Pages

Friday, 29 April 2022

“Daphne’s Law”: The European Commission introduces an anti-SLAPP initiative


 


Professor Justin Borg-Barthet, University of Aberdeen*

*Advisor to a coalition of press freedom NGOs on the introduction of SLAPPs, co-author of the CASE Model Law, lead author of a study commissioned by the European Parliament, and member of the Commission's Expert Group on SLAPPs and its legislative sub-group

 

Background

 

When Daphne Caruana Galizia was assassinated in Malta on 16th October 2017, 48 defamation cases were pending against her in Maltese and other courts. Daphne was at the peak of her journalistic powers when she was killed, producing a seemingly endless exposé of criminality involving government and private sector actors. Naturally, those she was exposing did not take kindly to the intrusion on the enjoyment of the fruits of their labour. Courts which offered few meaningful safeguards against vexatious litigation presented a nominally legitimate forum in which they would seek to exhaust and punish Daphne and to ensure that others did not engage in similar investigations. Most of these cases were inherited by her sons, whose grief was interrupted constantly by a need to appear in court in defence of their mother’s work.

 

The scale of abusive litigation which Daphne endured prompted several NGOs to look more closely at the phenomenon of SLAPPs. Strategic Lawsuits Against Public Participation, a term coined in American academic circles, are lawsuits intended not to serve the legitimate purpose of pursuing a claim against a respondent, but instead to use court procedure to suppress scrutiny of matters of public interest. The direct costs, psychological strain, and opportunity costs of defending oneself in court are intended to coerce retraction of legitimate public interest activity, and to have a chilling effect on others who might show an interest. While most SLAPPs are framed as defamation claims, there is also a growing body of abusive litigation which suppresses public participation using the pretext of other rights such as privacy and intellectual property.

 

In response to the growing SLAPP phenomenon, several US States, Canadian provinces and Australian states and territories have introduced anti-SLAPP statutes. Typically, these statutes provide for the early dismissal of cases, and include cost-shifting measures to compensate SLAPP victims and to dissuade claimants. No EU Member State has yet adopted similar laws. Prompted by Daphne’s experience, European NGOs and MEPs became increasingly aware of the alarming incidence of SLAPPs throughout Europe. They then set out to identify and advocate for legal solutions in the European Union.

 

Initially, the European Commission resisted calls for the introduction of anti-SLAPP legislation, citing a lack of specific legal basis. As the legal and statistical research bases for NGO advocacy evolved further, and following a change in the Commission’s political leadership, the Commission’s assessment changed. This culminated in the introduction of a package of anti-SLAPP measures on 27th April 2022, including a proposed anti-SLAPP Directive which Vice-President Jourova dubbed “Daphne’s Law”.

 

The legislative proposal is based, in part, on a Model Law which was commissioned by the Coalition Against SLAPPs in Europe (CASE), a grouping of NGOs established to further the research basis and advocacy for anti-SLAPP laws in Europe. That Model Law is itself inspired by anti-SLAPP statutes adopted in the United States, Canada and Australia, but accounts for divergent continental legal traditions, and benefits from extensive consultation with experts and practitioners in Europe and elsewhere.

 

Legal Basis and Scope

As noted above, the key barrier for NGOs and MEPs to persuade the Commission to initiate anti-SLAPP legislation was disagreement about whether the EU had competence to act in this area. Subsequently, however, the Commission recognised the internal market relevance of SLAPPs, as well as adopting a more strident approach to the rule of law and human rights implications of SLAPPs. Arguments concerning a legal basis included an approach based on numerous treaty articles (as in the Whistleblowers’ Directive), reliance on the internal market effects of SLAPPs (Article 114 TFEU) as in the Model Law, and the potential use of treaty provisions on cross-border judicial cooperation. Ultimately, in view of Member States’ expected resistance to intervention in domestic procedural law, the Commission’s draft proceeds on the basis that Article 81 TFEU confers competence in respect of judicial cooperation in civil matters.

 

The orthodox view of Article 81 TFEU presupposes an international element to matters falling within its scope. It was therefore incumbent on the drafters to constrain the scope of the proposed directive to cases having a cross-border dimension. The Commission’s proposal begins with a classic private international law formulation which refers to the domicile of the parties. A case lacks cross-border implications if the parties are both domiciled in the Member State of the court seised. This, however, is subject to a far-reaching caveat in Article 4(2):

 

Where both parties to the proceedings are domiciled in the same Member State as the court seised, the matter shall also be considered to have cross-border implications if:

a)      the act of public participation concerning a matter of public interest against which court proceedings are initiated is relevant to more than one Member State, or

b)      the claimant or associated entities have initiated concurrent or previous court proceedings against the same or associated defendants in another Member State.

