Zakrzewski v The Regional Court in Lodz, Poland

Where an application is made for the extradition of a convicted person to a category 1 territory, ie pursuant to a European arrest warrant, the warrant is required by section 2(6)(e) of the Extradition Act 2003 to include “particulars of the sentence which has been imposed under the law of the category 1 territory in respect of the offence, if the person has been sentenced for the offence.” The purpose of this requirement is to enable the court to apply section 65(2)(c), (3)(c), (4)(c) and (5)(c). These provide minimum sentences of imprisonment or detention which must have been imposed in order to disclose an extradition offence. The minimum periods are 12 months in the case of offences on the European Framework list or four months for offences which are not on the European Framework list but satisfy the relevant requirement of double criminality. In the present case, the relevant provision is section 65(3)(c), which applies to offences committed in the category 1 territory which would constitute an offence under the law of the relevant part of the United Kingdom if it occurred there, provided that “a sentence of imprisonment or another form of detention for a term of four months or a greater punishment has been imposed.”

In many states of the European Union the criminal law provides for the aggregation of successive sentences imposed by criminal courts on different occasions so as to produce a single sentence reflecting the totality of the course of criminality disclosed. This will commonly result in a reduction of the total period of imprisonment imposed, by comparison with the period arrived at by adding up each of the original sentences. Poland’s aggregation procedure is contained in articles 85-86 of the Penal Code and articles 569-577 of the Criminal Procedures Code, which require a court to aggregate successive sentences to produce a single “cumulative penalty”. The effect of this procedure has been considered in a number of cases in which a European arrest warrant has given particulars of the cumulative penalty but not of the individual sentences which were aggregated so as to produce it. The question whether this satisfies sections 2(6)(e) and 65(3)(c) of the Act was finally settled in Pilecki v Circuit Court of Legnica, Poland [2008] 1 WLR 325. The House of Lords held that, at any rate in a case where each of the original sentences was for conduct satisfying all the other requirements for an extradition offence, it was enough for the warrant to specify the cumulative sentence. If it exceeded four months it was irrelevant that some of the original sentences might have been less than that. The present appeal concerns the converse situation. What happens if the warrant specifies only the original sentences, but after it has been issued they have been aggregated and their totality reduced?

Lukasz Zakrzewski, was convicted on four occasions in Poland of various offences of dishonesty or violence. On 10 December 2003, he was convicted by the District Court in Grudziadz of assault and robbery committed on separate occasions in February 2003, for which he received a combined sentence of 14 months imprisonment. On 18 March 2004, he was convicted by the same court of two distinct offences of theft, and received a further combined sentence of 15 months imprisonment. On 28 May 2004, he was convicted of theft by the District Court of Swiecle and sentenced to six months imprisonment. On 14 January 2005, he was back before the District Court of Grudziadz, which convicted him of theft and sentenced him to a further ten months imprisonment. All of these sentences of imprisonment were initially suspended, but all of them were subsequently activated either by the commission of further offences during the period of probation which followed conviction, or by breaches of the probation terms. On 24 February 2010, Mr Zakrzewski having absconded, the Regional Court of Lodz issued a European Arrest Warrant against him, based on his conviction on these four occasions. The warrant specified the sentence passed on each occasion.

Mr Zakrzewski was arrested in England on 28 September 2010 and brought before City of Westminster Magistrates’ Court on the same day. At that time, he was facing further criminal charges in the United Kingdom. The extradition proceedings were therefore adjourned pending the resolution of proceedings arising from them. During the adjournment, Mr Zakrzewski applied to the District Court of Grudziadz to have the four sentences aggregated. The court duly aggregated them, and on 19 April 2011 imposed a cumulative sentence of 22 months, as opposed to the aggregate of 45 months under the original sentences. When Mr Zakrzewski came back before Westminster Magistrates on 20 May 2011, it was submitted on his behalf that the aggregation order meant that the warrant no longer gave the particulars required by section 2(6)(e) because the only relevant sentence was now the cumulative sentence. It followed, so it was said, that the warrant had become invalid, or that the court should exercise an inherent jurisdiction not to proceed with the extradition on the ground that it no longer gave proper, fair or accurate particulars: see Criminal Court at the National High Court, First Division v Murua [2010] EWHC 2609 (Admin). It will be noted that each of the original sentences was for conduct in Poland which would have been criminal if it had occurred in England, and that the original sentences and the cumulative sentence all exceeded four months. The argument advanced on Mr Zakrzewski’s behalf is therefore hardly overburdened with merit. It is about as technical as it could possibly be. It is common ground that a further warrant giving the same particulars but specifying the cumulative sentence would be good.

