On 25 October 2013, in its judgment in the R v Mohammed Gul case, the Supreme Court of the United Kingdom tackled two important issues: the definition of terrorism in times of armed conflict and the relationship between domestic legislation and international rules criminalizing certain behaviours. On both issues, the judgment is rather unsatisfying and may be considered a step back from the stand previously taken by the Court of Appeal in the same case (upon which I commented in the Journal of International Criminal Justice, vol. 11(2), 2013, pp. 425-440, some time ago; see also the excellent post by Kimberley Trapp here on EJIL:Talk!). In particular, the Supreme Court found the terrorism definition in UK law to be both unwise and undesirable but then relied on it to confirm the defendant’s conviction.
The defendant, a law student of British nationality, was accused of having disseminated terrorist publications, an offence under Section 2(3) of the UK Terrorism Act 2006. Actually, his conduct consisted in uploading onto the Internet, and particularly on Youtube, videos of attacks against military targets in Chechnya, Iran and Afghanistan. The videos were accompanied by prayers and praises for the attackers. One legal element of the offence is that the publication – in this case the videos – concerns actual terrorist attacks. The bone of contention here, then, is whether attacks against military targets in the context of non-international armed conflicts (NIACs) can be labelled as terrorist attacks.
The definition of terrorism in UK legislation is contained in Section 1 of the Terrorism Act 2000. It basically foresees three requirements: (1) an act or threat which involves serious violence or danger to the life of persons, serious damage to property, or serious interference with or disruption of electronic systems; (2) the “purpose of advancing a political, religious, racial or ideological cause”; and (3) the fact that the act or threat is “designed to influence the government or an international governmental organization, or to intimidate the public or a section of the public”. The act or threat need not to be designed to influence a government or an international organization or to intimidate the public when it involves the use of firearms or explosives. This means that any threat or use of firearms or explosives motivated by a political or ideological cause is an act of terrorism, as long as it involves serious danger to persons or serious damage to property and regardless of its purpose.
The definition is practically very broad (as recently noted by K.J. Heller). It seems to label as terrorist most acts of warfare in a NIAC, regardless of whether they are lawful or unlawful under International Humanitarian Law (IHL) and whether they are carried out by the armed forces of a State or by a non-State armed group. Indeed, most hostile acts in an armed conflict are likely to cause serious violence to persons or serious damage to property, and all of them are motivated by a political or ideological cause. Arguably, any hostile act in an armed conflict is designed to influence a government or involves the use of firearms or explosives. According to this definition, every person embracing weapons in a NIAC is considered a terrorist. The Prosecution in the Gul case argued that such a wide definition is counterbalanced by the requirement that prosecutions for terrorism are authorized by the Director of Public Prosecution if the activity occurred in the UK, or by the Attorney General if it occurred abroad, thus ensuring that criminal charges are formulated only in the appropriate cases (§ 30). This contention, far from solving the problem, seems to raise even more concerns, as I shall explain below.
The defendant contended instead, in my opinion correctly, that international law defines terrorism more narrowly. Leaving to one side that a definition of terrorism in peace-time might have crystallized in customary international law (as affirmed by the Special Tribunal for Lebanon in its 16 February 2011 decision, and agreed by the Court of Appeal of England and Wales at the previous stage of R v Gul), the terrorism label should not be stuck on attacks against military targets in the context of an armed conflict. Indeed, among the international conventions on terrorism binding on the UK, those most relevant to the attacks which Gul uploaded onto the Internet contain an exclusion clause for actions carried out by the parties to an armed conflict, regulated by IHL (see e.g. art. 19(2) of the Terrorist Bombings Convention. The combination of art. 2(1)(a) and art. 21 of the Terrorist Financing Convention has the same effect). Arguably, a definition of terrorist attacks in war-time – as ‘acts or threats of violence the primary purpose of which is to spread terror among the civilian population’ – is envisaged by art. 51(2) of Additional Protocol I (AP I) to the Geneva Conventions and art. 13(2) of Additional Protocol II (AP II), both deemed to have reached customary status (See the ICRC Study on customary IHL, Rule 2).
The Supreme Court nevertheless found no reason to read the definition narrowly, based on two strands of argument. The first strand (§§ 44-51) stems from the consideration that there is no accepted definition of terrorism in international law, and there is certainly no agreement as to whether attacks by so-called freedom fighters (i.e. peoples fighting against a racist regime, an alien occupation or a colonial domination) can be exempted from the terrorism label. The negotiations for a Draft Comprehensive Convention on Terrorism have stopped precisely for this reason. Moreover, in any case, there would be no combatant immunity for those who have taken arms against their state in non-international armed conflicts, so these would be punishable under domestic law even for attacks against military targets.