 

The Commission’s proposal adopts an innovative formulation, the breadth of which is commensurate to the internal market and EU governance implications of SLAPPs. Given the EU’s interconnectedness, it is paramount that the law account for the fact that cross-border implications do not flow only from the circumstances of the parties but also from transnational public interest in the underlying dispute.

 

The broad scope could be extended further if and when Member States come to transpose the proposed directive in national law. It is hoped, and indeed recommended as good practice, that Member States will take the view that national transposition measures will not be restricted to matters falling within the scope of the Directive but would apply also to purely domestic cases. This would avoid the prospect of reverse discrimination against SLAPP victims in domestic disputes. It would also minimise opportunistic litigation concerning the precise meaning of ‘[relevance] to more than one Member State’ in Article 4(2)(a).

 

Defining SLAPPs

Other than in the title and preamble, the proposed directive does not deploy the term ‘SLAPPs’. Discussions preceding the drafting process noted a number of difficulties associated with the term, not least (i) its unfamiliarity to a European legal audience, and (ii) the potential confusion resulting from the word ‘strategic’, which could be understood to require evidence of said strategy. In keeping with the Model Law, the Commission’s draft Directive deploys familiar language and focuses on the abusive nature of the proceedings. Rather than referring to SLAPPs, therefore, the text of the draft directive uses the term ‘abusive court proceedings against public participation’.

 

In identifying matters falling within the scope of the draft directive, it is first necessary to establish that a matter concerns ‘public participation’ on a matter of ‘public interest’. The Commission’s draft accounts for the fact that SLAPPs do not only target journalistic activity, but also seek to constrain legitimate action of civil society, NGOs, academics, and others. Public participation and public interest are therefore defined broadly as follows in Article 3:

 

‘public participation’ means any statement or activity by a natural or legal person expressed or carried out in the exercise of the right to freedom of expression and information on a matter of public interest, and preparatory, supporting or assisting action directly linked thereto. This includes complaints, petitions, administrative or judicial claims and participation in public hearings;

‘matter of public interest’ means any matter which affects the public to such an extent that the public may legitimately take an interest in it, in areas such as:

a)      public health, safety, the environment, climate or enjoyment of fundamental rights;

b)      activities of a person or entity in the public eye or of public interest;

c)       matters under public consideration or review by a legislative, executive, or judicial body, or any other public official proceedings;

d)      allegations of corruption, fraud or criminality;

e)      activities aimed to fight disinformation;

 

If a case concerns public participation in matters of public interest, it is then necessary to establish that the proceedings are abusive in accordance with the definition in Article 3:

‘abusive court proceedings against public participation’ mean court proceedings brought in relation to public participation that are fully or partially unfounded and have as their main purpose to prevent, restrict or penalize public participation. Indications of such a purpose can be:

a)      the disproportionate, excessive or unreasonable nature of the claim or part thereof;

b)      the existence of multiple proceedings initiated by the claimant or associated parties in relation to similar matters;

c)       intimidation, harassment or threats on the part of the claimant or his or her representatives.

 

There are therefore two key elements to the notion of abuse: (i) claims may be abusive because they are fully or partly unfounded, or (ii) they may be abusive because of vexatious tactics deployed by claimants. The implications of a finding of abusiveness will vary depending on the type of abuse identified in the proceedings, with more robust remedies available where the claim is manifestly unfounded in whole or in part.

 

Main legal mechanisms to combat SLAPPs

Once a court has established that proceedings constitute SLAPPs falling within the directive’s scope, three key remedies will be available to the respondent in the main proceedings: (i) the provision of security for costs and damages while proceedings are ongoing, (ii) the early dismissal of proceedings, and (iii) payment of costs and damages.

 

Speedy dismissal of claims is considered the cornerstone of anti-SLAPP legislation. Accelerated dismissal deprives the SLAPP claimant of the ability to extend the financial and psychological costs of proceedings to the detriment of the respondent. Early dismissal of cases must, of course, be granted only with great caution given it is arguable that this restricts the claimant’s fundamental right to access to courts. The solution provided in the draft directive is to restrict the availability of this remedy to claims which are manifestly unfounded in whole or in part. It is for the claimant in the main proceedings to show that their claim is not manifestly unfounded (Art 12).

 

Early dismissal is not available where the claim is not found to be manifestly unfounded, even if the its main purpose is ‘to prevent, restrict or penalize public participation’ (as evidenced by ‘(i) the disproportionate, excessive or unreasonable nature of the claim…the existence of multiple proceedings [or] intimidation, harassment or threats on the part of the claimant’). This differs from the Model Law which envisages early dismissal in cases which are not manifestly unfounded but which bear the hallmarks of abuse. The Model Law’s authors reasoned that a court should be empowered to dismiss a claim which is designed to abuse rather than vindicate rights. This would not, in our view, constitute a denial of the right to legitimate access to courts but would dissuade behaviour which is characterised as abusive in the Commission’s own draft instrument. While the Commission’s reasoning and caution are understandable, the high bar set by the requirement of manifest unfoundedness allows for significant continued abuse of process.