District Judge Rose rejected the argument in both its forms and made the extradition order. But it was accepted in both forms by Lloyd Jones J on appeal to the High Court. He allowed the appeal against the extradition order on 7 February 2012. In summary, he held that the information in the warrant must

“relate to the current operative sentence and not to earlier sentences which have been subsumed in an aggregated order. In determining whether the requirement of section 65 is satisfied, the court needs to know the total length of time which the court of the requesting state has ordered must be served in prison. In the present case, that is the aggregated order.”: [2012] 1 WLR 2248, para 26.

The basic features of the scheme for the execution of a European arrest warrant under Part 1 of the Extradition Act 2003 are too familiar to need extensive restatement here. It has often been pointed out that the contents of the warrant are critical to the operation of the scheme of both the Council Framework Decision 2002/584/JHA of 13 June 2002 and the United Kingdom Act. Extradition under Part 1 of the Act is by way of direct execution of the warrant. To fall within the definition of a “Part 1 warrant” and be capable of initiating extradition proceedings, it must contain the statements and information required by section 2 of the Act, which reflect the mandatory contents provided for by article 8 of the Framework Decision. The procedure operates at each stage by reference to the prescribed particulars contained in it. Thus, under section 10, the court must decide whether “the offence specified in the Part 1 warrant” is an extradition offence as defined by section 64 (in an “accusation” case) or section 65 (in a “conviction” case). Both sections require the court to consider whether the “offence constituted by the conduct” satisfied the requirements of those sections. “The conduct” for this purpose means that specified in the warrant, and it is not permissible to conduct an independent examination of the elements of the offence under the law of the requesting state: Office of the King’s Prosecutor, Brussels v Cando Armas [2006] 2 AC 1, paras 16 (Lord Bingham of Cornhill) and 30 (Lord Hope of Craighead). Under section 64(2)(b) and (c), the questions whether “the conduct” falls within the European Framework list and whether it is punishable under the law of the requesting state by a sentence of imprisonment of three years or more are to be determined by reference to information certified by the requesting authority, which may be (and commonly is) certified in the warrant itself: see Dabas v High Court of Justice in Madrid, Spain [2007] 2 AC 31. The same applies to the corresponding provisions of section 65(2)(b) and (c). If the warrant contains the prescribed particulars and these disclose an extradition offence, the court must extradite the defendant, unless one of the limited exceptions specified in the Act applies. The exceptions to the otherwise mandatory extradition of the defendant are dealt with by sections 10 to 21 and 25 of the Act. Some of these also operate by reference to “the conduct”, which must in the circumstances mean the conduct specified in the warrant: see sections 15 and 19B (as inserted by section 42 of, and para 4(2) of Schedule 13 to, the Police and Justice Act 2006).

All of these provisions reflect the underlying purpose of the Framework Decision and Part 1 of the Extradition Act to create a simplified and accelerated procedure based on the mutual recognition by the requested state of the antecedent decision to issue the warrant by the judicial authority in the requesting state. Recital (10) of the Framework Decision records that “the mechanism of the European arrest warrant is based on a high level of confidence between member states.” Or, as Lord Phillips put it in Assange v Swedish Prosecution Authority (Nos 1 & 2) [2012] 2 AC 471, para 79, “under the scheme of the Framework Decision the safeguard against the inappropriate issue of an EAW lies in the process antecedent to the issue of the EAW.”

It follows that the scheme of the Framework Decision and of Part 1 of the 2003 Act is that as a general rule the court of the executing state is bound to take the statements and information in the warrant at face value. The validity of the warrant depends on whether the prescribed particulars are to be found in it, and not on whether they are correct. It cannot be open to a defendant to challenge the validity of a warrant which contains the prescribed particulars by reference to extraneous evidence tending to show that those statements and information are wrong. If this is true of statements and information in a warrant which were wrong at the time of issue, it must necessarily be true of statements which were correct at the time of issue but ceased to be correct as a result of subsequent events. Validity is not a transient state. A warrant is either valid or not. It cannot change from one to the other over time.

