted in absentia is unable subsequently to obtain from a court which has heard him a fresh determination of the merits of the charge, in respect of both law and fact, where it has not been established that he has waived his right to appear and to defend himself” (at [82]).

Article 8 (4) and (5) are both made “without prejudice to national rules” to permit temporary exclusion of a suspect from their trial or proceedings being conducted in writing if certain conditions are met. It is unhelpful to refer back to national law in a document which aims to consolidate rights at European level.

Finally, it should be noted that the Directive attempts to set standards within national jurisdictions, a situation that should be distinguished from that in Melloni where national standards potentially impeded the uniform application of cross-border obligations. In dealing with a question of trial in absentia, the CJEU in the case C-399/11, Stefano Melloni v. Ministerio Fiscal, 26.2.13 appeared to suggest that the Charter embodies a maximum rather than a minimum standard of human rights protection (save where the EU has not fully harmonised the field). This was to prevent the efficacy of the EAW Framework Decision (as amended by a later Framework Decision concerning in absentia proceedings and mutual recognition) being compromised. The Directive on the presumption of innocence, however, is specifically articulated as a set of minimum standards meaning that Melloni does not prevent higher standards being established nationally. Indeed, Article 12 explicitly prohibits the Directive from limiting the law of any Member State which provides a higher level of protection.   

Article 9: Right to a new trial

This permits the right to a retrial only where the conditions in Article 8 (2) have not been met. In any retrial, “Member States shall ensure that the persons concerned have the right to be present, to participate effectively, in accordance with procedures under national law, and to exercise their rights of defence”. A previous proposal stipulating that retrials could be denied to those who fail to request a retrial or appeal “within a reasonable amount of time has been removed.  But this provision remains weak.

Under the ECHR, if a person is not present at trial, they cannot exercise their defence rights under Article 6 (3) ECHR. The ECtHR has held that without a clear demonstration of the accused’s actual knowledge of the proceedings, a court may not commence a trial in absentia unless the defendant has a right to retrial, Krombach v. France, No. 29731/96, 13.2.01 Such retrial or appeal must fully comply with the demands of Article 6 of the ECHR, including the right to confront previous evidence including the cross-examination of witnesses. This provision should have reflected the law more closely.

Article 10: Remedies

Article 10 confirms that “Member States shall ensure that suspects and accused persons have an effective remedy if their rights under this Directive are breached”.  Further, “without prejudice to national rules and systems on the admissibility of evidence, Member States shall ensure that, in criminal proceedings, in the assessment of statements made by suspects or accused persons or of evidence obtained in breach of their right to remain silent or their right not to incriminate themselves, the rights of the defence and the fairness of the proceedings are respected”. Recital 26 confirms that “as far as possible” the suspect should be placed “in the same position” they would have been in but for the breach.

This could have been a much more robust provision. It could, as a minimum, have reflected the current state of ECHR/EU law. The right to an effective remedy is set out in Article 13 ECHR and Article 47 EU Charter. The primary requirement is that the remedy should be “effective in practice as well as in law”. The type of remedy required will depend on the circumstances of the case but some core principles have been developed to determine effectiveness. For example, an effective remedy must be: (i) accessible; (ii) capable of providing redress in respect of the applicant’s complaints; and (iii) offer reasonable prospects of success (Selmouni v. France, No. 25803/94, 28 July 1999). In terms of unlawfully obtained evidence, the question is whether the proceedings as a whole, including the way in which the evidence was obtained, were fair. This involves an examination of the alleged unlawfulness in question (e.g. Khan v. the United Kingdom, No. 35394/97, 12.5.00). Particular considerations apply in respect of the use in criminal proceedings of evidence obtained in breach of Article 3. The use of such evidence always raises serious issues as to the fairness of the proceedings, even if not decisive in securing a conviction (e.g. Jalloh v. Germany, No. 54810/00, 11.7.06).  Recital 26a merely notes that “regard should be had” to the case law on Article 3. This is disappointingly weak phrasing for such an essential protection.


Governments are constantly struggling to balance security fears with a respect for individual human rights. This is complicated further by the growing disconnect between citizens and their governments which, in itself, presents a challenge to the legitimacy of political actions. The result is that we have begun to lose faith in the capacity of our criminal justice systems to tackle crime and so we start to look for shortcuts to restore an order we believe has been lost. All too often, crime control has become mired in the kind of political debate which looks for simple answers. This makes it easy to overlook the fact that our criminal justice systems do not exist solely to churn out speedy convictions. It allows us to forget that suspects are individuals who have not yet been proved guilty of any offence.

In this context, an EU Directive restating the importance of the presumption of innocence is a welcome step.  But does this instrument go far enough? Based on the analysis above, my answer would be no.  There are three key reasons for this conclusion: first, standard setting has been built on assumptions rather than based on an empirical understanding of the operation of criminal justice systems and the reasons why current standards fail; second, the Directive does not consistently shore up the basic requirements of the ECHR and its case law despite the non-regression clause in Article 12; and third, the Directive fails to reference effectively previously agreed EU instruments to create a holistic framework for the protection of fundamental rights.

Under Article 13, the Directive is to be transposed 24 months after its publication. The true test of its efficacy will lie in its capacity to challenge unlawful practice.

Photo credit: The
Barnard & Peers: chapter 9, chapter 25

[1] Fair trials: civil liberties MEPs back new EU rules on presumption of innocence, LIBE Press release, 10.11.15.
[2] For example, see the criminal courts charge Guardian Editorial, 18 October 2015. See also Access to Justice: A Comparative Analysis of Cuts to Legal Aid, Report of the Monash Warwick Legal Aid Workshop, 2014.
[3] See also, FTI, Presumption of innocence directive agreed, 11 November 2015.