 

This shortcoming is mitigated somewhat by the other remedies, namely the provision of security pendente lite (Article 8) and liability for costs, penalties, and compensatory damages (Articles 14-16), which are available regardless of whether the SLAPP is manifestly unfounded or merely characterised by abuse of rights. These financial remedies are especially useful insofar as they give the respondent some comfort that they will be compensated for the loss endured through litigation. They are also expected to have a dissuasive effect on SLAPP claimants who would be especially loathe to the notion of rewarding the respondent whose legitimate exercise of freedom of expression they had sought to dissuade or punish. Nevertheless, it bears repeating that in all cases these remedies, designed to compensate harm, should supplement the principal remedy of early dismissal which is intended to prevent harm.

 

In addition to these main devices to dissuade the initiation of abusive proceedings against public participation, the draft directive includes a number of further procedural safeguards. These include restrictions on the ability to alter claims with a view to avoiding the award of costs (see Recital 24 and Article 6), as well as the right to third party intervention (Article 7) which will enable NGOs to submit amicus briefs in proceedings concerning public participation. While this may appear to be a minor innovation at first blush, it could have substantial positive implications insofar as it would equip more vulnerable respondents (and less expert courts) with valuable expertise and oversight.

 

London Calling: Private International Law Innovation

While the provisions discussed above would limit the attractiveness of SLAPPs in EU courts, there would remain a significant gap if EU law did not provide protection against the institution of SLAPPs in third countries. London, with its high litigation costs and somewhat claimant friendly defamation laws, is an especially attractive forum for claimants who wish to suppress public scrutiny. Equally, other States could be attractive to claimants who wish to circumvent EU anti-SLAPP law, whether simply as a function of the burden of transnational litigation, or because of the specific content of their substantive and/or procedural laws. The draft directive therefore proposes to introduce harmonised rules on the treatment of SLAPP litigation in third countries.

 

Article 17 provides that the recognition and enforcement of judgments from the courts of third countries should be refused on grounds of public policy if the proceedings bear the hallmarks of SLAPPs. While Member States were already empowered to refuse recognition and enforcement in such cases, the inclusion of this article ensures that protection against enforcement of judgments derived from vexatious proceedings is available in all Member States.

 

Article 18 provides a further innovation by establishing a new harmonised jurisdictional rule and substantive rights to damages in respect of SLAPPs in third countries. The provision confers jurisdiction on the courts of the Member State in which a SLAPP victim is domiciled regardless of the domicile of the claimant in the SLAPP proceedings. This would provide an especially robust defence against the misuse of third country courts and reduce the attractiveness of London and the United States as venues from which to spook journalists into silence.

 

While the limitation of forum shopping in respect of third countries is, of course, welcome, there does remain a significant flaw insofar as EU law and the Lugano Convention facilitate forum shopping within the European judicial area. The cumulative effect of EU private international law of defamation is to provide mischievous litigants with ample opportunity to deploy transnational litigation as a weapon to suppress freedom of expression. NGOs have therefore requested amendment of two EU private international law instruments:

 

In the first instance, and as a matter of urgency, the Brussels I Regulation (recast) requires amendment with a view to grounding jurisdiction in the domicile of the defendant in matters relating to defamation. This would remove the facility for pursuers to abuse their ability to choose a court or courts which have little connection to the dispute;

The omission of defamation from the scope of the Rome II Regulation requires journalists to apply the lowest standard of press freedom available in the laws which might be applied to a potential dispute. We recommend the inclusion of a new rule which would require the application of the law of the place to which a publication is directed;

 

These changes have not yet been forthcoming. It is hoped that ongoing reviews of these instruments will yield further good news for public participation in the EU.

 

Concluding remarks

Daphne’s Law will now have to be approved by the Council of Ministers and the European Parliament. The legislative process may see a Parliament seeking more robust measures pitted against Member States who may be inclined to protect their procedural autonomy. The Commission has considered these competing demands in its draft and sought to propose legislation which strikes a balance between divergent institutional stances. Nevertheless, it must be expected that the draft will be refined as it makes its way through the approval process. As noted above, the draft would be improved if those refinements were to include the extension of early dismissal to cases beyond the narrow confines of manifest unfoundedness. Equally, the draft directive should be viewed as a first welcome step in the pushback against SLAPPs in Europe and that reviews of private international law instruments will follow soon after.

 

Photo credit: ContinentalEurope, on Wikicommons

No comments:

Post a Comment