It does not, however, follow from this that there is nothing to be done about it if the prescribed particulars in the warrant are or have become incorrect. It only means that the remedy must be found at the stage when the court is considering whether to extradite. Neither the Framework Decision nor Part 1 of the Act provides in terms for non-extradition on the ground of a factual error in the warrant. There are, however, two safeguards against an unjustified extradition in those circumstances.

The first and main one is the mutual trust between states party to the Framework Decision that informs the entire scheme. The requesting judicial authority has a right, recognised by article 15.3 of the Framework Decision, to forward additional information at any time. These are receivable in evidence by an English court under section 202 of the Act on the same basis as the warrant itself. If necessary, further information may be requested by the executing court under article 15.2. The Framework Decision proceeds on the assumption that requesting states can be trusted to ensure that statements and information in a European arrest warrant are true. By the same token, if they subsequently cease to be true, either the warrant will be withdrawn or the statements and information in it will be corrected by the provision of further information, with or without a request for it.

The second safeguard lies in the inherent right of an English court, as the executing court, to ensure that its process is not abused. One form of abuse of process is the fortunately rare case in which the prosecutor has manipulated the process of the executing court for a collateral and improper purpose: see R (Government of the United States of America) v Bow Street Magistrates’ Court [2007] 1 WLR 1157. We are not concerned with anything of that kind on this appeal. Another category comprises cases, rather less rare, in which the prescribed particulars are given in the warrant but they are wrong. In Caldarelli v Judge for Preliminary Investigations of the Court of Naples, Italy [2008] 1 WLR 1724, para 24, Lord Bingham observed that "it might in some circumstances be necessary to question statements made in the EAW”, notwithstanding the general rule. The question is in what circumstances is the power envisaged by Lord Bingham exercisable.

The clearest statement of the principle is to be found in the decision of Sir Anthony May, President of the Queen’s Bench Division of the High Court, in Criminal Court at the National High Court, First Division v Murua [2010] EWHC 2609 (Admin), which has been followed by the High Court on a number of occasions. Murua was an accusation case. The warrant alleged serious terrorist offences involving danger to life and concealment of identity. Both of these were significant aggravating factors under Spanish law, warranting imprisonment upon conviction for up to 48 years. The particulars of the offence specified the aggravating factors, and the maximum sentence associated with them. However, at the trial in Spain of seven other defendants for the same conduct, the prosecution had accepted that these aggravating factors could not be proved. The charges were reformulated, and the co-defendants convicted of lesser offences carrying a maximum term of imprisonment of three years. Sir Anthony May said, at paras 58-59:

“58.

The court's task -- jurisdiction, if you like -- is to determine whether the particulars required by section 2(4) have been properly given. It is a task to be undertaken with firm regard to mutual co-operation, recognition and respect. It does not extend to a debatable analysis of arguably discrepant evidence, nor to a detailed critique of the law of the requesting state as given by the issuing judicial authority. It may, however, occasionally be necessary to ask, on appropriately clear facts, whether the description of the conduct alleged to constitute the alleged extradition offence is fair, proper and accurate. I understood Ms Cumberland to accept this, agreeing that it was in the end a matter of fact and degree. She stressed, however, a variety of floodgates arguments with which in general I agree, that this kind of inquiry should not be entertained in any case where to do so would undermine the principles to be found in the introductory preambles to the Council Framework Decision of 13 June 2002.

59.

Ms Cumberland submitted that an argument of the kind which succeeded before the District Judge can be raised, but not with reference to section 2 of the 2003 Act. She said that the proper approach was to deal with it as an abuse argument, and this ties in with the appellant's third ground of appeal, to which I shall come in a few moments. I do not agree that the respondent's case could only be advanced as an abuse argument. It can properly be advanced, as it was, as a contention that the description in the warrant of the conduct alleged did not sufficiently conform with the requirements set out in section 2 for the reasons advanced by Mr Summers with reference to Dabas v High Court of Justice in Madrid, Spain [2007] 2 AC 31 and Pilecki v Circuit Court of Legnica, Poland [2008] 1 WLR 325. If that is shown, it is not a valid Part 1 warrant.”