These assumptions are problematic. The finding on the absence of a comprehensive definition of terrorism admittedly contradicts theconclusion of the Court of Appeal, which had agreed that a customary definition existed at least for acts of terrorism in peace-time, based on the STL decision cited above (see §§ 33-35 of the Court of Appeal’s judgment). The reference to freedom fighters does not match the issue at stake in the Gul case: the negotiations over a Comprehensive Convention seem to have stopped because some countries wanted to exempt attacks by freedom fighters even when launched against civilians. There was never a question as to whether attacks against military targets could be considered to be terrorist attacks. And indeed they are not, according to the ICTY in Galić (Trial Judgment, § 133) and the ICRC commentary (§ 1940) on art. 51(2) AP I. The reference to the absence of a combatant immunity in NIACs is not conclusive either: discriminate and proportionate attacks against governmental forces might well be punishable under domestic law, but they do not constitute terrorist attacks according to the applicable international rules.
This final consideration is clearly linked to the second line of arguments put forward by the Supreme Court (§§ 52-58). The Justices indeed affirm that, even if a number of international rules binding on the UK provide a definition of terrorism, the Parliament is not bound by them when defining terrorism at the domestic level. Domestic legislation could go a great deal further than what would normally be required by the international rule to be implemented, as long as the former does not contradict the latter. This practice, known as ‘gold plating’, would not be prohibited as such. The Supreme Court bases its reasoning on the famous Lotus principle (S.S. Lotus, France v. Turkey, 1927 PCIJ Series A, No. 10., § 44), according to which States are free to do whatever they like, as long as their behaviour is not explicitly prohibited by international law.
It appears, however, that this principle does not give the Supreme Court a sound basis for its argument. The Lotus principle dates back to 1927, when international law was far less developed compared to present day. At that time, the comparatively smaller number of applicable international rules conceded a wider margin of manoeuvre to States. After almost 90 years, however, the domaine réservé of States has been significantly reduced. The emergence of human rights law, international criminal law, and the law of non-international armed conflicts, inter alia, has significantly changed the setting. Nowadays the behaviour of States is regulated by the interplay of a vast number of sources, not all of them containing explicit prohibitions. The international regulatory framework on terrorism, in particular, gives a sufficient indication as to what can – and cannot – be considered terrorism. If a State can totally disregard this indication in its domestic law, what is the point of negotiating and signing all these conventions?
These considerations are particularly relevant given that, under the umbrella of counter-terrorism, some governments have in the past fought political dissent. I am not at all suggesting that the UK definition of terrorism is anti-democratic. However, it certainly raises deep concerns about respect for the principle of legality. The definition is so broad that, despite the contrary stand taken by the Supreme Court (§ 36), it ultimately delegates to the Prosecution the decision as to what constitutes a criminal act of terrorism. The case law of regional human rights bodies has already warned against such wide definitions of criminal offences. The Inter-American Court, in Castillo Petruzzi (30.05.1999, § 121), has made clear that
crimes must be classified and described in precise and unambiguous language that narrowly defines the punishable offense, thus giving full meaning to the principle of nullum crimen nulla poena sine lege praevia in criminal law. This means a clear definition of the criminalized conduct, establishing its elements and the factors that distinguish it from behaviors that are either not punishable offences or are punishable but not with imprisonment.
The European Court of Human Rights took a similar position in Kokkinakis, 25.05.1993, § 52.
Moreover, one should not forget that the entire body of UK counter-terrorism laws is based on this wide definition. Limitations of other human rights, including the right to liberty and security (because of increased police powers) and the right to freedom of expression (because of the prohibition on disseminating ‘terrorist publications’), risk becoming violations of the rules established by applicable international instruments, precisely because of the wide meaning given to the word ‘terrorism’.
Despite its questionable reasoning, the Supreme Court at least acknowledged its concerns over the wording of the definition of terrorism in UK law (§§ 63-64). By doing so, the Justices implicitly accepted that Mr. Gul’s conviction might be based on a vitiated piece of legislation. It is therefore surprising that they decided to uphold that same conviction, though accompanying it with a few tears of sorrow.