I agree with this statement, subject to four observations. The first is that the jurisdiction is exceptional. The statements in the warrant must comprise statutory particulars which are wrong or incomplete in some respect which is misleading (though not necessarily intentionally). Secondly, the true facts required to correct the error or omission must be clear and beyond legitimate dispute. The power of the court to prevent abuse of its process must be exercised in the light of the purposes of that process. In extradition cases, it must have regard, as Sir Anthony May observed, to the scheme and purpose of the legislation. It is not therefore to be used as an indirect way of mounting a contentious challenge to the factual or evidential basis for the conduct alleged in the warrant, this being a matter for the requesting court. Third, the error or omission must be material to the operation of the statutory scheme. No doubt errors in some particulars (such as the identity of the defendant or the offence charged) would by their very nature be material. In other cases, the materiality of the error will depend on its impact on the decision whether or not to order extradition. The fourth observation follows from the third. In my view, Ms Cumberland was right to submit to Sir Anthony May in Murua that the sole juridical basis for the inquiry into the accuracy of the particulars in the warrant is abuse of process. I do not think that it goes to the validity of the warrant. This is because in considering whether to refuse extradition on the ground of abuse of process, the materiality of the error in the warrant will be of critical importance, whereas if the error goes to the validity of the warrant, no question of materiality can arise. An invalid warrant is incapable of initiating extradition proceedings. I do not think that it is consistent with the scheme of the Framework Decision to refuse to act on a warrant in which the prescribed particulars were included, merely because those particulars contain immaterial errors.

I now return to the facts of Mr Zakrzewski’s case. The warrant issued against him was undoubtedly a valid warrant when it was issued. It was therefore effective to authorise the commencement of extradition proceedings in the United Kingdom against him. It did not become invalid when the aggregation order was made. It follows that the only basis on which Mr Zakrzewski could object to his extradition was that the conduct of the requesting judicial authority in persisting with extradition proceedings after the aggregation order was an abuse of those proceedings. The short answer to this contention in the present case is that the particulars of the sentence in the warrant, although no longer complete, were not wrong. This is because the Regional Court of Lodz supplied further information about the effect of the aggregation order in the following terms:

“The consequence of a composite sentence having been passed is that the single penalties imposed for each of the offences are replaced by that single composite sentence. In such a case, the penalties imposed for each of the offences are not to be enforced separately, but replaced with the new composite sentence that is to be enforced in respect of the convict... It should be noted that the issue of a composite sentence does not invalidate any of the individual sentences.”

This answer was effective to explain the contents of the warrant. Its effect is that the original sentences remain valid but the cumulative sentence determines what period of imprisonment will be treated as satisfying them. Therefore, the information in the warrant about the original sentences did not cease to be true when the cumulative sentence was passed.

Although true, the information in the warrant about the sentence imposed became incomplete when the cumulative sentence was passed. The prosecution of extradition proceedings on a warrant containing prescribed particulars which are (or have become) incomplete is capable of being an abuse of process, but only if the information omitted is material to the operation of the statutory scheme. In this case the fact that the period of imprisonment which would satisfy the four original sentences had been shortened was wholly immaterial, because even the shorter cumulative sentence was substantially longer than the minimum of four months. As Lord Hope observed in Pilecki [2008] 1 WLR 325, para 29, “all the executing court needs to know in these circumstances is whether or not the sentence was one for at least four months.” The position would be different if the composite sentence was below the four-month threshold, because there would then be no extradition offence.

I cannot agree with Lloyd Jones J [2012] 1 WLR 2248, para 26 that the failure of the warrant to specify the “current operative sentence” was fatal. The sentence of the court will rarely be the “current operative sentence”, since the period to be served will commonly be affected by a variety of factors, such as remission or parole. As the cases on aggregation procedure show, they may also be affected by aspects of criminal procedure which will vary from one jurisdiction to another without affecting the application of the ordinary criteria for extradition or undermining the purpose of the Framework Decision or Part I of the Act.

It follows that in the ordinary course the appeal would have been allowed and the order of the District Judge restored. However, just before this judgment was due to be delivered, the Court was informed that Mr. Zakrzewski had returned voluntarily to Poland after the argument on the appeal and been arrested there. Accordingly, the warrant has been withdrawn by the court which issued it. This does not affect the issue which the Court has to decide. But it does mean that, formally, the appeal must now be dismissed: see section 43(4).